A SPATIAL DYNAMIC FRAMEWORK TO INTEGRATE REGIONAL WATER USE EFFICIENCY AND ENERGY CONSUMPTION Aftab Ahmad B.Sc. (Agricultural Engineering) University of Agriculture, Faisalabad M.Eng. (Water Resource Engineering & Management) Asian Institute of Technology, Bangkok A thesis submitted to Charles Sturt University for the degree of Doctor of Philosophy School of Environmental Sciences, Faculty of Science, Charles Sturt University August 2013
455
Embed
A SPATIAL DYNAMIC FRAMEWORK TO INTEGRATE REGIONAL WATER …
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
A SPATIAL DYNAMIC FRAMEWORK TO INTEGRATE
REGIONAL WATER USE EFFICIENCY AND ENERGY
CONSUMPTION
Aftab Ahmad
B.Sc. (Agricultural Engineering)
University of Agriculture, Faisalabad
M.Eng. (Water Resource Engineering & Management)
Asian Institute of Technology, Bangkok
A thesis submitted to Charles Sturt University for the degree of Doctor of Philosophy
School of Environmental Sciences, Faculty of Science,
Charles Sturt University
August 2013
i
Table of contents
Certificate of authorship ............................................................................. xvii Acknowledgements ....................................................................................... xix Ethics approval ............................................................................................ xxi Glossary ..................................................................................................... xxiii Acronyms and abbreviations ....................................................................... xxv
Research publications and contributions ............................................................................ xxvi Refereed conference proceedings............................................................... xxvi Journal papers .......................................................................................... xxvii Abstract ...................................................................................................... xxix
Chapter 1 : Introduction ......................................................................................................... 33 1.1 Background and Problem Overview .................................................. 33 1.2 Setting the Scene: The Context for This Research .............................. 34 1.3 Research Objectives ........................................................................... 39 1.4 Research Scope and Limitations ........................................................ 42
2.1.1 Irrigation in Australia ....................................................................... 44 2.2 Exploring Energy and Water Nexus ........................................................ 46
2.2.1 Water and Energy Indicators ............................................................ 50 2.2.2 Water Footprints of Energy Production/Use .................................... 50 2.2.3 Environmental Footprints of Crop Production ................................. 54 2.2.4 Water Market as a Driver in Water-Energy Nexus .......................... 58 2.2.5 Implications of Introduction of ‘Cap’ .............................................. 61
2.3 Greenhouse Gas Emissions from Agriculture ......................................... 62 2.3.1 Direct and Indirect Emissions .......................................................... 62
2.4 Water Efficiency in Irrigation ................................................................. 65 2.4.1 Irrigation Project Efficiency ............................................................. 66 2.4.2 Whole-of-System Approach ............................................................. 68
2.5 Water-energy nexus for irrigation supply systems .................................. 71 2.6 Conversion to efficient irrigation systems ............................................... 75
2.6.1 Efficient Irrigation Technologies and Controlling Groundwater Rise 77
2.7 Water-energy nexus for horticulture in Australia ................................... 78 2.8 Energy availability and food security ..................................................... 80 2.9 Fertigation – a better way of saving energy input .................................. 82 2.10 Irrigation Management Strategies ........................................................ 84
2.11 Application of System Dynamics in Agriculture ................................... 86 2.12 Up-scaling Water and Energy Use ....................................................... 86 2.13 Testing economic viability of irrigated systems .................................... 87 2.14 Reliability of Irrigation Supply ............................................................. 87
Chapter 3 : Methodology ........................................................................................................ 89 3.1 Description of Study Region .................................................................... 89
3.1.1 The Murrumbidgee River Catchment .............................................. 90 3.1.2 Study Area Selection ........................................................................ 94 3.1.3 The Case Study Site ....................................................................... 108 3.1.4 Data Collection/Collation and Analysis ......................................... 110
3.2 The Overall Approach ........................................................................... 115
ii
3.2.1 Application of System Dynamics Approach .................................. 118 3.3 Node-link model Development .............................................................. 119
3.3.1 Modules of the Developed Node-link model ................................. 122 3.4 Node-link model Mass Balance Test ..................................................... 174 3.5 Demand-based verses fixed interval scheduling for different irrigation methods ....................................................................................................... 175 3.6 Calculating water and energy efficiency and productivity indicators .. 176 3.7 Structure of the Thesis Report ............................................................... 180 3.8 Chapter Summary ................................................................................. 182
Chapter 4 : Water and Energy Nexus for Demand Based Irrigation Methods and Conveyance Systems .............................................................................................................. 183
4.1 Rationale of this chapter ....................................................................... 183 4.2 Scenario 1 - Flood irrigation with open channel supply system........... 188
4.2.1 Irrigation demand versus irrigation delivery .................................. 189 4.2.2 Estimation of water losses.............................................................. 190 4.2.3 Effect on crop yield ........................................................................ 191 4.2.4 Irrigation Application Rate ............................................................ 193 4.2.5 Accounting for Energy Use and GHG Emissions in Crop Production for Scenario 1 ....................................................................... 194
4.3 Scenario 2 - Furrow irrigation with open channel supply system ........ 200 4.3.1 Irrigation demand versus irrigation delivery .................................. 201 4.3.2 Water losses estimation .................................................................. 202 4.3.3 Effect on crop yield ........................................................................ 203 4.3.4 Irrigation application rate ............................................................... 203 4.3.5 Accounting for energy use and GHG emissions in crop production for Scenario 2 ........................................................................ 204
4.4 Scenario 3 - Flood irrigation with pipe supply system ......................... 211 4.4.1 Optimization of pipe diameters and why ....................................... 211 4.4.2 Irrigation supply, losses and irrigation application rates ............... 213 4.4.3 Accounting for energy use and GHG emissions in crop production for Scenario 3 ........................................................................ 214
4.5 Scenario 4 - Furrow irrigation with pipe supply system ...................... 217 4.5.1 Optimization of pipe diameters ...................................................... 217 4.5.2 Irrigation supply, losses and irrigation application rates ............... 218 4.5.3 Accounting for energy use and GHG emissions in crop production for Scenario 4 ........................................................................ 219
4.6 Scenario 5 - Sprinkler irrigation with pipe supply system .................... 221 4.6.1 Irrigation demand versus irrigation delivery .................................. 222 4.6.2 Water losses estimation .................................................................. 223 4.6.3 Effect on crop yield ........................................................................ 224 4.6.4 Irrigation application rate ............................................................... 224 4.6.5 Accounting for energy use and GHG emissions in crop production for Scenario 5 ........................................................................ 225
4.7 Scenario 6 – Drip irrigation with pipe supply system .......................... 234 4.7.1 Irrigation demand versus irrigation delivery .................................. 235 4.7.2 Water losses estimation .................................................................. 236 4.7.3 Effect on crop yield ........................................................................ 237 4.7.4 Irrigation application rate ............................................................... 237 4.7.5 Accounting for energy use and GHG emissions in crop production for Scenario 6 ........................................................................ 238
iii
4.8 Comparison of the demand-based irrigation scenarios ........................ 248 4.8.1 Comparison of water and energy use rates .................................... 249 4.8.2 Comparison of efficiency and productivity indicators for water and energy ............................................................................................... 250 4.8.3 Comparison of greenhouse gas emissions for modelled scenarios 260
4.9 Sensitivity analysis ................................................................................ 262 4.9.1 Sensitivity of energy use in irrigation ............................................ 262
4.10 Chapter summary ................................................................................ 266 Chapter 5 : Water and Energy Nexus for Supply Based Irrigation Methods and Conveyance Systems .............................................................................................................. 269
5.1 Description of modelled scenarios ........................................................ 269 5.1.1 Scenario 1: Flood irrigation supplied with an open channel system 269 5.1.2 Scenario 2: Furrow irrigation supplied with an open channel system 269 5.1.3 Scenario 3: Sprinkler irrigation system connected with communal piped supply ............................................................................................ 270 5.1.4 Scenario 4: Drip irrigation system connected with communal piped supply ............................................................................................ 270
5.2 Modifications made in the node-link model .......................................... 270 5.2.1 Modifications in crop water use module ........................................ 270 5.2.2 Modifications in irrigation supply/conveyance module ................. 271 5.2.3 Modifications in irrigation application rate and irrigation interval 273
5.3 Determining irrigation application rate ............................................... 274 5.4 Water use and yield comparison of supply-based and demand-based irrigation ..................................................................................................... 276
5.4.1 Comparison of total irrigation water use ........................................ 277 5.4.2 Comparison of net irrigation rate ................................................... 278 5.4.3 Comparison of crop yield ............................................................... 278 5.4.4 Comparison of water losses ........................................................... 279
5.5 Energy and GHG emissions for the supply-based scenarios ................ 281 5.5.1 Comparison of energy use and energy output ................................ 281 5.5.2 Energy efficiency and energy productivity indicators ................... 286 5.5.3 Comparison of greenhouse gas emissions ...................................... 289
5.6 Sensitivity analysis of pressurized irrigation scenarios ........................ 291 5.6.1 Sensitivity of irrigation supply, pumping energy and yield to irrigation interval ..................................................................................... 291 5.6.2 Sensitivity of crop yield and energy use to irrigation water use .... 295
5.7 On-farm storages: water-energy analysis ............................................. 302 5.7.1 Function of on-farm storages ......................................................... 302 5.7.2 Incorporating on-farm storages into supply-based model .............. 303 5.7.3 Comparison of with and without on-farm storage scenarios ......... 304
5.8 Chapter summary .................................................................................. 315 5.8.1 Summary of the key variables ........................................................ 317 5.8.2 Pros and cons of demand-based versus supply-based irrigation strategy 320
Chapter 6 : Up-scaling Water and Energy Linkages from Case Study to Irrigation Scheme Level 323
6.1 Prerequisites for up-scaling demand-based irrigation system ............. 323 6.1.1 Data preparation and approach for up-scaling ............................... 325 6.1.2 Limitations regarding up-scaling water and energy use................. 328
iv
6.2 Node-link model run for representative area unit ................................ 333 6.3 Up-scaling the model results using mosaic approach .......................... 334
6.3.1 Water and energy use at representative area unit scale .................. 334 6.3.2 Water and energy use at MIA scale ............................................... 338 6.3.3 Water and energy use under different climatic conditions ............ 342
6.4 Estimating and mapping water and energy savings for MIA – using GIS-Based distributed approach ................................................................. 343 6.5 Estimating water and energy use at different levels of technology adoption ...................................................................................................... 352 6.6 Chapter Summary ................................................................................. 354
Chapter 7 : Is Irrigation Conversion Worthwhile? ............................................................ 357 7.1 Need for water saving irrigation technologies ..................................... 358
7.1.1 Water availability ........................................................................... 359 7.1.2 Water markets ................................................................................ 360 7.1.3 Crop yield improvement ................................................................ 363
7.2 Representative node-link model ............................................................ 364 7.2.1 Modelled water and energy use ..................................................... 364
7.3 Capital cost for conversion to pressurized irrigation system ............... 365 7.3.1 Assumptions for the economic analysis ......................................... 366 7.3.2 Capital costs of the irrigation systems ........................................... 367 7.3.3 Capital costs of pressurized pipe irrigation supply system ............ 369
7.4 Economic analysis of conversion to sprinkler or drip system for citrus370 7.4.1 Operating costs for furrow irrigation with citrus ........................... 371 7.4.2 Operating costs for low head sprinkler irrigation with citrus ........ 372 7.4.3 Operating costs for surface drip irrigation with citrus ................... 373 7.4.4 Financial benefits/returns from citrus with the three irrigation systems 375 7.4.5 Discounted payback period and financial viability of the three irrigation systems for citrus..................................................................... 376
7.5 Economic analysis of conversion to sprinkler or drip system for wine grapes .......................................................................................................... 382
7.5.1 Operating costs for furrow irrigation with wine grapes ................. 382 7.5.2 Operating costs for sprinkler irrigation with wine grapes.............. 383 7.5.3 Operating costs for drip irrigation with wine grapes ..................... 384 7.5.4 Financial benefits/returns from wine grapes irrigated with the three irrigation systems ........................................................................... 385 7.5.5 Discounted payback period and financial viability of the three irrigation systems for growing wine grapes ............................................ 386
Chapter 8 : Integrated Analysis, Discussion and Policy Implications ............................... 397 8.1 Understanding and representing the dynamics of the system ............... 397
8.1.1 Water availability versus water saving feedback loop ................... 398 8.1.2 Water savings versus energy use feedback loop ............................ 399 8.1.3 Water savings versus environmental benefits feedback loop ........ 400 8.1.4 Analysis of the feedback dynamics of the integrated system ........ 401
8.2 Discussion on main findings and policy implications ........................... 403 8.2.1 Modelling of water and energy for irrigation systems ................... 403 8.2.2 Water and energy nexus for irrigation strategy .............................. 404 8.2.3 Water and energy nexus for irrigation methods ............................. 404
v
8.2.4 Up-scaling modelled water and energy use.................................... 406 8.2.5 Effectiveness of on-farm storages versus centralized irrigation supply 407 8.2.6 Long-term viability of irrigation conversion .................................. 408 8.2.7 View from system dynamics lens .................................................. 411
Chapter 9 : Conclusions and the Way Forward .................................................................. 413 9.1 Major recommendations ....................................................................... 416
9.1.1 Recommendations for policy makers ............................................. 416 9.1.2 Recommendations for irrigators ..................................................... 417 9.1.3 Recommendations for irrigation providers .................................... 418
9.2 The Way Forward ................................................................................. 418 9.3 Changes in Developed Model for Application in Other Areas ............. 419
References 421 Appendix A: Excerpts from Vensim code for calculation of different model variables .. 447 Appendix B: A snapshot of developed Vensim model in dynamic simulation mode ....... 450 Appendix C: Fertilizer and chemicals input costs .............................................................. 451 Appendix D: Tractor operating costs as per 2008 .............................................................. 452
vi
vii
List of figures Figure 1.1: Water and energy efficiency feedback loop diagram ........................................ 39
Figure 2.1: Energy use and energy intensity by each sector in Australia in 2009-10 (Source: ABS, 2011) .......................................................................................................................... 47
Figure 2.2: Water use by each sector in Australia (Source: ABS, 2012) ............................. 48
Figure 2.3: Natural and regulated average monthly flows in Murrumbidgee River recorded at Balranald station before it joins the Murray River ........................................................... 57
Figure 2.4: Relative distribution of Australia’s direct greenhouse gas emissions by economic sector for 2009-10 (Source: DCC&EE, 2012)..................................................... 65
Figure 2.5: Monthly irrigation application rates to citrus using drip irrigation and low-level micro-sprinklers (Source: Falivene et al., 2006) .................................................................. 76
Figure 3.1: Major rivers and their tributaries in the Murray Darling Basin. (Source: www.mdba.gov.au) .............................................................................................................. 90
Figure 3.2: Dominant land uses of the Murrumbidgee region and its location in MDB (Source: CSIRO, 2008) ........................................................................................................ 92
Figure 3.3: Location of Murrumbidgee Irrigation Area in MDB and its five irrigation districts (Source: Murrumbidgee Irrigation Ltd.) ................................................................ 96
Figure 3.4: Irrigation supply and drainage network of MIA in its five irrigation districts (Source: Murrumbidgee Irrigation Ltd.) .............................................................................. 96
Figure 3.5: Rainfall zones of the Murrumbidgee catchment (Khan at al., 2005) ................. 97
Figure 3.6: Average annual rainfall for each decade since 1950 (Source: Patched Point Dataset from Silo at: http://www.longpaddock.qld.gov.au/silo/) ....................................... 98
Figure 3.7: Monthly Potential Evapotranspiration in the Murrumbidgee Catchment .......... 99
Figure 3.8: Soil groups and their percentage area in MIA (Source: Geoff Beecher’s soils database, unpublished) ....................................................................................................... 100
Figure 3.9: Percentage of irrigation area used by different irrigation systems in the Murrumbidgee Valley (Source: Hope and Wright, 2003) ................................................. 103
Figure 3.10: Irrigation systems in use as per cent of total irrigated area in MIA (Source: Ahmad and Khan, 2009) .................................................................................................... 104
Figure 3.11: Soil types map of the study area (Downloaded from http://www.irrigateway.net/tools/soilmaps/) ...................................................................... 109
Figure 3.12: Daily observed rainfall and evaporation and calculated potential evapotranspiration at Griffith CSIRO gauge for 2007-08 .................................................. 112
Figure 3.13: Daily observed maximum and minimum temperature at Griffith CSIRO gauge for 2007-08 ........................................................................................................................ 112
Figure 3.14: An inventory of factors involved in water and energy consumption and greenhouse gas emissions in irrigation supply systems: open channel network (left), pressurized pipe network (right) (Variables in dotted box are optional). .......................... 117
Figure 3.15: Hypothetical curves of water savings and associated energy use .................. 118
Figure 3.16: Schematic of farm nodes and supply channels/pipes (in parenthesis: channel/pipe length in metres) ........................................................................................... 120
Figure 3.17: Flowchart of interaction among different modules of the node-link model .. 124
Figure 3.18: Steps involved in calculation of crop cover fraction for citrus and stonefruit132
Figure 3.19: Steps involved in calculation of ETc using dual crop coefficient as implemented in the model ................................................................................................. 137
Figure 3.20: Causes tree for ETc adjusted for water stress for Farm No. 6 ....................... 138
viii
Figure 3.21: Schematic of root zone with water balance components (Adapted from Allen et al. (1998). ....................................................................................................................... 139
Figure 3.22: Vensim screen for setting optimisation parameters including optimisation decision variables .............................................................................................................. 141
Figure 3.23: Flowchart of parameter optimisation process as setup in Vensim optimisation framework .......................................................................................................................... 144
Figure 3.24: Setup screen for the objective function definition in Vensim ....................... 145
Figure 3.25: User interface of the developed dynamic model in Vensim model development environment ....................................................................................................................... 152
Figure 3.26: Causes Tree for flow volume at Node 9 of the open channel supply system 156
Figure 3.27: Schematic of supply pipe with outlet pipes to two farms .............................. 160
Figure 3.28: Flowchart of steps to account for energy use, productivity indicators as well as carbon footprint of energy use in irrigation and crop production ...................................... 167
Figure 3.29: Plot between applied water (including rainfall) and yield for citrus crops in South Australia (Source: Skewes, 2010) ........................................................................... 171
Figure 3.30: Feedback loops identified and quantified through integration of modelled variables ............................................................................................................................. 174
Figure 3.31: Flowchart of supply based irrigation strategy as implemented in the node-link model (n=days since start of simulation, d=days since crop gone in water stress) ............ 177
Figure 3.32: Water use accounting components at field scale (Adapted from Molden et al., 2003). ................................................................................................................................. 178
Figure 4.1: Seepage and evaporation losses from channel system of Murrumbidgee Irrigation Area (MIA) ........................................................................................................ 184
Figure 4.2: Comparison of daily irrigation demand and supply and cumulative shortage in irrigation supply (ML shown on y-axis) for Scenario 1 ..................................................... 190
Figure 4.3: Cumulative irrigation water losses (ML shown on y-axis) for Scenario 1 ...... 191
Figure 4.4: Normal and water deficit affected cumulative evapotranspiration (mm shown on y-axis) for the three crops for Scenario 1 ........................................................................... 192
Figure 4.5: Comparison of daily irrigation demand and supply and cumulative shortage in irrigation supply (ML shown on y-axis) for Scenario 2 ..................................................... 202
Figure 4.6: Cumulative irrigation water losses (ML shown on y-axis) for Scenario 2 ...... 203
Figure 4.7: Daily number of parallel pumps turned on to supply irrigation water for Scenario 3 .......................................................................................................................... 214
Figure 4.8: Comparison of daily irrigation demand and supply and cumulative shortage in irrigation supply (ML/day on Y-axis) for Scenario 5 ........................................................ 223
Figure 4.9: Time series of the daily number of pumps turned on in parallel configuration to supply irrigation water for Scenario 5 ............................................................................... 233
Figure 4.10: Comparison of daily irrigation demand and supply and cumulative shortage in irrigation supply (ML/day on Y-axis) for Scenario 6 ........................................................ 236
Figure 4.11: Daily number of pumps turned on in parallel configuration to supply irrigation water for Scenario 6 ........................................................................................................... 247
Figure 4.12: Irrigation application rates (ML/ha) for each crop for the six scenarios ....... 249
Figure 4.13: Energy use per hectare (KWh/ha) for each crop for the six scenarios ........... 250
Figure 4.14: Total greenhouse gas emissions per hectare (kg-CO2e) of each crop for the six scenarios (line graph shows GHG emissions from irrigation only and not other factors of crop production) ................................................................................................................ 261
ix
Figure 4.15: Cumulative probability distribution plots for the delivery pressure head for sprinkler (left) and drip system (right) ............................................................................... 263
Figure 4.16: Sensitivity of cumulative energy use (kWh) for sprinkler irrigation pumping to ±10% change in delivery pressure head (m) ...................................................................... 263
Figure 4.17: Sensitivity of cumulative energy use (kWh) for drip irrigation pumping to ±10% change in delivery pressure head (m) ...................................................................... 264
Figure 4.18: Cumulative probability distribution plots for the irrigation deficit factor for sprinkler (left) and drip system (right) ............................................................................... 265
Figure 4.19: Sensitivity of cumulative energy use (kWh) for sprinkler irrigation pumping to ±50% change in deficit factor ............................................................................................ 265
Figure 4.20: Sensitivity of cumulative energy use (kWh) for drip irrigation pumping to ±50% change in deficit factor ............................................................................................ 266
Figure 5.1: Process of triggering irrigation application events for a given irrigation method ........................................................................................................................................... 272
Figure 5.2: Layout of the module for optimization of the irrigation application rate for each crop .................................................................................................................................... 275
Figure 5.3: Maximum-minimum range and the optimized rates of irrigation for the three crops under the four scenarios ........................................................................................... 277
Figure 5.4: Percentage exceedance plots of total duty flow for pumps for demand-based and supply based irrigation (top plot: drip system, bottom plot: sprinkler system) .................. 285
Figure 5.5: Sensitivity graphs of cumulative irrigation supply (ML) for sprinkler (top) and drip (bottom) systems to irrigation interval (days) ............................................................ 293
Figure 5.6: Sensitivity graphs of cumulative pumping energy use (kWh) for sprinkler (top) and drip (bottom) to irrigation interval (days) ................................................................... 294
Figure 5.7: Sensitivity graphs of yield (t/ha) for sprinkler (left) and drip (right) to irrigation interval (days) for the three crops ...................................................................................... 295
Figure 5.8: Sensitivity of crop yield to irrigation water use for the three modelled crops with drip system ................................................................................................................. 297
Figure 5.9: Sensitivity of irrigation pumping energy consumption (kWh/ha) to irrigation water use (ML/ha) for the three modelled crops with drip system .................................... 298
Figure 5.10: Sensitivity of crop yield to irrigation water use for the three modelled crops with sprinkler system ......................................................................................................... 300
Figure 5.11: Sensitivity of irrigation pumping energy consumption to irrigation water use for the three modelled crops with sprinkler system ........................................................... 301
Figure 5.12: Flowchart of steps to execute supply-based model with on-farm storages ... 306
Figure 6.1: Map showing horticultural farm boundaries and their soil textural classes in the Murrumbidgee Irrigation Area........................................................................................... 325
Figure 6.2: Map of the five soil groups in the MIA horticultural area ............................... 327
Figure 6.3: Sensitivity of cumulative water use (ML) when the irrigated area is uniformly varied by ±50% for the node-link model with a drip system (blue line refers to the baseline value) ................................................................................................................................. 329
Figure 6.4: Sensitivity of cumulative energy use (KWh) when the irrigated area is uniformly varied by ±50% for the node-link model with a drip system (blue line refers to the baseline value) ............................................................................................................. 330
Figure 6.5: Water use (ML) and energy use (kWh) up-scaled from the model results for the whole MIA horticulture area for different climatic conditions .......................................... 342
Figure 6.6: Map of water use rate (ML/ha) for each horticultural farm in MIA for flood irrigation ............................................................................................................................ 349
x
Figure 6.7: Map of water use rate (ML/ha) for each horticultural farm in MIA for sprinkler irrigation ............................................................................................................................ 349
Figure 6.8: Map of water use rate (ML/ha) for each horticultural farm in MIA for drip irrigation ............................................................................................................................ 350
Figure 6.9: Map of water savings (ML/ha) for each horticultural farm in MIA for sprinkler irrigation ............................................................................................................................ 351
Figure 6.10: Map of water savings (ML/ha) for each horticultural farm in MIA for drip irrigation ............................................................................................................................ 351
Figure 6.11: Total water use (ML) and total energy use (MWh) for the two irrigation systems at various level of roll out in MIA horticultural area ........................................... 354
Figure 7.1: Per cent exceedance plot of announced allocation in MIA from 1993-94 to 2009-10 .............................................................................................................................. 359
Figure 7.2: Time series of announced allocation in MIA from 1993-94 to 2009-10 ......... 360
Figure 7.3: Announced percentage allocation versus water trade price ($/ML) in market for MIA from 1998-99 to 2010-11 .......................................................................................... 362
Figure 7.4: Per cent exceedance plots of water trade price ($/ML) and announced allocation (%) for MIA ....................................................................................................................... 363
Figure 7.5: Net present value plots of furrow irrigation with citrus over a period of 30 years ........................................................................................................................................... 379
Figure 7.6: Net present value plots of sprinkler irrigation with citrus connected with an integrated supply system over a period of 30 years ........................................................... 380
Figure 7.7: Net present value plots of drip irrigation with citrus connected with an integrated supply system over a period of 30 years ........................................................... 381
Figure 7.8: Net present value plots of drip, sprinkler and furrow irrigation with wine grapes connected (excluding furrow) with integrated supply system over a period of 30 years ... 388
Figure 8.1: Water availability, investment and water savings negative feedback loop ..... 399
Figure 8.2: Feedback loop between water savings and energy use .................................... 400
Figure 8.3: Positive feedback loop between water savings and environmental benefits ... 401
Figure 8.4: Representation of the integrated system and the constituent causal feedback loops .................................................................................................................................. 402
Figure 8.5: Annual costs and returns for the three irrigation systems with citrus on per hectare basis (capital cost includes the cost of integrated irrigation supply system, except for furrow irrigation).......................................................................................................... 410
Figure 8.6: Annual costs and returns for the three irrigation systems with wine grapes on a per hectare basis (capital cost includes cost of integrated irrigation supply system, except for furrow irrigation).......................................................................................................... 411
xi
List of tables Table 2.1: Final (end-of-water-year i.e. June) percentage general security irrigation allocations for Murrumbidgee valley ................................................................................... 46
Table 2.2: Distribution of water use (ML/year) by each industry under agriculture in Australia during 2009-10 (Source: ABS, 2012)................................................................... 48
Table 2.3: Global average water footprint of primary energy carriers (Gerbens-Leenes, et al., 2008) .............................................................................................................................. 52
Table 2.4: Water footprint of electricity generation in Australia in 2004‐05 (adapted from ABS, 2006) .......................................................................................................................... 53
Table 2.5: Global warming potential of major greenhouse gases (Source: DCC&EE, 2010) ............................................................................................................................................. 63
Table 2.7: Potential water saving options to improve water use/irrigation efficiencies (adapted from Khan et al., 2005a) ....................................................................................... 68
Table 2.8: Terms and definitions of irrigation efficiency at different scales as proposed by different researchers............................................................................................................. 69
Table 2.9: Accounted losses and potential water savings in MIA (Source: Khan et al., 2004) ............................................................................................................................................. 70
Table 2.10: Length of earthen irrigation channels in irrigation areas of Australia (Source: ANCID 2000) ...................................................................................................................... 72
Table 2.11: Crop water use (ML/ha) for horticultural crops and water saving potential by high tech irrigation technologies (Source: Khan et al., 2004) .............................................. 76
Table 2.12: Area and economic output of different agriculture industries in MIA (Source: Singh et al., 2005) ................................................................................................................ 80
Table 2.13: Rice and maize production by modern, transitional and traditional methods ... 82
Table 2.14: Approximate nutrient removals based on tonnes of grapes removed per hectare (Source: Giddings, 2004) ..................................................................................................... 83
Table 3.1: Land use distribution in the Murrumbidgee Valley in the year 2000 (Source: BRS, 2005) .......................................................................................................................... 94
Table 3.2: Water entitlements (licenses) in MIA and the Murrumbidgee Valley .............. 101
Table 3.3: Water balance for irrigation delivery system of MIA (all values in GL. source: MIA 2010) ......................................................................................................................... 106
Table 3.4: Information on basic features of the case study area ........................................ 110
Table 3.5: Details of Horticultural farms in the case study area ........................................ 111
Table 3.6: Summary of climatic data used in this study (Griffith CSIRO) ........................ 113
Table 3.7: Average irrigation application data for the three crops in the case study area .. 114
Table 3.8: Soil-water characteristics of WSL and LCL for the case study area ................ 115
Table 3.9: Monthly basal crop coefficients (Kcb) for modelled horticultural crops (Allen et al., 1998) ............................................................................................................................ 129
Table 3.10: Soil water characteristics used in calculation of soil evaporation reduction coefficient, Kr .................................................................................................................... 130
Table 3.11: Values of wetted and vegetative covered soil fraction for irrigation methods and crops modelled in this study .............................................................................................. 132
Table 3.12: Effective root zone and depletion fraction values used for the case study area ........................................................................................................................................... 136
xii
Table 3.13: Calibration variables and their calibrated model values for years 2003-04, 2004-05 and 2005-06 ......................................................................................................... 146
Table 3.14: Comparison of irrigation application rates (ML/ha) between the actual and the calibrated model ................................................................................................................ 146
Table 3.15: Soil-water availability parameters using calibrated model data for the three crops .................................................................................................................................. 147
Table 3.16: Model validation by comparing actual and modelled drip irrigation application rates (ML/ha) for horticultural crops on 13 farms in the study area (Figure in brackets is total number of irrigation days) ......................................................................................... 148
Table 3.17: Reported water use (ML/ha) for fruit and vines (Figures in braces are total crop area in ha) (Sources: MIA 2005, 2006, 2007, 2008, 2009). ............................................... 150
Table 3.18: Physical features of the open channel system in the case study area .............. 152
Table 3.19: Maximum flow capacities of the open channels in the case study area .......... 154
Table 3.20: Main characteristics of the pipe system .......................................................... 157
Table 3.21: Indicative pressure head requirement at each farm outlet ............................... 159
