ORIGINAL PAPER Simulation of snowmelt runoff in ungauged basins based on MODIS: a case study in the Lhasa River basin Linghua Qiu • Jinjun You • Fei Qiao • Dingzhi Peng Published online: 21 December 2013 Ó Springer-Verlag Berlin Heidelberg 2013 Abstract It is theoretically and practically significant to conduct snowmelt runoff simulations and hydrological research for high-elevation regions. The Lhasa River basin, an ungauged basin, is a typical alpine headwater region where snowmelt runoff contributes significantly to its stream flow. In this study, the snowmelt period, defined by the snow cover curves obtained at different altitudinal zones based on Moderate-Resolution Imaging Spectrora- diometer (MODIS) and Digital Elevation Model data, occurred from March 6 to July 12 in the basin. The snowmelt processes were simulated with the Snowmelt Runoff Model (SRM) in 2002 and 2003 for calibration and validation, respectively. The coefficients of determination (R 2 ) were 0.86 and 0.87 for calibration and validation, respectively, and the Nash–Sutcliffe coefficients were both 0.80, which indicate reasonable performances in simulating hydrological processes in the Lhasa River basin. The simulated snowmelt at altitudes below 5,000 m accounts for most of the snowmelt. And the simulated snowmelt runoff contributed 3–6 % to the total runoff. The sensitivity of individual parameters was analysed and ranked as fol- lows: a and c [ C S [ C R [ T crit . In short, the SRM based on MODIS remotely sensed data performed well for the ungauged Lhasa River basin. Keywords Lhasa River basin  MODIS  SRM  Snowmelt runoff 1 Introduction In high-elevation regions, most hydrological modelling studies were based on empirical or semi-empirical theo- retical models. Moreover, thorough studies of important snowmelt factors, i.e., essential processes that affect runoff generation and confluence, are lacking (Wang et al. 2009). Snowmelt and ice melt represent important headwater sources in snowmelt areas (Ferguson 1999) and can be indicators of climate change (Stone et al. 2002). Such watershed hydrological processes as snowfall, snow cover and snowmelt, are important characteristics in high-eleva- tion areas. And snowmelt runoff is a critical element for hydrological simulations in these regions (David and Danny 2005). Various snowmelt simulation methods have been applied, e.g., the dynamic-stochastic approach (Gelfan 2010) to assess extreme snowmelt floods, or the module framework method to predict snowmelt runoff (Smith and Marshall 2010) as well as the method that consider climate with runoff to forecast melt runoff (Archer and Fowler 2008). Combination of the degree-day method and energy balance approach (Su et al. 2011), or one of them combined with another approach (Kuchment and Gelfan 1996), have been utilised for snowmelt detection. Snowmelt models are important patterns for simulating the melting process. Johnson et al. (2003) applied two models, the Hydrological Simulation Program—FORTRAN and the Soil Moisture L. Qiu  D. Peng (&) College of Water Sciences, Beijing Normal University, Beijing, China e-mail: dzpeng@bnu.edu.cn J. You State Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, China Institute of Water Resources and Hydropower Research, Beijing, China F. Qiao Chinese Research Academy of Environmental Sciences, Beijing, China 123 Stoch Environ Res Risk Assess (2014) 28:1577–1585 DOI 10.1007/s00477-013-0837-4