Testing the CLEANED framework in Lushoto, Tanzania

• 1. Testing the CLEANED framework inLushoto, TanzaniaMats Lannerstad (ILRI), An Notenbaert (CIAT), Ylva Ran (SEI), Simon Fravel (ILRI), Birthe Paul(CIAT), Simon Mugatha (ILRI), Edmund Githoro (ILRI)CLEANED validation, synthesis and planning workshopMachakos, Kenya, 30-31 October 2014

2. The Lushoto pilot Aim = to provide a proof of concept From generic framework to practical implementation Livestock and Fish CRP connection: transforming thedairy value chain in Tanzania Tanga and Morogoro Presented here = the Lushoto case2 3. The step-wise procedureA. Setting the baseline Stratification Description Land use and management practices Stocks and flows Value chains Vulnerable and limiting resourcesB. Ex-ante assessment Intervention description Local impact assessment Out-scaling Flagging of risks 4. Describe systems, practices and VCs4 5. Primary dataData sources Participatory GIS Expert consultations Household surveys Secondary data HH-level Spatial data Literature Expert knowledge5 6. Aim:Participatory GIS Collect and calibratespatially-explicit data Explore scenariosof change Assessments producedaligned to and rooted in local understanding Resulting maps (with qualitative descriptions): Different dairy production systems Areas of dedicated feed production Environmental resources (status and risk)6 7. Stratification of Lushoto (TZ)Forest 3 main production systems: Intensive (cut&carry) Semi-intensive (some grazing) Extensive (pastoralism) Feed baskets:Fodder% grass/residues/otherArea%Milk yieldl/yr/LUExtensive 75/20/5 11 400Semi-intensive 65/22/12 5 1250Intensive 50/35/15 29 1250 8. Vulnerable and limiting resources 9. Losses along the VC9Waste/loss as a multiplying factor0% 2% 10% 2% 2%15.3% 10. Full description10Intensive Semi Extensive Waste %Area (ha) 119000 20500 43600 Feed 0 100.0Grassland fraction 0.15 0.3 0.85 LS 2 98.0Cropland fraction 0.75 0.6 0.05 Transport/processing 10 88.2Crop residue removal (fraction) 0.9 0.6 0.85 Distribution 2 86.4Grass removal (fraction) 0.5 0.5 0.4 Consumption 2 84.7Cattle (number) 10000 1500 3000 Overall loss 15.3Livestock density (head/ha) 0.08 0.07 0.07Manure production (kg/head/day) 4 3 3Total manure available (kg/ha/year) 122.69 80.12 75.34Milk yield (l/head/year) 1250 1250 450Manure to cropland (fraction) 1 0.75 0.33Nmanure (kg/kg) 0.03 0.03 0.03Manure N loss from volitilization0.7 0.7 0.7(ratio)Nfertiliser (kg/kg) 0.18 0.18 0.18Annual precipitation (mm/yr) 1100 1000 900Soil type (FAO) Acrisol Acrisol AcrisolSoilN (g/kg) 2.5 1.5 1.1SoilC (g/kg) 35 30 21Soilclay (%) 40 40 40Soil depth (m) 0.2 0.2 0.2Bulk density (g/cm3) 1230 1230 1230LS factor 3.20 3.00 3.00K factor 0.2 0.2 0.2P factor 1 1 1.00Soil loss (kg/ha) 21,403 18,920 13,273Waste (% milk) 15 15 15 11. Intervention description11 12. Scenario of change: intensification no land use changeScenario1Fodder%grass/residues/otherLivestockpopulationLUMilk yieldl/yr/livestock unitExtensive 75/10/15 12,500 1250Semi-intensive 70/20/10 1,875 2750Intensive 64/23/13 3,750 2750 25% increase in animal numbers Increase in fodder, concentrates and rice straw 100% increase in fertilizer input, with an associated yieldincrease of 50% 50% reduction of waste at the transport/processing stage 13. Intervention description The level of detail ~ the assessment methods Changes in relevant input parameters Suitability to the different livestock productionsystems and VCs13 14. Quantification of impacts14 15. Pathways and key indicators151. Water availability and quality: Appropriation of available resources Change in soil water holding capacity Change in water quality2. Soil and land health: Soil erosion Change in soil organic matter Nutrient3. GHG emissions: Total emissions of methane, nitrous oxide, carbon dioxide4. Biodiversity loss: Species diversity Landscape multi-functionality 16. Soil and Land16 17. Soil and Land Soil erosion Removal of valuable topsoil: Disturbance of seeds and plants Loss of nutrientsImpacts on crop emergence, growth and yield Deposited downstream: Disturbance of plant growth downstream Filling up and/or contaminating reservoirs and rivers17 18. Soil and Land Soil erosion Revised Universal Soil Loss Equation (RUSLE):Annual soil loss (kg/ha/yr)=R * K * LS * C * PR = ErosivityK = ErodibilityLS = Slope length and steepness factorsC = Cover management factorP = Support practice factor18 19. Preliminary results: soil loss1. Absolute:Small increase in soil loss2. Efficiency (compared to milkgain): Gains across the boardKg /ha/yr2500020000150001000050000Kg/1000l300025002000150010005000 