Abstract:
Within Kesem watershed land use is undergoing major changes due to pressures of human
activities. Changes in land use have potentially large impacts on water resources by causing
more surface runoff, decreased water retention capacity, loss of wetland and drying of river.
In this study, both the Soil and Water Assessment Tool (SWAT) model and Mann-Kendall
(MK) statistical time series analysis for measured and simulated streamflow were applied to
understand the streamflow variability and land use dynamics effect on hydrology of Kesem
watershed. Land use change detection was done using remote sensing techniques and the maps
were processed using ERDAS Imagine 2014 and ArcGIS10.1 software. From the land cover
change analysis results it was found that there has been a substantial decline of forest lands,
shrub lands, grass lands and drastic expansion of agricultural land. The SWAT modeling
results showed that an increase of streamflow by 23.2% comparing the two land use maps
(1993 versus 2005). The analysis also revealed that flow during the wet months has increased
by 36.4% while the flow during the dry season decreased by 33.6%. The MK test has been
applied to mean annual, seasonal, 1- and 7-days annual minimum and maximum flows. The
MK- test demonstrates that in the case of 1-day maximum flow, no significant trend is
noticeable; however, the extreme low flows indicators (e.g. 1-day & 7-day minimum) and dry
seasonal flows exhibited statistically significant decreasing trends. Generally, the combined
results of the SWAT model and the statistical tests revealed that land use change has caused a
significant increase on mean annual streamflow and decreasing dry season flows of the studied
watershed during the last three decades. The identified result is important to inform optimal
water resource management and to plan and manage water resources development within the
watershed in a sustainable manner.
