Assessment of the hydrological impact of land use/cover changes in a semi-arid basin using the SWAT model (case of the Oued Saïda basin in western Algeria)

Anouar Hachemaoui,Abdelkader Elouissi,Benali Benzater,Sahnoun Fellah

DOI: https://doi.org/10.1007/s40808-022-01422-6

2022-05-16

Modeling Earth Systems and Environment

Abstract:Algeria is considered a country at risk in the index of vulnerability to climate change. In terms of water resources, there has been a significant decrease in water supply in the western part for three decades. This has led to changes in agricultural practices and the viability of natural species, as well as socio-political conditions and anthropogenic action at the watershed scale, which directly affect the water cycle. This study aims to quantify the impact of land use/land cover change on the hydrological response of the Oued Saida basin using the Swat model from 1998 to 2005. Satellite image time series of 1987 and 2002 were used for land use and cover from. A supervised classification approach, using the maximum likelihood classification method was used for the land cover of 1987 and 2002. In addition, post-classification was made to detect changes in land use/land cover. FAO soil data 30 m spatial resolution ASTER DEM data and 1998 to 2005 climate data set were combined as input data to the SWAT model land cover map (2002), with. First for the calibration phase, then those from 1987 was incorporated into the model as the scenario used for the assessment of land use change impacts. Evaluation of the accuracy of classified images was made with the error matrix. The global precision (the kappa coefficient) found for the 1987 (2002) image is 89.5% (95.9%). As a result, evolutionary analysis of land use change from 1987 to 2002 showed a decrease in agricultural area (9.20%), an increase in forest land (5.42%) and an increase in urbanized areas (2.77%). The results of the model calibration using the SWAT-CUP software’s SUFI-2 algorithm were satisfactory with a Nash Sutcliffe (NSE) efficiency coefficient of 0.56 and an R2 of 0.57. In addition, parameters related to soil properties and land use are the most sensitive. Changes in land cover have affected the runoff, increasing the average annual (monthly) peak flow by 2.5 to 3.5% (0 to 6%) at the expense of the seepage water and groundwater recharge. These changes are explained by socio-economic, political and anthropogenic conditions. The results obtained provide useful information about observed and land use trends and affected hydrological behaviors and can be a tool to assist decision-making in the development and management of the catchment area in a global vision.