Spatial GR4J Conceptualization of the Tamor Glaciated Alpine Catchment in Eastern Nepal: Evaluation of GR4JSG Against Streamflow and MODIS Snow Extent (2016)

Please fill the following information to request the publication in hardcopy. We will get in touch with you shortly.

* are required.

Snow and glacial melt processes are an important part of the Himalayan water balance. Correct quantification of melt runoff processes is necessary to understand the region's vulnerability to climate change. This paper describes in detail an application of conceptual GR4J hydrological model in the Tamor catchment in Eastern Nepal using typical elevation band and degree-day factor approaches to model Himalayan snow and glacial melt processes. The model aims to provide a simple model that meets most water planning applications. The paper contributes a model conceptualization (GR4JSG) that enables coarse evaluation of modeled snow extents against remotely sensed MODIS snow extent. Novel aspects include the glacial store in GR4JSG and examination of how the parameters controlling snow and glacial stores correlate with existing parameters of GR4J. The model is calibrated using a Bayesian Monte Carlo Markov Chain method against observed streamflow for one glaciated catchment with reliable data. Evaluation of the modelled streamflow with observed streamflow gave Nash Sutcliffe Efficiency of 0.88 and Percent Bias of <4%. Comparison of the modelled snow extents with MODIS gave R2 of 0.46, with calibration against streamflow only. The contribution of melt runoff to total discharge from the catchment is 14-16% across different experiments. The model is highly sensitive to rainfall and temperature data, which suffer from known problems and biases, for example due to stations being located predominantly in valleys and at lower elevations. Testing of the model in other Himalayan catchments may reveal additional limitations. This article is protected by copyright. All rights reserved.
Year: 2016
Language: English
In: Hydrological Process, Online p.

Related links:


 Record created 2016-09-23, last modified 2016-09-23