Table 3.22: Pipe size variations and the corresponding sudden contraction loss coefficient Cc values ............................................................................................................................ 161
Table 3.23: Energy equivalent values for different farm inputs and outputs ..................... 166
Table 3.24: CO2 equivalent emissions factors for various farm inputs .............................. 170
Table 3.25: and values for the modelled crops ..................................................... 172
Table 3.26: Mass balance components as computed by model run for 2007-08 ............... 175
Table 3.27: Indicators of water and energy use efficiency and productivity ..................... 178
Table 3.28: Summary of key topics of the thesis ............................................................... 180
Table 4.1: Details about the crops in the modelled case study area ................................... 186
Table 4.2: Wetted area (m2/ha) for the modelled irrigation methods and the crops .......... 188
Table 4.3: Effect of water deficit due to channel capacity constraint on ETc (transpiration only) and crop yield ........................................................................................................... 192
Table 4.4: Average irrigation application rate for the three crops for the modelled Scenario 1 ......................................................................................................................................... 193
Table 4.5: Estimated time and fuel expended by channel operators to manage the irrigation orders for the farms in the case study area in a year .......................................................... 195
Table 4.6: Accounts for Energy inputs and GHG emissions on per hectare basis in the yearly production cycle of citrus crop for Scenario 1 ........................................................ 196
Table 4.7: Nutrient contents in major fertilizers and their application rates to supply 1kg of N, P or K ............................................................................................................................ 197
Table 4.8: Accounts for Energy inputs and GHG emissions on per hectare basis in the yearly production cycle of stone fruit crop for Scenario 1 ................................................. 198
Table 4.9: Accounts for Energy inputs and GHG emissions on per hectare basis in the yearly production cycle of wine grape crop for Scenario 1 ............................................... 199
Table 4.10: Average irrigation application rates for the three crops for the modelled Scenario 2 .......................................................................................................................... 204
Table 4.11: Estimated time and fuel expended by channel operators to the manage irrigation orders for the farms in the case study area in a year .......................................................... 205
Table 4.12: Accounts for Energy inputs and GHG emissions on per hectare basis in the yearly production cycle of citrus crop for Scenario 2 ........................................................ 206
xiii
Table 4.13: Accounts for Energy inputs and GHG emissions on per hectare basis in the yearly production cycle of stone fruit crop for Scenario 2 ................................................. 208
Table 4.14: Accounts for Energy inputs and GHG emissions on per hectare basis in the yearly production cycle of wine grapes crop for Scenario 2 .............................................. 209
Table 4.15: Original and optimized diameters for supply pipe network ............................ 212
Table 4.16: Comparison of losses and irrigation application rates for Scenario 3 and Scenario 1 .......................................................................................................................... 213
Table 4.17: Irrigation water applied and corresponding energy consumption in irrigation pumping for Scenario 3 ..................................................................................................... 215
Table 4.18: Accounts for Energy inputs and GHG emissions on per hectare basis in the yearly production cycle of citrus, stone fruit and wine grapes for Scenario 3 ................... 216
Table 4.19: Original and optimized diameters for supply pipe network for Scenario 4 .... 218
Table 4.20: Comparison of losses and irrigation application rates for Scenario 4 and Scenario 2 .......................................................................................................................... 219
Table 4.21: Irrigation water applied and corresponding energy consumption in irrigation pumping for Scenario 4 ..................................................................................................... 220
Table 4.22: Accounts for energy inputs and corresponding greenhouse gas emissions on per hectare basis in the yearly production cycle of citrus, stone fruit and wine grapes for Scenario 4 .......................................................................................................................... 221
Table 4.23: Average irrigation application rates for the three crops for the modelled Scenario 5 .......................................................................................................................... 225
Table 4.24: Accounts for Energy inputs and GHG emissions on per hectare basis in the yearly production cycle of citrus crop for Scenario 5 ........................................................ 228
Table 4.25: Accounts for Energy inputs and GHG emissions on per hectare basis in the yearly production cycle of stone fruit crop for Scenario 5 ................................................. 229
Table 4.26: Accounts for Energy inputs and GHG emissions on per hectare basis in the yearly production cycle of wine grapes crop for Scenario 5 .............................................. 230
Table 4.27: Irrigation water applied and corresponding energy consumption in irrigation pumping for Scenario 5 ..................................................................................................... 232
Table 4.28: Accounts for energy inputs and corresponding greenhouse gas emissions on per hectare basis in the yearly production cycle of citrus, stone fruit and wine grapes for Scenario 5 .......................................................................................................................... 234
Table 4.29: Average irrigation application rates for the three crops for the modelled Scenario 6 .......................................................................................................................... 238
Table 4.30: Accounts for energy inputs and GHG emissions on per hectare basis in the yearly production cycle of citrus crop for Scenario 6 ........................................................ 241
Table 4.31: Energy inputs and GHG emissions on per hectare basis in the yearly production cycle of stone fruit crop for Scenario 6 .............................................................................. 242
Table 4.32: Energy inputs and GHG emissions on per hectare basis in the yearly production cycle of wine grapes crop for Scenario 6 ........................................................................... 244
Table 4.33: Irrigation water applied and corresponding energy consumption in irrigation pumping for Scenario 6 ..................................................................................................... 246
Table 4.34: Energy inputs and corresponding greenhouse gas emissions on per hectare basis in the production cycle of citrus, stone fruit and wine grapes for Scenario 6 .................... 248
Table 4.35: Computed overall/project level irrigation efficiency for the six scenarios ..... 254
Table 4.36: Water productivity (kg/m3) indicators for the six scenarios ........................... 255
Table 4.37: Energy productivity (kg/kWh) indicators for the six scenarios ...................... 256
xiv
Table 4.38: Energy efficiency (kWh/kWh) indicators for the six scenarios ...................... 258
Table 4.39: Specific energy (kWh/kg) indicators for the six scenarios ............................. 259
Table 4.40: Water – energy productivity (g/m3/kWh) indicators for the six scenarios ...... 259
Table 4.41: Water – energy ratio (kWh/kWh) for the six scenarios .................................. 260
Table 5.1: Irrigation intervals used in the model for the four supply-based irrigation scenarios ............................................................................................................................ 274
Table 5.2: Comparison of total irrigation water use (ML) between supply-based and demand-based irrigation scenarios ..................................................................................... 278
Table 5.3: Net irrigation rate (ML/ha) for three crops for demand-based and supply-based scenarios ............................................................................................................................ 278
Table 5.4: Comparison modelled crop yield (t/ha) between supply-based and demand-based irrigation systems ............................................................................................................... 279
Table 5.5: Comparison of total water losses (ML) for supply-based and demand-based irrigation scenarios ............................................................................................................ 280
Table 5.6: Energy use for the three crops under demand-based scenarios and the computed energy use for the corresponding supply based scenarios ................................................. 283
Table 5.7: Total equivalent energy output (kWh/ha) from each crop for supply-based and demand-based irrigation scenarios ..................................................................................... 286
Table 5.8: Energy indicators for supply-based irrigation scenarios ................................... 287
Table 5.9: Greenhouse gas emissions rates (kgCO2-Eq/ha) for the three crops under demand-based and supply-based (computed) scenarios .................................................... 290
Table 5.10: Comparison of drip and sprinkler system in terms of yield response to water use ........................................................................................................................................... 299
Table 5.11: Irrigation rates for 4-day irrigation interval and on-farm storage capacity for each farm with sprinkler system ........................................................................................ 307
Table 5.12: Computation of final capacity of each on-farm storage for sprinkler system . 308
Table 5.13: Key variables for with and without on-farm storage scenarios for sprinkler system ................................................................................................................................ 309
Table 5.14: Irrigation rates for 4-day irrigation interval and on-farm storage capacity for each farm with drip system ................................................................................................ 311
Table 5.15: Computation of final capacity of each on-farm storage for drip system ......... 312
Table 5.16: Key variables for with and without on-farm storage scenarios for drip system ........................................................................................................................................... 313
Table 5.17: Comparison of use of on-farm storages and the common piped supply ......... 314
Table 5.18: Summary of important variables for all scenarios modelled in Chapter 5 under supply-based irrigation strategy ......................................................................................... 318
Table 5.19: Comparison of demand-based and supply-based irrigation strategies (the “high” or “low” refers to comparison with each other) ................................................................. 320
Table 6.1: Soil groups and their equivalent USDA soil types ........................................... 327
Table 6.2: Comparison of increase in pumping energy use with increase in total irrigated area for a supply-based drip irrigation strategy ................................................................. 330
Table 6.3: Distribution of different soil classes in MIA horticultural farms and percentage distribution of area of each horticultural crop for each soil type ....................................... 331
Table 6.4: Distribution of different soil groups in MIA horticultural farms and percentage distribution of area of each horticultural crop for each soil group ..................................... 332
xv
Table 6.5: Distribution of number of farms in the representative unit for each model run using a given soil group and crop area (in parentheses, ha) ............................................... 333
Table 6.6: Water and pumping energy uses for different soil groups with sprinkler irrigation system as computed by the demand-based node-link model for the representative unit area (300 ha) for 2007-08 .......................................................................................................... 336
Table 6.7: Water and pumping energy uses for different soil groups with drip irrigation system as computed by the demand-based node-link model for the representative unit area (300 ha) for 2007-08 .......................................................................................................... 337
Table 6.8: Water use for sprinkler and drip irrigation system for five soil groups at model scale and at the scale of the MIA irrigated horticulture area ............................................. 340
Table 6.9: Energy use in irrigation pumping and conveying for sprinkler and drip irrigation system for five soil groups at model scale and at the scale of the MIA irrigated horticulture area .................................................................................................................................... 340
Table 6.10: Water and Energy use for flood irrigation at the model scale for each crop for average climatic conditions ............................................................................................... 345
Table 6.11: Water and Energy use for sprinkler system at the model scale for each crop for average climatic conditions ............................................................................................... 346
Table 6.12: Water and Energy use for drip system at the model scale for each crop for average climatic conditions ............................................................................................... 346
Table 6.13: Total and unit area based water and energy use for sprinkler and drip systems for average climatic conditions for MIA horticultural area ............................................... 353
Table 6.14: Comparison of the two up-scaling methods for water and energy use over 28,970 ha area of MIA ....................................................................................................... 355
Table 7.1: Yield (t/ha) of citrus and wine grapes for various irrigation systems ............... 363
Table 7.2: Node-link model output for a modelled area of 550 ha .................................... 365
Table 7.3: Assumed values of various parameters for economic analysis ......................... 366
Table 7.4: Capital cost for furrow irrigation system (baseline case) ................................. 367
Table 7.5: Capital cost for conversion to low head sprinkler irrigation system ................. 368
Table 7.6: Capital cost for conversion to drip irrigation system ........................................ 368
Table 7.7: Capital costs of pressurized irrigation supply system (Source: MIA per. com.) ........................................................................................................................................... 369
Table 7.8: Values of common cost items for the three irrigation systems ......................... 371
Table 7.9: Annual operating costs per hectare for citrus with furrow irrigation ................ 372
Table 7.10: Annual operating costs per hectare for citrus with low head sprinkler irrigation ........................................................................................................................................... 373
Table 7.11: Annual operating costs per hectare for citrus with surface drip irrigation system ........................................................................................................................................... 374
Table 7.12: Annual financial returns per unit area per for the three irrigation systems growing citrus .................................................................................................................... 376
Table 7.13: Summary of initial and annual costs and annual returns for the three irrigation systems for citrus ............................................................................................................... 377
Table 7.14: Annual operating costs per hectare for wine grapes with furrow irrigation .... 383
Table 7.15: Annual operating costs per hectare for wine grapes with low-head sprinkler irrigation system ................................................................................................................ 384
Table 7.16: Annual operating costs per hectare for wine grapes with surface drip irrigation system ................................................................................................................................ 385
xvi
Table 7.17: Annual financial returns per unit area for the three irrigation systems growing wine grapes ........................................................................................................................ 386
Table 7.18: Summary of initial and annual costs and annual returns for the three irrigation systems for wine grapes ..................................................................................................... 387
Table 7.19: Profitability indicators for the three irrigation systems irrigating wine grapes over the case study area of 550 ha ..................................................................................... 389
Table 7.20: Sensitivity of present values of benefits, costs and B-C ratio to different cost and revenue items for citrus crop (–ve sign shows decrease with respect to original value) ........................................................................................................................................... 390
Table 7.21: Sensitivity of present values of benefits, costs and B-C ratio to different cost and revenue items for wine grapes crop (–ve sign shows decrease with respect to original value) ................................................................................................................................. 391
Table 7.22: Summary of selected profitability indicators for the three irrigation systems 394
Table 8.1: Greenhouse gas emissions cost as percentage of the total annual operational cost per hectare.......................................................................................................................... 411
xvii
Certificate of authorship
CERTIFICATE OF AUTHORSHIP OF THESIS & AGREEMENT FOR THE RETENTION &
USE OF THE THESIS
DOCTORAL AND MASTER BY RESEARCH APPLICANTS
I Aftab Ahmad
Hereby declare that this submission is my own work and that, to the best of my knowledge and belief, it contains no material previously published or written by another person nor material which to a substantial extent has been accepted for the award of any other degree or diploma at Charles Sturt University or any other educational institution, except where due acknowledgment is made in the thesis. Any contribution made to the research by colleagues with whom I have worked at Charles Sturt University or elsewhere during my candidature is fully acknowledged.
I agree that the thesis be accessible for the purpose of study and research in accordance with the normal conditions established by the University Librarian for the care, loan and reproduction of the thesis.*
Signature Date
* Subject to confidentiality provisions as approved by the University
xviii
xix
Acknowledgements
First and foremost, I am grateful to Allah, the most beneficial and the most
merciful, for granting me courage and intellect to undertake and complete
this study.
I also tender great respect and profound gratitude to my supervisor Professor
Shahbaz Khan for intelligible guidance, research training, moral support and
never-ending encouragement during the course of this study. His valuable
persistence, technical support, timely guidance and positive reinforcement
during my prolonged study are acknowledged and greatly appreciated.
I should also acknowledge the useful discussions with Associate Professor
John Louis from Charles Sturt University and Dr Kumar Narayan from
CSIRO regarding this PhD research. I am also grateful to my employer
Murray-Darling Basin Authority for granting me study leave to meet my
supervisor and to attend conferences when needed. Particularly, I am
thankful to Mr Andy Close for his continued support in this regard. Also big
thanks to my colleagues Mr Nadeem Samnakay and Dr Tariq Rana for their
support through the course of this study.
I gratefully acknowledge the financial support received from (former) Land
and Water Australia for this research. The success in bringing this thesis
into current form is also attributable to the officials of Murrumbidgee
Irrigation, who provided free access to data and useful field information.
Thanks to my late mother, may her soul rest in peace, for her love and
unreserved prayers for my success. Thanks to my lovely boys Rayyan and
Raed for their patience and my brother, sister, nephews, nieces, brothers-
xx
and-sisters-in-law and their family members for benevolent prayers and
sacrifices to achieve my study objectives.
Honest thanks to all my friends and well-wishers in Australia, New Zealand,
England, Thailand, Nepal, Middle East, USA, Canada and Pakistan who are
keen to see me achieve this milestone.
Finally, I would like to thank the special woman in my life, my wife Fozia,
who in every way has provided me with the inspiration, love and care,
which is vital for the long journey of a PhD.
xxi
Ethics approval
This doctoral research work did not involve any direct communications with the farmers or any other group and therefore ethics approval was not sought.
xxii
xxiii
Glossary
The meaning of terms can differ across disciplines. This glossary clarifies
the use of specific terms within the thesis
Channel seepage: Channel seepage can be defined as “loss of water
from a channel via infiltration through micro-pores
and soil processes (i.e. not via preferential flow
through macropores).
Energy efficiency: Energy efficiency of the agricultural production
system can be defined as the ratio of total energy
output from agricultural produce to the total energy
input to engender that produce.
Energy intensity: Measure of the energy consumed by an industry to
produce one unit of economic output
Evapotranspiration: The combined net effect of two processes:
evaporation and transpiration
Fertiliser use (or
recovery) efficiency:
The ratio of the amount of nutrient removed with
the crop to the amount of nutrient applied.
CO2-equivalent
emissions:
A universal measurement of greenhouse gas
emissions; the concentration of CO2 that would
cause the same amount of radiative forcing as a
given mixture of another greenhouse gas. It is
normally expressed in tonnes of CO2.
Net present value: The difference between the present value of cash
inflows (returns) and the present value of cash
outflows (costs) and is widely used for analysing
profitability of long-term projects.
Productivity: The ratio between agricultural output and resource
inputs, e.g. tonnes of product/ML water applied
xxiv
Runoff: Runoff is the movement of water, usually from
precipitation or gravity based irrigation, across the
soil surface towards stream channels, lakes,
depressions, or low points in the soil surface. It
affects the rate of runoff include rainfall duration
and intensity as well as the ground's slope, soil type
and ground cover.
Specific energy: The specific energy of an agricultural production
system is defined as the total energy input per unit
of marketable yield and is expressed as kWh/kg.
Total dynamic
head:
The total dynamic head is the total equivalent
height that a fluid is to be pumped, taking into
account all friction losses in the pipe.
Water intensity: Measure of the water consumed by an industry to
produce one unit of economic output
Water footprint
(WF):
The WF of energy generation is the amount of
water used to produce a unit of energy (m3/GJ).
Watertable: The upper surface of the saturated zone in aquifers
that are not confined by impermeable geologic
material where the water pressure is equal to
atmospheric pressure.
Water use
efficiency:
Water use efficiency is linked with consumptive use
(i.e. evapotranspiration) of water by a given crop. It
is defined as “the ratio between volume of water
consumptively used in evapotranspiration and the
volume of water actually applied.
xxv
Acronyms and abbreviations
ABARE Australian Bureau of Agricultural and Resource Economics ABS Australian Bureau of Statistics ASTM American Society for Testing and Materials B-C ratio Benefit – Cost ratio CIA Coleambally Irrigation Area CoAG Council of Australian Governments CO2-e Carbon dioxide equivalent
CSIRO Commonwealth Scientific and Industrial Research Organisation
DCC&EE Department of Climate Change and Energy Efficiency ET Evapotranspiration ETo Reference Evapotranspiration ETc Crop Evapotranspiration FAO Food and Agriculture Organization GHG Greenhouse Gas GJ Giga joules 109 joules) GL Giga litres (109 litres) ICID International Commission on Irrigation and Drainage KWh Kilowatt-hours MWh Megawatt-hours (1000 KWh) MDB Murray-Darling Basin MDBA Murray-Darling Basin Authority MIA Murrumbidgee Irrigation Area MIL Murray Irrigation Limited ML Mega litres (106 litres) NPB Net Present Benefit NPV Net Present Value NSW New South Wales NVIRP Northern Victoria Irrigation Renewal Project NWC National Water Commission PVC Polyvinyl chloride RAW Readily Available Moisture SILO Specialised Information for Land Owners SD System Dynamics
SEWPaC Sustainability, Environment, Water, population & Communities
SWAGMAN Salt Water and Groundwater MANagement WF Water footprint
xxvi
Research publications and contributions
The following papers were either fully or partly based on analysis of my
collected data, methods and approaches used in this research and developed
water-energy node-link model.
Refereed conference proceedings
1. Ahmad, A., Khan, S., and Louis, J. (2010). Water–energy nexus in
irrigation supply systems using a demand based dynamic node-link
model. In ed. Khan, S., Savenije, H., Demuth, S., Hubert, P. (2010).
Hydrocomplexity: new tools for solving wicked water problems;
proceedings of the Xth Kovacs Colloquium, Paris, 2-3 July 2010;
International Association of Hydrological Sciences (IAHS) Publ. 338;
(2010) ISBN 978-1-907161-11-7, 272 pp.
2. Ahmad, A., and Khan, S. (2009). Comparison of water and energy
productivities in pressurized irrigation systems. Proceedings of
MODSIM, International Congress on Modelling and Simulation, 13-17
July 2009, Cairns, Australia. ISBN: 978-0-9758400-7-8
3. Ahmad, A., Khan, S., and Rana, T. (2007). System dynamics approach
for modelling seasonality of river flows. Proceedings of MODSIM,
International Congress on Modelling and Simulation, 10-13 December
2007, Christchurch, New Zealand.
4. Jackson, T.M., Khan, S., and Ahmad, A. (2007). Exploring energy
productivity for a groundwater dependent irrigated farm using a system
dynamics approach. MODSIM, International Congress on Modelling
and Simulation, 10-13 December 2007, Christchurch, New Zealand.
5. Ahmad, A., and Khan, S. (2008). Systems approach for modelling
dynamics of water and energy inputs in groundwater dependent
In 1971 there were 935 horticultural farms in the MIA. The total area of
permanent plantings on these farms was 10,405 ha (Kennedy, 1973). In
2003 there were more than 1,000 horticultural farms with a total area of
24,800 ha. Grapes and citrus are the two major horticultural enterprises that
107
accounted for 97 per cent of the total area under fruit crops with 37 per cent
under citrus and 60 per cent area under grapes. The rest of the area was
under prunes and other fruits like apricots, peaches, plums nectarines, nuts
(Singh et al. 2005). Currently about 29,237 ha in MIA is covered with
horticulture crops (MIA 2010). Irrigation data for MIA indicates that water
use for irrigating perennial crops (citrus, vines, other fruits) can be as high
as 65% of total delivered irrigation water when general security allocation to
seasonal crops (cereals, vegetables etc.) is restricted due to low water
availability. Therefore this research project is particularly focused on water
savings in different on-farm irrigation methods for perennial crops. Some of
the irrigation methods are more energy intensive than others.
The level of irrigation system modernization especially, for horticulture
crops in MIA is depicted by the level of irrigation technology adoption as
shown in Figure 3.10. A review of the previous studies indicates that up to 4
ML/ha can be realized in water savings by high-pressure drip irrigation
which is being rapidly rolled out for horticulture areas of MIA. Further
water savings can be achieved by improving conveyance efficiency by
lining the leaky channels and replacing some open-channels with pipes. On
one hand such initiatives can result in significant amount of water savings
while on the other hand they require significant energy inputs in various
forms as compared to traditional methods and practices. In addition, broad
acre farms are gradually being converted into horticultural farms in MIA
that exacerbates the abovementioned issues and opportunities.
Murrumbidgee Irrigation Area was selected as a focus study area for this
PhD research as it is experiencing the abovementioned changes and
challenges. The latest challenge for the irrigators in MIA is to find water
savings to achieve a reduction of up to 320 GL in their annual diversions as
part of the Murray-Darling Basin Plan (MDBA, 2012). One of the real
challenges for water managers is to understand the dynamics and feedback
between water savings and energy use and to strike a balance between water
savings, energy consumption and their environmental footprints.
108
3.1.3 The Case Study Site
A communal irrigation site within the larger irrigated area of MIA was
selected as a case study. It is located in Yanco Irrigation District of MIA.
The case study area mainly consists of horticulture farms with citrus as the
dominant crop. The soil map of the case study area is shown in Figure 3.11.
Although there are 10 soil types in the case study area, they were aggregated
into two representative soil types based on irrigated area. The two soil types
are Sandy Loam (SL) and Clay Loam (CL). Taylor and Hooper (1938)
described SL group profile as, “the sandy loam surface and the somewhat
shallower clay subsoil that appears at about 45 cm. The change to the
medium clay may be more rapid, but it is only occasionally met with above
120 cm. The subsoils have a sticky feel and are apparently slowly
permeable”; and the CL group as, “they are mapped at Leeton, Wamoon and
Stanbridge”. The profile always contains a light clay band, frequently
continuing from 100 to 180 cm without change. The surface loam is always
shallow and is probably the cultivated zone of the original clay loam
surface. The subsoil clay bands are variable in thickness with heavy clay
sometimes absent altogether; the light clay occurs between 90 cm and 123
cm. Occasionally, there is a sandy loam surface, the deep subsoil may go to
a sandy clay approaching the Jondaryan clay loam type”. These soils are
placed into Red Brown Earth group. Both soil types are common and
suitable for horticulture crops in the MIA (Taylor and Hooper, 1938).
109
Figure 3.11: Soil types map of the study area (Downloaded from http://www.irrigateway.net/tools/soilmaps/)
The basic information about the case study area is given in Table 3.3. The
irrigation supply system in the area is being gradually converted from
gravity-fed open channels to high pressure piped supply to each farm. This
conversion will allow farmers to switch from flood or furrow irrigation on
their farms to pressurised irrigation systems, which includes sprinkler and
drip systems. The conversion to pressurised pipe irrigation supply also
eliminates the need for farmers to construct on-farm storages which are
highly inefficient and eliminate the need for the installation of pumps on
individual farms which are expensive to run and maintain.
The irrigation system modernisation of both the supply system and on-farm
irrigation technology can realise potential water savings by preventing
losses through seepage, evaporation and run-off. Under the current roster
system the farmers have to wait up to four days for their turn to irrigate due
to access and channel capacity constraints. This limits the frequency of
access to irrigation water and potentially affects crop yield during critical
crop lifecycles and during extreme hot days. The new system under
investigation ensures timely and on-demand supply of water to farms that
potentially leads to improved crop quality, quantity and better farm
110
management. It also minimises conveyance losses including seepage and
evaporation from the unlined open channels. These upgrades however have
an increased energy cost which this thesis is seeking to analyse.
Table 3.4: Information on basic features of the case study area
Item Data Total area (ha) 291 Vines (ha) 22.6 Citrus (ha) 248.6 Stonefruit (ha) 19.8 Total number of farms 13 Vines 2 Citrus 9 Stonefruit 2 Number of trees per hectare Citrus 550 Stonefruit 225 Average radius of a mature plant canopy (m) Citrus 2.0 Stonefruit 1.95 Average row-to-row distance for wine grapes (m) 2.0 Total length of unlined main supply channels (m) 4,074 Average roster time of irrigation for a farm (days) 4
3.1.4 Data Collection/Collation and Analysis
The data used for this research project was mainly collected from
Murrumbidgee Irrigation, Bureau of Meteorology and other public sources.
It includes data on climate variables, water use, landuse maps, soil types,
irrigation infrastructure etc. Due to the commercial sensitivities of some
data, the identifiable data fields like Farm Identification Number are not
disclosed here. The farms in the study area are rather identified by
alphabetical notation as given in Table 3.5.
111
Table 3.5: Details of Horticultural farms in the case study area
Farm No. Farm ID Crop Farm Area
1a A Citrus 54.26
3 B Citrus 35.4
4 C Citrus 28.18
5 D Citrus 35.3
6 E Citrus 27.7
7 F Citrus 28.76
7a G Citrus 11.24
8 H Vine 10.17
9 I Vine 12.43
10 J Citrus 6.87
11 K Citrus 16.32
12 L Stonefruit 4.57
13 M Stonefruit 19.77
Total 291
In the Murrumbidgee catchment and in fact the whole of New South Wales
the water year starts in July when the initial announcement of water
availability for general security entitlement holders for the rest of the year is
made by State Water and subsequently revised if water resource conditions
improve. By November farmers get a very good idea of irrigation water
allocation and plan their annual crops accordingly. The water year 2007-08
(July 2007 to June 2008) was selected as study period which was used to
test the developed model. The hydro-climatic data for the study period is
shown in Figure 3.12 and Figure 3.13. A complete series of available
climatic data is given in Appendix A. The average daily potential
evapotranspiration (ETo) for 2007-08 is 4.1 mm/day.
112
Figure 3.12: Daily observed rainfall and evaporation and calculated potential evapotranspiration at Griffith CSIRO gauge for 2007-08
Figure 3.13: Daily observed maximum and minimum temperature at Griffith CSIRO gauge for 2007-08
Table 3.6 summarizes hydro-climatic variables for the daily data for period
from water year 2003-2004 to water year 2008-2009 at Griffith CSIRO
weather station. This data was sourced from SILO patched point database
(http://www.longpaddock.qld.gov.au/silo/). Out of these six years 2006-
2007 was the driest and hottest year with only 186.6 mm of total rain and
0
15
30
45
1/07/2007
15/07/2007
29/07/2007
12/08/2007
26/08/2007
9/09/2007
23/09/2007
7/10/2007
21/10/2007
4/11/2007
18/11/2007
2/12/2007
16/12/2007
30/12/2007
13/01/2008
27/01/2008
10/02/2008
24/02/2008
9/03/2008
23/03/2008
6/04/2008
20/04/2008
4/05/2008
18/05/2008
1/06/2008
15/06/2008
29/06/2008
Rain/Evaporation (m
m/day)
Date
Rain (mm)
Evap (mm)
FAO56 (ETo) (mm)
‐5
0
5
10
15
20
25
30
35
40
45
1/07/2007
15/07/2007
29/07/2007
12/08/2007
26/08/2007
9/09/2007
23/09/2007
7/10/2007
21/10/2007
4/11/2007
18/11/2007
2/12/2007
16/12/2007
30/12/2007
13/01/2008
27/01/2008
10/02/2008
24/02/2008
9/03/2008
23/03/2008
6/04/2008
20/04/2008
4/05/2008
18/05/2008
1/06/2008
15/06/2008
29/06/2008
Daily Temperature (oC)
Date
Max. Temperature (oC)
Min. Temperature (oC)
113
average daily temperature of 18.2oC and highest average daily
evapotranspiration rate of 4.3 mm/day.
The water use data as given in Table 3.7 shows that the water application
rate for citrus and stonefruit is relatively higher for 2006-2007 as compared
to other years with the exception for vine crops which may have gone under
deficit irrigation in 2006-2007. Deficit irrigation in vine crops is a modern
practice to regulate vegetative growth and improve fruit quality while
achieving high irrigation efficiency (CRCV, 2005; Kriedemann and
Goodwin, 2003; Goodwin, 1995; Goodwin and Boland, 2002).
Table 3.6: Summary of climatic data used in this study (Griffith CSIRO)
Water-year (Jul –
Jun)
2003-
2004
2004-
2005
2005-
2006
2006-
2007
2007-
2008
2008-
2009
Long-
term
average
Average maximum
daily temperature
(oC)
24.3 24.8 24.4 25.5 24.8 24.4 24.7
Average minimum
daily temperature
(oC)
9.9 10.6 10.1 10.9 11.1 11.0 10.6
Average daily
temperature (oC) 17.1 17.7 17.3 18.2 17.9 17.7 17.65
Average
evaporation
(mm/day)
5.5 5.4 5.5 5.9 5.4 5.7 5.57
Total potential
evapotranspiration
(mm)
1469 1493 1460 1558 1493 1501 1497
Average potential
evapotranspiration
(mm/day)
4.0 4.1 4.0 4.3 4.1 4.1 4.1
Total rain
(mm/year) 333 263.1 369.3 186.6 336.4 313 300.2
114
The average water use data expressed as megalitres per hectare (ML/ha) for
citrus, stonefruit and vines for the 13 farms in the case study area is given in
Table 3.7. The data is reported for 2003-04 to 2008-09. All farms are either
irrigated with drip/trickle systems. Previous studies indicate that average
water use by flood irrigated citrus and vines in MIA ranged from 9 ML/ha
to 12 ML/ha and 7 ML/ha to 9 ML/ha, respectively (Khan and Abbas, 2007;
Khan et al. 2005a).
Table 3.7: Average irrigation application data for the three crops in the case study area
Crop 2003-
04
2004-
05
2005-
06
2006-
07
2007-
08
2008-
09
Average
(ML/ha)
Citrus
(ML/ha) 5.4 5.5 5.0 6.0 4.2 5.0 5.2
Stonefruit
(ML/ha) 5.6 5.5 5.7 6.0 4.5 5.7 5.5
Vines (ML/ha) 4.1 4.2 4.5 3.5 4.0 3.6 4.0
The soil-water characteristics given in Table 3.8 for both WSL and LCL soil
types were reported in Hornbuckle and Christen (1999) and determined by
Loveday et al. (1978) and Talsma (1963). These values of soil-water
characteristics for the USDA soil textural classes are similar to those
reported in Rawls et al., (1982) and Allen et al., (1998). The amount of
water stored in the soil profile is the difference between field capacity and
wilting point for a given soil texture. This is the total water storage capacity
of the soil. Plant root system extracts water from different depths depending
on crop type, irrigation frequency and weather conditions. Therefore, an
effective root zone depth is used for each crop and multiplied with soil total
storage capacity, which gives the total plant available water. However, only
a portion of the total plant available water can be extracted by plants without
becoming stressed.
115
Table 3.8: Soil-water characteristics of WSL and LCL for the case study area
Soil Type
Moisture content at field capacity, θfc (10Kpa)
(m3/m3) (A)
Moisture content at wilting point, θwp (1500Kpa)
(m3/m3) (B)
Depth of soil
profile (m)
Total soil water
storage (m3/m3) (A – B)
Sandy loam
0.23 (0.18 – 0.28)
0.11 (0.06 – 0.16)
0.8 0.12
Clay loam
0.34 (0.30 – 0.37)
0.18 (0.15 – 0.21)
1.2 0.16
As given in Table 3.8, sands have less water storage capacity than clays but
most of it is available to plants. Therefore, low but frequent irrigation
should be more effective for sandy soils and vice versa for clays in terms of
water availability to plants.
3.2 The Overall Approach
While the development of a biophysical model for a given area is a complex
and time consuming process; its repeated application to test new scenarios
and interpretation of results requires even more time and expertise.
Furthermore, usually individual models are developed to address specific
aspects of a given area. For example, a groundwater model only simulates
groundwater movement and does not account for any crop-water
interactions. A separate crop-water model is needed to understand crop
water use; another one to simulate water movement in unsaturated zone,
another one to account for energy inputs and so on. Although there are some
modelling platforms available that can integrate all such processes, they are
complex, time consuming and sometimes area specific. The main focus of
this PhD research is to addresses this issue by devising an integrated
framework as an alternative approach to biophysical models. This
framework involves development of a simple yet dynamic node-link model
that is based on general principles and mathematical relationships that are
derived, and lumped to certain extent, from application of complex
biophysical models previously developed for the area. For example, if
groundwater model developed for an area indicates that 20% of its recharge
116
is contributed from surface irrigation for a range of irrigation application
rates then we do not have to run that groundwater model again and again to
find recharge for different crops in the same or similar areas.
Figure 3.14 provides an overview of different factors and processes
involved in understanding the water, energy, and greenhouse gas emissions
nexus. The first step in development of the framework is to gain an insight
of biophysical models for the study area and derive mathematical
relationships among various variables by observing behaviour and response
of key output variables to possible changes in input variables. An important
aspect that requires careful consideration in developing robust integrated
framework is to identify and incorporate any feedback mechanisms that
control interplay and non-linearity among various components and explain
the dynamic behaviour of a system as a whole. For example, Khan et al.
(2009c) described the interaction between evapotranspiration (ET) and
capillary rise for shallow watertable situation: the larger the ET, the larger
the capillary rise, then the larger the soil water content and the water stress
coefficient, which in turn increases ET, completing the positive feedback
loop. However, this mechanism cannot be explained if watertable
information from groundwater model is not taken into account. After
studying and analysing various models and their results for the study area,
an integrated node-link model is developed in this study that links data and
processes including climate, soil, crop water use, irrigation application
(methods and rates), irrigation scheduling, irrigation water supply system,
soil-water movement and groundwater response. Each simulation period
covers one year from July 1st to June 30th with a daily time step. However,
both the simulation time and computation time step can be varied relatively
easily. Each node is a irrigation supply point to the adjacent farm. The
model is developed in the development environment software called
VenSim (Ventana Systems, 2004). The model is relatively simple to use and
dynamic in nature where users can vary any parameter during run-time and
see the model response instantaneously both visually and numerically.
