MATHEMATICAL MODELING OF RAINWATER HARVESTING … · RAINWATER HARVESTING SYSTEM FOR UNGAUGED CATCHMENT AREA Mohammad Amaar Hassan ... Mathematical Modeling of Rainwater Harvesting

http://www.iaeme.com/IJCIET/index.asp 823 [email protected]

International Journal of Civil Engineering and Technology (IJCIET)

Volume 9, Issue 11, November 2018, pp. 823–837, Article ID: IJCIET_09_11_078

Available online at http://www.iaeme.com/ijciet/issues.asp?JType=IJCIET&VType=9&IType=11

ISSN Print: 0976-6308 and ISSN Online: 0976-6316

© IAEME Publication Scopus Indexed

MATHEMATICAL MODELING OF

RAINWATER HARVESTING SYSTEM FOR

UNGAUGED CATCHMENT AREA

Mohammad Amaar Hassan

Graduate Student, University of Diyala , College of Eng., Dept. of Civil Eng

Dr. Qassem H. Jalut

Associate Professor, University of Diyala , College of Eng., Dept. of Civil Eng

ABSTRACT

The purpose of this study is to estimate the amount of surface runoff volume that

can be harvested during flood seasons for a selected ungagged area in Diyala

governorate near the Iraqi-Iranian border. The study area (the upper part of

Khanaqin city) was modeled using the mathematical model (SWAT), where the

modeling results showed that the area of the catchment is (701.836 km2) by using

remote sensing techniques and satellite images include DEM maps, land cover maps

and soil maps (FAO) covering the study area (upper Khanaqin). The results showed

that the volume of surface runoff for 21-year period from 1990 to 2010 is estimated to

be 3,242,268.13 m3. Furthermore, the predicted value of the volume of surface runoff

for 18 years during a period extended from 2018 to 2035 is estimated to be

(76,352,348.53m3). It Furthermore, has been proposed site for the location of the dam

for design system of irrigation relying on remote-sensing techniques and

understanding the elevations and slopes of the topographic maps by the GIS program.

Key words: Water Harvesting, Ungauged Catchment area, remote-sensing

Cite this Article: Mohammad Amaar Hassan and Dr. Qassem H. Jalut, Mathematical

Modeling of Rainwater Harvesting System for Ungauged Catchment Area,

International Journal of Civil Engineering and Technology, 9(11), 2018, pp. 823–837

http://www.iaeme.com/IJCIET/issues.asp?JType=IJCIET&VType=9&IType=11

1. INTRODUCTION

Rainwater harvesting (RWH) is a method of collecting the runoff that occurs due to rainfall

storms and flow on the ground surface to store it in earth, tanks, or small dams where it is

utilised for a range of purposes. RWH is a free and clean source of water and it only

requires minor treatment before it is suitable for use for domestic purposes. (Khoury-Nolde,

Mohammad Amaar Hassan and Dr. Qassem H. Jalut


N., 2016). There are three components of RWH, which are detailed as follows (Kareem, I.

R., 2013):

1. Catchment area: this refers to the area from which the runoff will flow over and be

collected.

2. The surface characteristics: the runoff will be high where the surface is hard and

smooth.

3. The utilization system: this is the system used for the management and use of the

collected water. The utilization system varies depending on the purpose for which

the harvested rainwater will be used.

The aims of the study are firstly to estimate the runoff volume from the ungauged

catchment area using the SWAT model, and secondly, to suggest a location for the storage

structure (i.e. a small dam) in upper Khanaqin region.

2. APPLICATION OF GIS AND REMOTE SENSING IN

HYDROLOGICAL MODELLING

The technology of the GIS gives a wide flexible field of environment for dealing with the

input data, including analysis and displaying through different resources. Substantial research

will be conducted into GIS and remote sensing technology for spotting the land use changes

based upon the soil conservation service (SCS) model (Nayak et al., 2012). The GIS uses

different combinations of data which causes it to be outstanding and plays a vital role in the

field of hydrological modelling. Khaddor et al. (2015) studied a rainfall-runoff simulation of

flooding and predicted the peak discharge for return periods of 5, 10, 20, 50, 100, and 200

years respectively. They used GIS and remote sensing technology with hydrological model

HEC_HMS for this study in the Meghougha watershed in the north-west of Morocco.