20. Soil and Land Nutrient balance Soil fertility decline Impacts on crop yields Losses to air and waterWater quality and GHG need to find a good balance!20 21. Soil and Land N NUTMON calculations(IN1 + IN2 + IN3 + IN4) (OUT1 + OUT2 + OUT3 + OUT4 + OUT5)21INPUT/OUTPUT ID NAME FORMULAIN1 Mineral fertilizer Amount of fertilizer (kg/ha)*fertilizer N(kg/kg)*area (ha)IN2 Animal manure Amount of manure manure N*area (ha)IN3 Atmospheric deposition 0.14*p*area (ha)IN4 Biological N fixation Non-symbiotic N fixation {2+(p-1350)*0.005} +Symbiotic N-fixation (% uptake attributed to Nfixation*total N uptake) *area (ha)OUT1 Harvested crop products N content in harvested product(kg/kg)*yield ofcrop (kg/ha)OUT2 Crop residue N content in crop residues(kg/kg)*yield of crop(kg/ha) *area (ha)OUT 3 N leaching (if clay35% and55%) (Soil N +Fertilizer N)*(2.1*10*p+5.4) *area(ha)OUT 4 Gaseous losses (Soil N +Fertilizer N )*(-9.4+0.13*clay+0.01p)*area (ha)OUT 5 Soil erosionSoil loss (kg/ha/year)*Soil N*1.5 22. Preliminary results: N balance1. Absolute:Increase in nutrient mining,leaching, gaseous losses2. Efficiency (compared to milkgain): Gains across the boardKg /ha/yr Kg/1000l0.00-10.00-20.00-30.00-40.00-50.00-60.00-70.00-80.000.00-1.00-2.00-3.00-4.00-5.00-6.00 23. Soil and Land Next steps Improve calculations and feedback loops, e.g.: Add organic fertiliser Link manure production to DM feed intake Add Soil organic matter calculations Triangulate assumptions with HH-level data, valuesfrom literature and expert opinion Convert quantitative calculations into qualitativeassessment Link to GIS and produce maps User-friendly tool23 24. Water24 25. Water Why?Water scarcity is a rising global problemVital for humans and functioning ecosystemsIn livestock production: Provides drinking and servicing water Supports growth of animal feed and grazingBut - water resource use is highly complex to analyze Considers a moving resource in a landscape Variability of time and space Hidden in animal feed consumption25 26. Water quantity Calculation of actual evapotranspiration (ET) per systemusing the Aquacrop modelSoil water holding capacity (SWHC) Long term perspective of water availability for crops Comparing different land use management practices forinterventionsWater quality Change in water quality due to management practices Combined risk index based on fertilizers , chemicals and soilerosionWater How?26 27. Water How?Underlying assumptions: Area proportion per crop in the system reflects feedcomposition Modelled ET is indicative of actual ET Two cropping/growth seasons corresponding to rainperiods Same growing conditions are assumed across thestudy area An average harvest index of 35 % leaves 65 % ofcrop biomass as residues entirely used for fodder27 28. Water - Results28Waterquantity706050403020100I S-I EET/MAR (m3/m3, %)Baseline Scenario180016001400120010008006004002000I S-I EET/feed (m3 /ton)Baseline Scenario180160140120100806040200I S-I EET/milk (m3/tton)7006005004003002001000I S-I EET mm/year & system)ET/milkET/feedET/systemET/MARI = IntensiveS-I = Semi-intensiveE = Extensive 29. Water - Results29ScenarioSystem(Lushoto)SWHCratingWaterqualityratingIntensive Low LowSemi-intensive Low LowExtensive Low LowWater quality:- Little impact- Low levels of fertilizers &chemicals applied mosttaken up by plantsSWHC calculation:- Organic mulch- Fertilizer- Cropping patterns and tillage- Impact in Lushoto is very low especiallycompared to increase WP 30. Water Next steps Weight the 3 result components into a single score Enables to capture small impacts and flag themeven though they are not significantly changing thefinal score Create water score maps for each component, andthe single score Indicating the difference between components andthe overall water score in red-green light over thelandscape30 31. Water Lessons learned Water for livestock is complex! Everything isinterlinked! The results for water use changes depending onthe lens you are looking throughDecreased SWHC indicates a negative result. Butmanagement radically increases yield and WP -thus leaving water quantity positive How do we properly capture that in weighting thecomponents in a final score? Results will be equally complex and need to bevisualized, component for component, but alsotogether to provide a water impact measure31 32. Biodiversity32 33. Biodiversity: Rationale Agriculture depends on biodiversity Gene pool of crops and animals: risks and missedopportunities Future generations33 34. Biodiversity: Loss drivers Long history of Agriculture:Species