117
Figure 3.14: An inventory of factors involved in water and energy consumption and greenhouse gas emissions in irrigation supply systems: open channel network (left), pressurized pipe network (right) (Variables in dotted box are optional).
The overall goal of this research is to understand the water-energy nexus
and find an optimum match between water saved and energy used, as shown
Total number, volume and
timing of water
Type of irrigation supply system
Gravity based open Pressurized supply pipe
Conveyance & delivery losses
Channel seepage
Channel bed material
Evaporation loss
Climate conditions
Total irrigation demand
Total number, volume and
timing of water orders
Farm crop & soil
Farm irrigation method
On-farm storage
On-farm pumping
Total energy consumed
Energy consumed in construction
Channel length
Channel capacity
constraint
End-of-channel outflow
Rain rejections
Channel pre-filling
Conveyance & delivery losses
Pipe leakage
Age of the fittings
Pipe system capacity
Total irrigation
Farm crop & soil
Farm pressure irrigation system
Total irrigation volume
delivered
Total irrigation demand
Energy consumed
in pumping
Number of pumps in operation
Energy loss in pipe friction
Energy loss due to
Pipe network characteristics
Total energy
consumed
Minimum pressure
head requirement
Steady flow rate
Climate conditio
Greenhouse gas emissions
Greenhouse gas emissions
118
hypothetically in Figure 3.15, within the environmental and economic
constraints for four irrigation systems including flood, furrow, sprinkler and
drip. The approach is similar to Humphreys et al., (2005) where these
irrigation systems were compared side-by-side in terms of net irrigation
water use, net water productivity and yield.
Figure 3.15: Hypothetical curves of water savings and associated energy use
3.2.1 Application of System Dynamics Approach
System dynamics is a system modelling technique. A system level holistic
approach is required to understand the complex interactions amongst use of
water saving irrigation solutions, energy consumption and associated
greenhouse gas emissions and economic rationalization. To achieve
robustness in an integrated model it is vital to ensure that all interactions and
feedback mechanisms are well-understood. The system dynamics approach
will help us conceptualise discover and explain the underlying feedback
mechanisms at the scale of an irrigation scheme in this research. The use of
this approach is elaborated more in the later parts of the thesis.
119
3.3 Node-link model Development
It is possible to measure water savings and energy use of different irrigation
methods by conducting a comprehensive survey of the farms practicing
those irrigation methods and collecting extensive data. But those farms may
be operating at sub-optimal level and therefore, a daily simulation model is
considered a more appropriate, rapid and versatile tool that can be used to
explore maximum water and energy savings potential of different irrigation
supply and application methods. The case study area modelled in this
research consists of 13 horticulture farms (Table 3.5) covering an area of
about 291 hectares in MIA. Each farm grows a single crop only. All farms
are connected to a common water source which is located roughly at the
middle of upper side of the area. Water used to be conveyed under gravity to
the farms via a main earthen open channel which splits into two branch
channels. Those channel structures still exit but now water is conveyed to
these farms via pressurised pipes which are buried parallel to those open
channels and connected to a large water pumping station. A schematic of the
modelled case study area is given in Figure 3.16. The total length of the
distribution channels and also the adjacent irrigation pipes is 4,074 metres.
It is a branched water distribution system supplying water to farms in a line
and is best represented by a nodal network as used by Xevi and Khan
(2005). The node-link model was developed in system dynamics
environment using Vensim software (Ventana Systems Inc., 2004). Vensim
provides very basic building blocks, logical tools and mathematical
functions that can be used to model inter-linked processes and feedback
loops to develop a dynamic model and provides greater flexibility and
portability.
120
Figure 3.16: Schematic of farm nodes and supply channels/pipes (in parenthesis: channel/pipe length in metres)
In the node-link model, a node represented by Ni is created at each point on
the supply channels where an inlet structure is located for farm ‘i’. While a
farm may have more than one inlet, they are represented by only one node
in the model for computation simplicity. The model executes on daily time
steps and all calculations are carried out at the start of the day. Hence the
value of most of the variables for current day depends on previous day
calculations. The whole system can be driven in real-time by water demand
where each farm acts as a demand unit. For the real-time case, the water
demand of a farm is less than or equal to the calculated crop water
requirement depending on the availability of irrigation water which
sometimes may be limited due to constraints on capacity of the conveyance
system. For the on-farm water storage case, the temporal pattern of water
demand and supply will change. More details on schema of irrigation
N2
N3
N4 N5 N6
N7
N8
N9
N10 N11 N12 N13
WaterSource
N1a
N7a
(273) (552)(291)
(102)
(446) (150)
(368)
(100)
(585)
(508)
(271)
(324) (4) (94)
121
scheduling will follow in latter sections. The model structure is designed in
a way that it can easily be configured to model any crop on any soil type,
irrigation systems and irrigation supply networks. Nodes and links are the
building blocks of a nodal network. Further details about the nodes and links
of the developed model are given in the following sections.
Characteristics of a Node
Each inlet point from where water is supplied to a farm along the supply
channel is designated as a node, Ni, where ‘i’ represents the farm ID. The
farm ID is a number given to each farm for this research as listed in Table
3.5. The real farm numbers are not disclosed. At a given node all outflows
are balanced by inflow. The total flow demand at a given node is sum of
flow demands of all downstream nodes and is given by Equation 3.1:
∑ Equation 3.1
Where,
‘n’ represents the total number of nodes downstream of current node
At node, N2, which is a branching node (see Figure 3.16); the total flow is
the sum of the total flow demands on both the left and right branch.
Similarly, the total hydraulic head at a given node is the sum of hydraulic
heads required at the downstream nodes including head losses. When it is a
pressurised piped network then hydraulic head at a point also includes
pressure head. Other characteristics of a node include its elevation and
chainage. The termination node at each branch of the supply system
accumulates any surplus or shortfalls in daily irrigation supply.
Characteristics of a Link
The model considers each farm as one lumped demand unit. Each demand
unit (farm) is hydraulically connected to the supply system at a node. Each
node is hydraulically connected at its upstream and downstream node (if
any). The hydraulic connection between two nodes is called a ‘link’. The
length of each link is given in Figure 3.16. For the open channel supply
system, a link has the characteristics of an open channel while for a piped
122
supply system a link has the characteristics of a pressurised pipe. The
hydraulics of the two types of links is incorporated in the model. The
slope/grade of each link is determined by elevation and chainage difference
between its nodes. A link representing an open channel may have seepage
and evaporation losses. For a pipe link such losses are assumed to be zero.
Flow through a link is limited by its flow capacity.
3.3.1 Modules of the Developed Node-link model
The node-link model developed using Vensim software consists of the
following computational modules.
Crop water demand module
Calculates daily crop evapotranspiration for normal and water
stressed conditions. It is also capable of calculating crop irrigation
use under different irrigation application systems which includes
drip, sprinkler, furrow, and flood irrigation.
Irrigation application system module
The furrow system consists of 1.0m wide furrows with 0.5m on each
side of the tree or vine. The sprinkler system used in this study is
described as a non-overlapping under-canopy irrigation sprinkler
system. The drip system is a surface drip system with one drip line on
each side of a tree or vine. The variation in applied irrigation depth due
to non-uniform distribution, pressure variation, and irrigation time are
ignored for the sake of simplicity in modelling these systems in this
thesis.
Irrigation supply network (conveyance) module
This module can be configured to model either the open channel
irrigation supply system or the pressurised pipe irrigation system.
Energy use module
123
This module computes energy used in pumping and delivering the
irrigation water to the farms via a piped system. It also computes energy
used in operating the pressurised irrigation application systems
including sprinkler or drip. A separate spreadsheet model was developed
to account for other energy inputs in the annual crop production cycle.
Greenhouse gas emissions accounting (separate spreadsheet model)
Greenhouse gas emissions resulting from various energy inputs are
accounted in a separate spreadsheet model which is linked with the
energy accounting spreadsheet model.
Crop yield module
The crop yield module estimates reduction in crop yield resulting from
water shortages. The main reason of shortage of irrigation water is the
capacity constraint of the irrigation delivery system.
Economics module
The economics module is also a spreadsheet model, which gets input
data from the other modules. It includes crop annual budgets, financial
analysis of investment in water saving technologies and calculation of
indicators like water productivity.
Integration module (System dynamics module)
The integration module links all abovementioned modules. It targets to
identify feedback loops based on output of these modules. The purpose
of the integration module is to develop a water-energy policy framework
for irrigation systems based on interactions between different aspects of
the system considered in this thesis.
124
Figure 3.17: Flowchart of interaction among different modules of the node-link model
Each computational module is set up for a specific output that depends on
input from another module. The interaction among various modules of the
node-link model is graphically depicted by arrows in Figure 3.17.
3.3.1.1 Crop water demand module for Calculation of Crop
Evapotranspiration (ETc) for Various Irrigation Techniques
To accurately estimate energy use by a given irrigation method it is
important to first make an accurate estimation of water supplied to the crops
using that irrigation method. Furthermore, water application by different
irrigation methods needs to be estimated to compare water savings for a
given crop. This section corresponds to “crop water demand” module in
Figure 3.17. It explains the procedure adopted in the model for daily
calculation of evapotranspiration for crops irrigated with flood, furrow,
sprinkler or drip irrigation systems. The developed model has flexibility to
be configured to almost any irrigation system. The crop evapotranspiration
Irrigation supply system
Crop water demand
Crop yield
Field irrigation application system
Energy use
GHG emissions
Economic analysis
System dynamics
Water diverted
Water delivered
Start
125
is computed by the model for each crop grown on the 13 farms in the case
study area.
Evapotranspiration consists of two components; evaporation from the
wetted area of soil and the crop transpiration, a process by which water
acquired via root systems is lost through the leaves of a plant. Soil
evaporation is controlled by the amount of solar energy absorbed by the soil
surface, which depends on canopy cover of the crop and the soil moisture
level, which is maximum following rain or irrigation application. Each crop
has different rate and amount of evapotranspiration depending on the crop’s
physiological characteristics, development stage and on climate and the soil
type. To calculate crop evapotranspiration (ETc), the first step is to calculate
reference crop evapotranspiration (ETo) for the given area. For the study
area of this research, the daily ETo values were directly downloaded from
SILO website (http://www.longpaddock.qld.gov.au/silo/) for Griffith
CSIRO weather station (station ID 75174). The ETo calculation is based on
the FAO Penman-Monteith method (Allen et al., 1998). FAO Penman-
Monteith is the preferred method for ETo calculation as it closely
approximates reference grass ETo and explicitly incorporates both
physiological and aerodynamic parameters that control evapotranspiration.
The ETc for each crop is then calculated by (Equation 3.2):
Equation 3.2
Where, Kc is referred to as “crop coefficient”.
Determining Crop Coefficient (Kc)
In the Penman-Monteith method (Ellen et al., 1998) most of the effects of
the various weather conditions are incorporated into the ETo estimate.
Therefore, as ETo represents an index of climatic water demand, Kc varies
predominately with the specific crop characteristics and only to a limited
extent with climate. This enables the transfer of standard values for Kc
between locations and between climates. This has been a primary reason for
the global acceptance and usefulness of the crop coefficient approach and
126
the Kc factors developed in past studies. The Kc in Equation 3.2 predicts
ETc under standard conditions where no limitations are placed on crop
growth or evapotranspiration due to water stress, crop density, disease, or
salinity pressures.
Using Dual Crop Coefficient (Kc = Kcb + Ke) for Different Irrigation
Methods
Evapotranspiration comprises of two phenomena: transpiration from plants
and evaporation from soil. The single crop coefficient Kc, combines the
effect of crop transpiration and soil evaporation into a single time and space
averaged value. The amount of crop evapotranspiration obtained from
multiplication of the single crop coefficient and the reference crop
evapotranspiration for a given crop assumes soil evaporation from the entire
crop area. This is a more valid assumption for high plant density crops, for
example; cereal crops including wheat, rice and corn or pastures but not
fully applicable for estimation of evapotranspiration for horticultural crops.
The soil evaporation component of evapotranspiration is regulated by the
extent of wetted soil area and the uncovered (bare) soil area. Soil wetting
events like irrigation or rainfall affect the value of the crop coefficient due
to varying rates of evaporation from soil surface on a day-to-day basis. The
single mean crop coefficient value does not account for these varying
evapotranspiration rates resulting from wetting events. This research
postulates that aforementioned discrepancy in single crop coefficient is
more observable for the horticultural crops especially when irrigated with
high efficiency irrigation system and computed daily. Therefore, contrary to
the traditional single crop coefficient approach, this research adopts a dual
crop coefficients approach for estimation of crop evapotranspiration for
various irrigation application methods used for irrigating horticulture crops.
Each irrigation method has different frequency and different rates of
irrigation application which results in different extent of soil wetting (partial
wetting) under and around the plants’ canopy. Therefore, it was preferred to
use a dual crop coefficient that separates the effect of soil evaporation and
127
crop transpiration. The two coefficients are: the basal crop coefficient (Kcb)
which represents plant transpiration and the soil water evaporation
coefficient (Ke) which represents evaporation from the top soil surface. The
dual crop coefficient approach is relatively complex and computation
intensive but more precise for daily estimation of evapotranspiration in
horticultural crops. A similar approach was adopted by Johnson et al.,
(2004) for different irrigation systems for peach orchards. Similarly, Allen
and Robison (2007) revised ET estimations for 125 weather stations in
Idaho by employing the dual crop coefficient procedure and the ASCE
standardized Penman-Monteith method as the preferred method for water
transfer and administration by using dual crop coefficient method as
summarized by Allen et al., (2005) from Allen et al, (1998). For the dual
crop coefficient approach the Equation 3.2 for ETc is revised as Equation
3.3:
Equation 3.3
As indicated by Allen et al., (1998), the value of Ke is large following a rain
event or irrigation but the sum of Ke and Kcb can never exceed maximum
value of Kcmax which is determined by the energy available for
evapotranspiration at the soil surface. The value of Ke even drops to zero
when no further water can evaporate from the soil surface. The value of Ke
depends on remaining evaporable water content in the soil profile
represented by soil evaporation reduction factor, Kr, and hence a daily
continuous soil water balance computation is made for each farm in the
model developed for this research. This also involves estimation of deep
percolation which occurs when rainfall or irrigation is in excess of
prevailing soil moisture depletion. Apart from the soil evaporation reduction
factor (Kr), the other factor that impacts Ke is few, which is defined as the
extent to which soil surface of a given crop area is both wet and exposed to
sunlight and air ventilation. Mathematically, the value of Ke is determined
by Equation 3.4:
, Equation 3.4
128
Where,
Kr is soil evaporation reduction coefficient dependent on the
cumulative depth of water depleted (evaporated) from the top soil
(dimensionless),
few represents the fraction of the soil that is both exposed and wetted,
i.e., the fraction of soil surface from which most evaporation occurs
and,
Kcmax is the maximum value (upper limit) of Kc for a given growth
stage.
Equation 3.4 indicates that value of evaporation coefficient, Ke, depends on
two factors; amount of remaining water in top soil that can evaporate, and
the extent of surface area of top soil with evaporable water. The values of
Kcb for initial, middle and final growth stages of the crops used in this study
were taken from Table 17 in Allen et al., (1998). The Kcb values were
assigned to months according to local crop calendar based on Meyer (1996)
and are given in Table 3.9. Due to their deciduous nature there is a greater
variation in Kcb values for stonefruit. The values of Kcbmid and Kcbend in
Table 3.9 were corrected for local climatic conditions as analysed by Pereira
et al., (1999) for minimum relative humidity and average wind speed
differing from 45% and 2 m/s, respectively, using the following formula
(Equation 3.5) in the model.
0.04 2 0.004 45.
Equation 3.5
Where,
Kcb (Table) the value for Kcb mid or Kcb end (if ≥ 0.45) taken from
Table 17 in Allen et al., (1998),
u2 the mean value for daily wind speed at 2 m height over grass
during the mid or late season growth stage (m s-1) for 1 m s-1 ≤ u2 ≤
6 m s-1,
129
RHmin the mean value for daily minimum relative humidity during
the mid- or late season growth stage (%) for 20% ≤ RHmin ≤ 80%,
h the mean plant height during the mid or late season stage (m).
Table 3.9: Monthly basal crop coefficients (Kcb) for modelled horticultural crops (Allen et al., 1998)
Crops Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun
The value of upper limits on evapotranspiration, the Kcmax, used in Equation
3.4 is given by Equation 3.6 (adopted from Allen et al. 1998):
0.04 2 0.004 45.
,
0.05 Equation 3.6
Equation 3.6 ensures that Kcmax is always greater or equal to the sum of
Kcb and 0.05.
Following irrigation application or a rain event the evaporation from the soil
surface is unrestricted hence Kr is set to 1 for such events in the model. As
the soil surface dries, Kr becomes less than one and soil evaporation begins
to reduce. Kr becomes zero when no water is left for evaporation in the
upper soil layer. For the latter case, Kr in the model is determined by
Equation 3.7:
, , Equation 3.7
Where,
De,i-1 is the cumulative depth of evaporation (depletion) from the soil
surface layer at the end of day i-1 (the previous day) (mm)
TEW is total evaporable water. It is the maximum cumulative depth
of evaporation (depletion) from the soil surface layer (mm),
130
REW is the readily evaporable water as a fraction of TEW. It is the
cumulative depth of soil-water evaporation when depletion = REW
(mm)
TEW is the amount of water that can be depleted by evaporation during a
complete wetting to halfway between complete drying, and was estimated
by Equation 3.8:
1000 0.5 Equation 3.8
Where,
θfc is the soil water content at field capacity (m3/m3)
θwp is the soil water content at wilting point (m3/m3)
Ze is the effective depth of the surface soil layer that is subject to
drying by way of evaporation (m)
The values of the abovementioned soil-water parameters for the soil types
used in this study are given in Table 3.10 (Hornbuckle and Christen, 1999;
Allen et al. 1998). The value of Kr remains unchanged from 1 until
cumulative depth of evaporation (De) exceeds REW. In the model De gets
adjusted for rainfall or irrigation and even reduced to zero when rainfall or
irrigation application depth is greater than or equal to De.
Table 3.10: Soil water characteristics used in calculation of soil evaporation reduction coefficient, Kr Soil type
(texture)
θfc
(m3/m3)
θwp
(m3/m3)
Ze
(m)
TEW
(mm)
REW
(mm)
Sandy loam 0.23 0.11 0.15 26.25 13.65
(52%)
Clay loam 0.34 0.18 0.10 25.00 9.5 (38%)
Referring back to Equation 3.4, the value of few, the fraction of the soil that
is both exposed and wetted, the second factor which determines value of Ke,
is calculated as follows.
Where the complete soil surface is wetted, as by precipitation or centre
pivots or flood irrigation, the fraction of soil surface from which most
131
evaporation occurs, few, is defined as (1- fc), where fc is the average fraction
of soil surface covered by vegetation and (1-fc) is essentially the
approximate fraction of soil surface that is effectively exposed to sun light.
However, for irrigation systems where only a fraction of the ground surface
is wetted like drip systems, few is limited to fw, the fraction of the soil
surface wetted by applied irrigation water. Therefore, in the model
developed, few, is calculated as Equation 3.9:
1 , Equation 3.9
The model developed for this study can simulate soil-water evaporation
from both partially wetting and fully wetting irrigation methods for crops
with different canopy covers. The values of fc the crop cover fraction for the
modelled citrus/stonefruit crops is computed using the steps listed in Figure
3.18. The average number of trees per hectare and the average canopy radius
were computed from collected data. The total canopy area per hectare for
wine grape was computed from the number of rows per hectare, number of
plants in a row, the distance between two plants in a row and average height
and width of the foliage. The values of fc computed this way for the three
crops are given in Table 3.11 and were used in the model. The soil wetted
area for different irrigation methods and crops was estimated through model
calibration and is explained under that section.
132
Figure 3.18: Steps involved in calculation of crop cover fraction for citrus and stonefruit Table 3.11: Values of wetted and vegetative covered soil fraction for irrigation methods and crops modelled in this study Crop cover fraction
fc Wetted soil surface fraction fw Comments
Citrus 0.69 Flood All crops 1 Whole field is wetted Stonefruit 0.62 Furrow
Table 3.13: Calibration variables and their calibrated model values for years 2003-04, 2004-05 and 2005-06
Variable Range 2003-
04
2004-
05
2005-
06Average
Zr_citrus (m) 0.33 –
1.0 0.72 0.71 0.80 0.74
p_citrus 0.4 –
0.6 0.50 0.40 0.52 0.47
Wetted_area_citrus (m2/ha) 4000 –
7000 6200 5900 6000 6000
Zr_stone_fruit (m) 0.38 –
1.13 0.44 0.47 0.46 0.46
p_ stone_fruit 0.4 –
0.6 0.52 0.48 0.40 0.47
Wetted_area_stone_fruit
(m2/ha)
4000 –
7000 5300 5000 5000 5100
Zr_vines (m) 0.38 –
1.13 0.88 1.04 1.06 0.99
p_vines 0.35 –
0.55 0.41 0.49 0.45 0.45
Wetted_area_vines (m2/ha) 4000 –
6000 5100 5100 5000 5100
Irrigation adjustment factor 0.02 –
0.2 0.024 0.038 0.069 0.040
Fitness value (difference
between actual and optimised
solution)
0.0019 0.0012 0.0022
Table 3.14: Comparison of irrigation application rates (ML/ha) between the actual and the calibrated model
Year Citrus Stone
fruit Vines
2003-
04
Model 5.4 5.6 4.09
Data 5.4 5.6 4.1
2004- Model 5.50 5.50 4.199
147
05 Data 5.5 5.5 4.2
2005-
06 Model 5.499 5.700 4.801
Data 5.5 5.7 4.8
The soil-water availability parameters are calculated using the calibrated
data on effective rootzone depth and the depletion fraction for the modelled
crops and the soils and given in Table 3.15.
Table 3.15: Soil-water availability parameters using calibrated model data for the three crops
Clay Loam Sandy Loam
TAW
(mm)
RAW
(mm)
TAW
(mm)
RAW
(mm)
Citrus 118.4 55.6 88.8 41.7
Stone fruit 73.6 34.6 NA*- NA*-
Wine
grape
158.4 71.3 118.8 53.5
NA* not applicable as model only includes stone fruit on clay loam
Once the model was calibrated for crop water use, it was used for simulating
various scenarios which are discussed in the following chapters of this
thesis.
Validation of the Crop Water Use Module
The crop water use module of the node-link model computes daily irrigation
requirement for each crop on each of the modelled 13 farms for a specified
irrigation application method. Irrigation application is driven by soil water
depletion from the effective rootzone. The soil water depletion is computed
on a daily time step by the rootzone water balance approach. Rootzone
water balance components including loss of water due to soil evaporation,
crop transpiration, deep percolation, effective rainfall and irrigation depth
(when applied) are accounted for in the calculation of daily soil moisture
depletion. Irrigation is applied when soil moisture depletion is equal to the
readily available moisture times the ‘deficit factor’.
148
In contrast to traditional fixed-interval irrigation scheduling models, the
developed model is essentially a demand based irrigation model with
variable irrigation intervals. During hot and dry seasons the soil water
depletes faster and crop irrigation demand is higher and more frequent than
relatively cold and wet periods. This approach more closely represents
actual irrigation practices in the study area. Also the model assumes a non-
deficit irrigation practice. In this case the depth of each irrigation application
is such that the current depletion is reduced to zero and thus effectively
bring soil moisture back to the readily available moisture level provided that
no incidents of constraints on water delivery system are experienced. A
value of the deficit factor greater than 1 in the model gives effect to deficit
irrigation practice which reduces the irrigation application depth as well as
the irrigation application frequency and in this case both the crop
transpiration (Ks value becomes less than 1) and soil evaporation (Kr
becomes less than 1) rate begin to reduce and are limited by available soil
water once the soil water depletion exceeds readily available moisture level.
Table 3.16: Model validation by comparing actual and modelled drip irrigation application rates (ML/ha) for horticultural crops on 13 farms in the study area (Figure in brackets is total number of irrigation days)
Crop 2006-07 2007-08 2008-09 Average (ML/ha)
Citrus (ML/ha) Data 6.0 4.9 5.0 5.30
Model 6.01 (212) 5.55 (198) 5.43 (153) 5.66
Stonefruit (ML/ha) Data 6.0 5.0 5.7 5.57
Model 5.97 (159) 5.53 (159) 5.50 (158) 5.66
Vines (ML/ha) Data 4.5 4.0 4.1 4.2
Model 4.59 (81) 4.10 (59) 4.19 (60) 4.29
The validation of the calibrated crop irrigation demand model was carried
out for three years from 2006-07 to 2008-09. Table 3.16 presents actual and
modelled water use in megalitres per hectare for citrus, stonefruit and vines
for the three years from 2006-07 to 2008-09 with all modelled farms
irrigated with drip irrigation system. There is a reasonable agreement
between the actual and modelled irrigation application rates using the
calibrated model, especially between three-year modelled and field data
149
averages. The values of water use for each crop are averaged for all the
farms growing that crop. The model does not consider physical capacity
constraints on irrigation supply and generally suggests higher irrigation rates
than the actual ones. The actual irrigation rates are lower due to channel
capacity constraints during peak irrigation demand times or due to deficit
irrigation.
The number in brackets in Table 3.16 represents the total number of days
the irrigation was turned on for that crop. The average irrigation days for
citrus are higher than the other crops due to the fact that citrus are non-
deciduous and evapotranspiration continues through all seasons. In contrast,
both the stone fruits and wine grapes are deciduous plants and lose their
foliage in autumn.
The other reason for least number of irrigation days for wine grapes is the
fact that those farms are dominated by clay loam soils which exhibit a wider
range of soil moisture retention capacity than sandy clay soils of the other
farms. Moreover, some wine grape farms practice regulated deficit irrigation
which is not implemented in the developed model. As indicated in Table
3.6, 2006-07 was the driest year of the modelled period which resulted in
the highest number of irrigation days and highest irrigation application rates
for all crops. Another reason for lower water use on these perennial crops in
the MIA is the fact that water traded out of the area doubled in 2007-08 (MI
AR 2008) as compared to the previous year (about 100.8 GL in 2007-08 as
compared to 51.2 GL in 2006-07) owing to high water prices in the water
trade market due to continued drought conditions throughout the Murray-
Darling Basin since 2002-03. A more detailed account of the water being
traded out of MIA is given by ACIL Tasman (2009).
Moreover, the following assumptions that were made in the developed crop
water use model should be noted while comparing actual and modelled
water use for each crop.
The model assumes the same irrigation scheduling rules for all farms
for a given crop regardless of soil type.
150
The water use of a crop grown on different soil types across the
modelled farms was averaged regardless of soil type.
Once started, an irrigation event may last for at least 24 hours.
The average irrigation application rate (ML/ha) for over 900 horticulture
farms in MIA is listed in Table 3.17 from 2004-05 to 2008-09. Although for
horticultural crops, the difference in crop variety, irrigation system type,
irrigation management (such as regulated deficit irrigation in vines) and the
age of plantings makes an average water use figure not so representative;
there is a reasonable agreement between average water use figures for the
whole MIA as reported in Table 3.17 and the modelled ones for the case
study area within MIA as given in Table 3.16. The reported water use by
“other fruits” in Table 3.17 includes all fruits grown in MIA except for
citrus and vine and is not limited to stonefruit only. Therefore it is not
entirely comparable to modelled water use for stonefruit.
Table 3.17 also indicates that an additional 2,445 ha have been converted
into vineyards from 2004 to 2009. This was mainly due to the reason that
vines were favoured crops under tight water availability and the ever-
increasing demand for Australian wine locally and overseas.
Table 3.17: Reported water use (ML/ha) for fruit and vines (Figures in braces are total crop area in ha) (Sources: MIA 2005, 2006, 2007, 2008, 2009).
Crop 2004-
05
2005-
06
2006-
07
2007-
08
2008-
09
Average
(ML/ha)
Citrus
(ML/ha)
6.1
(8364)
5.5
(8423)
5.7
(8434)
5.0
(8357)
5.4
(8216) 5.5
*Other
fruit
(ML/ha)
3.8
(1881)
3.9
1953)
4.1
(2197)
3.5
(2546)
4.6
(2538) 4.0
Vines
(ML/ha)
5.5
(16798)
5.1
(17151)
4.5
(18160)
3.9
(18866)
4.0
(19243)4.6
151
It can be concluded from the above discussion that the developed irrigation
water use model is valid for the study area and reasonably extendable to
model irrigation water use by the whole horticultural area of MIA.
To explore the full extent of water and energy linkages, all scenarios
modelled and discussed in the following chapters are modelled with non-
and the transport of agricultural inputs and outputs. Solar energy is the
biggest direct energy input but it is not considered in this analysis because
of zero cost. Direct energy inputs primarily include various primary and
secondary fuel sources that are used to operate farm machinery and
165
irrigation pumps. A list of major direct energy inputs considered in this
study includes the following:
Electricity,
Fuel (diesel, petrol, gas) and lubricants,
Machinery use,
Irrigation,
Human labour,
Farmyard manure
Indirect Energy Inputs
Indirect energy inputs are in the form of sequestered energy in fertilizers,
herbicides, pesticides, and insecticides. Indirect energy consumption refers
to energy inputs in manufacturing the equipment and other goods and
services that are used on-farm (Pimental, 1992). This includes energy used
in production of fertilisers, tractors, agrochemicals, irrigation equipment and
harvesting equipment. Other indirect energy inputs include seeds, and
energy used in installation of water supply infrastructure and construction of
on-farm storages.
Different inputs in farm operations contain different levels of intrinsic
energy. For that reason, all forms of direct and indirect energy inputs as
used on the horticulture farms in the case study area were converted into a
common equivalent energy unit of kilo-watt-hour (KWh) to account for
total energy use and to make them comparable based on the current
literature and given in Table 3.23. To compute the total use for a given
energy input, the actual input quantity is multiplied with its equivalent
energy unit.
The developed node-link model only computes energy consumed in
irrigation water pumping and that used in operating the pressurised
irrigation systems. A separate spreadsheet model was developed for energy
and greenhouse gas emissions accounting.
166
Table 3.23: Energy equivalent values for different farm inputs and outputs
Input Sub-type Unit Equivalent energy (KWh/unit)
Reference or source
Human labour
Hour 0.64 Ozkan et al., 2004; Hatirli et al., 2006; FAO, 2000
Diesel Litre 10.73 Dept. Climate Change & Energy Efficiency, 2010
Lubricants Litre 10.78
Farm machinery (tractor)
Hour 161.38 Falivene, 2003
Electricity KWh 1.0 Fertilizer Nitrogen Kg 18.38 Hatirli et al., 2006 Phosphorous Kg 3.46 Hatirli et al., 2006 Potassium Kg 3.10 Hatirli et al., 2006
Farmyard manure
Tons 84.26 Hatirli et al., 2006; Canakci et al., 2005
Chemicals Herbicide Kg 66.72 FAO, 2000 (Falivene, 2003 reported 33.36)
Fungicide Kg 28.58 FAO, 2000 Pesticide Kg 55.60 FAO, 2000 Irrigation water
m3 0.18 Ozkan et al., 2004
Yield (output)
Orange Kg 0.53 Ozkan et al., 2004
Stone fruit (peach)
Kg 0.61 Johansson & Liljequist, 2009
Grapes Kg 3.28 Ozkan et al., 2007; Singh, 2002
As shown in Table 3.23, the energy sequestered in one kilogram of grapes is
significantly higher than that of peach or citrus due to high sugars and
carbohydrates in grapes.
167
Figure 3.28: Flowchart of steps to account for energy use, productivity indicators as well as carbon footprint of energy use in irrigation and crop production
The mechanical energy used on the farm mainly includes a tractor which
consumes diesel and lubricating oil (other tractor oil inputs are ignored for
their negligible magnitude). An 86 HP tractor which is commonly used in
MIA consumes diesel at the rate of 15 l/h and 10 litres of engine oil is
Identify energy use related processes in crop life cycle within study area
(including irrigation)
Identify energy inputs in each process
List all direct energy inputs
List all indirect energy inputs
Convert all energy input types to a single equivalent
(sequestered) energy unit
Convert each energy input to equivalent
CO2 emission
Compute total energy input for all
considered processes
Compute total equivalent CO2
emissions
Compute total system output (i.e. yield)
Compute carbon footprint of system output (CO2
emissions per unit output)
Compute energy footprint of system output (KW
consumed per unit output)
Repeat abovementioned steps for different scenarios (combinations of irrigation methods and conveyance systems)
Convert total system output into equivalent energy
Compute other relevant energy use and carbon emission indicators
Compare those scenarios
168
replaced every 250 hours of running time (i.e. 0.04 l/h). Therefore, the
energy input in one hour of tractor operations is equal to 161.38 KWh. The
energy consumption in practices like soil preparation, growing cover crops,
harvesting the cover crop etcetera, are not considered in this study.
