Iranian (Iranica) Journal of Energy & Environment Watershed Hydrological Responses to Changes in Land Use and Land Cover at Hangar Watershed, Ethiopia

The various water resources project planning and implementation need the knowledge of factors influencing watershed hydrology. Therefore, this study analyzed Hangar watershed hydrological responses to land use/land cover change (LULCC) from 1987 to 2017. In addition to the LULC data, the input data used for the soil and water assessment tool (SWAT) model simulation were the DEM data, soil data, and climatic data. The model calibrated and validated using measured streamflow data of 13 years (1990-2002) and nine years (2003-2011), respectively. The result after simulation indicated that the annual total water yield of the watershed decreased. The annual simulated stream flow through the study period is increased for wet and short rainy season whereas, decreased for dry season. An increase of wet season flow may result in flooding, and the reduction of dry season flow may affect water scheme practice. Therefore, this study enables the concerned body to curve the changes in LULC towards increasing vegetation cover so that, surface runoff that contributes to wet season flow will be reduced and infiltration that supply groundwater from which dry season/base flow contributed will be increased.


INTRODUCTION 1
Land use/land cover change affects the different hydrological components like; interception, infiltration, and evapotranspiration, thereby influencing soil moisture content, runoff generation (both process and volume), and streamflow regimes [1,2]. Climate models have even shown that the land use and land cover change affect global precipitation and temperature patterns [3], which influences the hydrological process. The spatial and temporal variability of watershed resources (particularly land cover change and climatic change) have a significant influence on the quantity and quality of river water flow [4]. Many studies performed in different parts of the country, for instance, [5] in Headstream of Abbay Watershed; [6] in northern Ethiopian highlands addressed a common concern as the water resource degradation brought about by the decrease in the area under natural vegetation and its conversion into other types of land use that are human-managed systems. Human-induced landuse changes such as deforestation, afforestation, and agricultural and urban development within the river basin can affect the hydrological cycle [7]. Human health and welfare, *Corresponding Author Email: abdiwak7@gmail.com (A. W. Galata) food security, and industrial developments are dependent on adequate supplies of suitable water; however, water resources affected by many parameters [8]. Both conversion and modifications of land use and land cover have critical environmental consequences through their impacts on soil, water, biodiversity, and microclimate, and hence, contribute to watershed degradation [4]. Both climate and land use and land cover change have a great influence on the hydrological response of a watershed [1]. Land-use changes in a watershed can influence water supply by altering hydrological processes such as infiltration, groundwater recharge, base flow, and runoff [9]. Its influence is direct on climate and water resources on the ground. Land under little vegetation cover [10], is subjected to high surface runoff, low water retention, and low infiltration rate. The performed studies about the factors that could affect the hydrological process at the watershed level are not much as in the largest basins of the country. Studies of LULC dynamics at the subwatershed level are rare in Ethiopia [11]. To predict the demands for different water resources schemes, enough studies should carry out about the factors affecting the watershed. However, no study carried out about factors that affect the hydrology of the watershed behavior and their relation to land use/land cover changes of Hangar Watershed, which can be a relevant consideration in the design of integrated watershed management and of appropriate sustainable land management practices, strategies, and policies. Since the study watershed located in an agricultural area, the change in land use and land cover continued unless the factors facilitating these changes identified and measures need to take recommended. However, at the study area, no software-based land use and land cover changes determination carried out. To fill this gap, the hydrological responses of the Hangar watershed to the changes in land use/land cover evaluated by using the SWAT model.