Specifically, they employed the curve number soil conservation service (SCS), the spatial

input data like land cover/land use, soil map, and DEM, which is reflected in the accuracy of

the result. For the present study the SWAT model and ArcGIS 2016-V10.4.1 has been used

(SWAT is installed as extension with GIS). Steps of implementation have been illustrated in

Figure(1).

Mathematical Modeling of Rainwater Harvesting System for Ungauged Catchment Area


Figure 1 ArcGIS and SWAT model implementation diagram

3. CFSR CLIMATE DATA

Climate Forecast System Reanalysis (CFSR) is a modern American technology that provides

accurate climate data. It accomplished in excess of a 35-year range from 1979-2014

completed by the National Centre for Environmental Prediction (NCEP). The CFSR and high

resolution can provide data for any region in the world. It has become a leading hydrological

model, replacing the conventional rain gauge station, as it provides the best model results.

Fuka et al. (2014) and Kadhim N.K,2018 made a comparison between the CFSR and

traditional rain gauge station, in terms of the adequacy and performance of the model by

using the two input climate data (CFSR & traditional rain gage station). Tomy, T. and

Sumam, K. S. (2016) proposed using CFSR climate data and traditional rain gauge station

data individually in the SWAT model for the Karuvannur watershed in the Thrissur region of

India. After running the model, they calculated the discharge by comparing the CFSR and

traditional rain gauge station. They found the value of peak discharge remains the same for 6

stations and more, and significantly decreases when the number of gauge station is less than

6. The CFSR is employed in the present study.

4. SWAT MODEL APPLICATION FOR RUNOFF MODELLING

The Soil Water Assessment Tool (SWAT) is a physically-based watershed model used for the

prediction and simulation of runoff, sediment and agriculture chemical yield. It is recognised

and utilised throughout the world as an effective runoff model for watershed modelling

(Neitsch et al., 2011). It is evident that RWH must be adopted in the Diyala province to

avoid the expected present and future water crisis. Collecting water throughout the wet

season for multipurpose in the dry season is critical. Previous studies offer valuable

information about choosing the right input data and its’ references for the implementation of



the SWAT model in ungauged catchments are highly beneficial. The research focuses on

using remote sensing technology for preparing all required maps such as soil maps (FAO soil

map), DEM maps with 30-meter resolution and land use/land cover maps. Also, it clarifies

the method for calibrating the ungauged catchment by using the regionalization approach

(rational method).

5. FEATURES OF THE STUDY AREA

The location of the study area is in the north-eastern part of Iraq, at the border that this

country shares with Iran and including a portion of the latter country as well. The area is part

of the Diyala province, which has the following coordinates: 45°29'07" - 45°51'44" E

longitude and 34°73'54" _34°37'46" N latitude. Rainfall in the mountains represents the

primary source of water in this area. In terms of size, the area under investigation, which is

referred to as the catchment area, spans a surface that measures 701.836 m2, incorporating

portions of land from the territory of Iraq as well as from the territory of Iran. Furthermore,

with regard to ground surface elevation, the study area is not uniform but varies significantly,

with elevation of between 248 meters and 2569 meters above sea level, as was previously

mentioned. In order to gain a comprehensive understanding of the study area under

investigation and to generate a detailed picture of it, data were gathered not only in relation to

the climatic conditions in the area, but also in relation to the land cover and types of soils.

The site of the study area is shown in Figure 2.

Figure 2 location of the study area (upper Khanaqine).



From a topographical perspective, the area that is the focus of the present investigation is

predominantly mountainous in character, although the height of elevation differs

considerably; more specifically, in the portion situated on the territory of Iran, the area

consists of high mountains that reach up to 2569 meters above sea level, whereas in the

portion situated on the territory of Iraq, the area consists of small hills with a height of no

more than 240-250 meters above sea level. The digital elevation model (DEM) was employed

for the purposes of representation of the topography of the area in question; this

representation is illustrated in Figure 3 with a resolution of 30 meters. The DEM was

downloaded from the website of the United States Geological Survey (USGS)

(https://earthexplorer.usgs.gov/). This website is a valuable resource for any research such as

the one undertaken in the present study, due to the fact that it can provide the topographical

map of any region on earth on the basis of the satellite search that it incorporates.