Figure 3.28 provides a generic procedure to quantify energy and carbon
footprints of crop production for a given scenario. The right hand side boxes
describe steps to quantify energy footprint of crop production expressed as
KW consumed in producing a unit output. The left had side boxed describe
steps to compute various quantities required to quantify
carbon/environmental footprint of crop production expressed as CO2
emissions exhausted to produce a unit output. Other relevant energy and
CO2 emissions indicators as defined in Chapter 2 are also computed using
the computed values for different steps/boxes in Figure 3.28.
Calculating CO2 equivalent emissions
Agriculture produces greenhouse emissions in Australia at 15.7% of the net
national emissions in 2005 (AGO, 2007). Here the term 'agriculture' is
generic to cover agricultural, livestock, forestry and fishery activities.
Agriculture is the single dominant source of methane and nitrous oxide
emissions. However, methane and nitrous oxide are mainly associated with
livestock, rice cultivation, and field burning of agricultural residuals
etcetera. Carbon dioxide equivalents (CO2-e) is a unit of measurement that
allows the effect of different greenhouse gases and other factors to be
compared using CO2 as a standard unit for reference. The emissions of
different greenhouse gases can be aggregated by converting them to carbon
dioxide equivalents. The conversion is done by multiplying the mass of
emissions by the appropriate global warming potentials (GWPs). GWPs
represent the relative warming effect of a unit mass of the gas when
compared with the same mass of CO2 over a specific period (IPCC, 2001;
OECD, 2001).
This research does not examine the CO2 mitigation function that agriculture
provides in the forms of carbon storage in forestry/trees or carbon
169
sequestration in soil. The focus of this research is limited to irrigation
systems and production of the horticultural crop rather than the whole
agriculture sector. The energy inputs relevant to this study are listed in
Table 3.23. The conversion factors for equivalent CO2 emissions expressed
as kilogram of CO2-e per KWh of energy contents for the inputs related to
this study are given in Table 3.24. A similar approach for computation of
CO2 equivalent emissions was employed by Wells, (2001); Dept. CC&EE,
(2010); and Barber, (2004). Those conversion factors were applied to
calculate and compare the carbon footprint of irrigation conveyance
systems, irrigation application methods and the horticultural crops
production. The conversion factors for one kilowatt of electricity purchased
from the national grid are different for different States connected to the grid
due to varying transmission losses. For example the electricity purchased
from NSW/ACT has a conversion factor of 0.9. For this study the CO2
emissions associated with energy inputs including human labour and
machinery, are also considered. However, the CO2 equivalent emissions
from the consumption of fuel by farm machinery are considered while any
heat radiated by their use is not considered. It should also be noted that for
the sake of convenience in aggregating the CO2 equivalent emissions, the
CO2 equivalent factor for inputs like fertilizers is expressed in units of
KgCO2-e/KWh (i.e. CO2-e emissions per KWh of sequestered energy)
instead of KgCO2-e/Kg. Human work hour is computed to be equivalent to
GHG emissions of 0.426 KgCO2-e based on the assumption that human
energy is sourced from meat intake. For example, the production of 1 kg of
beef, results in GHG emission with global warming potential of 36.4
KgCO2-e (Ogino et al., 2007). The CO2-equivalent emissions from use of a
tractor are computed for the diesel and oil consumed by the tractor engine.
The emissions from an 86 HP tractor as mentioned above are estimated to
be 40.22 KgCO2-e per hour of operation.
170
Table 3.24: CO2 equivalent emissions factors for various farm inputs
Input Sub-type CO2-e emissions (KgCO2-e/KWh of input energy)
Source reference
Diesel 0.249
Dept. Climate Change & Energy Efficiency, 2010
Petrol 0.240 Electricity NSW/ACT 0.90 Victoria 1.23 Queensland 0.89 South Australia 0.72 South West
Barber, 2004 Potassium 0.216 Chemicals Herbicides 0.216 Fungicides 0.216 Insecticides 0.216 Human 0.426 Computed based
on: Ogino et al., 2007
3.3.1.4 Crop Yield Module
Crop yield is affected by water shortage/deficit which can be due to limited
water availability, capacity constraints or inadequate irrigation scheduling.
Crop-water production functions are developed for a given crop from the
field data on water use and the yield obtained. No data covering a sufficient
range of water and crop was available for any crops in the case study area
for this purpose. Some crop-water production functions for other regions of
Australia were available. For example, Figure 3.29 shows a production
function for citrus crops from monitoring sites in South Australia. The line
in Figure 3.29 is statistically fitted and represents the boundary of dataset
171
and approximates the potential yield at a given depth of applied water
(Skewes, 2010). In the absence of yield-water use data for the study area, a
simpler and linear crop-water production function developed by Doorenbos
and Kassam, (1979) was used for this study. The FAO crop-water
production function (Equation 3.32) predicts the reduction in crop yield
when crop stress is caused by a shortage of soil water.
Figure 3.29: Plot between applied water (including rainfall) and yield for citrus crops in South Australia (Source: Skewes, 2010)
1 1
: 1 1 Equation 3.32
Where,
, is the actual crop yield (t/ha),
, is maximum expected or agronomically attainable crop yield under no
water stress (t/ha),
, is the adjusted evapotranspiration for water deficit. It is calculated
by the crop water demand module of the developed node-link model.
172
, is the crop evapotranspiration for standard conditions, i.e. no water
stress,
1 , is the relative yield decrease due to water shortage,
1 , is water stress or relative evapotranspiration deficit. In this
study, the magnitude of water deficit refers to the deficit in relation to crop
water requirements over the total growing period of the crop.
K , is the yield response factor. It is the relative decrease in yield per unit
relative water deficit. Different crops have different Ky value. The values of
K for the whole growing season and for the three crops used in the
model are given in Table 3.25. The Ky values in Table 3.25 are based on
Doorenbos and Kassam, (1979) and Stewart et al. (1977). The value of
in Table 3.25 are the reported maximum yields (Khan et al., 2005a; Hope
and Wright, 2003; Singh et al., 2005). The agronomic potential yield may be
even higher than these values. Higher values of Ky indicate that the crop
yield is more sensitive to water deficiency than that of lower Ky values. In
other words if Ky < 1 then the decrease in yield is proportionally less with
the increase in water deficit and while the decrease in yield is proportionally
greater with the increase in water deficit for crops with Ky > 1. Crops like
citrus have a Ky value greater than 1 while wine grapes have Ky less than 1
as shown in Table 3.25.
Table 3.25: and values for the modelled crops
Crop Citrus Stone fruit Wine grape
K * 1.3 1.0 0.85
**(t/ha) 50 20 25
*Source: Doorenbos and Kassam, (1979). **Source: Khan et al., (2005)
The crop water demand model performs daily calculation of and water
stress coefficient (K ) for computing adjusted where,
. The daily values of and are aggregated by
the Crop Yield Module at each time step and the actual yield ( ) is
173
estimated after the final time step of the model using Equation 3.32. At each
computation time step of the model and for each crop, a composite value of
water stress coefficient (Ks) for adjusting ETc in Equation 3.32 is computed
by multiplying water stress coefficient values for the individual farms
growing the same crop.
3.3.1.5 Economic Analysis Module
The economic analysis module covers the financial aspects of different
scenarios. Financial analysis is required to determine whether increased
water or energy efficiency and yield are financially beneficial to farmers in
the long term as recommended by O’Neill et al. (2006). It includes the
analysis of the capital and running costs of different irrigation application
systems and irrigation infrastructure over their working life. The economic
analysis module is developed in MS Excel spreadsheet but it extensively
utilises various outputs of the nodal network modules. The methodology of
economic analysis is similar to others including Giddings, (2005); Giddings,
(2004); Falivene, (2003); and Sing et al. (2005). It performs the following
analyses:
Annual farm budget of each modelled crop by incorporating variable
costs,
Cost-benefit analysis of water saving options including energy costs
involved,
Financial analysis of different capital investment scenarios,
Calculation of economic performance indicators.
The methodology of the economic analysis is presented in more detail in the
relevant chapter in the later part of this thesis.
3.3.1.6 Integration Module
The integration module primarily presents all computed indicators for water,
energy, greenhouse emissions and economics. It provides a holistic
overview of different aspects of any scenario being considered. It also
applies a system dynamics approach to identify feedback loops between
174
different model variables within the boundaries of the system under
consideration. The feedback loops identified by this approach are shown in
Figure 3.30. These feedback loops can be quantified using model outputs
and some external data.
There are six positive feedback loops identified in Figure 3.30. For example,
the feedback loop shown in orange colour can be described as follows:
Greater the ‘energy use’, higher will be the ‘water savings’ which will result
in further ‘capital investment’ to convert more area to ‘pressurised
irrigation’. This will result into even higher ‘energy use’. This completes a
positive feedback loop starting from ‘energy use’. Other feedback loops can
be explained in similar fashion.
Figure 3.30: Feedback loops identified and quantified through integration of modelled variables
3.4 Node-link model Mass Balance Test
To test the robustness of the developed model and for the proof of concept
the model was executed in flood irrigation mode with the open channel
supply system for the year 2007-08. This model run was made under no
constraints on supply capacity. Therefore, total irrigation supply should be
IrrigationEfficiency
GroundwaterRecharge
WatertableRise
WaterSaving
CapitalInvestment
PressurisedIrrigation
Salinity
EnvironmentalBenifits
EnergyUse
+
+
-
+
+
- +
+
GHGEmissions
+
+
-
+
+
Water Traded toEnvironment
+
-
+
ProductionCost+
Yields
+
+
175
equal to total demand if the conveyance system is 100% efficient, otherwise
the total conveyance loss should make up the difference between demand
and supply. The mass balance components as computed by the model are
given in Table 3.26. The total conveyance loss which is the sum of the
channel seepage and channel evaporation was calculated by the model as
16.42 ML for the whole irrigation season in the modelled case study area.
The difference between total irrigation demand and total irrigation water
supply is calculated to be 17 ML. Thus the percentage mass balance error as
computed by Equation 3.33 is 0.03% (0.58ML). Such a minuscule error in
mass balance could be due to the rounding-off phenomenon in the model
computations.
∗ 100 Equation 3.33
Table 3.26: Mass balance components as computed by model run for 2007-08
Total
volume
diverted
(ML)
I
Total
volume
delivered
(supply)
(ML)
II
Total
irrigation
requirement
(demand)
(ML)
III
Demand –
Supply
(ML)
IV
Conveyance
losses (seepage
+ evaporation)
(ML)
V
% Mass
balance
error
VI = (IV –
V) / IIx100
2,302 2,285 2,302 17.0 16.42 0.03
3.5 Demand-based verses fixed interval scheduling for different
irrigation methods
The level of soil moisture content available to plants in the rootzone is the
key factor that triggers the need for moisture replenishment by irrigation in a
demand-based irrigation application system. The rootzone moisture
depletion is regulated by climatic factors and plant growth stage. Rootzone
soil moisture content depletion can be monitored by different methods
including moisture probes, rootzone water balance etcetera. Then irrigation
is applied when a certain level of rootzone moisture depletion is reached.
176
This implies more frequent irrigation during hot and dry seasons than the
wet and cooler seasons. This approach of irrigation management is called
demand-based irrigation scheduling. On the other hand, the fixed interval
irrigation or supply based scheduling does not involve any complicated
equipment and does not require any soil moisture monitoring. This
traditional method of irrigation scheduling is adopted due to channel
capacity constraints where the supply channel is not big enough to serve all
users at a time or due to the absence of any environmental impact
considerations or any water restrictions. The demand based irrigation
scheduling is an advanced method of irrigation that supposedly results in
minimum groundwater accessions and minimum runoff (return flow) from
irrigation areas (Khan et al., 2004). Demand-based irrigation is discussed in
detail in Section 3.6. A separate model is developed for the supply based
irrigation scheduling where both the irrigation application rate and irrigation
interval are fixed for each crop. The optimisation framework on Vensim was
used to optimise both the irrigation interval and irrigation rate. The supply
based irrigation approach does not take into account any climatic influences
on crop water demand and presents more risk, especially in case of
horticultural crops. However, it is included in this research to compare and
justify any cost and benefits associated with demand-based hi-tech irrigation
approaches. The procedure for supply-based irrigation as implemented in
the developed node-link model is shown in Figure 3.31. It also includes
calculation of number of days (d) of continuous water stress (cumulative
ETc readily available moisture) as well as any water lost due to excess
irrigation.
3.6 Calculating water and energy efficiency and productivity
indicators
The water and energy efficiency and productivity indicators were computed
for the farming operations for the whole case study area. The definitions of
these well-established and commonly used indicators are given in Table
3.27 and are based on Khan et al., (2009); Koctürk and Engindeniz, (2009),
177
Pereira (2006), Pereira (2007) and others. Water accounting involves
estimation of water use and losses to compute water productivity indicators.
According to Molden et al., (2003), water accounting can be applied at all
scales of interest, and requires the definition of a domain bounded in three-
dimensional space and time. For example, at the field scale, this could be
from the top of the plant canopy to the bottom of the root zone, bounded by
the edges of the field, over a growing season.
Figure 3.31: Flowchart of supply based irrigation strategy as implemented in the node-link model (n=days since start of simulation, d=days since crop gone in water stress)
The task in water accounting is to estimate the flows across the boundaries
of the domain during the specified time period. At the field scale, water
enters the domain by rain, by subsurface flows and, when irrigation is
available, through irrigation supplies. Water is depleted by the processes of
Apply irrigation on day n = 1
No. of days since last irrigation, n
n = specified irrigation interval?
Apply irrigation (I) & set n = 0
Supply available?
ETc for day n
Cumulative ETc = Cumulative ETc + ETc
Cumulative ETc = Cumulative ETc – I
AND: d = 0
IF: Cumulative ETc > RAW
THEN: Cumulative ETc = RAW
AND: d = d+1
IF: Cumulative ETc < I
THEN:
Loss = Loss + I - Cumulative ETc
n = n + 1
178
growing plants: transpiration and evaporation. The remainder flows out of
the domain as surface runoff or subsurface flows or is retained as soil-
moisture storage. In estimating irrigation water productivity, we are
interested in water inflows (rain plus irrigation) and water depletion
(evaporation and transpiration) as shown in Figure 3.32. At irrigation
system scale, as in the case study area of this research, the water losses due
to channel seepage, channel evaporation or pipe leakage are also considered.
A similar approach for segregation of water accounting components is
adopted in this study. These indicators are computed for each scenario
considered in this study where applicable to capture the water and energy
use footprints.
Figure 3.32: Water use accounting components at field scale (Adapted from Molden et al., 2003).
Table 3.27: Indicators of water and energy use efficiency and productivity
Indicator Unit Definition Description
Energy
efficiency Ratio
The ratio of
total energy
output to total
energy input
Water
productivity Kg/m3
Yield of
marketable
produce per unit
Rain
Irrigation
Out
flow
Percolation
Runoff
Evaporation
Transpiration
179
of water used.
Energy
productivity Kg/kWh
Yield of
marketable
produce per unit
of energy input.
Specific
energy kWh/kg
Energy input
per unit of
marketable
yield.
Water and
energy
productivity
Kg/
m3kWh
Yield per unit
of energy and
water inputs. It
captures the
effect of these
inputs on yield.
Lower values
may indicate
lower efficiency
and higher
environmental
footprint.
Net energy
return KWh/ha
Absolute
difference
between energy
output and
energy input
Water
energy ratio Ratio
The ratio of
energy input
from irrigation
to total energy
input. Higher
ratio may imply
180
higher water
footprint.
3.7 Structure of the Thesis Report
In order to assist readers, the major topics and key scenarios discussed in
this thesis are summarised in Table 3.28.
Table 3.28: Summary of key topics of the thesis
Serial No.
Title Brief description
4.2 Scenario 1 - Flood irrigation with open channel supply system
Water and energy analysis of demand-based open channel flood irrigation.
4.3 Scenario 2 - Furrow irrigation with open channel supply system
Water and energy analysis of demand-based open channel furrow irrigation.
4.4 Scenario 3 - Flood irrigation with pipe supply system
Water and energy analysis of demand-based piped supply flood irrigation.
4.5 Scenario 4 - Furrow irrigation with pipe supply system
Water and energy analysis of demand-based piped supply furrow irrigation.
4.6 Scenario 5 - Sprinkler irrigation with pipe supply system
Water and energy analysis of demand-based piped supply sprinkler irrigation.
4.7 Scenario 6 – Drip irrigation with pipe supply system
Water and energy analysis of demand-based piped supply drip irrigation.
4.8 Comparison of the demand-based irrigation scenarios
Comparison of water, energy and GHG emissions of above listed scenarios.
5.1.1 Scenario 1: Flood irrigation supplied with an open channel system
Water and energy analysis of supply-based flood irrigation supplied with an open channel system
5.1.2 Scenario 2: Furrow irrigation supplied with an open channel system
Water and energy analysis of supply-based furrow irrigation supplied with an open channel system.
181
5.1.3 Scenario 3: Sprinkler irrigation system connected with communal piped supply
Water and energy analysis of supply-based sprinkler irrigation connected with piped supply system.
5.1.4 Scenario 4: Drip irrigation system connected with communal piped supply
Water and energy analysis of supply-based drip irrigation connected with piped supply system.
5.2 Modifications made in the node-link model
Details of the modifications made in the node-link model to change it from demand-based to supply-based model.
5.4 Water use and yield comparison of supply-based and demand-based irrigation
Water use and yield comparison of corresponding scenarios in Chapter 4 and Chapter 5.
5.5 Energy and GHG emissions for the supply-based scenarios
Accounting of energy and GHG emissions for the supply-based scenarios.
5.7 On-farm storages: water-energy analysis
Water and energy analysis of using on-farm storages and comparison with piped supply system.
6.3 Up-scaling the model results using mosaic approach
Up-scaling the model water and energy results using mosaic approach based on representative area.
6.4 Estimating and mapping water and energy savings for MIA – using GIS-Based distributed approach
Estimating and mapping water and energy use/savings for whole MIA – using GIS-Based distributed processing.
7.3 Capital cost for conversion to pressurized irrigation system
Capital cost for conversion to pressurized irrigation systems including pipe network and sprinkler and drip irrigation systems.
7.4 Economic analysis of conversion to sprinkler or drip system for citrus
Full economic analysis of conversion to sprinkler or drip system and piped supply network for citrus.
7.5 Economic analysis of conversion to sprinkler or drip
Full economic analysis of conversion to sprinkler or drip
182
system for wine grapes system and piped supply network for wine grapes.
8.1 Understanding and representing the dynamics of the system
A holistic view of the system using system dynamics approach.
3.8 Chapter Summary
This chapter starts with an introduction of the Murrumbidgee River basin
which is the geographic focus in this thesis. It is then followed by the
description of the salient features of the Murrumbidgee Irrigation area
(MIA) which is the specific case study area with particular spotlight on
irrigated horticulture. After introducing the study area and the rationale for
choosing it for this purpose, the case study site located in MIA is described
for the purpose of a test case for development of a node-link model. A major
part of this chapter is dedicated to the explanation of the methodology used
for the node-link model development. The mathematical equations, data and
procedures followed in each module of the model are explained in detail.
The results of these modules are validated against observed data.
This chapter also explains the methodology used for energy input / output
accounting and greenhouse gas emissions estimation for various irrigation
methods, crops and irrigation strategies. Finally, various commonly known
indicators for water and energy efficiency and productivity are explained in
their mathematical forms. These indicators are the most useful and valid
tools to test the effectiveness of improving water and energy systems.
183
Chapter 4: Water and Energy Nexus for Demand Based
Irrigation Methods and Conveyance Systems
Water resources in the sub-catchments of Murray-Darling Basin are in
instances either fully and in some catchments over-allocated for
consumptive use to the detriment of the environment. For example, it is
estimated that in New South Wales, licences and water allocations equal 120
per cent of total available water resources (Melville and Broughton 2004).
To ensure that we have enough water for irrigation development, the water
we have should be used more efficiently at both farm and catchment scales.
Water can be saved through better management of its delivery and
application (Khan et al., 2004, Khan et al., 2005b). The focus of this chapter
is to investigate how water losses in irrigation can be minimized by looking
at both the delivery and the application sides. However, the water savings
are realized only at the expense of high inputs and potentially contribute to
other aspects of environmental deterioration. This chapter puts forward the
argument that unless the energy requirement aspects are equally considered,
the improvement in irrigation efficiency is a partial solution for minimizing
the environmental footprint of consumptive use of water and to tackle water
shortage.
4.1 Rationale of this chapter
The efficiency in transport of water from its source to the farm is referred to
as conveyance efficiency. The efficiency in application of water in the field
is called irrigation application efficiency. Irrigation conveyance losses
which impact upon conveyance efficiency can be caused by evaporation,
seepage, leakage and operational losses but by far the greatest losses are to
seepage (Meyer, 2005). Seepage and leakage from water supply channels
contribute substantially to ground water accessions creating salinity
concerns from rising groundwater which is mostly highly saline. Average
weighted conveyance losses for 46 irrigation districts nationally was
184
reported as 27.8% where Murrumbidgee irrigation region sits at 22.3% for
the year 1999-2000 (Marsden Jacob Associates, 2003).
Figure 4.1 shows the ten year accounts of seepage and evaporation losses
from the open channel system of Murrumbidgee Irrigation Area (MIA)
(Murrumbidgee Irrigation, 2008). It is evident from Figure 4.1 that
conveyance losses fluctuate with seasonal climate conditions and supply and
diversion volume. Another important operational factor that causes high
conveyance losses in MIA is the fact that all open channels are constantly
kept filled with water during the irrigation season to maintain the reliability
and timeliness of irrigation supply. The conveyance losses can be entirely
eliminated by replacement with piped system. As identified previously, two
systems of irrigation water conveyance or delivery are considered in this
research including:
Open channel system
Pressurized pipe supply system
Figure 4.1: Seepage and evaporation losses from channel system of Murrumbidgee Irrigation Area (MIA)
The “high-tech” irrigation application systems like high pressure subsurface
drip irrigation system, sprinkler system etcetera are remarkably water
growth and decreased yield recorded in flood irrigated trees (Loveys et al.,
1999).
Figure 4.4: Normal and water deficit affected cumulative evapotranspiration (mm shown on y-axis) for the three crops for Scenario 1
The effect on crop yield and other related variables as output by model are
reported in Table 4.3. The biggest impact is predicted on the citrus crop with
1.54 t/ha reduction in yield resulting from 23 mm reduction in ETc and least
impact on wine grape yield with a reduction of just 0.09 t/ha resulting from
4 mm reduction in ETc over the whole year. The soil water stress in this
scenario is neither very high nor prolonged; therefore, the impact on crop
yield is not significant. However, severe water shortage can result in
detrimental effects on crop yield (Doorenbos and Kassam, 1979).
Table 4.3: Effect of water deficit due to channel capacity constraint on ETc (transpiration only) and crop yield Variable Citrus Stone fruit Wine grape Cumulative ETc without water deficit for whole year (mm)
963 1,118 852
Cumulative ETc adjusted for water deficit (ETc adj) for whole year (mm)
940 1,100 848
Yield without water deficit (Ym) (t/ha) 50 20 25 Yield with water deficit (Ya) (t/ha) 48.46 19.64 24.91
, is the rated flow capacity of a single pump operating at maximum
efficiency,
, is the instantaneous duty flow rate depending on the irrigation demand
for the whole irrigation system for a given day. The ratio is rounded up or
rounded down to a whole number as appropriate by the model.
The node-link model determines the number of active pumps on a given
irrigation day by using Equation 4.2. The node-link model tweaked for
simulating this particular scenario indicates that up to 11 pumps installed at
a pumping station near the water source, each with a peak discharge rate of
0.08 m3/s, are simultaneously operated in parallel configuration to supply
irrigation water as shown in Figure 4.7. In the parallel configuration of the
pumps the flow rate is added up and hence pumps are turned on or turned
off by the electronic control system software depending on the current duty
flow. The model reports the maximum and average pumping system duty
flow rates of 0.916 m3/s and 0.115 m3/s, respectively.
The model computes that a total of 481.2 MWh (megawatt-hour) of
electrical energy is consumed by the electrical motors to drive the pumps to
supply irrigation water with a total volume of 3,619 ML to the three crops
during one complete year of simulation. The energy consumed in irrigating
individual crops is assumed to be proportional to the irrigation volume
applied to that crop. The total energy consumption is divided among the
three crops based on their proportional water use as reported by the model
and is given in Table 4.17.
Table 4.17: Irrigation water applied and corresponding energy consumption in irrigation pumping for Scenario 3
Total Citrus Stone fruit Wine grapeIrrigation (ML) 3,619.0 3,034.0 331.2 253.5 Pumping energy (MWh) 481.2 350.6 38.3 29.3 Pumping energy (KWh/ha) 1653.8 1436.7 1573.5 1296.5
All energy inputs for each of the three crops for Scenario 3 are assumed to
be similar in magnitude as that of Scenario 1 with the exclusion of energy
consumption in the form of electricity which is used in pumping the
216
irrigation water at the pumping station. The electricity consumed in
pumping irrigation water depends on a range of factors including flow rate,
flow volume, type of pump, pump efficiency, electric motor efficiency, pipe
size and pipe material. All these factors are incorporated in the node-link
model for accurate estimation of energy consumed by the pumping system.
The theoretical background and the governing equations that are
implemented in the energy module of the developed node-link model to
compute energy consumption in irrigation pumping on a daily basis are
discussed in Chapter 3 in greater detail. In the current model, a value of 70%
for the centrifugal pumps and a value of 80% for the electric motors were
used based on the specifications for the installed equipment.
The modelled energy/electricity use for irrigation pumping, other energy
inputs and corresponding greenhouse gas emissions in the form of
equivalent carbon dioxide emissions on a per hectare basis are given in
Table 4.18. The electricity energy consumption was converted in to
equivalent kilograms of carbon dioxide emissions per kilowatt of electricity
using the conversion factor 0.9 for electricity generated within NSW.
Among the three crops, stone fruit requires the highest amount of pumping
energy per hectare followed by citrus and wine grapes. However, citrus
stays at the top when the total energy requirement per hectare from all
inputs is compared for the three crops. Similarly, the total greenhouse gas
emissions per hectare of crop associated with the energy inputs are also
highest for citrus crop followed by stone fruit and wine grapes.
Table 4.18: Accounts for Energy inputs and GHG emissions on per hectare basis in the yearly production cycle of citrus, stone fruit and wine grapes for Scenario 3
Citrus Stone fruit
Wine grape
Energy input excluding electricity for pumping (KWh/ha)
7,889.21 7,195.27 6,728.07
Electricity consumed in irrigation pumping (KWh/ha)
1,436.7 1,573.5 1,296.5
Total energy input (KWh/ha) 9,325.91 8,768.77 8,024.57 Total energy sequestered in yield (KWh/ha)
18,550 10,980 65,600
217
GHG emissions, excluding electricity for pumping (KgCO2-e/ha)
1,832.74 1,634.52 1,532.50
GHG emissions for electricity consumed in irrigation pumping (KgCO2-e/ha)
1,293.03 1,416.15 1,166.85
Total GHG emissions (KgCO2-e/ha) 3,125.77 3,050.67 2,699.35
The total GHG emissions from electricity consumption are as high as 70%
to 87% of the GHG emissions from energy used excluding electricity by
each crop. This indicates that the electricity consumed by irrigation delivery
and application systems is likely to result in a profound environmental
footprint. Therefore, one of the prime questions addressed in this research is
how to design and operate irrigation system that is both environmentally and
economically balanced.
4.5 Scenario 4 - Furrow irrigation with pipe supply system
This scenario is similar to Scenario 2 except that the irrigation water is
delivered under pressure through pipe from the water source to the farm
inlets. The pipe flow model used to simulate this scenario is described in
Chapter 3. Unlike flood irrigation, water is not delivered under atmospheric
pressure; instead, a minimum pressure head of 3 m is maintained at the
supply pipe outlets (farm inlets) by installing pressure regulating valves at
each outlet. Each supply pipe outlet is connected with on-farm riser pipes
which deliver water at the top end of each furrow through taps. The use of
pressurized riser pipes and taps eliminates the need for priming the siphons.
It also improves the water application rate. Hence under this scenario energy
is required to pressurize and move water through the pipe against the pipe
friction, and in some sections against the slope, by the pumps. Except for
the addition of energy required for irrigation pumping and delivery, all other
energy inputs are assumed to be the same as for Scenario 2. The only
modification made in the energy inputs for Scenario 2 is that the human
labour energy for irrigation is reduced to half due to elimination of the need
for priming of siphons.
4.5.1 Optimization of pipe diameters
218
The magnitude of flow velocity through a pipe is inversely proportional to
the diameter of the pipe. Therefore, to avoid flow velocity to exceed the
limit of 3 m/s the pipe diameter should be increased. The same optimization
module which is set up for Scenario 3 for the same reasons was executed to
find the optimum diameter of the flow pipes for the furrow irrigation case.
The optimization results are shown in Table 4.19. The average diameter of
the pipe system is increased by 15 mm as compared to Scenario 3.
Table 4.19: Original and optimized diameters for supply pipe network for Scenario 4
4.5.3 Accounting for energy use and GHG emissions in crop
production for Scenario 4
Since the irrigation application method and the total water use do not
change significantly from Scenario 2, the magnitude of energy inputs for the
three crops in Scenario 4 are also assumed to be the same as that of Scenario
2. The only modification made in the energy inputs for Scenario 2 is that the
human labour energy for irrigation is reduced to half due to elimination of
the need for priming of siphons to each furrow. However, the energy
consumption (especially diesel) in irrigation monitoring trips by channel
operators is not changed from Scenario 2. In addition, a significant amount
of energy is consumed in running pumps to move water from its source to
the farm outlets through the pipe system for Scenario 4. Also energy is
required to generate the pressure head of 3 m for the current scenario at each
pipe outlet. Hence the total energy required for Scenario 4 should be higher
than that of Scenario 2. However, when compared with Scenario 3 (flood
irrigation with piped supply), the optimized diameter of supply pipes is
relatively increased (340 mm versus 333 mm) while the total flow volume
pumped for irrigation is decreased (2841 GL versus 3619 GL). These two
factors contribute to a remarkable decrease in pumping energy consumption
for the current scenario.
The node-link model determines the number of active pumps on a given
irrigation day depending on the duty flow rate. The node-link model
220
tweaked for simulating this particular scenario indicates that as high as 11
pumps installed at a pumping station near the surface water source, each
with a peak discharge rate of 0.08 m3/s, are simultaneously operated in
parallel configuration to supply irrigation water. The model reports the
maximum and average duty flow rates of 0.916 m3/s and 0.093 m3/s,
respectively, for the communal pumping system. The model computes that a
total of 388.9 MWh of electrical energy is consumed by the electrical
motors to drive the pumps to supply irrigation water with a total supply
volume of 2,841 ML to the three crops during one complete year of
simulation. The total energy consumption is divided among the three crops
based on their proportional water use of the total irrigation volume as given
in Table 4.21. The pumping energy consumed per hectare of a crop is given
by dividing the total pumping energy consumption of that crop by its total
area in hectares.
Table 4.21: Irrigation water applied and corresponding energy consumption in irrigation pumping for Scenario 4
Total Citrus Stone fruit Wine grape Irrigation (ML) 2,841.0 2,450.0 222.1 168.8 Pumping energy (MWh) 388.9 335.4 30.4 23.1 Pumping energy (KWh/ha) 1336.6 1374.4 1249.0 1022.1
The modelled use of energy/electricity for irrigation pumping, other energy
inputs and the corresponding greenhouse gas emissions in the form of
equivalent carbon dioxide emissions on per hectare crop area basis are given
in Table 4.22 for the three crops for the current scenario.
Among the three crops, citrus requires the highest amount of pumping
energy on a per hectare basis followed by stone fruit and wine grapes. Citrus
is also highest when the total energy requirement per hectare from all energy
inputs is compared for the three crops. Similarly, the total greenhouse gas
emissions per hectare of crop associated with the energy inputs are also
highest for citrus crop followed by stone fruit and wine grapes.