Description of the study area
Hangar River watershed located in west-central Ethiopia. The river emerges from the Horo Guduru Wollega zone near Jardaga Jarte district, and it flows south-west to join Didessa River, which is a tributary of the Blue Nile (also called the Abbay River basin in Ethiopia). Hangar enters the Didessa approximately halfway between the town of Nekemt and the village of Cherari at a latitude and longitude of 9°35′N and 36°2′E, respectively. It has several tributaries that cover an area of nearly 7673.87 km 2 . The topography or elevation of the watershed ranges from 844 to 3207 m above mean sea level. Generally, the Hangar River watershed geographically located between 36 o 31' 41" to 37 o 06' 50" East longitude and 9 o 41'58" to 9 o 59' 56" North Latitude ( Figure 1). The regional geology of the study area developed from three types of geological terrains. These are Quaternary sediments, Paleozoic to Mesozoic rock, Precambrian rock (from youngest to oldest). Most of the study area is covered with intrusive Precambrian rocks mainly granite with coarsegrained texture and massive, which is overlaid by thick black to brownish cotton soil. Climatic elements like rainfall, temperature, relative humidity, sunshine, and wind can affect by geographic location and altitude. As per the data collected from the National Metrological Service Agency (NMSA), the study area receives heavy rainfall from June to September and experiences a limited amount of rainfall for the left seven months. In the study area, the average maximum temperature experienced in the February, March, and April, whereas the average minimum temperature occurred in the September, October, and November.

Data collections and sources
The dataset collected from primary and secondary sources. Primary data are the ground truth data about the LULC and gained from the study area by using different methods such as; interviewing with those who are living at the site, discussing with others who have information about the field and collected with GPS for the recent period during field observation. Whereas, secondary data are recorded data, collected from different sources. All-weather data collected from the National Meteorological Service Agency (NMSA) of Ethiopia. Land use and land cover data of 1987 (Landsat-5

Data analysis
Since the study area could not be covered with one image, more images of Digital Elevation Model (DEM) of 12.5 by 12.5 m resolution downloaded and mosaicked with the aid of Arc GIS 10.6 before extracting the area of interest. The SWAT model needs full daily weather data to analysis and generates the result. The collected missed daily rainfall and temperature data from the National Metrological Service Agency filled by Xlsat 2018 program, where multiple leaner regression used to fill missed daily rainfall data from neighboring stations and missed maximum and minimum daily temperature data filled by average multiple imputation methods. Inconsistency of climatic data could happen during record because of changes in conditions, changes in instrumentation, changes in gauge location, and changes in observation practices. Before using any weather data, it is necessary to analyze and checks whether it is consistent or not. For this particular study, the consistency of recorded data for four stations checked by double mass curve and no need for corrections because they correlated. The three stations (Alibo, Hangar Gute, and Gelila) contain only precipitation and temperature (minimum and maximum) data. However, Nekemte station contains all climatic data such as precipitation, temperature (minimum and maximum), sunshine, relative humidity, and wind speed. Therefore, sunshine, relative humidity, and wind speed data generated for Alibo, Hangar Gute and Gelila stations from Nekemte station. The parameters required for weather generator calculated using software programs PCP STAT.exe and dew02.exe. The program PCP STAT.exe using daily precipitation calculated the statistical parameters of daily precipitation data. Whereas, the program dew02.exe calculated the average daily dewpoint temperature per month using daily air temperature and humidity data. The calculated parameters for weather generator adjusted and added into the SWAT weather database table.

SWAT model setup
The SWAT model designed to predict the impact of land management practices on water, sediment, and agricultural chemical yields in large complex watersheds with varying conditions over long periods [12]. The SWAT model proceed by sequential procedures to give output for which past procedure is an input for the next one. Looking for the next task without properly completing one of these steps is impossible. After completion of SWAT database preparation, the first procedures in the SWAT model is to create a new project or DEM set up of having identified folder in which the whole work could executed. The watershed delineation interface in Arc SWAT separated into five sections, including DEM Set Up, DEM-based Stream Definition (flow direction and accumulation and drainage network generation), Outlet, and Inlet Definition, Watershed Outlet(s) Selection and Definition and Calculation of Sub-basin parameters. In this study, the smaller area (7600 ha) provided to get 61 subbasins of the Hangar river basin, and outlet is defined, in which it later taken as a point of calibration of the simulated flows. The multiple scenarios that account for 15% land use, 15% soil, and 15% slope threshold combination give a better estimation of streamflow [13]. As the percentage of land use, slope, and soil threshold increases, the actual evapotranspiration decreases due to eliminated land-use classes [14]. Taking the objective of the study into consideration and paying attention to characteristics of HRUs as the key factors affecting the streamflow, land use, soil and slope class threshold of 10%, 15%, and 15% used respectively. Hence, the Hangar River basin results in 196 HRUs in the whole basin. The prepared soil layers, classified LULC and slope layers, and delineated Watershed by Arc SWAT overlapped 100%. The input to the model finalized, and the output generated and read after running the model in the SWAT simulation. For this study, the SWAT model was run with the meteorological data of 1987 to 2017 by keeping three years (1987)(1988)(1989) for the warm-up period to avoid the impacts of the initial conditions of the model.

