Benchmarking land use change impacts on direct runoff in ungauged urban watersheds Hasan Ozdemir a,⇑ , Emre Elbas ßı b a Istanbul University, Geography Department, Physical Geography Division, 34459 _ Istanbul, Turkey b Istanbul University, Institute of Social Sciences, Geography Division, 34459 _ Istanbul, Turkey article info Article history: Received 15 January 2014 Received in revised form 7 July 2014 Accepted 4 August 2014 Available online xxxx Keywords: SCS-CN model Land use change Runoff Urban watershed abstract This paper describes the results of benchmark testing of land use change impact on direct runoff using Soil Conservation Service-Curve Number (SCS-CN) model in two ungauged neighbouring urban water- sheds (Çınar and Kadıyakuplu) in Istanbul, Turkey. To examine this impact, the model was applied to daily rainfall data using three different dated (1982, 1996 and 2012) hydrological soil groups and land use of the two ungauged urban watersheds. Finally, the impact of land use change and model perfor- mance were evaluated with the rainfall-runoff regression, the coefficient of determination and the NSE test using benchmark runoff data based on 1982 land use conditions. The results of the analysis indicate that the changing of land use types from natural surfaces to impervious surfaces has a significant impact on surface runoff. Additionally, remarkable spatial variations of the land use changes and their impact on the runoff in 1996 and 2012 were more detected in the Çınar watershed compared with the Kadıyakuplu watershed. The planning decision on land use of the watersheds, has vital role in these differences. The results of this research also reveal that change to intensive land use in urban watersheds has a signifi- cantly larger impact on runoff generation than those rainfall. Ó 2014 Elsevier Ltd. All rights reserved. 1. Introduction Urban flood events are increasing in frequency and severity as a consequence of: (1) reduced infiltration capacity due to increasing impervious surfaces; (2) increased construction in flood prone areas due to population growth; (3) the possible amplification of rainfall intensity due to climate change; (4) sea level rise which threatens coastal development and (5) poorly engineered flood control infrastructure (Hsu et al., 2000; Brown et al., 2007; Mason et al., 2003; Gallegos et al., 2009). These factors will contrib- ute to an increase in urban flood risk in the future, and as a result, an understanding of and improved modelling for urban flooding have been identified as a research priority (Wheater, 2002). Land use and its change have significant impact on the nature of the runoff and related hydrological characteristics of a watershed. This runoff process is especially vital in urban areas due to the increase in impervious surfaces. Land use changes can affect runoff generation and flow patterns by altering hydrological factors such as interception, infiltration and evaporation and thus causes changes in the frequency and intensity of surface runoff and flood- ing (Weng, 2001; Ali et al., 2011). Therefore, a better understand- ing and assessment of land use change impacts on the watershed hydrologic process, is of great importance for predicting flood potential and the mitigation of hazard, and has become a crucial issue for the planning, management, and sustainable development of the watershed (Vorosmarty et al., 2000; Defries and Eshleman, 2004; Wang et al., 2007; Chen et al., 2009). The impact of land use on runoff has been documented in the literature from different perspectives, including the analysis of runoff changes in response to land use changes, the prediction of runoff for future climatic and land use conditions, and the study of the spatial variability of urbanization and its effect on runoff generation, among others (Tang et al., 2005; Burns et al., 2005; Bari et al., 2005; Lin et al., 2007; Chen et al., 2009; Gholami et al., 2010; Ali et al., 2011; Zhang et al., 2012). In most of the stud- ies, past and present land use conditions are used as input to event-scale hydrological models to determine the watershed’s hydrological response to observed or hypothetical rainstorms (De Roo et al., 2003; Camorani et al., 2005; Olivera and DeFee, 2007). All these models are generally calibrated and validated with observed data in the watersheds studied. However, many river catchments are ungauged for stream flow data even in urban areas. The Soil Conversation Service-Curve Number (SCS-CN) method (SCS, 1956, 1964, 1971, 1985, 1993) is one of the most popular methods for computing the direct surface runoff for given rainfall http://dx.doi.org/10.1016/j.pce.2014.08.001 1474-7065/Ó 2014 Elsevier Ltd. All rights reserved. ⇑ Corresponding author. E-mail address: ozdemirh@istanbul.edu.tr (H. Ozdemir). Physics and Chemistry of the Earth xxx (2014) xxx–xxx Contents lists available at ScienceDirect Physics and Chemistry of the Earth journal homepage: www.elsevier.com/locate/pce Please cite this article in press as: Ozdemir, H., Elbas ßı, E. Benchmarking land use change impacts on direct runoff in ungauged urban watersheds. J. Phys. Chem. Earth (2014), http://dx.doi.org/10.1016/j.pce.2014.08.001