221
Table 4.22: Accounts for energy inputs and corresponding greenhouse gas emissions on per hectare basis in the yearly production cycle of citrus, stone fruit and wine grapes for Scenario 4
Citrus Stone fruit
Wine grape
Energy input excluding electricity for pumping (KWh/ha)
7,762.1 7,388.9 6,623.7
Electricity consumed in irrigation pumping (KWh/ha)
1374.4 1249.0 1022.1
Total energy input (KWh/ha) 9,136.5 8,637.9 7,645.8 Total energy sequestered in yield (KWh/ha)
21,200 11,590 72,160
GHG emissions, excluding electricity for pumping (KgCO2-e/ha)
1,806.8 1,683.7 1,507.1
GHG emissions for electricity consumed in irrigation pumping (KgCO2-e/ha)
1,237.0 1,124.1 919.9
Total GHG emissions (KgCO2-e/ha) 3,043.8 2,807.8 2,427.0
The line to line comparison of Table 4.18 for Scenario 3 (flood irrigation
with piped supply) and Table 4.22 for Scenario 4 (furrow irrigation with
piped supply) indicates that the two cases are not much different from each
other in terms of total energy use and total greenhouse gas emissions for
each crop. However, Scenario 4 performs marginally better, chiefly due to
the reduced volume of total irrigation water pumped as compared to
Scenario 3. However, Scenario 4 can perform even better if the minimum
pressure head of 3m is not to be maintained at each outlet.
4.6 Scenario 5 - Sprinkler irrigation with pipe supply system
Scenario 5 refers to a fixed-head sprinkler irrigation system connected with
a pressurized pipe water supply system from a common water source to the
farm outlet. The system is assumed to be installed on each of the 13 farms
of the case study area. Unlike flood or furrow irrigation, the sprinkler
system requires a minimum pressure at each sprinkler head to operate. An
operating pressure less than the minimum required pressure will result in
lesser coverage of irrigation area and higher drainage loss; similarly, a very
high operating pressure may result in mist formation by sprinkler heads and
hence less distribution uniformity and greater evaporation losses. The
222
sprinkler system simulated under Scenario 5 works on a commonly used
operating pressure of 35 PSI or 25 m water head. That’s why; the node-link
model of the pipe supply system is setup with a fixed pressure head of 25 m
at each supply node (farm inlet).
The pressure required to be produced by the pump(s) at the pumping station
is much higher than the abovementioned fixed required pressure at each
outlet. This is due to the fact that as water moves through a pipe it loses
pressure due to a phenomenon called "friction loss" as explained in Chapter
3. The amount of friction loss is determined by the type of pipe, the
diameter of the pipe, the amount/speed of water flowing through the pipe,
and the length of the pipe. These factors are plugged into the Williams-
Hazen formula that is implemented in the node-link model to calculate the
friction loss in terms of meters of water head loss. In addition to friction
loss, pressure is also lost as water passes through a valve, pipe bend or
change in pipe diameter. These losses are termed as minor losses. The total
friction loss and minor losses are added to the fixed pressure head at each
farm inlet required to operate the sprinkler system, to get the total operating
pressure head required at the pump outflow pipe.
4.6.1 Irrigation demand versus irrigation delivery
The daily total irrigation demand and daily total actual supply time series
are shown in Figure 4.8. The irrigation demand computed by the crop ET
model and the daily irrigation volume supplied via supply pipe by the
pumping system are same as shown in Figure 4.8. The cumulative shortage
in irrigation supply remains zero throughout the simulation period; hence
the condition for a demand-based irrigation system is fulfilled. The total
daily irrigation demand is peaked at 48 ML/day as compared to 78.57
ML/day for furrow irrigation under Scenario 2. The cumulative irrigation
demand and the cumulative irrigation supply over the whole year remain
2,312 ML.
223
Figure 4.8: Comparison of daily irrigation demand and supply and cumulative shortage in irrigation supply (ML/day on Y-axis) for Scenario 5
4.6.2 Water losses estimation
Since this scenario consists of the piped irrigation supply system, no
channel seepage and channel evaporation loss occurs. The field losses
include deep percolation and the surface runoff. The cumulative deep
percolation losses at the end of the simulation are the highest at 47.19 ML
followed by the cumulative surface runoff (or surface drainage) of 30.66
ML. The total non-productive losses (sum of deep percolation and surface
runoff losses) of irrigation water amounts to 77.85 ML which is 3.4 per cent
of the total irrigation supply at the end of a one-year simulation and roughly
0.27 ML of water loss per hectare for the overall irrigated area of the case
study. The irrigation water loss rate for the current sprinkler irrigation
system is significantly lesser than the flood (Scenario 1) and furrow
(Scenario 2) irrigation system. There are two main reasons for lower losses
under the current scenario, firstly, the more precise and adequate application
of irrigation water where and when needed and elimination of transmission
Table 4.28 is the fact that the greenhouse gas emissions from the single
energy input for irrigation pumping operation are almost equal (17% to 22%
less) in magnitude to the total greenhouse gas emissions from all other
energy inputs for each of the three crops. This signifies the link between the
irrigation modernization and its environmental footprint that possibly
contributes toward exacerbation of phenomenon of climate change.
Table 4.28: Accounts for energy inputs and corresponding greenhouse gas emissions on per hectare basis in the yearly production cycle of citrus, stone fruit and wine grapes for Scenario 5
Citrus Stone fruit
Wine grape
Total energy input excluding electricity for pumping (KWh/ha)
6,924.7 6,192.4 5,618.7
Electricity consumed in irrigation pumping (KWh/ha)
1508.4 1528.3 1123.9
Total energy input (KWh/ha) 8,433.1 7,720.7 6,742.6 Total energy sequestered in yield (KWh/ha)
23,320.0 12,810 75,440
GHG emissions, excluding electricity for pumping (KgCO2-e/ha)
1,639.1 1,430.5 1,294.3
GHG emissions for electricity consumed in irrigation pumping (KgCO2-e/ha)
1,357.6 1,375.5 1,011.5
Total GHG emissions (KgCO2-e/ha) 2,996.7 2,806.0 2,305.8
The rates of energy (electricity) consumed per hectare of the three crops for
irrigation pumping using a communal pumping station are reasonably
comparable to what is reported in Giddings (2004) and Giddings (2005).
Other than a difference in pump size, pumping efficiency etcetera,
additional energy is required to pump water from the off-farm central
location (communal pumping station) as compared to the on-farm pumping.
4.7 Scenario 6 – Drip irrigation with pipe supply system
Scenario 6 refers to the surface drip/trickle irrigation system connected with
a pressurized pipe water supply system from a common water source to the
farm outlet. The common pipe supply network provides for the minimum
hydraulic pressure required for operating drippers on each drip line in the
field. This scenario closely replicates the current field setup of the 13 farms
235
of the case study area. The drip system for each farm as simulated under
Scenario 6 works on a commonly used operating pressure of 45 PSI or 32 m
water head. Therefore, the node-link model of the pipe supply system is
setup with a fixed pressure head of 32 m at each supply node (farm inlet).
Due to their differing approach for application of irrigation to the trees, the
operating pressure for drip system is higher than that of sprinkler system;
however, the rate of irrigation application volume is the other way around
(i.e. lesser).
The pressure required to be produced by the pump(s) at the pumping station
has to be much higher than the abovementioned fixed required pressure at
each outlet. This is due to "friction loss" as explained in Chapter 3 and
Scenario 5. The total friction loss and minor losses are added to the fixed
pressure head to get the total operating pressure head required at the
pump(s) outflow pipe.
4.7.1 Irrigation demand versus irrigation delivery
The daily total irrigation demand and daily total actual supply time series
for the case study area are shown in Figure 4.10. The irrigation demand
computed by the crop ET model and the daily irrigation volume supplied via
supply pipe by the operation of the pumping system are the same as shown
in Figure 4.10. Therefore, the cumulative shortage in irrigation supply
remains zero throughout the simulation period; hence the condition for a
demand-based irrigation system is fulfilled. The total daily irrigation
demand is peaked at 28.4 ML/day as compared to 48 ML/day for sprinkler
irrigation under Scenario 5. The cumulative irrigation demand and the
cumulative irrigation supply over the whole year remain 1,789 ML as
compared to 2,312 ML for the sprinkler system under Scenario 5.
236
Figure 4.10: Comparison of daily irrigation demand and supply and cumulative shortage in irrigation supply (ML/day on Y-axis) for Scenario 6
4.7.2 Water losses estimation
No channel seepage and channel evaporation loss occurs for this scenario as
it consists of the piped irrigation supply system. The field losses include
deep percolation and the surface runoff. The cumulative deep percolation
losses over the simulation period are higher than the cumulative surface
runoff as 34.7 ML and 22.0 ML, respectively. The total non-productive
losses (sum of deep percolation and surface runoff losses) of irrigation water
amounts to 56.7 ML which is 3.2 per cent of the total irrigation supply at the
end of a one-year simulation and roughly one-fifth (0.19 ML) of each mega-
litre of irrigation water applied per hectare is lost for the irrigated area of the
case study. The irrigation water loss rate for the drip irrigation system is
significantly lesser than all other irrigation systems discussed so far. There
are three main reasons for lower losses under the current scenario, firstly;
the more precise and controlled application of irrigation water where needed
and when needed, secondly; lesser field irrigation evaporation losses due to
smaller wetted area and thirdly; the elimination of transmission losses due to
piped supply. However, the frequency of irrigation for drip irrigation system
is higher than other irrigation methods due to least rootzone storage due to
least size of the bulk of wetted region around the trees. For example, the
total number of irrigation days for Scenario 6 is 307 days as compared to
176 days for Scenario 5 (sprinkler system).
4.7.3 Effect on crop yield
The model shows that the demand-based irrigation supply system under
Scenario 6 does not encounter any supply constraint. Therefore, there is no
reduction in crop evapotranspiration due to the absence of any water stress
and hence no reduction in yield of the three crops. The drip irrigation
system has the advantage of the timely and precise application of irrigation
which results in an increase in yield as compared to flood and furrow
irrigation. As mentioned by Dasberg (1995) and supported by field trials by
others the drip system helps manipulate irrigation application during water
stress sensitive periods such as during the crop growth cycle. This helps
control quantity as well as quality of yield.
4.7.4 Irrigation application rate
The developed model takes into account soil type in determining the
irrigation demand and irrigation timing for individual farms using a soil-
water balance approach. However, the irrigation application rate is defined
as the number of megalitres of water applied per hectare for a given crop
averaged for all farms with that crop regardless of the soil type of the
individual farms. The irrigation application rates for the three crops as
computed by the node-link model for the current scenario are given in Table
4.29. The irrigation application rate computed by the model are very similar
to those reported by Khan et al., (2005); Khan and Abbas (2007); Giddings
(2005); and Giddings, (2004) and other published literature. The analysis of
information given in Table 4.29 shows that on average there is potential for
further water savings of at least 0.97 ML/ha which, can be achieved by
minimizing irrigation water loss in the form of soil evaporation, deep
percolation and surface runoff, by efficient management of the irrigation
system and by adopting water saving practices and technology, for example,
238
replacing surface drip system with subsurface drip system. The average
modelled irrigation application rate is also the lowest and application
efficiency highest among the scenarios discussed so far. The total irrigation
water use combined for the three crops in the case study area for a complete
one-year cycle for Scenario 6 is 1,789 ML as estimated by the model, which
is 523 ML less than that of Scenario 5 indicating the water savings potential
of improved irrigation technology and supply system. In drip irrigation
system coupled with piped supply, less water is pumped but more efficiently
applied to the plants. The drip irrigation systems are potentially the most
water efficient than any other irrigation options, if managed properly, by
minimizing water loss through deep drainage, surface runoff and
evaporation from the soil surface. The drip systems may involve more
maintenance requirement than gravity based irrigation systems however; the
maintenance can also be automated to some extent for drip system.
Table 4.29: Average irrigation application rates for the three crops for the modelled Scenario 6
CitrusStone fruit
Wine grape
Average
Irrigation application rate (ML/ha)
6.26 6.30 4.77 5.78
Cumulative soil evaporation component of ETc (ML)
221.89 19.65 13.82 255.38 (total)
Soil evaporation component of ETc (ML/ha)
0.91 0.81 0.61 0.78
Deep percolation (ML/ha) - - - 0.12 Surface runoff or drainage (ML/ha)
- - - 0.07
4.7.5 Accounting for energy use and GHG emissions in crop
production for Scenario 6
In this section different direct and indirect energy inputs and their
greenhouse gas emissions are discussed for Scenario 6. Similar to the
sprinkler irrigation system discussed in Scenario 5, irrigation supply is made
via pipes from a central pumping station at certain fixed pressure for
Scenario 6. The water is pumped and moved under hydraulic pressure
239
through irrigation supply pipes from the source (pumping station), to the
supply points i.e. the farm irrigation system inlets. The pipe flow rate is
varied by increasing/decreasing the number of pumps depending upon the
irrigation demand for a given day, however, the pressure head at each pipe
outlet (farm inlet) is kept almost constant by use of pressure regulating
valves to operate the drip irrigation system on each farm. Since irrigation
pumping is an energy intensive operation; the energy use for irrigation
supply has to be incorporated in the energy use analysis. The drip lines,
once installed on either side of the tree line, are fixed and no energy is
required to roll the drip lines. However, some human hours are expended in
inspecting and maintenance of the drippers and drip lines at each farm. It
also involves random trips by irrigation operators/inspectors on four-wheel
drive vehicles to inspect and monitor the piped water supply for any
leakages or unauthorized access. Given that there is relatively lesser need
for monitoring, both the human hours spent by irrigation inspectors and the
fuel consumed by their vehicles are assumed to be halved as compared to
what is reported in Table 4.11 (Scenario 2) for citrus, stone fruit and wine
grape farms of the case study area. Based on this data and energy conversion
factors for diesel consumption and human hours given in Chapter 3, the
equivalent energy consumed in monitoring operations amounts to 4.4
KWh/ha for citrus, 29.25 KWh/ha for stone fruit and 14.7 KWh/ha for wine
grape crop.
The energy consumed in monitoring operations also includes periodic trips
by the technical personnel to check pumping system at the pumping station
for any potential faults and scheduled maintenance.
4.7.5.1 Energy inputs and GHG emissions for citrus
Energy inputs, energy outputs and corresponding greenhouse gas emissions
per hectare of drip irrigated citrus crop with pipe supply system are given in
Table 4.30. The labour hours under “irrigation” in Table 4.30 also includes
time spent by irrigators on the farm to manage irrigation, for example
monitoring and servicing of drippers in this case, as well as the time
240
expended by irrigation inspectors. Similar to the preceding scenarios, the
rates of direct and indirect energy inputs are either based on data collection,
personal communications or publications by local agencies including NSW
Department of Primary Industries; Falivene (2003); Giddings (2005);
Giddings (2004); and Crean et al., (2004). The fertilizer application rates are
relatively conservative to those of Scenario 5 (Sprinkler system) owing to
improved fertilizer use efficiency for suitably managed drip irrigation
system.
In order to compute the total energy use, each type of input energy is
essentially converted into the equivalent energy sequestered in that input
and expressed as kilowatt hours using the conversion factors given in
Chapter 3. With drip irrigation system, through fertigation, continuous small
applications of soluble nutrients are made which overcome the fertilizer loss
through runoff or leaching problems, save labour, reduce compaction in the
field, result in the fertilizer being placed around the plant roots uniformly
and allow for rapid uptake of nutrients by the plant. This allows easier,
controlled, more effective and more precise application of fertilizers
especially Urea which can quickly leach out of the root zone due to its high
solubility. Therefore, quantity of the fertilizers used for Scenario 6 is much
less than that of Scenario 1 (flood) or Scenario 2 (furrow) as given in Table
4.30. Fertigation also eliminates the use of a tractor to spread fertilizer in the
field. These factors contribute in a decrease in both the direct (diesel,
labour) and the indirect (fertilizer) energy inputs. The fertilizers are usually
dissolved in water with the ratio of 1:5, i.e. 100 kg of fertilizer in 500 litres
of water (Giddings, 2004) and applied during irrigation using methods like
suction injection, pressure differential injection or pump injection (NSW
DPI, 2000) and (Treeby et al., 2011).
As given in Table 4.30, the total energy use from all considered energy
inputs (excluding electricity consumption in irrigation pumping) to the drip
irrigated citrus farms connected with piped irrigation supply system is
aggregated to 6,897.1 KWh per hectare and the total output energy
241
sequestered in the resulting citrus yield at the rate of 48 t/ha is 25,440 KWh
per ha. Similarly the total GHG emissions from energy use for citrus
farming operations excluding irrigation pumping are estimated to be
1,637.75 Kg of CO2-equivalent emissions per hectare.
Table 4.30: Accounts for energy inputs and GHG emissions on per hectare basis in the yearly production cycle of citrus crop for Scenario 6 Input Quantity
pumping operations are 17% to 34% less than the total greenhouse gas
emissions from all other energy inputs for each of the three crops. This
signifies the link between irrigation modernization and its environmental
footprint that possibly contributes toward exacerbation of phenomenon of
climate change. However, the greenhouse gas emissions from drip irrigation
system in Scenario 6 are much less than those for the sprinkler system in
Scenario 5.
Table 4.34: Energy inputs and corresponding greenhouse gas emissions on per hectare basis in the production cycle of citrus, stone fruit and wine grapes for Scenario 6
Citrus Stone fruit
Wine grape
Total energy input excluding electricity for pumping (KWh/ha)
6,897.1 6,000.9 5,739.5
Electricity consumed in irrigation pumping (KWh/ha)
1,210.9 1,212.0 1,212.4
Total energy input (KWh/ha) 8,108.0 7,212.9 6,951.9 Total energy sequestered in yield (KWh/ha)
25,440.0 15,250.0 85,280
GHG emissions, excluding electricity for pumping (KgCO2-e/ha)
1,637.8 1,396.7 1,314.7
GHG emissions for electricity consumed in irrigation pumping (KgCO2-e/ha)
1,089.8 1,090.8 1,091.2
Total GHG emissions (KgCO2-e/ha) 2,727.6 2,487.5 2,405.9
4.8 Comparison of the demand-based irrigation scenarios
This section draws comparisons among the abovementioned six irrigation
scenarios with respect to water, energy and greenhouse gas emissions. A
range of established indicators and key variables are calculated and
discussed in this section to cover different aspects and viewpoints for the
simulated scenarios. All six scenarios consist of the same crops with the
same representative irrigated area and wit the only difference from one
scenario to the other being the irrigation method and/or irrigation supply
system (open channel or pipes). Therefore, the indicators discussed here
reflect purely the water and energy aspects of irrigation methods and
irrigation conveyance systems than the crop themselves.
249
4.8.1 Comparison of water and energy use rates
The irrigation application rates (ML/ha) to the three selected crops in the
case study area, for each of the six scenarios, are shown in Figure 4.12.
Similarly the energy use rate per hectare (kWh/ha) of the three crops can be
compared from Figure 4.13. Since each scenario is simulated as a demand-
based irrigation system, the irrigation rates tend to be higher than fixed
interval irrigation scheduling systems. However, as mentioned earlier,
Scenario 1 is not fully simulated as a demand-based system because of the
capacity constraint of the open channel supply system. That is why; the
irrigation rate for Scenario 3 is rather higher than Scenario 1 despite water
savings from conveyance losses.
Figure 4.12: Irrigation application rates (ML/ha) for each crop for the six scenarios
The comparison of irrigation rates for the given six scenarios in Figure 4.12
is in fact comparison of water savings from different irrigation application
methods as well as those resulting from open channel to piped supply
system. Based on the results shown in Figure 4.12, drip irrigation connected
with pressurized pipe supply system offers the highest water savings.
It should be clarified that there is only a slight difference in the rates of
energy use rate for flood irrigation based system (Scenario 1) compared to
the drip irrigation based system (Scenario 6). But in the context of energy
use for irrigation pumping alone, Scenario 6 involves as much as 1,212
250
kWh/ha of energy while energy use for irrigation pumping for Scenario 1 is
effectively zero. A similar magnitude of total energy use for both scenarios
is the fact that the energy inputs in the form other than irrigation pumping
are significantly lesser for Scenario 6 than that of Scenario 1. However,
these energy savings are off-set by the energy used for irrigation pumping
for Scenario 6.
Figure 4.13: Energy use per hectare (KWh/ha) for each crop for the six scenarios
From the comparison of Figure 4.12 and Figure 4.13, it is suggested that
both the energy and the water saving aspects of conversion from open
channel to the piped supply system should be duly considered side-by-side.
The water and energy analysis of using on-farm storages are carried out in a
later chapter of this thesis.
4.8.2 Comparison of efficiency and productivity indicators for
water and energy
Efficiency and productivity indices or indicators are a well-adopted
approach to compare different scenarios which deal with similar problems
and also to compare scenarios against some standard or acceptable
benchmark values of those indices. The domain of the current discussion
encompasses water and energy linkages in different irrigation supply and
application systems for a given study area. These indices/indicators are
251
widely used in scientific and research discussions about water-energy nexus
in irrigation and crop production systems. Their theory is discussed in
Chapter 3.
4.8.2.1 Comparison of irrigation efficiency
The definition of “irrigation efficiency” as endorsed by the Irrigation
Association of Australia is based on an approach suggested by the
International Commission on Irrigation and Drainage (ICID) as per the
paper by Bos et al., (1993). One of the essential elements of this approach is
that it tracks and accounts for water use from the point of supply all the way
through to the crop and provides the following (Equation 4.3) overall
definition of “irrigation project efficiency”. This definition is suitable for all
irrigation systems at an irrigation case study/scheme/district level and
above.
Equation
4.3
Another definition of irrigation efficiency which is closest to the above
equation is given by Israelsen (1932) as “the ratio of irrigation water
transpired by the crops of an irrigation farm or project during their growth
period to the water diverted from a river or other natural source into the
farm or project canal or canals during the same period of time.” It is usually
expressed in percentage terms.
The term defined in Equation 4.3 is further broken up into sub-components
including conveyance efficiency, distribution efficiency, and field
application efficiency. For this study, however, more emphasis is given to
the irrigation project efficiency as the focus is to compare water and energy
use at the representative case study scale rather than the individual farms
scales.
Despite two soil types with somewhat different irrigation thresholds a single
value of irrigation efficiency is computed for the case study area for
simplicity. High levels of irrigation efficiency translate into lower operating
252
costs and energy use, improved production per megalitre of water used and
improved environmental management. A comparison of computed irrigation
efficiency at the case study area level for the six scenarios is given in Table
4.35. The total transpiration is from the three crops in the case study area
and similarly total irrigation supplied refers to water volume extracted from
the communal water source over the course of one year. The transpiration
depth is duly converted to water volume by multiplying with the product of
irrigation wetted area (m2/ha) and the crop area. The conveyance losses
which include channel seepage and channel evaporation (open channel
supply only); field losses including evaporation from wet soil surface;
surface runoff and deep percolation constitute the difference between
irrigation supply and total transpiration volume.
It is evident from Table 4.35 that pressurized irrigation i.e. sprinkler and
drip system with a piped supply system result in the highest irrigation
efficiency. As discussed for the individual scenarios earlier, the sprinkler or
the drip system has lower water supply requirements due to minimal field
losses, zero conveyance losses and precise and controlled application of
irrigation water to the plants. It can also be concluded from Table 4.35 that
the only water savings from conversion of open channels (Scenarios 1 and
2) to a piped supply (Scenarios 3 and 4) for flood and furrow irrigation
systems leads to savings through reductions in conveyance losses. Since the
magnitude of conveyance losses is much lesser than the field losses, as long
as the irrigation efficiency is concerned, there is no significant improvement
in it for the case study area with these two irrigation systems even if the
piped supply is used. However, this may not be the case for large irrigation
areas with vast network of open channels where conveyance losses can be
significant. Therefore, Scenario 3 and Scenario 4 do not make any
improvement as far as irrigation efficiency is concerned, but result in
conveyance loss savings of 4.6 ML/km of supply channel. It is also evident
from Table 4.35 that there is more efficient water use by plants under
sprinkler and drip irrigation systems and hence relative improvements in
yields. Moreover, in terms of field losses, the extent of wetted area is a key
253
determinant. The larger the wetted area, the larger will be the field losses.
Hence the field losses for flood and furrow irrigation are much higher than
those of sprinkler and drip systems. The irrigation efficiency ranges from
76.1% (flood irrigation) to 92.6% (drip irrigation) as given in Table 4.35 for
the six scenarios. These irrigation efficiency values are much high than what
is attainable with conventional systems and can be attributed to the demand-
based irrigation strategy. The irrigation amount and irrigation interval are
varied as per field conditions under demand-based irrigation to minimise
water losses and achieve maximum irrigation efficiency.
254
Table 4.35: Computed overall/project level irrigation efficiency for the six scenarios
The comparison of water productivities given in Table 4.36 manifests that
Scenario 6 which represents drip irrigation installed on all farms and
connected with a communal pipe supply system has the highest water
productivity for each of the three modelled crops. Significant improvement
in water use efficiency for drip irrigation as compared to the other scenarios
is the main contributing factor to the highest water productivity than the
fruit yield which just marginally improves among these scenarios.
Moreover, the productive use of 1 cubic meter of irrigation water for drip
irrigated crop is significantly higher than that of flood irrigation, mainly due
to the fact that a given quantity of irrigation is applied more frequently in
smaller amounts for drip irrigation and hence more water remains available
to the plants for a longer time than that of flood irrigated crops.
4.8.2.3 Comparison of energy productivity
Energy productivity in agriculture refers to the quantity of marketable yield
per unit of input energy and is expressed as kg/kWh. It includes all direct
and indirect energy inputs which are regularly applied. Optimum energy use
is vital for agricultural production systems (Ommani, 2011). Energy
256
productivity reflects the performance of an agricultural production system
(irrigated horticulture in current case) especially when total energy use is a
particular concern and reflects the utilization of energy by a given
agricultural system. The higher the energy productivity of a system, the
greater the production per unit of energy input.
The energy productivity values are given in Table 4.37 and are calculated
from simulated results discussed earlier for the six scenarios. Drip irrigation
is usually considered an energy intensive system mainly due to the high
energy requirements for pumping the irrigation water. However, a
significant portion of the required pumping energy to operate drip system is
offset by energy savings from reduced volumes of water required, reduced
application of fertilizers and other chemicals and of course yield
improvements due to proper and timely irrigation management as compared
to other irrigation systems. Therefore, drip system (Scenario 6) exhibits the
highest energy productivity among all scenarios.
Table 4.37: Energy productivity (kg/kWh) indicators for the six scenarios
Scenario No.
Citrus (kg/kWh)
Stone fruit (kg/kWh)
Wine grape (kg/kWh)
Average (kg/kWh)
1 4.44 2.50 2.97 3.30
2 5.13 2.56 3.32 3.67
3 3.75 2.05 2.49 2.76
4 4.38 2.20 2.88 3.15
5 5.22 2.72 3.41 3.78
6 5.92 3.47 3.74 4.38
The comparisons of energy productivity and water productivity given in
Table 4.37 for each crop for Scenario 1 with Scenario 3 and that of Scenario
2 with Scenario 4 indicate that despite some water savings by conversion
from open channel to pipes to supply water to the flood or furrow irrigation
system, the additional energy required for pumping water through piped
supply is huge and results in significant reduction in energy productivity
with relatively small increases in water savings from channel seepage and
257
channel evaporation. Therefore, Scenario 3 and Scenario 4 will have to be
rejected as viable irrigation practices based on their low energy productivity.
4.8.2.4 Comparison of energy efficiency
Energy efficiency of the agricultural production system can be defined as
the ratio of total energy output from agricultural produce to the total energy
input to engender that produce. For the current study, the energy inputs
include fertilizers, chemicals, pruning, thinning, fruit picking, use of
machinery and labour and electricity for irrigation pumping for three
horticulture crops. Energy from the sun is also a major input which is
usually not considered in energy analysis for crop production as it is not
purchased. The only output energy accounted in this analysis is in the form
of fruit yield. Energy sequestered in the remaining biomass e.g. trunk,
branches, leaves and fruit waste are not considered. As explained by the
laws of thermodynamics, the useful energy extracted from an energy store
(fruit yield in this case) is always less than the energy put into that energy
store. It means energy efficiency of a production system can never be
greater than unity. However, we do not consider free energy inputs like solar
in this analysis and hence the energy efficiency of each scenario is expected
to be greater than unity.
Crop production is an energy sequestration process, mainly through
photosynthesis, and therefore energy efficiency of a given crop should be at
least higher than unity. The energy efficiency indicators as computed for the
three crops grown on the farms in the case study area for each of the six
scenarios are given in Table 4.38. It is evident from energy efficiency values
in Table 3.38 that conversion from open channel to piped supply for gravity
irrigation systems namely flood and furrow, does not improve energy
efficiency of the selected crops as long as the irrigation demand is fully met
by either supply system. Therefore, Scenario 3 and Scenario 4 are
inefficiently high energy demanding options given the use of piped supply
which has to be pumped. However, there could be other justifiable reasons
for this form of conversion, such as where conveyance losses are high,
258
limited and untimely supply of irrigation water and inaccurate metering of
farm water use. All these issues are addressed by use of piped supply.
Table 4.38: Energy efficiency (kWh/kWh) indicators for the six scenarios
Scenario No.
Citrus (kWh/kWh)
Stone fruit (kWh/kWh)
Wine grape (kWh/kWh)
Average (kWh/kWh)
1 2.35 1.53 9.75 4.54
2 2.72 1.56 10.88 5.05
3 1.99 1.25 8.17 3.80
4 2.32 1.34 9.44 4.37
5 2.77 1.66 11.19 5.21
6 3.14 2.11 12.27 5.84
The energy efficiency of sprinkler and drip irrigation systems, both
connected with pressurized piped supply systems, is quite comparable.
However, the drip system performs best due to relatively low energy inputs
and marginally higher or equal yield for each crop as compared to the
sprinkler system. Wine grapes contain the highest amount of energy in the
fruit and therefore highest energy efficiency among all three selected crops.
However, energy efficiency of grain crops is usually higher than horticulture
as reported by Khan et al., (2009).
4.8.2.5 Comparison of specific energy
The specific energy of an agricultural production system can be defined as
the total energy input per unit of marketable yield and is expressed as
kWh/kg. It is essentially the reciprocal of energy productivity. As
mentioned earlier, the total input energy for specific energy estimation does
not include free solar energy. The specific energy is the amount of energy
used in different forms through different processes to produce a unit of
marketable yield rather than actual energy that is ultimately sequestered in
the yield. Specific energy calculated for the six scenarios in the current
study are given in Table 4.39. The lower the value of specific energy of an
agricultural production system the more efficient that system is in producing
259
that output as is the case for the pressurized pipe driven drip system
represented by Scenario 6.
Table 4.39: Specific energy (kWh/kg) indicators for the six scenarios Scenario
No.
Citrus
(kWh/kg)
Stone fruit
(kWh/kg)
Wine grape
(kWh/kg)
Average
(kWh/kg)
1 0.23 0.40 0.34 0.32
2 0.19 0.39 0.30 0.29
3 0.27 0.49 0.40 0.36
4 0.23 0.45 0.35 0.32
5 0.19 0.37 0.29 0.26
6 0.17 0.29 0.27 0.23
4.8.2.6 Comparison of water – energy productivity
Water-energy productivity refers to yield per unit of energy and water inputs
and expressed as g/m3/kWh. This indicator captures the effect of these major
inputs on yield. Lower values of water-energy productivity may indicate
lower efficiency and higher environmental footprint of the system under
consideration.
Table 4.40: Water – energy productivity (g/m3/kWh) indicators for the six scenarios Scenario
No.
Citrus
(g/m3/kWh)
Stone fruit
(g/m3/kWh)
Wine grape
(g/m3/kWh)
Average
(g/m3/kWh)
1 0.36 0.19 0.27 0.27
2 0.51 0.29 0.45 0.42
3 0.30 0.15 0.22 0.23
4 0.44 0.24 0.39 0.35
5 0.64 0.33 0.53 0.50
6 0.95 0.55 0.78 0.76
Water-energy productivity as computed for the six simulated scenarios is
given in Table 4.40. The values are expressed as g/m3/kWh in Table 4.40.
The water-energy productivity of Scenario 6 is the highest amongst all
scenarios. The relatively higher magnitude of this indicator for Scenario 6
indicates that Scenario 6 has the highest yield and the lowest energy
260
footprint of both water use and energy input. Citrus especially outperforms
the other two crops.
4.8.2.7 Comparison of water – energy ratio
The water – energy ratio is the ratio of energy input from irrigation to total energy
input. It is the fraction of the total input energy that is expended in irrigation
operations. A higher ratio may imply higher input energy for irrigation and thus
higher environmental footprint of irrigation. Each irrigation method involves use of
energy in different forms including human labour, machinery and fuels
(diesel/electricity etc.). The modern irrigation technologies including sprinkler,
centre pivot and drip systems are more energy intensive methods of irrigation
which require significant amount of direct energy for pumping operations as
compared to the conventional gravity based irrigation methods. Water energy ratios
for the three selected crops for each of the six scenarios are given in Table 4.41
where the last column represents average values for the overall case study area.