selection, cultivation practices andconverting natural vegetation Drivers in Tanzania:Agriculture, unsustainable harvesting, mining,built environments, contamination of soil andwaterways.34 35. Biodiversity: Scope and methodScope: Vulnerable, threatened and endangered speciesMethod: Intersect of IUCN Redlist and study area Investigate Source of threat, Geographical extent Group species Mitigating strategies for all relevant species35 36. Biodiversity: Preliminary results In Lushoto: 18 speciesthreatened byagriculture Average extent globally:4,300 km2 Causal link with dairydevelopment? If only minor driver, stillan opportunity to raiseawareness36 37. Biodiversity: Next steps Develop management strategies.Potential groups:Group 1) Birds 6 speciesGroup 2) Other insectivores, similar location - 8 speciesGroup 3) Other reptiles 6 species Incorporate botanical and aquatic species Links with water pathway on water quality Indicators for landscape multi-functionality37 38. Biodiversity: Lessons Limited data for insect and agricultural biodiversity Individual species easier to analyse than ecosysteminteractions Causal links challenging LUC can indirectly increasehabitat pressure. Identifying risks and mitigation options can be morepractical than quantified impact.38 39. Greenhouse Gas Emissions39 40. GHGs: rationale A long term global issue of global warming climate variability and sea level rise. Relevance to farmers lost energy and nutrients Linking with environmental Donors have a long term view of development andpotential risks40 41. GHGs: scope and methodScope: On farm emissions Livestock emissions Ruminant / IPCC CH4 enteric fermentation guidelines (eq 15) IPCC manure management emissions (eq 10.23, 10.25,10.27) Land management changes IPCC rice cultivation guidelines (Cool farm tool) IPCC guidelines on cropland (Cool farm tool) Land use change IPCC land use change guidelines / PAS:205041 42. GHGs: Lushoto results42Emissions in CO2 equivalent(annual) BaselineExtensive Intensive/Semi IntensiveEnteric fermentation (head) 1152 1838Manure management (head) 878 1092Fertiliser emissions (head) 0.57 52FPCM yield (l) 421 1315Emission intensity FPCM 4.8 2.2Net emissions* CO2-e (head) 2031 2982ScenarioExtensive Intensive/Semi IntensiveEnteric fermentation (head) 1816 2882Manure management (head) 1092 1520Fertiliser emissions 1 93FPCM yield (l) 1315 2892Emission intensity FPCM 2.3 1.6Net emissions* CO2-e (head) 2909 4495 Net emissionsincrease in Lushoto bycirca +35% Emission intensitydecreases from 3 to1.7 kg CO2-e / 1l*Background N2O emissions were excluded, but would be consistent FPCMbetween baseline and scenario 43. GHGs: next steps Test more complex scenarios incorporating age atfirst calving and manure management Test accuracy of results with more detailed data andcomplex modeling Review allocation of emissions over the lifecycle43 44. GHGs: lessons Post farm gate scenarios and emission estimates Scenarios have to be fleshed out liveweights,milk yields by season, feed baskets.44 45. First Reflections I Pathways / impact categories: capture the most important issues true? Pathway indicators: Subjectively selected can we do better? Should other indicators be possible to use in othercontexts, e.g. aquaculture? Pastoral vs. mixed? Absolute vs. efficiency??? Pathway calculations: How to better capture seasonality? Not all VCs are the same: for now only captures in thewaste assessment how to improve How to indicate some kind of confidence level?45 46. First Reflections II Intervention descriptions: Based on lots of assumptions and expert knowledge - Isit possible to make this user-friendly? Single interventions as well as more systemic changes Is it a rapid tool? Calculations difficult to set up But as soon as set should be quick Visualisation Aggregate how far? Indicator/pathway and trade-offs vs. Overall impact Per system, VC, study area Traffic light on map feels good Same spatial unit all pathways46 47. First Reflections III Implementation of the framework: Participatory approach (e.g. Through PGIS) mightincrease trust in/use of results Use of the framework Useful for comparing interventions WITHIN a study area Only environment But it captures several dimensions, not only GHGs Other assessments will answer other questions Income, productivity, equality Follow up assessment might be required (e.g. In redflagged areas/pathways) How to ensure use by different of stakeholders?47 48. First Reflections IV Appears to work for dairy needs strenghtening and refining- Needs further testing in other VCs and other systems48