Figure 3 DEM of the study area (USGS 2016).

https://earthexplorer.usgs.gov/



6. THE PROCESS OF CONSTRUCTION OF THE DEM

The DEM that was obtained from the USGS took the form of an image that was generated on

the basis of the data gathered by two satellites. Those two satellites encompassed the entire

area of interest for the present investigation. Geographic information system (GIS) methods

were subsequently employed in order to combine the data that were captured by the two

satellites and thus to create a single image. The use of these kinds of methods involved the

creation of a raster dataset, which then made it possible to obtain one map through the merger

of two maps. Figure 4 provides an illustration of this process.

Figure 4 Merging two DEM maps in one



7. CLIMATIC CONDITIONS, LAND COVER AND SOIL PROFILES IN

THE RESEARCH AREA

The Climate Forecast System Reanalysis (CFSR) was the climate-related system that was

employed as the source of all the data pertaining to climate that were employed for the

purposes of the present study. These data were downloaded from the website

https://globalweather.tamu.edu/ and were subsequently generated through the means of the

SWAT. The main advantage of the CFSR that sets it apart from the conventional rain gauge

station is the fact that it is much more accurate and precise, which is the reason why it is

widely suggested for use in conjunction not only with the SWAT model, but also with other

models. This is also the reason why the CFSR was chosen as the source of climate data in this

study. An additional reason was that, within the area of concern in the present study, there are

two stations of CFSR; more specifically, one of these stations of CFSR is situated in the

portion of the area on the territory of Iraq, while the other of the two CFSR stations is

situated in the portion of the area on the territory of Iran. The related coordinates are as

follows: 34.5247 south latitude, 45.2164 west longitude, 34.985 north latitude, and 45.7822

east longitude. With respect to the climatic conditions that are dominant in the area under

investigation, it is rainy and cool during the winter months, while it is dry and warm to very

warm during the summer months. The period of the summer season begins in the month of

June and ends in the month of September; during this period, the temperature can reach

extremely high levels and there is no rainfall. Meanwhile, the period of the winter season

begins in the month of September and ends in the month of May or the start of June. August

is the month when the temperature reaches its highest level, which can be up to 45°C. On the

other hand, January is the month when the temperature is at its lowest level, ranging between

0°C and 10°C maximum.

GIS methods were employed in a controlled way in order to create an accurate

representation of the manner in which the land was used as well as the manner in which the

vegetation cover was distributed throughout the land in the area in question, with bare soil

accounting for a proportion of 92.2% of the land while water accounted for a proportion of

8% of the land.

The catchment area that was investigated in the present study is divided into two major

parts; the first part is found on the territory of Iraq, while the other part is found on the

territory of Iran. DIVA-GIS provided the shapefiles for these two countries, which were

downloaded from the website http://www.diva-gis.org/. On the basis of those shapefiles, data

related to the parts of the study area located in Iraq and Iran were pooled together with the

help of a GIS method in order to gain a comprehensive picture of the characteristics of the

entire area under consideration. Furthermore, in order to ensure compatibility with and

implementation in the SWAT model, the pooled data from the two countries spanned by the

catchment area and the map of the soil distribution within the catchment area were clipped.

Moreover, in keeping with the database created by the FAO with regard to types of soils, the

catchment area under investigation was established to be made up of more than one type of

soil, ranging from clay to clay loam.

8. THE DELINEATION OF THE WATERSHED

The system for maps and geographic information known as ArcGIS was employed in

conjunction with the interface access GIS and spatial analyst in order to try to achieve the

projection and thus to form the watershed. To that end, the approach that was adopted

involved setting the drainage line based on establishing the direction of flow between the

pixels. With regard to the creation of sub-basins, there were two particular factors that were

https://globalweather.tamu.edu/

http://www.diva-gis.org/



influential in deciding how many sub-basins to be formed; one of those two factors was the

inlet and outlet that were supplied, known as the monitoring point, while the other of the two

factors was associated with the topographical properties of the area under investigation,

namely, the elevation above sea level. Figure 5 illustrates the manner in which the sub-

basins that were identified were distributed.