The water energy ratios reconfirm that the irrigation systems which require
pumping of water have conspicuously higher energy and thus exhibit higher energy
footprint. The water energy ratio for the drip system is marginally lower than that
of the sprinkler system mainly due to lower volumes of irrigation pumping for the
former.
Table 4.41: Water – energy ratio (kWh/kWh) for the six scenarios Scenario
No.
Citrus
(kWh/kWh)
Stone fruit
(kWh/kWh)
Wine grape
(kWh/kWh)
Average
(kWh/kWh)
1 0.01 0.01 0.01 0.01
2 0.01 0.01 0.01 0.01
3 0.16 0.19 0.17 0.17
4 0.16 0.16 0.14 0.15
5 0.18 0.21 0.17 0.19
6 0.15 0.18 0.18 0.17
4.8.3 Comparison of greenhouse gas emissions for modelled
scenarios
During the process of energy conversion and energy consumption, different
greenhouse gases are emitted. Similarly, greenhouse gases are emitted
261
during the crop production process as a result of the use of different forms
of energy as discussed previously. Generally, higher energy and water use is
linked with higher greenhouse gas emissions. The rate of greenhouse gas
emissions produced from expending of different energy sources is different.
Nevertheless, the water and energy indicators discussed in the previous
section are directly linked with greenhouse gas emission rates and hence can
be used as a surrogate indicator for the environmental footprint of water and
energy use.
Figure 4.14: Total greenhouse gas emissions per hectare (kg-CO2e) of each crop for the six scenarios (line graph shows GHG emissions from irrigation only and not other factors of crop production)
The total greenhouse gas emissions from irrigation and non-irrigation
energy inputs on a per hectare basis for each crop in the case study area are
plotted in Figure 4.14. The line graphs show GHG emissions from energy
inputs for irrigation operations only (water supply and delivery) and exclude
other production factors. For Scenario 1 and Scenario 2, the emissions rates
are lower as there is no irrigation water pumping involved. The sprinkler
and drip systems operate under high pressure which expends high pumping
energy and are thus categorized as high environmental footprint options.
262
4.9 Sensitivity analysis
Sensitivity analysis of selected modelled scenarios is carried out to
understand how a particular output variable responds to the variation in the
selected inputs within a specific range. The sensitivity analysis can be uni-
variate or multivariate. In this study only a uni-variate approach was taken,
where sensitivity of a given output is gauged against variability/uncertainty
of a single input variable at a time. The sprinkler and drip system as
discussed before involve some prominent characteristics including
significant amounts of energy use in irrigation pumping, GHG emissions,
water savings, yield improvement and increasing rate of technology
adoption. Therefore, sensitivity analysis of key variables mainly “total
energy use” is carried out in this study for sprinkler (Scenario 5) and drip
systems (Scenario 6). Since all scenarios discussed in this chapter are
demand-based irrigation system, crop water shortages are assumed to be
non-existent for each scenario and therefore total water use remains
unchanged negating the need for sensitivity analysis of water use.
4.9.1 Sensitivity of energy use in irrigation
Energy consumed in irrigation pumping constitutes the single major
component of total energy inputs in crop production with pressurized
irrigation systems. Therefore, sensitivity analyses of energy use for
irrigation pumping each for sprinkler (Scenario 5) and drip system (Scenario
6) are carried out to determine how energy use responds to variation in
different factors. For example, the delivery pressure head at the irrigation
outlets (sprinkler heads for sprinkler system and drippers for drip system) is
assumed to be constant and a single value is used for the entire simulation of
a given irrigation method. However, in the field situation, despite
installation of pressure compensating devices, the delivery pressure is likely
to vary within a certain range around its mean value and less likely to take
extreme values. Therefore, normal distribution of probability of delivery
pressure head values was assumed to capture this uncertainly. Then the
original irrigation simulation model developed in Vensim for each of these
263
two irrigation systems scenarios was setup to execute in sensitivity analysis
mode. For each scenario, the sensitivity module randomly generates the
irrigation delivery pressure head within 10% of the original model values
using “normal distribution” for up to 500 iterations of the model to compute
the consequent energy required for irrigation pumping for the whole case
study area. The cumulative probability of normal distribution functions used
for varying the delivery pressure in the model for the sprinkler and drip
system scenarios are plotted in Figure 4.15.
Figure 4.15: Cumulative probability distribution plots for the delivery pressure head for sprinkler (left) and drip system (right)
Figure 4.16: Sensitivity of cumulative energy use (kWh) for sprinkler irrigation pumping to ±10% change in delivery pressure head (m)
0.0
0.2
0.4
0.6
0.8
1.0
22.5 23.5 24.5 25.5 26.5 27.5Cumulative probability distribution
Table 5.3: Net irrigation rate (ML/ha) for three crops for demand-based and supply-based scenarios
Irrigation regime
Crop Scenario 1 Scenario 2
Scenario 3 (Scenario
5 for demand based)
Scenario 4 (Scenario
6 for demand based)
Demand based
Citrus 12.38 10.03 8.1 6.26 Stone fruit
13.38 8.87 8.20 6.30
Wine grapes
11.13 7.38 6.40 4.77
Supply based
Citrus 6.22 5.68 4.87 3.18 Stone fruit
6.31 8.91 7.23 6.61
Wine grapes
5.44 6.05 5.52 4.42
5.4.3 Comparison of crop yield
Comparison of the reduction in irrigation water use alone is not that
meaningful unless the effects on crop yield are analysed simultaneously.
279
Moreover, the supply-based irrigation system is more likely to have long
dry spells therefore the crop yield should be affected as given in Table 5.4.
For supply-based systems the yield of citrus crop is reduced by 49 per cent
for flood to 66 per cent for drip irrigation. The yields of stone fruit and wine
grapes are not reduced by similar magnitude. Hence, it can be concluded
that citrus crops are more sensitive to irrigation management strategies as
compared to stone fruit and wine grapes. This also raises the question of
return on capital investment for hi-tech (drip) irrigation systems under
supply based scheduling, particularly for irrigated citrus areas, where water
supply is as low as 58% of crop water demand.
Table 5.4: Comparison modelled crop yield (t/ha) between supply-based and demand-based irrigation systems
Irrigation regime
Crop Scenario
1 Scenario
2
Scenario 3 (Scenario 5
for demand based)
Scenario 4 (Scenario
6 for demand based)
Demand based
Citrus 35 40 44 48 Stone fruit
18 19 21 25
Wine grapes
20 22 23 26
Supply based
Citrus 18.0 (49%)
22.8 (43%)
20.8 (53%) 16.5 (66%)
Stone fruit
8.8 (51%)16.4
(14%) 16.7 (21%) 21.1 (16%)
Wine grapes
13.9 (30%)
21.3 (3%) 21.0 (9%) 24.2 (7.1%)
5.4.4 Comparison of water losses
The comparison of different types of water losses is a key component of
analysis of the demand-based and supply-based irrigation strategies. Table
5.5 lists different water loss components which are outputs of the developed
model for each scenario. When compared with demand-based scenarios,
there is a drastic reduction (up to 93%) in deep percolation and surface
runoff for the supply based scenarios due to reduction in irrigation volumes.
However, the total soil evaporation reduction is only limited to 9% for the
280
supply-based scenarios mainly due to the fact that the soil wetted area is
kept identical for both irrigation strategies. It is also evident from Table 5.5
that total water losses for high water use scenarios (flood and furrow) are
more sensitive to irrigation management strategy than the low water use
(sprinkler and drip) scenarios. It can also be concluded that the supply-based
irrigation strategy has effectively worked in reducing deep percolation and
surface runoff for the flood irrigation scenario.
The conveyance loss from open channels is reduced from 18.7 ML for
demand-based to just 6.8 ML for the supply-based irrigation, mainly due to
reduced number of irrigation days for the latter system. The model assumes
that the supply channels are filled with water at the time of irrigation supply
and hence the conveyance loss takes place only when irrigation is being
supplied. The conveyance loss would be of significantly higher magnitude if
supply channels remain pre-filled during the irrigation season.
Table 5.5: Comparison of total water losses (ML) for supply-based and demand-based irrigation scenarios
Irrigation regime Loss Type Scenario
1 Scenario
2
Scenario 3 (Scenario 5 for demand
based)
Scenario 4 (Scenario 6 for demand
based)
Demand based
Conveyance loss 18.74 18.75 0.0 0.0
Soil evaporation
368.43 307.31 313.14 255.38
Deep percolation
448.85 273.14 47.19 34.7
Surface runoff 296.68 184.57 30.66 22.0
Total water Loss 1132.7 783.77 390.99 312.08
Supply based
Conveyance loss 6.84 6.69 0.0 0.0
Soil evaporation
376.35 309.50 300.29 231.54
Deep percolation
33.07 48.70 39.58 40.18
Surface runoff 21.69 38.70 28.53 31.49
Total Water Loss
437.95 403.59 368.4 303.21
281
The comparison of yields given in Table 5.4 and total water losses given in
Table 5.5 for each scenario for the two irrigation management strategies
indicates that there is negative feedback or inverse relationships between the
two variables. The supply based irrigation strategy results in more water
savings but at the expense of crop yield and vice versa for the demand based
case. This brings in the need for an intermediary approach that optimizes the
two quantities. This observation will be further explored in a separate
chapter on system dynamics.
5.5 Energy and GHG emissions for the supply-based scenarios
This section provides a detailed analysis of energy use and GHG emissions
for each of the scenarios being discussed in this chapter and also provides
their comparison with corresponding scenarios discussed in Chapter 4 on
demand-based irrigation.
5.5.1 Comparison of energy use and energy output
It can be envisaged that energy use is proportional to irrigation water use for
any crop production. The yield of a fruit crop also increases up to a certain
limit with increasing irrigation volume, which implies that more energy has
to be expended in pruning the trees and harvesting the produce. Similarly,
more fertilizer can be usefully applied and expended with more and frequent
irrigation. Likewise, the energy use in irrigation pumping for pressurized
irrigation systems is increased with increased volume of irrigation. The
major difference between demand-based and supply-based irrigation is that
a higher and more frequent volume of water is applied in the former case.
Due to the unavailability of full data on energy inputs in production of
individual crops for the supply-based irrigation scenarios discussed in this
chapter, it is assumed that the energy use (excluding energy consumed in
irrigation pumping where applicable) for the production of three crops for a
given scenario is factored from the energy used in corresponding demand-
based scenario by the ratio of the water use volume for the supply-based and
demand-based irrigation for that scenario. The node-link model is able to
282
simulate total energy consumed in pumping irrigation water for the piped
supply systems (Scenario 3 and Scenario 4). No energy is used in pumping
water for open channel supply system that feeds water to gravity based
irrigation (flood and furrow system) as no pumping is involved.
Based on the abovementioned assumptions and discussion, the total direct
and indirect energy inputs in crop production i.e. excluding irrigation
pumping, for the four scenarios for both irrigation strategies is given in
Table 5.6. The energy use for each crop under the supply based scenario is
computed from that of demand-based scenarios using the formula given in
Equation 5.1.
Equation 5.1
Where,
= energy use for a given crop under given supply-based scenario
(kWh/ha)
= energy use for a given crop under given demand-based scenario
(kWh/ha)
= total water use for a given crop under given supply-based scenario
(ML/ha)
= total water use for a given crop under given demand-based scenario
(ML/ha)
Similar to the approach followed for demand-based scenarios, the modelled
total energy consumption for irrigation pumping for each supply-based
scenario is distributed among the three crops in proportion to their irrigation
volume. The resulting pumping energy is converted to kWh/ha by dividing
it by the crop irrigated area and is reported in Table 5.6. The energy use for
the three crops under demand-based scenarios is also included in Table 5.6
for quick comparisons.
283
Table 5.6: Energy use for the three crops under demand-based scenarios and the computed energy use for the corresponding supply based scenarios
Irrigation regime
Crop Energy
form
Scenario 1
(kWh/ha)
Scenario 2
(kWh/ha)
Scenario 3 (Scenario
5 for demand based)
(kWh/ha)
Scenario 4 (Scenario
6 for demand based)
(kWh/ha)
Demand based
Citrus
Crop production
7,889.2 7,794.6 6,924.7 6,897.1
Irrigation pumping
0.0 0.0 1508.4 1,210.9
Stone fruit
Crop production
7,195.3 7,409.3 6,192.4 6,000.9
Irrigation pumping
0.0 0.0 1528.3 1,212.0
Wine grapes
Crop production
6,728.1 6,632.4 5,618.7 5,739.5
Irrigation pumping
0.0 0.0 1123.9 1,212.4
Total
energy use
21,812.6 21,836.3 22,896.4 22,272.8
Supply based
Citrus
Crop production
3,963.7 4,414.1 4,163.4 3,503.6
Irrigation pumping
0.0 0.0 1204.8 863.8
Stone fruit
Crop production
3,393.3 7,409.3 5,459.9 6,296.2
Irrigation pumping
0.0 0.0 1,788.6 1,795.5
Wine grapes
Crop production
3,288.5 5,437.1 4,846.1 5,318.4
Irrigation pumping
0.0 0.0 1,365.6 1,200.6
Total
energy use
10,645.5 17,260.5 18,828.4 18,978.1
284
The overall high energy intensity (kWh/ha) of demand-based irrigation
scenarios for crop production (excludes irrigation) is clearly evident from
Table 5.6 where Scenario 5 the sprinkler system connected with piped
irrigation supply sits at the top of the ladder with energy intensity of 22,896
kWh/ha of irrigated area. However, for the supply-based irrigation strategy,
Scenario 4, the drip system requires marginally higher amounts of overall
energy than the scenario with sprinkler system. It is also evident from Table
5.6 that the electrical energy consumed in pumping irrigation water for the
supply-based scenarios (sprinkler and drip only) is not necessarily lower
than corresponding demand-based scenarios despite the fact that lesser total
water volume has to be pumped in the former case. The reason for this
observation is evidenced from Figure 5.4. It is a hydraulically proven fact
that the dynamic head/energy required for pumping water through a pipe is
proportional to the squared magnitude of flow velocity as depicted by
Bernoulli’s energy equation (Daugherty et. al., 1985). In other words, if
flow rate through the pipe is doubled, the energy required to move the water
will increase by four fold, keeping other parameters constant.
Figure 5.4 shows that the occurrence of higher duty flow rates for supply
based scenarios are much more frequent than that of demand-based
scenarios, which is what is expected for a supply-based irrigation strategy
i.e. apply larger amounts of irrigation at longer intervals. The use of higher
flow rates, though not much often, results in higher head losses and hence
higher pumping energy consumption for supply-based scenarios. This is the
explanation for higher pumping energy consumption for supply-based
scenarios discussed in the current study. These results indicate that water
use and enormity of energy inputs and energy outputs in irrigated crops
(horticultural crops in this case study) are fundamentally and seamlessly
intertwined.
285
Figure 5.4: Percentage exceedance plots of total duty flow for pumps for demand-based and supply based irrigation (top plot: drip system, bottom plot: sprinkler system)
The total energy output, expressed as kWh/ha, for a given scenario is
calculated by converting yield of each crop into the equivalent energy
quantity using the conversion factors given in Chapter 3 and then summing
up the equivalent energy of the three crops. The total energy output
expressed as equivalent kilowatt hours for each crop under each scenario is
listed in Table 5.7. The energy sequestered in crop output for demand-based
irrigation regimes is higher than that of supply-based irrigation regimes
Table 5.7: Total equivalent energy output (kWh/ha) from each crop for supply-based and demand-based irrigation scenarios
Irrigation regime
Crop Scenario
1 (kWh/ha)
Scenario 2
(kWh/ha)
Scenario 3 (Scenario 5 for demand
based) (kWh/ha)
Scenario 4 (Scenario 6 for demand
based) (kWh/ha)
Demand based
Citrus 18,550 21,200 23,320 25,440 Stone fruit
10,980 11,590 12,810 15,250
Wine grapes
65,600 72,160 75,440 85,280
Total 95,130 104,950 111,570 125,970
Supply based
Citrus 9,540 12,084 11,024 8,745 Stone fruit
5,368 10,004 10,187 12,871
Wine grapes
45,592 69,864 68,880 79,376
Total 60,500 91,952 90,091 100,992
5.5.2 Energy efficiency and energy productivity indicators
The energy efficiency and productivity indicators are defined in Chapter 3
and discussed in detail in Chapter 4 for demand-based irrigation. Therefore,
the mathematical equations representing those indicators are not repeated
for the current chapter.
The indicators given in Table 5.8 are computed using modelled results for
supply-based scenarios. The energy efficiency of gravity irrigation scenarios
is higher than that of pressurized irrigation as net use of energy is higher in
the latter case due to extra energy required for irrigation water pumping
through the supply pipes despite application of lesser volume of irrigation
than the gravity based scenarios. Among the pressurized irrigation
scenarios, the drip system is more energy efficient than the sprinkler system.
This is mainly due to higher irrigation efficiency of the drip system than that
of the sprinkler system connected with a similar pressurized pipe supply
system.
287
Table 5.8: Energy indicators for supply-based irrigation scenarios
Indicator Crop Scenario
1 Scenario
2 Scenario
3 Scenario
4
Energy efficiency
(kWh/kWh)
Citrus 2.40 2.74 2.05 2.00
Stone fruit
1.58 1.35 1.41 1.59
Wine grapes
13.86 12.85 11.10 12.18
Average 5.95 5.65 4.85 5.26
Energy productivity (kg/kWh)
Citrus 4.54 5.17 3.87 3.78
Stone fruit
2.59 2.21 2.30 2.61
Wine grapes
4.23 3.92 3.38 3.71
Average 3.79 3.77 3.18 3.37
Specific energy
(kWh/kg)
Citrus 0.22 0.19 0.26 0.26
Stone fruit
0.39 0.45 0.43 0.38
Wine grapes
0.24 0.26 0.30 0.27
Average 0.28 0.30 0.33 0.30
Water – energy
productivity (g/m3/kWh)
Citrus 0.73 0.91 0.80 1.19
Stone fruit
0.41 0.25 0.32 0.39
Wine grapes
0.78 0.65 0.61 0.84
Average 0.64 0.60 0.58 0.81
Water – energy ratio (kWh/kWh)
Citrus 0 0 0.22 0.20
Stone fruit
0 0 0.25 0.22
Wine grapes
0 0 0.22 0.18
Average 0 0 0.23 0.20
288
Similarly, the energy productivity of the pressurized irrigation systems
(Scenario 3 and Scenario 4) is less than that of the gravity based irrigation
systems (Scenario 1 and Scenario 2) due the reasons explained above. As
given in Table 5.8, more energy is required to produce a kilogram of
produce using the pressurized irrigation systems than that of gravity based
irrigation systems. Again, the drip irrigation scenario generally used less
energy to produce a unit output among the pressurized irrigation systems.
So far we have compared irrigation systems based on energy input and
energy output only. This approach, which is solely energy based, ranked
gravity based system to be more favoured than the pressurized/modern
irrigation systems. However, the picture is not complete unless water use is
also compared for the four supply-based scenarios. The “water-energy
productivity” and “energy ratio” are two indicators which represent a
holistic and system wide perspective by considering both water and energy.
The cropping scenario using a drip irrigation system demonstrates the
highest value of 0.81 g/m3/kWh for the water-energy productivity indicator
which shows that more is produced from a given amount of water and
energy input for the drip system as compared to the other scenarios. It is
also revealed from Table 5.8 (energy ratio) that the energy consumed in
irrigation operations for crops with drip system ranges from 18 % to 22 %
of total input energy as compared to 22 % to 25 % of total input energy for
the sprinkler system. Therefore, holistically the horticulture crops irrigated
with a drip irrigation system which is connected with pressurized pipe
supply operated under a supply-based irrigation strategy have the least water
and energy footprint among the scenarios discussed in this chapter.
The water and energy indicators for the demand-based scenarios of Chapter
4 are also revisited here for comparison. The overall energy efficiency of
pressurized irrigation scenarios is marginally higher than that of gravity
irrigation due to the fact that, despite no pumping operations, the total use of
energy is higher in the latter case due to extra energy input in the form of
higher fertilizer application rates etcetera required to overcome nutrient
losses through leaching and surface runoff.
289
The comparison of energy efficiency for pressurized irrigation scenarios for
demand-based (Chapter 4) and supply-based (Chapter 5) settings leads to
some interesting findings. The energy efficiency of the supply-based
settings is lower than that of demand-based ones despite lesser irrigation
pumping and hence lesser energy input for the former case. This is in fact a
result of significant reduction (up to 66%) in yield (energy output) for the
pressurized irrigation scenarios for the supply-based irrigation regime.
However, for the holistic overview, the combined water and energy
productivity indicator, the water-energy productivity, should be compared.
The values of this indicator are higher for supply-based scenarios than that
of demand-based scenarios. This discussion concludes that supply-based
scenarios are less favored when only energy aspects are considered. But a
contrasting conclusion is reached when a holistic approach is adopted and
both the water and energy aspects are analysed. Similarly, the energy ratio,
which refers to the fraction of total input energy that is expended in irrigation
operations, is higher for the supply-based pressurized irrigation than that of
corresponding demand-based ones. The reason for this unexpected behavior can be
explained by plots of duty flow rates in Figure 5.3 and Figure 5.4, which indicate
that supply-based irrigation systems frequently involve significantly higher flow
rates through the pipe network thus consuming higher energy in irrigation pumping
operations.
Looking at the water-energy productivity indicator values for individual crops in
Table 5.8, it is evident that citrus under a drip irrigation system has the least water
and energy footprint and highest water-energy productivity; followed by wine
grapes and then stone fruit. Wine grapes under a drip system have the least water-
energy ratio followed by citrus and stone fruit, indicating that irrigation pumping
for wine grapes results in the lowest water and energy footprint while, in contrast,
stone fruit irrigation both under drip and sprinkler systems results in higher water
and energy impacts.
5.5.3 Comparison of greenhouse gas emissions
For supply-based scenarios, greenhouse gas emissions (GHG) data for
individual energy inputs for each crop is not available. The only information
290
which is modelled energy use in irrigation pumping (where applicable) is
available. Therefore, it is assumed that the GHG emissions, expressed as
kilograms of equivalent CO2 per hectare (kgCO2-eq/ha), resulting from
energy use in crop production for the scenarios discussed in this chapter, are
a factor of the ratio of the energy use for the supply-based and demand-
based irrigation scenarios. The resulting GHG emissions values for each
crop under each scenario are given in Table 5.9. The GHG emission values
are calculated from data in Table 5.6 and the GHG emission values
mentioned in Chapter 4 for Scenarios 1, 2, 5 and 6. The GHG emissions
values for the demand-based scenarios from Chapter 4 are also given in
Table 5.9 for ready reference.
Table 5.9: Greenhouse gas emissions rates (kgCO2-Eq/ha) for the three crops under demand-based and supply-based (computed) scenarios
Irrigation regime
Crop GHG
emissions from
Scenario 1
(kgCO2-Eq/ha)
Scenario 2
(kgCO2-Eq/ha)
Scenario 3
(Scenario 5 for
demand based)
(kgCO2-Eq/ha)
Scenario 4
(Scenario 6 for
demand based)
(kgCO2-Eq/ha)
Demand based
Citrus
Crop productio
n 1832.74 1820.66 2996.7 2727.6
Irrigation pumping
0 0 1373.36 1103.28
Stone fruit
Crop productio
n 1634.52 1667.51 2806.00 2487.5
Irrigation pumping
0 0 1395.92 1109.04
Wine grape
s
Crop productio
n 1532.5 1515.77 2305.8 2405.9
Irrigation pumping
0 0 1029.07 1105.23
Supply based
Citrus Crop
productio920.8 1031.0 1801.7 1385.6
291
n Irrigation pumping
0 0 1084.3 777.4
Stone fruit
Crop productio
n 770.8 1667.5 2474.1 2609.9
Irrigation pumping
0 0 1609.7 1616.0
Wine grape
s
Crop productio
n 749.0 1242.6 1988.7 2229.4
Irrigation pumping
0 0 1229.0 1080.5
Based on data in Table 5.9, it would be right to say that GHG emissions
from pressurized irrigation scenarios are higher than the gravity based
system mainly due to the fact that pressurized irrigation systems are more
energy intensive. Among the pressurized irrigation supply-based scenarios,
citrus has the least carbon footprint both from crop production and irrigation
operations.
5.6 Sensitivity analysis of pressurized irrigation scenarios
Sensitivity analysis is an effective tool to gauge the response of the
dependent variables of a developed model to a change in an independent
variable within a specified feasible range. It also provides modellers a lead
to identifying the most prominent variables in models that involve many
independent variables. The sensitivity of water use, pumping energy use and
crop yield to irrigation interval for sprinkler and drip system for a supply-
based model developed in Vensim is presented in this section. The reason
for choosing only pressurized irrigation scenarios is that they seem to be
more responsive in terms of changes to water and energy footprints.
5.6.1 Sensitivity of irrigation supply, pumping energy and yield
to irrigation interval
The reason for choosing irrigation interval is the observation that it is an
operational variable that seems to be varying in the model as well as in the
292
field both for supply-based and demand-based irrigation regimes and seems
to have significant impact on water use, energy use and yield. The irrigation
interval is varied by ±3 days from its default value of 7 days with the change
of ±1 day. For this purpose, “vector” distribution in Vensim’s sensitivity
analysis module is selected. The model selects a value of irrigation interval
and keeps it constant for the one complete simulation.
The sensitivity plots of cumulative irrigation supply (ML) to the irrigation
interval for sprinkler and drip irrigation can be compared from Figure 5.5.
The blue line shows the cumulative irrigation for the default value of
irrigation interval which is 7 days. A common observation from the two
plots is that irrigation supply response is not linearly proportional to
increase or decrease in irrigation interval from its default value. The
response to decrease in irrigation interval is more prominent than the
increase. This indicates that lower irrigation intervals are more suited to the
crop-soil combination of this study area. Moreover, irrigation use is almost
doubled for drip system and less than double for sprinkler system when the
irrigation interval is reduced from 10 days to 3 days. This shows that drip
irrigation system required more frequent irrigation. This is mainly due to
lower soil-water storage for drip irrigated crops than that of the sprinkler
system.
293
Figure 5.5: Sensitivity graphs of cumulative irrigation supply (ML) for sprinkler (top) and drip (bottom) systems to irrigation interval (days)
The sensitivity plots of cumulative irrigation pumping energy use (kWh) to
the irrigation interval for sprinkler and drip irrigation are given in Figure
5.6. The blue line shows the cumulative pumping energy use for the default
value of irrigation interval which is 4 days. The general trends for sensitivity
of energy use and irrigation use are the same. However, the range of
variation of pumping energy for sprinkler system is wider than that of drip
system. The pumping energy use for sprinkler system is increased by
213,427 kWh in response to an increase of 1017 ML in irrigation use (i.e.
210 kWh/ML increase in irrigation volume) while the pumping energy use
for the drip system is increased by 201,945 kWh in response to an increase
of 782 ML in irrigation use (i.e. 258 kWh/ML increase in irrigation
Sprinkler-supply-based50% 75% 95% 100%
Cumulative_Irrigation_Supplied4,000
3,200
2,400
1,600
800
01 92 183 274 365
Time (Day)
Drip-supply-based-sen50% 75% 95% 100%
Cumulative_Irrigation_Supplied2,000
1,600
1,200
800
400
01 92 183 274 365
Time (Day)
294
volume). The reason is most likely due to a higher operating hydraulic
pressure head (32 m) being maintained in the case of drip irrigation system
as compared to that of sprinkler system (25 m).
Figure 5.6: Sensitivity graphs of cumulative pumping energy use (kWh) for sprinkler (top) and drip (bottom) to irrigation interval (days)
Sensitivity graphs of yield (t/ha) of citrus (first row), stone fruit (second
row) and wine grapes (third row) for sprinkler system and drip system to
irrigation interval are given in Figure 5.7. The blue line on each plot shows
the yield of default irrigation interval of seven days. These plots show that
the yield of citrus crops is more sensitive to irrigation interval; hence to total
irrigation volume, and water shortage events, while wine grapes are
comparatively least sensitive.
295
Figure 5.7: Sensitivity graphs of yield (t/ha) for sprinkler (left) and drip (right) to irrigation interval (days) for the three crops
5.6.2 Sensitivity of crop yield and energy use to irrigation water
use
This sub-section explores sensitivity of energy use in irrigation pumping
and that of crop yield to various levels of irrigation water use on a per
hectare basis. This analysis is indirectly covered in the previous sub-section
as level of water use is linked with length of irrigation interval, but here it is
aimed to explore it in further detail.
To get a variety of model responses the irrigation interval (proxy of total
water use) is varied by a wider range of ±5 days from its default value of 7
days with the change of ±1 day, which indirectly changes the volume of
irrigation water pumped and the corresponding energy consumed. A new
model run is performed for each value of the irrigation interval. Hence a
296
total of ten supply-based model runs are performed for each of the drip and
the sprinkler systems. Then the model results are post-processed to compute
energy use in kWh/ML for each of the three crops. The model outputs water
use in ML/ha and yield in t/ha for each crop. The total water use increases
with decreasing irrigation interval and vice versa and so does the water use
per hectare.
Although, it is obvious that energy use and yield will increase with increase
in water use within a certain range, this analysis focuses on determining the
nature of this relationship as it can be a steep/flat linear or non-linear
relationship.
5.6.2.1 For drip irrigation system
Figure 5.8 shows the scatter plots and the polynomial models fitted between
water use rates (ML/ha) and the corresponding yield in tons/ha for each of
the three crops. A second degree polynomial fits well to the scatter data. It is
evident from Figure 5.8 that for wine grapes and stonefruit, there is no
further improvement in yield for irrigation rates higher than 5.9 ML/ha and
9.75 ML/ha, respectively. In case of citrus, it can be said that the crop yield
responds almost linearly for the given range of irrigation application rates.
However, similar to other crops, this response is likely to diminish for
higher irrigation application rates. Hence, it can be concluded that a higher
gain in yield can be achieved with a little increase in irrigation use for the
lower range of irrigation application rates. A similar approach is adopted by
Khan et al. (2005a) and Khan and Abbas (2007) to optimize crop yield and
water use in the Murrumbidgee Irrigation area.
297
Figure 5.8: Sensitivity of crop yield to irrigation water use for the three modelled crops with drip system
The modelled irrigation system is based on use of a communal line for
irrigation water supply; it does not have separate supply lines for each crop,
therefore the model is only capable of computing overall daily energy use in
irrigation pumping for the modelled area and cannot track down energy use
in irrigation pumping for supply to individual farms or crops. The energy
use per unit irrigated area (kWh/ha) is calculated by dividing the cumulative
y = 0.078x2 + 9.240x ‐ 13.607R² = 1.000
0.0
10.0
20.0
30.0
40.0
50.0
60.0
0 1 2 3 4 5 6 7
Crop
Yield (t/ha)
Water use (ML/ha)
Sensitivity of yield to water use (Citrus)
y = ‐1.020x2 + 12.368x ‐ 11.358R² = 0.982
0.0
5.0
10.0
15.0
20.0
25.0
30.0
0 1 2 3 4 5 6 7
Crop
Yield (t/ha)
Water use (ML/ha)
Sensitivity of yield to water use (Wine Grapes)
y = ‐0.310x2 + 6.251x ‐ 6.424R² = 0.988
0.0
5.0
10.0
15.0
20.0
25.0
30.0
0 2 4 6 8 10 12
Crop
Yield (t/ha)
Water use (ML/ha)
Sensitivity of yield to water use (Stonefruits)
298
energy use at system level (model output) by the total modelled irrigated
area with the assumption that energy use to pump a unit volume of water is
always the same regardless of which crop is irrigated with that water. This
assumption is valid only if the pipe supply network is not very large and the
peak irrigation demand for one crop is not significantly higher than the other
crops.
Figure 5.9 shows the scatter plot between modelled irrigation pumping
energy use (kWh/ha) and the water use (ML/ha) for a range of irrigation
intervals. The energy use varies from 600 kWh/ha to 1545 kWh/ha when
irrigation rate is varied from 2 ML/ha to 7 ML/ha. A second degree
polynomial fits best to the formulate relation between the two variables. It
should be noted that the plot in Figure 5.9 is based on total water and energy
use at a modelled system level and the relationship for individual crops may
vary from the one presented here, however the general trend should not vary
much.