Validation of the model
For the catchment with longtime series, split sample test is involved [15] for which one part used to calibrate the model, and the second part is used for testing (validating) if calibrated parameters produced simulations which satisfy goodness-of-fit tests. Therefore, since it has thirty-one years of data, a split-sample test was applied in this watershed for which measured streamflow data of 22 years was scaled 60%

Watershed hydrological responses to changes in LULC
The study indicated that (Table 1) area could be easily detached, and transported to downstream than covered land with vegetation, which would be resulted in increased sediment load. For this study, increased surface runoff has resulted in sediment load increment. Reduction of total aquifer recharge is resulted from increased surface runoff, which reduces infiltration capacity of the soil; thereby percolation of water from the soil to recharge deep aquifer decreased. The expansion of agricultural land and built-up area over other land covers results in the increase of surface runoff following rainfall events and causes alteration in soil moisture conditions and groundwater storage. The water infiltrated into the ground to recharge the shallow aquifer reduced. Therefore, the change in the components of streamflow due to LULCC expected to decrease dry season discharge, which mostly comes from base flow (shallow aquifer contribution) and increases discharge during the wet months, which supplied from surface runoff. The finding of the study is compatible with other studies carried out in different parts of the country for instance, by Mengistu [16] in  (Figure 6). The increased cultivated land and builtup area and extraction of vegetation covers also expected to become the reason for these changes. Since land cover such as forest, grassland, and rangeland decreased, surface runoff increased that contributed to the increment of wet and short rainy season streamflow. The infiltration rate of the watershed reduced due to the expansion of the built-up area. The reduction in infiltration rate decreased shallow aquifer from which dry season streamflow contributed. The low contributed shallow aquifer resulted in dry season streamflow reduction. The comparison of simulated streamflow for the LULC of the three periods summarized in Table.2. The finding of the study is consistent with other studies. For example, the result of a study by Mengistu [16] in Hare watershed indicated that the mean monthly discharge for wet months had increased by 12

CONCLUSIONS
The land use and land cover changes have significant impacts on the functioning of socioeconomic and environmental systems. In Ethiopia, most parts of the regions are vulnerable to problems concerning food production that mostly affects rural livelihood, mainly due to an increase in population on the one hand and inappropriate management of resources on the other hand. The SWAT model used the result of LULCCs to evaluate the hydrological responses of the watershed to changes in LULC. The SWAT-CUP used for sensitivity analysis of parameters, calibration, and validation. It found that CN2, SURLAG, and CANMX are the most three top sensitive parameters in the study area. For both calibration and validation, the SWAT model performed correctly, having the value of NSE, PBIAS, and coefficient of determination (R 2 ) in a very good range. Generally, the study revealed that the expansions of cultivated land and built-up area and the extraction of the forest, grassland, and rangeland during the 1987 to 2017 periods had decreased the average annual total water yield contribution of the watershed, lateral flow, percolation from the soil, evapotranspiration, aquifer recharge, and dry season streamflow. Conversely, the LULC changes had increased surface runoff, total sediment yield, wet and short rainy season streamflow.

ACKNOWLEDGMENT
I sincerely thank my principal advisor Dr. Ing Tamene Adugna and my co-advisor, Mr. Megersa Kebede, for their advice during the study and constructive comments on the manuscript.