Figure 5 Watershed delineation

Land use, soil map and slope were the fundamental parameters for input of data in the

SWAT model in an effort to define the HRU. ArcGIS was the system that was chosen for the

purpose of projection and predefinition of each parameter. Meanwhile, the database

associated with the SWAT model was kept up-to-date with the new set of data related to

soils, as well as the set of data related to use of land; at the same time, data regarding soil

map and use of land were uploaded in a successive manner and a new classification was

formulated. In addition, the slope was characterized on the basis of the DEM that was defined

in the context of the delineated watershed via a series of five steps, ranging from 0-10% to

10-100%. All the aspects discussed above are presented and summarized in Table 1 and

Table 2, as well as in Figure 6 and Figure 7. The model simulation of the area under

investigation spanned a period of 21 years and revealed that the curve number had a value of



94, that the amount of precipitation falling in one year was 304 mm, and that the average

depth of runoff in the watershed was 100.46 mm. The values of hydrologic balance

associated with the model can be seen in Figure 8.

Table 1 Slope type of the study area

DEM resolution

(m)

Type of slope %

Area km2

30 x 30

0-10 316.713

10-20 185.025

20-30 88.022

30-40 51.473

40-9999 60.6

Table 2 Soil hydrologic types of the catchment area

FAO soil name Area km

2

Soil type

I-Rc-Xk-c-3122 188.115 clay

Xk28-b-3300

513.721 Clay-loam

Figure 6 Slopes Reclassified of Study Area



Figure 7 Soil reclassify of the study area

Figure 8 The hydrologic cycle as SWAT model simulation display



9. RAINFALL AND SURFACE RUNOFF MODEL

Over the interval of time between 1990 and 2010, the values that were obtained for the

watershed with an area of 701.836 km2 and separated into a number of 115 zones were 94 for

the average curve number and 324,226,8.13 m3for the surface runoff over the watershed area.

The relation in the figure (9) which shows the relationship and trending between rainfall and

surface runoff for 21 years, it's demonstrated that the amount of runoff is directly

proportional to the amount of rain falling, where the floods associated with rainstorms are

significantly higher due to the nature of the soil (clay- clayey loam) and vegetation-cover,

which are almost non-existent. The amount of surface runoff as a volume generated during

the rainy season may cause a large risk of locally crop damage and soil erosion. And on the

other hand, lack during the dried season explains the importance of establishing a rainwater

harvesting structure to reserve water during the flood season and to prevent the risks

associated with them and to benefit from it the dried season for different important purposes.

Figure 9 Runoff depth and Annual precipitation relation through 21 years.

9. ESTIMATION OF AMOUNT OF RAINFALL AND VOLUME OF

SURFACE RUNOFF

In order to anticipate the amount of rainfall and the volume of surface runoff that will occur

in the future, the SWAT model relies on the input data that is supplied and that includes data

related to DEM, land cover, climate, and other aspects. In the present study, the period of

time over which the model has been run to anticipate the values of the parameters of interest

was 18 years. The findings obtained from running the model revealed that the average runoff

volume anticipated to occur over the stated period was 707,905,51.97 m3. An overview of

these findings is provided in Table 3.



Table 3 Runoff prediction volume of the study area

10. The catchment outlet

The application of the mathematical model yielded a series of values that pointed to the fact

that the river served as a thread that connected the sub-basins to one another, as well as that

every sub-basin was dependent on its topographical features and precipitation flows down the

river that connected all of the sub-basins. Furthermore, it was also observed that all the sub-

basins poured from one another according to how high they were compared to sea level and

every one of them moved in the direction of the outlet of the first sub-basin, which had the

coordinates of 45° 30' 18.59" and 34° 37' 25.85". Therefore, in keeping with these findings

and taking into account the engineering standards on the design, it was concluded that the

catchment outlet was the ideal site for building the water barrier. Figure 10.

