Citation

Fernando NS, Shrestha S, Kc S, Mohanasundaram S. Prog. Disaster Sci. 2022; 13: e100212.

Copyright

(Copyright © 2022, Elsevier Publishing)

DOI

10.1016/j.pdisas.2021.100212

PMID

unavailable

Abstract

In recent years, the damages caused by flooding has been severe globally, and several research studies indicated extreme precipitations and changes in land-use plays a crucial role. The hydrological and climate impact studies in data-scarce regions are relatively challenging, and several global dataset products have aided in overcoming them. However, their accuracy and reliability vary from climatic regions to the topography of the land surface. Therefore, this study employed global dataset products for precipitation and land use to identify the major flood driver in tropical, subtropical, and temperate regions where severe flooding had occurred in the past decade. The study evaluated the performances of the PERSIANN-CDR, PERSIANN-CCS and PERSIANN precipitation products, and ESACCI-LC land-use product to develop a statistical relationship among the land-use and extreme precipitation variables using the Multiple Linear Regression technique. The result shows that the PERSIANN-CDR estimates were more accurate than others in the selected study basins. The statistical model showed that the combined contribution of both land-use and precipitation to the flood (R2) are 73.9%, 66.7% and 37.4%, for the Mun River Basin (MRB), Thailand, the Bagmati River Basin (BRB), Nepal and the Missouri Little Sioux (MLSB) Basin, USA, respectively. Moreover, it correlated with the flood (R) by 85.9%, 81.7% and 61.1% in the MRB, BRB, and MLSB, respectively. Additionally, the results indicated that the major cause of flooding in MRB and BRB is likely to be the changes in precipitation, while land-use change is likely to be the major cause in the MLSB. The result from the study shall be useful for the researchers, practitioners, and decision-makers in determining the applicability of a suitable precipitation product in data-scarce regions, visualise the major cause of flooding and plan the flood risk management strategies accordingly by minimising the exposure and maximising the resiliency for possible future events.

Language: en

Keywords

ESACCI-LC; Flood driver; Multiple linear regression; PERSIANN