Figure 5.9: Sensitivity of irrigation pumping energy consumption (kWh/ha) to irrigation water use (ML/ha) for the three modelled crops with drip system
5.6.2.2 For sprinkler irrigation system
The same node-link model with the irrigation system changed to a sprinkler
system was executed with the irrigation interval varied between default
y = ‐26.912x2 + 456.056x ‐ 288.288R² = 0.994
0.00
200.00
400.00
600.00
800.00
1000.00
1200.00
1400.00
1600.00
1800.00
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00
Energy Use (kW
h/ha)
Water Use (ML/ha)
299
irrigation intervals of 7 ± 5 days. For each run, the irrigation interval was
increased or decreased by 1 day to impose an irrigation interval of a
maximum of 12 days and minimum of 2 days. The modelled irrigation use
(ML/ha), crop yield (t/ha) and pumping energy use (kWh/ha) were recorded
for each run. The scatter plots between different levels of modelled water
use (ML/ha) and corresponding yield for sprinkler system are shown in
Figure 5.10. The response of yield to a change in irrigation rate is very
similar to the ones for the drip system i.e. the marginal increase in crop
yield diminishes with increasing amount of water application. The water use
rate for sprinkler system varies over a wider range than that of drip system
due to higher irrigation demand to replenish a larger wetted area as
compared to drip irrigation.
Table 5.10: Comparison of drip and sprinkler system in terms of yield response to water use
6.1.2 Limitations regarding up-scaling water and energy use
As mentioned earlier the main objective of this chapter is to estimate water
and energy use by horticulture over the whole MIA irrigated area.
Generally, the up-scaling procedure involves determination of properties for
a unit area of a given soil type which are then linearly extrapolated to the
entire area. This approach is quite suitable for properties like water use rate,
water savings or crop yield which are mainly dependent on a single
parameter i.e. soil type.
In the current study we have also included a rather more complex variable;
the energy use in irrigation pumping and application. In fact, the case of
energy use for pressurized irrigation pumping is totally different from other
quantities. It cannot be linearly extrapolated from unit area to the larger area
of interest. The reason behind this up-scaling limitation on pumping energy
use is the physical reality that it is non-linearly related to the size of the
system (pipe lengths and diameters), pipe gradient, and to the flow rate
inside an irrigation supply system. This limitation is demonstrated by
running the supply-based node-link model with piped irrigation supply and
with total irrigated area increased to different levels and then noting the total
329
energy consumption to irrigate that area with drip irrigation system while
keeping the size of the pipe supply network unchanged. The results are
summarized in Table 6.2. It is evident from Table 6.2 that the energy use is
increased by 75 per cent with 50 per cent increase in irrigated area and by
177 per cent with 100 per cent increase in irrigated area. Furthermore,
Figure 6.3 shows that water use is increased and decreased by the same
proportions (equal distance from either side of blue line) as that of the
irrigated area. On the other hand, the increase in cumulative pumping
energy consumption is significantly higher than its decrease for the ±50%
variation in irrigated area, as shown in Figure 6.4. The water and energy
sensitivity plots given in Figure 6.3 and Figure 6.4, respectively, are based
on 500 simulation runs for each case.
Figure 6.3: Sensitivity of cumulative water use (ML) when the irrigated area is uniformly varied by ±50% for the node-link model with a drip system (blue line refers to the baseline value)
Drip_demand_based_sensitivity_run50% 75% 95% 100%
Cumulative_Irrigation_Supplied4,000
3,200
2,400
1,600
800
01 92 183 274 365
Time (Day)
330
Figure 6.4: Sensitivity of cumulative energy use (KWh) when the irrigated area is uniformly varied by ±50% for the node-link model with a drip system (blue line refers to the baseline value)
Table 6.2: Comparison of increase in pumping energy use with increase in total irrigated area for a supply-based drip irrigation strategy
Area Case Total pumping energy
consumed (kWh) Change in energy use
(±% of baseline) Baseline area 321,362 0 Area increased by 50 per cent
563,833 +75%
Area decreased by 50 per cent
600,285 142,689
-56%
Area increased by 100 per cent
892,944 +177%
The above discussion concludes that it will be highly inaccurate to compute
energy use at unit area (1 hectare) and uniformly upscale it to the entire
area. Similarly, it is inappropriate to extend the model domain to cover the
entire horticultural area of MIA by a single connected network of supply
pipes as it is practically impossible. To address this limitation the concept of
a “representative unit” was introduced and the whole MIA horticulture area
is assumed to be a mosaic of representative units. A representative unit here
is defined as a grouping of 300 hectares consisting of ten irrigated farms all
having the same size (30 ha), the same soil type/soil group and growing the
three horticultural crops distributed in the same proportion as to that of the
Drip_demand_based_sensitivity_run50% 75% 95% 100%
Cumulative_Energy_Use600,000
480,000
360,000
240,000
120,000
01 92 183 274 365
Time (Day)
331
three crops distribution with that soil type over the entire MIA horticultural
area. The node-link model was setup with the representative unit and run
with each set of soil types and the values of the parameters to be up-scaled
were recorded.
As shown in Figure 6.1 and Figure 6.2, most of the farms with the same soil
type/group are co-located therefore, the assumption of using the same soil
type for all farms in the representative unit is considered to be an
appropriate one. To find the relative distribution of the crop area for the
representative unit the attributes data of the GIS map in Figure 6.1 was
analysed. The attributes include soil type, crop name and crop area for each
horticultural farm in MIA as per year 2007-08. Table 6.3 provides a
summary of the analysis of this attribute data. It gives the area (both
hectares and percentage) of each soil type in MIA’s horticultural zones. The
table also provides the area of each of the three horticultural crops as
percentages of the total area of a given soil type. For example, Table 6.3
shows that citrus are grown at 42 per cent of the area under clay loam.
Similarly, for wine grapes and stone fruits (all other fruits) grown on clay
loam, the percentage area is 54% and 4%, respectively. It should be noted in
Table 6.3 that about 84% of the soils of horticultural farms are some sort of
clayey soils. Medium clay is the most common (34%) soil type followed by
light clay (18%) and light medium clay (11%) and then all others in the
MIA horticultural soils. This further supports the assumption of using the
same soil type for all ten farms of the representative unit.
Table 6.3: Distribution of different soil classes in MIA horticultural farms and percentage distribution of area of each horticultural crop for each soil type
Given these 15 soil types, the node-link model would have to run for times
respectively for each irrigation method thus generating excessive
information. Since we have already sorted those 15 soil types into five soil
groups as given in Table 6.1, only five model runs were produced i.e. one
set of results for each soil group. As shown in Table 6.4, 43 per cent of the
horticultural area in MIA is transitional red-brown earths followed by non
self-mulching clays at 29 per cent. The least area is covered with sandy soils
at 3 per cent.
Table 6.4: Distribution of different soil groups in MIA horticultural farms and percentage distribution of area of each horticultural crop for each soil group
Soil Group Area (ha)
Area as %age of total area
Citrus area (%)
Wine grapes
area (%)
Stone fruit area (%)
Self-mulching clay
3953 14 40 42 18
Non self-mulching clay
8395 29 21 76 3
Transitional red-brown earths
12457 43 24 69 6
Red-brown earths
3224 11 46 53 1
Sandy soils 941 3 58 46 1
333
6.2 Node-link model run for representative area unit
As mentioned earlier the node-link model is based on the same framework
which is discussed in previous chapters. The physical layout of the irrigation
supply system which is a branched piped supply system as represented by
the model is also unchanged. Running the node-link model at the
representative area scale to determine water-energy use is the first step. For
each of the two irrigation methods (i.e. sprinkler and drip), five model runs
(instead of 15 runs) were repeated constituted by one model run for each of
the five soils groups using the relative proportions of the three horticultural
crop areas as given in Table 6.4 over the representative unit. For a given
irrigation method, one model run differs from the other only by its soil type.
Since the total area of the representative unit (300 ha) is almost same as the
one used for the case study (290.97 ha) in the previous chapters, no change
in length, diameters or material of the supply pipes was assumed for the
model runs discussed here.
Table 6.5: Distribution of number of farms in the representative unit for each model run using a given soil group and crop area (in parentheses, ha)
Soil Group No. of citrus
farms No. of wine
grapes farms No. of stone fruit farms
Self-mulching clay 4
(120) 4
(120) 2
(60)
Non self-mulching clay
2 (60)
8 (240)
0 (0)
Transitional red-brown earths
2 (60)
7 (210)
1 (30)
Red-brown earths 5
(150) 5
(150) 0
(0)
Sandy soils 6
(180) 4
(120) 0
(0)
The number of 30 ha farms each growing one of the three crops for each soil
type model run using the representative area are given in Table 6.5. The
number of farms is rounded up to the closest whole number; therefore, there
are no stonefruit farms on some of the soil group due to their very small area
as compared to the other two crops. The total irrigated area for each soil
334
group is summed to be 300 hectares. A depletion factor of 80% was used for
all model runs i.e. an irrigation order is placed and irrigation delivered when
and if soil-water is depleted more than 80% of the readily available soil
moisture for the given soil group to a given farm.
6.3 Up-scaling the model results using mosaic approach
The node-link model set up as per the abovementioned configuration was
run five times; one run for each of the five soil groups under sprinkler
irrigation system with piped supply and then same number of model runs
were repeated for the drip irrigation system. Since it is a demand-based
model, both the irrigation quantity and irrigation delivery are regulated by
the crop-water demand by keeping a continuous account of the soil-water
depletion. It also assumes no constraints on availability of irrigation water.
6.3.1 Water and energy use at representative area unit scale
First, the model computes the water use per hectare for each crop and the
total water use for the entire model area i.e. 300 hectares and total energy
consumed in pumping and delivering this water. Then the modelled water
use and energy use are up-scaled to the entire horticultural area of MIA. The
results of the first step i.e. water and energy uses for the representative unit
for the sprinkler irrigation case are given in Table 6.6 and those for the drip
irrigation are given in Table 6.7. The energy use reported in Table 6.6 and
Table 6.7 is the total energy consumed in irrigation pumping for the
sprinkler and drip irrigation systems, respectively for the modelled
representative area and can be converted into values for individual crops
using the water use proportions method applied in previous chapters. It is
worth noting from the these two tables for sprinkler and drip system that
energy use for irrigation pumping and delivery for sandy soils is not highest
among the five soil groups despite the fact that the water volume pumped is
the highest for the sandy soils. The most commanding reason for this is that
sandy soils have the lowest water holding capacity among the five soil
groups and hence water is depleted relatively quickly in sandy soils thus
requiring more frequent irrigation but in lesser quantity.
335
As mentioned in the beginning of this chapter, energy use for irrigation
pumping is very sensitive to instantaneous flow rates through the supply
pipes. Therefore, less energy is consumed in irrigating sandy soils due to
smaller flow rates required. As expected, the total water use and energy use
are higher for the sprinkler irrigation system than that of the drip irrigation
system.
336
Table 6.6: Water and pumping energy uses for different soil groups with sprinkler irrigation system as computed by the demand-based node-link model for the representative unit area (300 ha) for 2007-08
Water use (ML/ha) Total Water Use for Representative Unit (ML/300ha)Total water
Table 6.7: Water and pumping energy uses for different soil groups with drip irrigation system as computed by the demand-based node-link model for the representative unit area (300 ha) for 2007-08
Water use (ML/ha) Total Water Use for Representative Unit (ML/300ha)Total water
The next step is to up-scale modelled water use from the representative unit
to the whole horticultural area of MIA. The procedure for up-scaling water
use is depicted by Equation 6.1. The water use amount computed by using
Equation 6.1 for each of the five soil groups is added up to get the total
water use for the entire MIA horticultural area irrigated with a given
irrigation method. The same procedure is followed for the second irrigation
method. In Equation 6.1, only the quantity is computed by the model.
Equation 6.1
Where,
i is one of the five soil groups listed in Table 6.5; is the total water use
for the MIA horticultural area with the soil group i; is the total MIA
horticultural area with the soil group i; is the modelled total water for the
representative unit area which has soil group i; and is the size of the
representative unit used in the model i.e. 300 hectares.
The total horticultural area in MIA is 28,970 ha while that of the model is
300 ha. Therefore, at least a total of 97 pumping stations will have to be
established to cover the entire MIA horticultural area with each station
servicing its command area of 300 hectares.
A formula similar to Equation 6.1 is applied for each soil type to up-scale
the modelled energy use in pumping irrigation water and in operating the
pressurized irrigation delivery systems from water source to each farm, for
the entire horticultural area of MIA. The formula used for energy up-scaling
is given in Equation 6.2. The requirement for the mosaic approach for
estimating pumping energy at the MIA scale by using the representative unit
area has been mentioned in the previous section. The sole purpose of this
approach is to avoid over-estimates of pumping energy consumption.
Equation 6.2
Where,
339
is the total pumping energy used in the MIA horticultural area with the
soil group i; and is the modelled energy consumed in irrigation pumping
for the representative unit area with the same soil group i.
The data for the variables used in Equation 6.1 and the total up-scaled water
use estimated for each soil group with horticultural crops in MIA are given
in Table 6.8. The modelled water use given in Table 6.8 is the output of the
model as mentioned above for each soil group with all farms irrigated with
sprinkler system as well as for the case of all farms irrigated with drip
system for the representative unit area of 300 ha. There are zero conveyance
losses as water is transmitted through pipes and the only losses constituting
the water use are on-farm water losses.
The up-scaled water use for each soil group is the result of Equation 6.1.
The total water use for each irrigation method, which is assumed to be
applied to the entire horticultural area of MIA (28,970 ha), is the sum of that
of the five soil types and is given on the last row of Table 6.8. The total
water use at MIA horticultural area scale is 23% higher for sprinkler system
than that of the drip system. As derived from Table 6.8, regardless of crop
type, the water use per hectare for sprinkler system is 6.1 ML/ha and that for
drip system is 4.7 ML/ha for the overall horticultural area of MIA. Hence,
water savings of 1.4 ML/ha/year can be achieved by converting all the MIA
horticultural area from sprinkler to drip system (both connected with
pressurized irrigation supply system from the water source).
To estimate the potential water savings, the node-link model was also run
with flood irrigation system for each soil type. The water use estimates for
the modelled area and the up-scaled values for the entire MIA under flood
irrigation are given in Table 6.8. It is evident from these results that the total
water use for MIA horticultural area with flood irrigation is 1.7 times higher
and 2.2 times higher than that for the sprinkler and drip irrigation systems,
respectively.
340
Table 6.8: Water use for sprinkler and drip irrigation system for five soil groups at model scale and at the scale of the MIA irrigated horticulture area
The total energy consumed to pump water at the communal pumping station
and then conveying it to individual farms via pressurized pipes for the
representative area and for the whole MIA area with horticultural crops, for
each of the five soil groups over the whole year, is given in Table 6.9.
Table 6.9: Energy use in irrigation pumping and conveying for sprinkler and drip irrigation system for five soil groups at model scale and at the scale of the MIA irrigated horticulture area
A comparison of yields of citrus and wine grapes for three irrigation
systems is given in Table 7.1. These yield values have been used in various
analyses in this thesis. The yield for citrus varies significantly depending on
the variety of a crop as given in Table 7.1. For water and energy analysis for
citrus in previous chapters, only the highest yield values are used.
Drivers for changing production practices to hi-tech include decline in
reliability and availability of irrigation allocation due to climate change,
high water buyback prices and high water trade prices supported by high
water trade demands during dry periods and finally, the potential
improvement in crop yield warrants the use of water efficient irrigation
technologies. This chapter is dedicated to the economic analysis of water
efficient irrigation technologies in the context of horticulture crops.
Irrigation water conveyance losses (seepage and evaporation) can also be
mitigated by using piped supply from canal/source to the individual farms.
7.2 Representative node-link model
The node-link model developed in Chapter 4 for the demand-based
irrigation system represents a horticultural area of around 300 hectares. The
integrated irrigation supply system (pumping station, pipe network, farm
delivery outlets, filtration system, computer control system etc.) for the case
study area is originally designed to irrigate a horticultural area of up to 550
hectares. However, only 13 farms with a total area of around 300 hectares
were connected to the system at the time of this study. To perform a credible
financial analysis of the system the whole 550 hectares were essentially
assumed connected and supplied with on-demand irrigation from the
integrated irrigation supply system. For this purpose, the node-link model
that was used in Chapter 4 was extended to represent the whole 550 hectares
consisting of citrus, stone fruit and wine grapes.
7.2.1 Modelled water and energy use
Table 7.2 summarizes the key results in terms of water use and energy use
for the node-link model representing the 550 hectare area for furrow,
365
sprinkler and drip irrigation systems. The furrow irrigation on each farm is
setup to be supplied with open channels while pressurized pipe supply from
a central pumping station to each farm is setup in the model for both
sprinkler and drip systems. The water and energy use results of the model
for an area of 550 hectares are summarized in Table 7.2 and will be used in
the benefit-cost analysis later on in this chapter.
Table 7.2: Node-link model output for a modelled area of 550 ha
Irrigation system
Crop
Total water used (ML)
Water use
(ML/ha)
Pumping energy use (KWh/ha)
Irrigation hours
(hr/ha)
Pumping energy
use (KW/ha)
Furrow Citrus 5459 9.93 0 40 0
Wine grapes
4084 7.43 0 40 0
Low-head sprinkler
Citrus 4451 8.09 1996 46 43.4
Wine grapes
3321 6.04 1489 44 33.8
Drip Citrus 3443 6.26 1664 60 27.7
Wine grapes
2621 4.77 1266 62 20.4
7.3 Capital cost for conversion to pressurized irrigation system
Conversion of an irrigation system requires high capital investment and
ongoing operating costs. Therefore, economic feasibility of such projects
has to be performed to justify the investment. As mentioned earlier in this
chapter, the financial profitability and economic viability of the conversion
from gravity based irrigation (furrow) to pressurized irrigation (sprinkler or
drip) is analysed by using benefit-cost analysis and net present value method
in the context of horticulture crops including citrus and wine grapes. The
analysis also includes replacement of open-channel supply system with
pressurized piped supply. Furrow irrigation is assumed as a
baseline/benchmark case in the economic analyses. This section only deals
366
with capital investment which is initially required for setting up an irrigation
system.
7.3.1 Assumptions for the economic analysis
A number of assumptions were made to carry out the economic analysis of
conversion from furrow to sprinkler or drip system. The assumptions which
are common between the three irrigation systems in the horticultural area of
MIA are listed in Table 7.3. The total area of the representative unit was
taken as 550 ha; however different cost items (capital or operational costs)
were computed on a per hectare basis.
Table 7.3: Assumed values of various parameters for economic analysis
Item Value Comments/source Irrigated crop area (ha) 550 Modelled area Water usage charges ($/ML) 8.67 MIA website Irrigation facilities charges ($/ML) 19.84 MIA website Landholding charges ($/ha) 3.48 MIA website Average temporary water trade price ($/ML)
7.4.5 Discounted payback period and financial viability of the
three irrigation systems for citrus
The irrigation systems namely furrow, sprinkler, and drip system; to be
analysed in this chapter have a working life of at least 30 years and impose
high initial investment costs, especially the sprinkler and drip systems.
Therefore the financial analyses of these systems are conducted based on the
concept of time value of money which essentially implies that for any given
amount, the value of future money is less than its present value. Therefore,
the future potential returns from the system are converted to their present
value (PV) using a discount rate of 10 per cent. The Net Present Value
(NPV) is defined as the difference between present value of cash
inflows/income/benefits (PVB) and the present value of cash outflows/costs,
including initial capital costs (PVC). The NPV approach is applied over the
life of the irrigation system(s) to determine the discounted payback period
(break even period on capital investment) and the overall value/profitability
of the system under consideration. The discounted payback period is defined
as the number of years by which the PVB equals the initial capital
investment.
377
The formula for NPV is given in Equation 7.1.
∑ Equation 7.1
Where,
n, is the total number of years of useful life of the system; Rt, is the net
return for a given year ‘t’, and r, is the discount/interest rate.
In the NPV calculation for the current study, the annual costs and annual
returns are assumed to be constant over an analysis period of 30 years for a
given irrigation system with a given horticultural crop. The NPV during
initial years is likely to be a negative value due to high initial costs but
improves each successive year. The number of years by which the NPV
becomes zero corresponds to the discounted pay-back period.
Another profitability indicator called benefit cost ratio (B-C ratio) is also
computed by dividing present value of benefits (PVB) by present value of
costs (PVC). A B-C ratio of greater than 1 indicates a profitable project.
A summary of initial investments, annual operating costs and annual returns
for the three irrigation systems with citrus is given in Table 7.13. It includes
figures on a per hectare basis as well as total for the case study area of 550
ha. The returns from water savings of 22 ML/year from channel seepage and
evaporation loss by piped supply system are also included in Table 7.13.
Table 7.13: Summary of initial and annual costs and annual returns for the three irrigation systems for citrus Furrow Item $/ha Total for 550 haInitial capital cost
Irrigation Supply System 0 0Irrigation System 2,200 1,210,000
Crop production 9,200 5,060,000Water Trade 353 194,183
Water saving by pipe supply 5,588 Total returns 5,259,771 Drip Initial capital cost
Irrigation Supply System 2,216 1,218,713Irrigation System 7,100 3,905,000
Total initial cost 5,123,713 Operating Costs Annual operating cost 3,161 1,738,586 Total returns
Crop production 10,400 5,720,000Water Trade 676 371,602
Water saving by pipe supply 5,588 Total returns 6,097,190
388
Figure 7.8 shows the plots of NPV values of the drip, sprinkler and furrow
irrigation systems over a period of 30 years. The drip and sprinkler systems
are connected with central integrated irrigation supply system for irrigating
550 ha of wine grapes. The capital cost of installing the integrated irrigation
supply system is also taken into account in the analyses. Figure 7.8 indicates
that NPV for both the drip system and sprinkler system turns positive at the
end of the second year of operation indicating that the system is paid off in
just two years as compared to the same irrigation system for citrus which
took up to 18 years to return a positive NPV. Unlike sprinkler and drip
systems the furrow irrigation with wine grapes returns a positive NPV at the
end of very first year of its operation. However, the furrow with citrus is
found not to be so profitable. Moreover, as shown in Figure 7.8, the NPV of
returns from drip exceeds that of furrow in first five years.
Figure 7.8: Net present value plots of drip, sprinkler and furrow irrigation with wine grapes connected (excluding furrow) with integrated supply system over a period of 30 years
‐10000
‐5000
0
5000
10000
15000
20000
25000
30000
35000
40000
0 5 10 15 20 25 30
NPV ($'000)
Year
Total NPV‐Drip ('000) Total NPV‐Sprinkler ('000) Total NPV‐Furrow ('000)
389
Table 7.19: Profitability indicators for the three irrigation systems irrigating wine grapes over the case study area of 550 ha
Irrigation System Indicator Value
Furrow PVB 8,321,202 PVC 3,339,279
B-C Ratio 2.49
Sprinkler PVB 9,042,896 PVC 3,558,445
B-C Ratio 2.54
Drip PVB 10,482,634 PVC 3,282,709
B-C Ratio 3.19 Values of the computed profitability indicators for the three irrigation
systems with wine grapes over the working life of 30 years are given in
Table 7.19. It indicates that all three irrigation systems with production of
wine grapes are highly profitable. The drip irrigation with wine grapes is the
most profitable among the three irrigation systems. When compared with
citrus irrigation, the wine grapes crop brings relatively higher returns for the
three irrigation systems.
From the analyses given in Section 7.4 and Section 7.5, it is evident that
conversion from furrow to drip system is the most worthwhile option for
both citrus and wine grapes; especially for the latter the B-C ratio being
higher than 3 indicates that the risk of financial loss is very low.
7.6 Sensitivity analysis
The economic analyses given in the above sections are based on the
assumption that the average value of operating costs (e.g. fertilizers,
electricity, labour etc.) and the average value of financial returns (e.g. sale
price of production, market price of water etc.) remains constant over the
entire analysis period of 30 years. This assumption may not remain valid if
there is a long-term shift in costs or returns. For example, there has been a
significant reduction in wine prices in Australia due to oversupply in the
market during the last 4 to 5 years. Such factors influence the long-term
financial viability of the system under consideration. To take into account
390
the effect of variation in the key variables of the financial analysis, the
sensitivity of the outcome is tested against those key variables.
In the context of this chapter, the sensitivity of financial viability of the
conversion of the irrigation system of the two crops is carried out here.
Table 7.20 lists changes in various costs and return items and their
corresponding effect on PVB, PVC and B-C ratio. It indicates that
profitability of all three irrigation systems are highly sensitive to labour
costs; where furrow irrigation is at the top of the sensitivity ladder due to
higher dependency on labour as compared to the more mechanized farming
using sprinkler or drip irrigation. The level of increase in costs for an
increase of 3 c/KWh (peak) to 5 c/KWh (off-peak) in electricity price and
doubling of the price of GHG emissions to 46 $/t CO2e is almost the same
among the three irrigation systems except that no electricity cost incurred
for furrow irrigation. On the benefits side, drip irrigation is more sensitive to
change in price of water trade and to change in the sale price of citrus than
those for sprinkler irrigated citrus. The movement in B-C ratio as a result of
change in sale price of citrus is highest among all variables.
Table 7.20: Sensitivity of present values of benefits, costs and B-C ratio to different cost and revenue items for citrus crop (–ve sign shows decrease with respect to original value)
Item Change in value
New value
Indicator Furrow Sprinkler Drip
Labour ($/hr)
5 25
PVB 0 0 0 PVC 392,893 321,501 269,493
Change in B-C Ratio
-0.07 -0.06 -0.06
Electricity (c/kWh)
Peak: 3 c/kWh,
Off-peak: 5 c/kWh
20 and 15
PVB 0 0 0
PVC 0 75,493 62,926 Change in B-C Ratio 0 -0.01 -0.02
GHG emission
price ($/T-CO2e)
23 46
PVB 0 0 0
PVC 39,604 74,713 68,188 Change in B-C Ratio
-0.02 -0.01 -0.01
391
Water usage
charges ($/ML)
3.33 12
PVB 0 0 0
PVC 31,255 25,483 19,712 Change in B-C Ratio
-0.005 -0.005 -0.005
Water trade price
($/ML) -54 200
PVB 0 -95,639 -189,235
PVC 0 0 0 B-C Ratio 0 -0.016 -0.034
Water trade price
($/ML) 46 300
PVB 0 81,470 161,201
PVC 0 0 0 Change in B-C Ratio
0 0.013 0.029
Citrus sale price ($/T)
-50 175
PVB -
1,323,828 -1,430,207 -1,631,145
PVC 0 0 0 Change in B-C Ratio
-0.225 -0.237 -0.294
Citrus sale price ($/T)
50 275
PVB 1,323,82
8 1,430,207 1,631,145
PVC 0 0 0 Change in B-C Ratio
0.225 0.237 0.294
Table 7.21 lists assumed changes in selected costs and return items and their
corresponding effect on PVB, PVC and B-C ratio. The response of the B-C
ratio to a $5 increase in labour cost is relatively higher for wine grapes than
citrus. This is due to the fact that labour cost is a major cost component for
wine grapes, mainly due to manual pruning and training of vines. The
marginal response to variation in other variables is similar to that of citrus
for the three irrigation systems.
Table 7.21: Sensitivity of present values of benefits, costs and B-C ratio to different cost and revenue items for wine grapes crop (–ve sign shows decrease with respect to original value)
Item Chang
e in value
New value
Indicator
Furrow Sprinkle
r Drip
Labour 5 25 PVB 0 0 0
392
($/hr) PVC 416,533 359,325 347,505
B-C Ratio
-0.276 -0.233 -0.306
Electricity (c/kWh)
Peak: 3 c/kWh,
Off-peak: 5 c/kWh
20 and 15
PVB 0 0 0
PVC 0 56,333 47,898
B-C Ratio
0 -0.01 -0.046
GHG emission
price ($/T-
CO2e)
23 46
PVB 0 0 0
PVC 32,971 57,295 53,365
B-C Ratio
-0.024 -0.04 -0.051
Water usage
charges ($/ML)
3.33 12
PVB 0 0 0
PVC 23,396 19,019 15,020
B-C Ratio
-0.017 -0.014 -0.015
Water trade price
($/ML) -54 200
PVB 0 -73,019 -137,867
PVC 0 0 0
B-C Ratio
0 -0.021 -0.042
Water trade price
($/ML) 46 300
PVB 0 62,201 117,442
PVC 0 0 0
B-C Ratio
0 0.017 0.036
Wine grapes
sale price ($/T)
-50 350
PVB -
1,040,150
-1,087,43
0
-1,229,268
PVC 0 0 0
B-C Ratio
-0.311 -0.306 -0.374
Wine grapes
sale price ($/T)
50 450
PVB 1,040,15
0 1,087,43
0 1,229,268
PVC 0 0 0
B-C Ratio
0.311 0.306 0.374
393
7.7 Chapter summary
In this chapter we analysed costs and benefits of individual irrigation
systems, item by item for the two crops in greater detail. Before doing a
proper financial analysis of the irrigation systems in question, we looked at
different factors that support the need for adoption of hi-tech irrigation
systems. These factors include planned cuts in irrigation water entitlements,
increasing pressure on farmers to realize irrigation savings, increasing price
of water in the water trade market and potential improvement in crop yield
by controlled irrigation. For example, in MIA, the case study area of this
research, the price of water temporarily traded in the open market exceeded
$1,100 during the drought of 2007-08 (Watermove, 2011). Data given in
Table 7.1 indicates that more than 25% improvement in citrus yield can be
obtained by converting from furrow to drip irrigation along with water
savings of more than 3.6 ML/ha.
To model the water and energy use and savings, the previously developed
node-link model was extended over a case study area of 550 ha. The model
computed water savings and energy consumption in irrigation pumping
while taking into account the operation of an integrated pump supply system
under a demand-based irrigation strategy. The node-link model results for
the three irrigation systems for each of the two crops are given in Table 7.2.
A 30 year working life of each irrigation system was assumed. To conduct
the financial viability analyses of the irrigation systems under consideration,
profitability indicators like present value of costs, present value of benefits,
benefit-cost ratio and payback period using net benefit approach were used.
All future costs and returns were discounted at the assumed interest rate of
10%. For these indicators, the inputs including capital investment, annual
operating costs, and annual benefits were prepared for each irrigation
system (including integrated piped irrigation supply system) for the two
crops. New cost items like tax on GHG emissions are also factored in. A
summary of key items related to the financial analyses conducted in this
chapter is given in Table 7.22.
394
Table 7.22: Summary of selected profitability indicators for the three irrigation systems
Indicator Crop Furrow Sprinkler Drip
Capital cost ($/yr/ha) 277 1,416 1,068
Operating cost ($/yr/ha) Citrus 6,222 6,045 5,559
Wine grapes 3,522 3,419 3,161
B-C ratio Citrus 1.01 1.14 1.48
Wine grapes 2.49 2.54 3.19
*Net payback period (yr) Citrus 30+ 18 3
Wine grapes 0 2 2
*Payback period also includes cost recovery of the integrated irrigation supply system
Conversion costs from furrow to sprinkler or drip irrigation systems is
$8,100/ha and $7,100/ha (Table 7.13), respectively. However, the
profitability indicators given in Table 7.22 indicate that conversion from
furrow to drip irrigation is likely to be more profitable and viable in the
long-term than that of conversion to sprinkler system. An important
assumption worth mentioning here is that average values of cost items and
those of returns are assumed to be unchanged over the analysis period of 30
years. To test the sensitivity of profitability indicators to assumed variation
in certain input variables, sensitivity analyses were carried out. The
sensitivity analyses indicate that profitability is highly sensitive to labour
cost, water trade price and crop revenues. It also indicates that due to high
profit margins the risk of unprofitability of drip irrigation is very low as
compared to the other two irrigation systems over the long-term.
As mentioned earlier, conversion to sprinkler irrigation for citrus is not very
economical due to a long payback period and non-attractive B-C ratios.
However, any government subsidies paid for conversion to sprinkler system
could make this an increasingly viable option. Also with the introduction of
the environment as a third user/competitor the future water trade prices may
395
be be higher than in the past which will make the conversion to water saving
irrigation technologies even more financially viable.
In the financial analyses separate annual operating costs were estimated for
each crop and each irrigation system due to varying levels of energy use and
other inputs. Also in this chapter the scenario of installing individual
irrigation pumping stations on each farm to operate their pressurized
irrigation systems is not analysed. The cost of individual pumping stations
could be even higher due to higher initial capital investments (e.g.
individual power supply poles and transformers) and higher operational
costs (e.g. higher electricity charges given as individual customer rates,
higher maintenance costs) which need to be investigated.
396
397
Chapter 8: Integrated Analysis, Discussion and Policy
Implications
Previous chapters discussed modelling and analysis of irrigation water,
energy and greenhouse gas emission linkages for two irrigation strategies,
namely demand-based and supply-based irrigation for three irrigation
methods including furrow, sprinkler and drip irrigation for the major
horticultural crops in the Murrumbidgee Irrigation Area (MIA). This study
has looked into benefits and energy implications of using a centralized piped
supply system to pump pressurized water from source to individual farms to
operate pressurized irrigation i.e. sprinkler and drip irrigation. The study
also analysed the water and energy use with regard to private on-farm
irrigation storages on individual farms and compared this option against the
centralized piped supply system. In the preceding chapter, a detailed
economic and financial analysis of conversion from furrow to pressurized
irrigation was conducted for selected options based on physical quantities
relating to water and energy use in irrigation of horticultural crops which
were determined through modelling.
