ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Evaluating rainfall datasets to reconstruct floods in data-sparse Himalayan region

Chawla, I and Mujumdar, P P (2020) Evaluating rainfall datasets to reconstruct floods in data-sparse Himalayan region. In: Journal of Hydrology, 588 .

[img] PDF
jou_hyd_588_2020.pdf - Published Version
Restricted to Registered users only

Download (5MB) | Request a copy
Official URL: https://doi.org/10.1016/j.jhydrol.2020.125090


Floods are widespread natural disasters having significant socio-economic impacts and require appropriate modelling and management strategies. Flood modelling is an enduring challenge, especially over the data-sparse mountainous regions such as the Himalayas, due to the lack of accurate rainfall measurements. Global datasets are often employed for flood modelling; however, their applicability over the Himalayan terrain is still uncertain. In this regard, the Weather Research and Forecasting (WRF) numerical weather prediction model is increasingly used to obtain reliable rainfall estimates. This work aims at evaluating the performance of different reanalysis datasets, satellite product, and the WRF model to represent heavy rainfall over the Himalayan terrain. Further, these datasets are used to drive the Variable Infiltration Capacity (VIC) hydrologic model to assess their ability to reconstruct floods in the Himalayan region. A Rainfall-Flood Skill Score (RFSS) is proposed in this work to rank different rainfall datasets in the order of their ability to represent flooding in the region. The range of RFSS can vary from -∞ to 1, with positive values as the desired value. The analysis is conducted over the upstream part of the Upper Ganga Basin, spanning from high elevation mountains to the foothills of the Himalayas for three heavy rainfall events that caused flooding in the region. Quantitative evaluation of rainfall datasets in terms of bias, RMSE, and scale errors shows high spatial variability with different datasets performing differently over various regions of the study area. The topography is observed to influence the performance of rainfall datasets. Based on the error metrics, it is found that the rainfall simulated using the WRF model exhibits least error in high elevation and valley regions of the Himalayas, whereas, Tropical Rainfall Measuring Mission (TRMM) Multisatellite Precipitation Analysis (TMPA) and Climate Forecast System Reanalysis (CFSR, in some cases) datasets can capture the rainfall in the foothills of the Himalayas. On forcing the VIC model with these datasets, the WRF model with a moderate resolution of 9 km is found to have an overall highest RFSS score of 0.97 among all the datasets and, therefore, considered most suitable for simulating floods in the study region. © 2020 Elsevier B.V.

Item Type: Journal Article
Publication: Journal of Hydrology
Publisher: Elsevier
Additional Information: The copyright for this article belongs to Elsevier.
Keywords: Climate models; Digital storage; Disasters; Errors; Floods; Landforms; Rain; Topography; Weather forecasting, Management strategies; Numerical weather prediction models; Quantitative evaluation; Rainfall measurements; Socio-economic impacts; Tropical rainfall measuring missions; Variable infiltration capacities; Weather research and forecasting, Rain gages
Department/Centre: Division of Interdisciplinary Sciences > Interdisciplinary Centre for Water Research
Division of Mechanical Sciences > Civil Engineering
Date Deposited: 29 Mar 2021 11:59
Last Modified: 29 Mar 2021 11:59
URI: http://eprints.iisc.ac.in/id/eprint/65572

Actions (login required)

View Item View Item