Figure 10 catchment's sub basin distribution



11. ANALYSIS OF THE SENSITIVITY OF THE SWAT MODEL

The four parameters that were employed for the purposes of this process of sensitivity

analysis were the Curve Number II, the Alpha Base Factor, the Threshold Water Depth in the

Shallow Aquifer for Flow, and the Groundwater Delay Time. Furthermore, these four

parameters were also used for the purposes of the computation of a number of subsequent

recession systems. The method that was employed in the present study in order to analyse

how sensitive the implemented SWAT model was in the context of the area under

investigation was the t-stat method. The analysis that was carried out with the help of this

method produced one result of particular significance. More specifically, this significant

result was that one of the four parameters that were used, namely, the Alpha Base Flow

Factor, presented two values, one maximum and one minimum; the maximum value was the

maximum absolute value of t, while the minimum value was the minimum absolute value of

p. What could be inferred from this result of the sensitivity analysis regarding the maximum

and minimum values was the fact that, out of the four parameters that were used, the

parameter that exhibited the highest degree of sensitivity and that therefore had the impact of

greatest magnitude was the Alpha Base Flow Factor. None of the other three parameters that

were employed were as sensitive as the Alpha Base Flow Factor in the context of the

sensitivity analysis of the SWAT model. Figure 11 and Table 4 provide more detailed

information about the aspects discussed above and about the results that were obtained.

Table 4 Global sensitivity analysis parameter values

parameter name t-state x10-2

p-valuex10-2

4:V__GWQMN.gw 40.23 68.75

3:V__GW_DELAY.gw 239.43 1.70

2:V__ALPHA_BF.gw 387.19 0.01

1:R__CN2.mgt -445.95 0.001



Figure 10 Global sensitivity analysis



12. CONCLUSIONS

From modelling the study area and analyzing the results the following can be concluded:

1. The GIS technique enables the understanding of the land slope and elevations,

which helps to predict and choose a suitable location for the dam and design the

irrigation system.

2. The volume of the runoff for the 21-year period from 1990 to 2010 is

3,242,268.13 m3.

3. The predicted runoff volume for 24-year period from 2018 to 2035 is

76,352,348.53m3.

4. The nature and topography of the study area ranged from the high mountains to

low hills, which reflects the suitability of the site for the construction of a small

dam in the proposed location at the outlet of the catchment, with coordinates 45°

30' 18.59", 34° 37' 25.85" for the collection of the water.

5. The quantity of harvested water is good, and can solve a part of the water crises in

the regions surrounding the study area specifically and for the province generally,

therefore it is necessary to construct the harvesting dam.

REFRENCES

[1] Fuka, D.R., Walter, M.T., MacAlister, C., Degaetano, A.T., Steenhuis, T.S. and Easton,

Z.M., 2014. Using the Climate Forecast System Reanalysis as weather input data for

watershed models. Hydrological Processes, 28(22), pp.5613-5623.

[2] Kareem, I.R., 2013. Artificial groundwater recharge in Iraq through rainwater harvesting

(Case Study). Engineering and Technology Journal, 31(6 Part (A) Engineering), pp.1069-

1080.

[3] Khaddor, I., Achab, M. and Alaoui, A.H., Simulation of Rainfall-Runoff using GIS,

Hydrologic Modeling System and SCS Curves Number: Application to the Meghougha

Watershed (Tangier, NW Morocco).

[4] Kadhim N. K "Evaluation of the Maximum Rainfall Magnitude in Mid-

Mesopotamian Plain by Using Frequency Factors Method” (IJCIET), Volume 9,

Issue 6, (Jun 2018)

[5] Kadhim Naief Kadhim (Estimating of Consumptive Use of Water in Babylon

Governorate-Iraq by Using Different Methods). (IJCIET), Volume 9, Issue 2, (Feb 2018)

[6] Khoury-Nolde, N., 2016. Rainwater harvesting. Zero M. Germany.

[7] Nayak, T., 2012. SCS curve number method in Narmada basin.

[8] Neitsch, S.L., Arnold, J.G., Kiniry, J.R. and Williams, J.R., 2011. Soil and water

assessment tool theoretical documentation version 2009. Texas Water Resources

Institute.

[9] Tomy, T. and Sumam, K.S., 2016. Determining the Adequacy of CFSR Data for Rainfall-

Runoff Modeling Using SWAT. Procedia Technology, 24, pp.309-316.

MATHEMATICAL MODELING OF RAINWATER HARVESTING … · RAINWATER HARVESTING SYSTEM FOR UNGAUGED CATCHMENT AREA Mohammad Amaar Hassan ... Mathematical Modeling of Rainwater Harvesting

Documents