This chapter is focused on bringing together the key learning from
interpretation of modelling results, sensitivity analyses and the economic
analyses by applying a system dynamics framework. It looks into
identifying inter-dependent variables and understanding dynamics of the
processes and exogenous factors that control those variables. The major
findings of this study are also summarized in this chapter.
8.1 Understanding and representing the dynamics of the system
To define a system we need to define its physical and conceptual
boundaries. For the purpose of this research the system under consideration
consists of the large irrigation area of MIA with particular focus on
horticulture, its crop growing and harvesting practices and irrigation
methods. The MIA is not a closed or isolated system because it is connected
to a bigger system. It responds to the quantity and seasonal availability of
398
the upstream water sources and has potential to impact/control downstream
users including river environment and consumptive users. The behavior of
the system is characterized by various practices (irrigation, pruning,
harvesting), processes (evapotranspiration, fruiting) and inputs and outputs
(mainly in the form of water and energy). All these characteristics of the
system have been in one way or the other considered and discussed in
previous chapters.
The aim of this chapter is to integrate and inter-relate what is found in
previous chapters to understand the overall dynamics of the system in a
holistic manner. In the following sub-sections we try to explore those
dynamic relationships and underlying feedback loops among the inter-
dependent variables. Vensim is a powerful tool to analyse the
interdependence and dynamics of the modelled variables. As an example of
integrated analysis, Appendix B shows a graphic view of the developed
Vensim model in “dynamic simulation” mode.
8.1.1 Water availability versus water saving feedback loop
Irrigation water availability can be affected by multiple factors including
system constraints, climate change, climate shift and/or changes in policy
settings such as changes to the limits on diversions from river system for
irrigation purposes. For the Murrumbidgee case study area irrigation
diversions may reduce by 320 GL if the new basin plan is implemented
(MDBA, 2010; MDBA 2012). Refer to the Figure 8.1 to understand this
feedback mechanism. It is demonstrated in water trade data discussed in
previous chapters that water trade price peaked during the driest years. It
exceeded the price mark of $1100/ML which was offered by downstream
water users in 2007-08 in the water trade market during the last drought
period in MDB. Hence the price of additional water purchased is determined
by the water trade markets driven by water availability. Hence a negative
causal relationship exists between water availability and market price of
water; i.e. the lower the availability of water, the higher the market price.
Furthermore, high water trade price leads to decisions of more investment in
399
water saving irrigation technologies in order to save irrigation water to sell
in the market or to save costs that would otherwise have been incurred to
meet irrigation demand; a positive causal feedback as represented
diagrammatically by a positive arrow in Figure 8.1.
A second positive causal feedback exists between adoption of irrigation
technology and the water savings achieved. The amount of water savings
achieved depends on the level of investment and the savings achieved per
dollar invested. In other words, highly efficient irrigation system may
require higher capital investment. For example, as mentioned in Section
7.3.2, the capital investment for drip irrigation is estimated as $7,100/ha as
compared to just $2,200/ha for less efficient furrow irrigation.
Figure 8.1: Water availability, investment and water savings negative feedback loop
Higher water savings, initiated by low water availability trigger investment
in water saving irrigation technologies, ultimately increasing water
availability. Hence a negative feedback loop exists between water
availability and the water savings as represented by the negative sign in red
colour inside the arrow representing the feedback loop.
8.1.2 Water savings versus energy use feedback loop
Exploring dynamic links between achieved water savings and increased
energy requirement is one of the major objectives of this thesis and is
discussed in detail in different parts of this thesis including Sections 4.8,
Water_Availability Water_Trade_Price
Investment_on_WaterSaving_Irrigation
-
+
Irrigation_WaterSavings
+
+ Water_savingsper_$_invested
Climate_change
Climate_shift
Change_inpolicy_settings
400
4.9, 5.5, 5.6, 5.7, 6.1, 6.3, 6.4, 6.5, 7.2 and 7.6 for various irrigation systems.
Model results summarized in Table 7.2 of Section 7.2 show that up to 3.7
ML/ha can be saved by drip irrigation when compared with furrow
irrigation for citrus. However, at the same time up to 1664 KWh/ha energy
is consumed just in pumping and pressurizing water for drip irrigation as
compared to zero energy requirements for furrow irrigation. Energy
consumption also results in greenhouse gas (GHG) emissions. For drip
system the GHG emissions only from additional energy consumption in
irrigation water pumping are estimated to be 1.5 CO2e t/ha for citrus which
requires payment of GHG emissions tax. Hence, as shown in Figure 8.2, the
higher the water savings, the greater the energy consumption which in turn
causes more emissions tax liability; hence capping and reducing the net
financial return from water savings. This completes a negative feedback
loop as shown in Figure 8.2.
Figure 8.2: Feedback loop between water savings and energy use
8.1.3 Water savings versus environmental benefits feedback
loop
Water savings achieved by adoption of more efficient irrigation
technologies decreases the need for drainage of saline water from irrigation
areas and increases the amount of fresh water available to the environment.
Irrigation_WaterSavings
+
PumpingEnergy_Use
GHGEmissions_Tax
Net_Returnfrom_Saved
Water
+
+
-
- Energy_use_perML_savings
Emissions_perKWh_energy_use
401
Water buyback programs by the government sector also encourage more
water savings and return saved water to the environment. This will result in
more water in riparian systems and hence provide benefit to the ecosystem
as well as offset impacts of GHG emissions from pumping energy use. The
financial incentives from the government (e.g. subsidies on water saving
infrastructure) to make water available for the environment as well as the
long-term intrinsic benefits (e.g. avoidance of climate change etc.) from the
improved environment also encourage the adoption of water saving
irrigation technology. In other words, high water savings result in more
water available to environment. This in turn brings more environmental
benefits and can serve to increase investment/incentives from the
government which ultimately encourages more water savings. Hence, as
shown in Figure 8.3, a positive causal loop exists between water savings and
the environmental benefits.
Figure 8.3: Positive feedback loop between water savings and environmental benefits
8.1.4 Analysis of the feedback dynamics of the integrated
system
Figure 8.4 provides a holistic view of the overall system under consideration
and integration of feedback mechanisms which are discussed above. It is a
representation of how the system components namely water, energy, returns
and the environment are integrated and how they interact interdependently.
Irrigation_WaterSavings
-
Water_forEnvironment
EnvironmentalBenefits
Long-termbenefits
++
+ +
Climate_changemitigation
Governmentsubsidies
Water_buyback_bygovernment
402
Figure 8.4: Representation of the integrated system and the constituent causal feedback loops
The variables shown in italics in Figure 8.4 represent the external factors
which impact a subset of or the entire system. It shows that water savings
are driven by water availability. However, the negative feedback loop
indicates that water savings are not always driven in one direction by water
availability. There can be a point when marginal increase in achieved water
savings becomes higher than the marginal decrease in water availability.
The second negative feedback loop between water savings and the energy
use indicates that the energy costs and associated GHG emissions cost also
limit water savings.
The third feedback loop is likely to be the driving force behind the adoption
of water saving irrigation technologies. There is a positive feedback
between water savings and the environmental benefits. The new knowledge
on importance of improving the environment by returning its share of the
water resource for long-term sustainability of the whole system is the main
spur for the need for defining sustainable diversions limits for consumptive
uses in the Murray-Darling Basin. Since there is no hard limit on share of
water for the environment, the need for water savings for the environment
will always be going in the positive direction. Hence, the positive feedback
Water_Availability Water_Trade_Price
Investment_on_WaterSaving_Irrigation
-
+
Irrigation_WaterSavings
+
+ Water_savingsper_$_invested
Climate_change
Climate_shift
Change_inpolicy_settings
PumpingEnergy_Use
GHGEmissions_Tax
Net_Returnfrom_Saved
Water
+
+
-
- Energy_use_perML_savings
Emissions_perKWh_energy_use
Water_forEnvironment
EnvironmentalBenefits
Long-termbenefits
++
+ +
Climate_changemitigation
Governmentsubsidies
Water_buyback_bygovernment
403
causal loop between water savings and environmental returns. Due to the
location of MIA in the upper part of the basin, the water savings can
potentially serve a dual role; i.e. generating economic return from trading
saved water to downstream users as well as the associated environmental
benefits (lower salinity, benefits to flora and fauna) from increased flows in
the river.
8.2 Discussion on main findings and policy implications
The main conclusions derived from this study are directly applicable to the
MIA case study area with reference to horticulture crops; however, the
developed modelling approach is generic and therefore can also be applied
in other irrigated areas.
As mentioned earlier, this study is focused around the idea of irrigation
conversion from furrow to pressurized irrigation methods namely sprinkler
and drip irrigation for large irrigation areas. Irrigation demand and supply,
water savings, water trade price, gross margins, and energy use (particularly
energy use in irrigation pumping), are the typical variables modelled in this
study for horticultural crops.
The main findings of this research work and envisaged policy implications
are given below.
8.2.1 Modelling of water and energy for irrigation systems
In this research a node-link model is developed which computes irrigation
demand, irrigation supply, soil water balance, water stress affected crop
yield, conveyance losses (if applicable) and energy consumed in irrigation
pumping (if applicable) on a daily time step. One model run covers one
complete yearly cycle of crop production. The node-link model is developed
using Vensim software from scratch and is unique in its ability to simulate
both water use and pumping energy consumption at the same time on a daily
time step. The model is also configured for simulating any of the irrigation
systems including flood, furrow, sprinkler or drip irrigation. The model is
also capable of simulating either open channel supply system or pressurized
404
pipe irrigation supply system. It accounts for conveyance losses in open
channel and head losses in pipe system. It is developed as a generic tool and
can be applied to any irrigation area if data is available.
8.2.2 Water and energy nexus for irrigation strategy
In this research two irrigation strategies are explored, namely, demand-
based and supply based irrigation. Demand-based irrigation strategy
requires constant availability of water which is pumped to irrigate a crop
when needed. On the other hand supply-based irrigation is driven by water
availability and usually involves fixed-interval irrigation. Water can be
supplied through open channels for supply based irrigation. It is found that
demand-based irrigation consumes higher energy but at the same time
produces higher yields due to stress free plant water availability as
compared to supply based irrigation. For citrus under supply-based
irrigation the water use per hectare is as low as 46% of that of demand-
based irrigation but at the same time the yield is found to be as low as 66%
of that of drip irrigation method. Similar trends prevail for wine grapes
production.
Although it is evident that demand-based irrigation produces more yields, at
the same time the cost of energy and its environmental impacts should not
be ignored. Demand-based irrigation involves less labour and relies more on
technological advances. The decision on whether to invest in demand-based
irrigation is to a large degree influenced by policy and economic factors and
their relationships on water use, energy consumption and crop yield. This
study has investigated these relationships through sensitivity analyses and
detailed economic analyses.
8.2.3 Water and energy nexus for irrigation methods
One of the objectives of this research is to explore the water and energy
nexus for various irrigation systems and analyse the water and energy use
implications of irrigation system upgrades. It is noted in this modelling
study that there are significant variations in water use and energy
consumption among various irrigation methods for a given crop and given
405
irrigation strategy. Gravity-fed irrigation like flood and furrow has wetted
area of up to 100 per cent of the crop area. These methods also apply large
amount of irrigation in short time. Hence, the water loss in evaporation from
soil surface and deep percolation are very high for gravity-fed irrigation.
The results indicate that there is significant difference in water use rate
between gravity-fed and pressurized irrigation systems. For example, the
water application rate for flood and furrow irrigated citrus is 12 ML/ha and
10 ML/ha, respectively. On the other hand it is around 8 ML/ha and 6
ML/ha for sprinkler and drip irrigation, respectively, representing 50%
water savings with conversion from flood to drip irrigation. The
corresponding water savings for wine grapes are as high as 60%.
There is almost zero energy consumption using surface water for flood and
furrow irrigation. Groundwater pumping for irrigation is not considered in
this study as it does not occur in the study area. In contrast to gravity-fed
irrigation, the simulation of pressurized irrigation shows that although it
saves irrigation water yet requires more energy to operate pumps. Moreover,
the timely and precise application of irrigation water ensures higher yields
which improve both water productivity and energy productivity. For
example, the model results given in Chapter 4 show that water productivity
of drip irrigation is 5.7 kg/m3 as compared to just 1.99 kg/m3 for flood
irrigated horticultural crops. Furthermore, the energy productivity of drip
irrigation is 4.38 kg/kWh as compared to 3.30 kg/kWh for flood irrigation
of horticultural crops in the case study area. Other key water and energy
indicators are computed and discussed in Chapter 4 and Chapter 5. The
water and energy indicators computed and discussed in this thesis provide a
basis for making informed policy and investment decisions in relation to
irrigation conversion.
It is interpreted from the results in this study that the conversion to hi-tech
irrigation is economically and environmentally justifiable as long as the
increased energy cost and environmental impacts due to greenhouse gas
406
emissions are offset by increased yield, lesser accessions to the saline
groundwater and more water returned to the environment.
It is also noted that drip irrigation outperforms sprinkler irrigation both in
terms of water use and energy consumption for the horticulture crops under
both irrigation strategies. This assertion may not hold true for the sprinkler
system for irrigating broad acre crops as their irrigation application pattern
is totally different from horticulture.
8.2.4 Up-scaling modelled water and energy use
A node-link model is developed in this study to model water and energy
consumption in various irrigation systems. This model represents a case
study area of around 300 hectares and computes water and energy use at the
model scale. One of the objectives of this study is to examine water and
energy dynamics at the irrigation scheme scale, in this case the MIA.
Keeping all other parameters same, soil type is a major factor that controls
irrigation water requirement. The developed model computes irrigation
water use for the major soil groups for given horticulture crops in MIA.
Using the information of soils and the corresponding model output on water
use rate for each horticultural crop, it is an acceptable approach to linearly
up-scale water use to the entire MIA horticultural area. However, this linear
up-scaling approach is not valid for pumping energy use because the
pumping energy consumption is not a linear function of irrigated area
because of non-linear relationship between head losses and the flow volume
in pipes.
It is noted from the model runs completed in this study that the pumping
energy use almost doubles with 50% increase in irrigated area. To overcome
this issue, two up-scaling approaches are proposed as discussed in Chapter
6. The first approach is based on some relatively crude lumping assumptions
but still gives reasonably accurate results. The second approach is GIS
based and involves intensive processing at each farm scale and is relatively
more accurate. However, both up-scaling approaches are physically based
407
on soil data. The water and pumping energy use are up-scaled for each
irrigation system at various levels of adoption.
It is estimated that given 100% conversion of the MIA horticultural area of
28,970 ha to drip irrigation technology would result in around 137.49 GL of
water use per annum, while around 45,400 MWh of electricity would be
consumed in pumping that irrigation water over the year. For sprinkler
irrigation at 100% adoption level the total water and total energy use are
roughly 30% and 64% higher than that of drip irrigation, respectively. These
results again emphasize the point that drip irrigation outperforms sprinkler
irrigation both in terms of water savings and energy consumption for
horticultural crops.
8.2.5 Effectiveness of on-farm storages versus centralized
irrigation supply
All the observations and results discussed in the above sub-sections are for
the irrigation systems of each farm connected with a centrally located water
supply source, typically an irrigation canal or en-route storage. Private on-
farm storages are also widely used in the study area, especially at the farms
which use some sort of pressurized irrigation system. Therefore, in this
study we have also modelled and compared the effectiveness of the on-farm
storages in terms of water savings and pumping energy consumption in
Chapter 5. The major function of on-farm storages is to ensure the timely
supply of irrigation when needed and when the total irrigation demand
exceeds the capacity of the regular irrigation supply system.
It is evident from results in Chapter 5 that on-farm storages are less efficient
both in terms of water savings and energy consumption. For example in the
case of the drip irrigation scenario, the on-farm storage option shows
additional evaporation and seepage loss of 362 ML for the case study area.
Interestingly the pumping energy consumption of on-farm storages option is
negligibly higher than that of the centralized irrigation option. Hence,
significant water savings can be achieved by adopting centralized irrigation
supply system for drip irrigated farms.
408
For sprinkler irrigated farms, the water losses from on-farm storage option
are as high as 564 ML as compared to the centralized system. But the
energy consumption for on-farm storages is significantly lower than the
alternative option. However, it is estimated that for each 1 ML of water
savings, an additional 0.26 MWh energy are consumed by the centralized
pumping system. The market value of 1 ML of water is much higher than
that of 0.26 MWh of additional required energy. Moreover, this analysis
does not consider the fact that operation and running cost for the centralized
irrigation system are significantly lower than the individual pumping
stations on each farm. In totality, the centralized integrated irrigation supply
system is more effective than on-farm storages.
8.2.6 Long-term viability of irrigation conversion
It is hard to justify the conversion of gravity based irrigation system to one
of the pressurized systems if it does not payback capital costs within a
reasonable period and remains profitable in the long run. Therefore, the
economic viability of each conversion option is tested thoroughly in this
study. The analysis also includes the capital cost of installing the centralized
pumping station and the distribution pipe network with at least one outlet to
each farm. The economic analysis is conducted for three irrigation methods
(furrow, low head sprinkler and drip) in the case study area of 550 ha for
each citrus and wine grape production. The water use and energy
consumption rates are computed by the developed node-link model. The
reason for using the size of 550 ha of the study area is the fact that the
centralized/integrated irrigation supply system is designed to service this
much area. Therefore, it is imperative to analyse economic efficiency of this
system for the design area. Since, the economic analyses are based on the
model results and the data related to the case study area in MIA, the
conclusions are directly linked to the MIA. However, the methods applied
are applicable to any area and the general conclusions are likely to remain
unchanged.
409
On the cost side, capital costs and running costs are taken into account. On
the benefits side, returns from the sale of the raw product and from the
selling of the saved water in the water trade market, are considered. All
costs (including interest on initial capital investment and equipment
depreciation) and benefits (including yield sale and water traded out) are
converted into annual values. The payback period is computed by
comparing cumulative annual present value of benefits with the cumulative
annual present value of costs using an interest rate of 10%. The working life
of each irrigation technology is assumed to be 30 years. The results indicate
that the drip irrigation system with wine grapes has the least payback period
of 2 years followed by 3 years for the drip system with citrus. The sprinkler
system and furrow irrigation with citrus have payback periods of 18 years to
over 30 years, respectively. The reason for long payback period for furrow
irrigation is the fact that its annual operational costs are higher than the
annual returns. Similarly, the longer payback period for sprinkler system
owes to higher initial capital costs, higher energy costs and relatively lower
annual returns compared to drip system. It is noticed that the profitability
indicator; the benefit-cost ratio, for citrus crop is highly sensitive to the sale
price of the yield obtained, followed by the trade price of water, followed by
the labour cost which is followed by the energy/electricity price. For wine
grapes the benefit-cost ratio is most sensitive to the sale price of the yield,
followed by labour cost. Although, both citrus and wine grapes farms are
highly mechanized and automated, a significant cost of manual labour is
incurred in pruning and fruit harvesting as reported in Chapter 7.
410
Figure 8.5: Annual costs and returns for the three irrigation systems with citrus on per hectare basis (capital cost includes the cost of integrated irrigation supply system, except for furrow irrigation)
Figure 8.5 and Figure 8.6 summarize the annual costs and annual returns for
the three irrigation systems for citrus and wine grape, respectively, on a per
hectare basis. The capital cost also includes the annual cost incurred in the
installation of a centralized irrigation supply system. The annual operational
cost includes the cost of all inputs, equipment use, water charges (fixed and
usage based) and most importantly the cost of greenhouse gas emissions
(carbon tax).
A comparison of greenhouse gas (GHG) emission costs for various
irrigation methods for the two crops is given in Table 8.1 based on numbers
reported in previous chapters. The GHG emissions are accounted for all
energy inputs including fertilizer, pumping etc. The greenhouse gas
emissions cost, also referred to as carbon tax, is reported as the percentage
of the total annual operational cost on a per hectare basis. Drip system
operated by the centralized irrigation supply system for growing wine
grapes has the highest GHG emissions cost of roughly 2% of the annual
operational cost. The furrow irrigation has the least GHG emissions cost due
to absence of irrigation pumping.
411
Figure 8.6: Annual costs and returns for the three irrigation systems with wine grapes on a per hectare basis (capital cost includes cost of integrated irrigation supply system, except for furrow irrigation) Table 8.1: Greenhouse gas emissions cost as percentage of the total annual operational cost per hectare
Crop Furrow Sprinkler Drip
Citrus 0.67% 1.31% 1.30%
Wine grapes 0.99% 1.77% 1.79%
As mentioned earlier and also shown in Figure 8.5 and Figure 8.6,
conversion to sprinkler irrigation for citrus production is not a very
economical option due to long payback periods, high cost low return and
consequently non-attractive B-C ratio. However, any government subsidies
on conversion to sprinkler system could make it a viable option. Also, with
the introduction of the environment as a third user/competitor, the future
water trade prices are likely to be higher than in the past which will make
the conversion to water saving irrigation technologies an even more
financially attractive option.
8.2.7 View from system dynamics lens
A detailed system dynamic analysis of the system under consideration is
carried out at the start of this chapter. Partial dynamic analyses have also
been conducted at appropriate parts of this thesis. In a nutshell, the system
dynamics analysis suggests that at the system scale the need to achieve
maximum water savings is driven by the water availability which in turn is
412
driven by natural factors and policy shifts. It is evident from the analyses of
the identified feedback loops that there is a conflict between maximizing
water savings to support the environment and the negative environmental
impacts of the means adopted to achieve those water savings.
This study has to a greater extent quantified the interacting variables and
exogenous factors for various developed scenarios to help establish better
understanding of the underlying feedback mechanisms. It also shows that
the water and energy nexus is a complex structure to comprehend. The
analysis also suggests that water and energy nexus should be looked at a
wider scale to support any policy decision making. The analysis at irrigation
scheme scale as conducted in this study seems to be appropriate if not best
scale. It is not appropriate to make a policy decision just by looking at farm-
scale water and energy results. For example, the amount of irrigation
pumping energy required at farm scale seems relatively low. However, if a
decision is made to provide assistance to convert all farms in the irrigation
scheme to pressurized irrigation then the amount of total energy required
can be equivalent to half the generating capacity of the Snowy Hydro
Scheme as mentioned in Chapter 6. Installing a new coal fired power
generation plant to fulfil this additional energy requirement would not be an
environmentally sustainable solution, to say the least. Therefore,
consideration of the scale of problem and taking a holistic approach is very
important to reach an environmentally and economically optimum decision.
413
Chapter 9: Conclusions and the Way Forward
This doctoral research thesis is an attempt to analyse the complex nexus
between water and energy that exists in irrigated systems. The basic
hypothesis behind this research has been to identify and realize water and
energy savings it is critical to adopt a system level thinking to explore the
water-energy nexus. The system level thinking implies just not a larger
physical scale of the problem; it also refers to the brining more and more
inter-related variables and processes into consideration. Therefore, in this
research we have focused on a large irrigation scheme, the Murrumbidgee
Irrigation Area (MIA), and all possibly inter-related variables like water use,
energy consumption, environment, and last but not the least the economic
factors. However, before considering the whole irrigation scheme scale, a
smaller case study is analysed first. The water, energy, greenhouse gas
emissions and economic indicators are explored by developing a node-link
model and other methods at the case study scale. Then the results are up-
scaled and critically analysed at the irrigation scheme level.
The overarching objectives of this research are as follows:
1. To synthesise knowledge and future challenges related to energy and
water use efficiency in large irrigation areas.
2. To quantify spatio-temporal trends in energy and water use
efficiency in a major irrigation area using a node-link model.
3. To develop a hydrologic-economic dynamic system framework for
testing the economic viability and for estimating the environmental
footprint of farming operations by exploring system-wide linkages
among water use efficiency and associated costs, irrigation
management strategies, energy-yield relationships, energy
consumption and associated greenhouse gas emissions.
414
To achieve these objectives this research has addressed a number of
questions and drawn conclusions from the results/answers to these following
questions:
1. What are the missing/unknown links between water, energy and
environment which could play out as huge challenges with future
irrigated systems?
The literature review indicates that in the past most of the emphasis has
been given to improving water use efficiency in irrigated systems. Use of
modern irrigation systems has been accepted as the most effective solution
to achieving high irrigation efficiency. However, the literature concludes
that very little attention has been rendered to the estimation of increased
energy that is required to operate modern irrigation systems. The literature
review highlights the knowledge gap to properly understand water
efficiency and energy consumption nexus. To understand water-energy-
environment nexus this study has focused on the environmental impacts of
water diversion from rivers for irrigation, the potential impacts of climate
change on water resources and the greenhouse gas emissions from increased
energy consumption by modern irrigation systems in the MIA.
2. How can a biophysical tool help understand and quantify water-energy-
environment interactions in a large irrigation area?
This question is addressed by developing a node-link model of the
horticultural area of MIA. The developed node-link model has various
modules that compute irrigation demand and supply, irrigation management
strategy (demand-based or supply-based), irrigation supply system and yield
for a given crop and given irrigation method on a daily time step. At the
same time, the model keeps track of electricity consumption in pumping
irrigation water by computing energy head requirements to overcome head
losses in the irrigation supply system and to provide the required pressure
head at each farm inlet to operate hi-tech irrigation systems. The model is
developed to represent a case study area and then results are up-scaled using
415
appropriate up-scaling techniques over the entire horticultural area of the
MIA. The other direct and indirect energy inputs and greenhouse gas
emissions are also estimated for each scenario. Hence the developed model
and other biophysical data provide an adequate information-base to
understand and quantify the water-energy-environment nexus.
3. The third and the most comprehensive question is, “What is the nature
of linkages between water use, energy consumption and greenhouse gas
emissions from irrigation conversion for a large irrigation area, and
what approach should be taken to understand those linkages”?
The major part of this thesis is dedicated to finding an answer to this
question. Different scenarios representing different irrigation methods, crops
and irrigation strategies were modelled using the developed node-link
model. A holistic and system dynamics approach was adopted to
simultaneously monitor behaviour of key variables including irrigation rate,
water losses, water savings, energy consumption in pumping, and
corresponding greenhouse gas emissions. Furthermore, economic analysis
and sensitivity was also conducted for the most promising scenarios. All
variables related to the water, energy, greenhouse gas emissions and
profitability indicators were put into a matrix. This matrix was analysed
through a system dynamics lens to identify underlying feedback loops
between the inter-dependent variables. This analysis concludes that there is
a strong inter-dependence between water savings, energy consumption and
environmental implications and that no decision should be made based on
just one of these key variables. If we do so, we will never get an optimum
solution. The analysis of the feedback mechanisms also shows that the
whole water and energy initiative is mainly driven by water availability and
environmental considerations. The overall framework developed to analyse
the water-energy-environment nexus in this study is not area specific and in
fact can be applied to any large irrigation area to achieve similar objectives.
416
9.1 Major recommendations
This thesis research adds knowledge that helps understand the water-energy-
environment nexus and make decisions regarding conversion to modern
irrigation technology in a large horticultural area in Australia. However, the
developed framework is applicable to any area with surface irrigation.
This study suggests the following recommendations for the stakeholders in
the irrigation industry in general and horticultural production industry in
particular, in Australia:
9.1.1 Recommendations for policy makers
Collect as much relevant data as possible to cover the length, breadth
and depth of an irrigation conversion problem. A well considered
problem definition will help devise more effective solutions.
Always widen the scope of problem to a possible extent that defines a
comprehensive irrigation scheme conversion objective so that an
effective and fit-for-all decision can be made. For example, improved
infrastructure alone may provide maximum water savings but may have
adverse economic, energy related and/or environmental consequences in
the long run. These issues are highlighted in Chapter 4, Chapter 5,
Chapter 6 and Chapter 7.
This study does not take into account soil carbon sequestration in
agriculture as mentioned in equivalent GHG emissions calculations in
Chapter 4 and Chapter 5. Detailed policy should be developed to offset
the GHG emissions tax on agriculture with the amount of carbon
sequestered by the crops.
Take a holistic and long-term view to devise a possible solution. For
example, it is found in Chapter 6 that if a decision is made to convert the
whole of MIA horticultural area to sprinkler system, an additional 50%
of Snowy hydro generation capacity is required to supply these energy
needs. The energy required to supply 100 adoption of drip system in
417
MIA requires an additional 20% of Snowy hydro generation capacity.
This high requirement of energy should be considered in decision
making on conversion of the entire MIA horticulture area.
If possible, visit the area or talk to the local farmers before making any
decision to support a particular water saving initiative.
9.1.2 Recommendations for irrigators
Time has come to change community preferences to favour
improvements in irrigation efficiency to help the environment as a
legitimate stakeholder in the water industry.
Make informed decisions on acceptable and viable tradeoffs on water
use, energy consumption and achievable yield. The framework
developed in this thesis as well as sample results obtained using real
farming data can help inform these decisions.
Should not blindly follow others as the each farm may have different
circumstances. For example, converting to drip irrigation may not
necessarily be economic if your soil/crop/environment is able to achieve
comparable water usage by furrow irrigation system.
Undertake a proper biophysical analysis and long-term economic
analysis of all alternatives and choose the most optimum alternative.
Also manage the key variables by conducting a sensitivity analysis.
Demand-based irrigation is better suited to modern irrigation systems
and supply-based irrigation strategy is more appropriate for
conventional gravity-based irrigation methods.
Connect your farms with centralized irrigation supply system to operate
your sprinkler or drip system. With a nominal service fee it saves time,
labour cost as well as operation and maintenance costs. Energy cost is
also reduced by avoiding fees for installation of electricity supply
418
equipment (transformer etc.) at each farm. Moreover, the irrigation
company may negotiate price of the electricity with the provider.
In developed countries like Australia, farm labour availability is low and
labour costs are very high. There are up to 31% reduction in cost of
overall labour-based operations for horticulture production with drip
irrigation as compared to furrows. The labour saving for irrigation alone
is estimated to be 80% less than furrow irrigation. For sprinkler based
production the overall labour savings are 18% as compared to furrow
system and 48% labour savings in irrigation only.
Use of on-farm storages is not recommended as the irrigator has to bear
evaporation losses from the storage. Also the energy savings as
compared to the centralized irrigation supply system do not offset the
increased operation and maintenance costs.
Where possible, be prepared to adapt to the potential future challenges
such as climate change and policy reforms.
9.1.3 Recommendations for irrigation providers
Adopt appropriate measures to minimize water conveyance losses from
“hot-spots” by lining the channels or by replacing open channels with
pipes.
Water trade has now become a significant part of the water industry.
Therefore, work with irrigators and policy makers to facilitate water
trade in an open water market.
Work with policy makers and irrigators to achieve environmental and
economic objectives.
9.2 The Way Forward
This research study is conducted by adopting available methods and tools to
link water, energy and environment in horticultural areas; however, there is
scope for improvement though further work in the following areas:
419
Modifying the developed node-link model to perform continuous
simulation over multiple years.
Currently the model uses a fixed proportion of applied irrigation being
lost through deep drainage. Furthermore, it does not take into account
the effects of a raised watertable. A proper biophysical model could be
developed to achieve a dynamic relationship to quantify surface-
groundwater interactions in irrigated areas.
The developed node-link model does not simulate groundwater pumping
as it does not occur in the study area. However, the model can be
modified to include groundwater pumping and energy use in
groundwater pumping.
Application of the developed water-energy-environment analysis methodology at the river basin scale.
9.3 Changes in Developed Model for Application in Other Areas
The following major changes will have to be made to apply developed
node-link model to other areas.
Soil parameters as per new soil types being modelled.
Crop parameters for the new crops.
Number and/or size of pipe system.
The layout of the model components to represent physical system.
Simulation period.
420
421
References
Australian Bureau of Statistics (ABS). (2006). Water Account Australia
2004-05. Australian Bureau of Statistics, Canberra. ABS Catalogue
No. 4610.0. ISBN 0 642 47942 9.
Australian Bureau of Statistics (ABS) (2009). Murrumbidgee sustainable
yields region socioeconomic profile, report for the Murray–Darling
Basin Authority by the Australian Bureau of Statistics, Australian
Bureau of Agricultural and Resource Economics & Bureau of Rural
Sciences, Adelaide.
Australian Bureau of Statistics (ABS) (2011). Energy Account Australia
2009-10. Australian Bureau of Statistics, Canberra. ABS Catalogue
No. 4604.0.
Australian Bureau of Statistics (ABS). (2012). Water Account Australia
2009-10. Australian Bureau of Statistics, Canberra. ABS Catalogue
No. 4610.0.
ACIL Tasman, (2009). Regional economic effects of irrigation efficiency
projects. Report of case studies prepared for the Crane Group.
Available online at: http://www.pc.gov.au/ (accessed in May 2011)
Ahlfeld, D. P. and Laverty, M.M. (2011). Analytical solutions for
minimization of energy use for groundwater pumping, Water