Satellite observations and modeling to understand the Lower Mekong River basin streamflow variability

Ibrahim Nourein Mohammed; John D Bolten; Raghavan Srinivasan; Venkat Lakshmi

doi:10.1016/j.jhydrol.2018.07.030

Satellite observations and modeling to understand the Lower Mekong River basin streamflow variability

J Hydrol (Amst). 2018 Sep:564:559-573. doi: 10.1016/j.jhydrol.2018.07.030. Epub 2018 Jul 24.

Authors

Ibrahim Nourein Mohammed¹, John D Bolten², Raghavan Srinivasan³, Venkat Lakshmi⁴

Affiliations

¹ Science Applications International Corporation, Hydrological Sciences Laboratory, NASA Goddard Space Flight Center, Mail Code 617.0, Greenbelt, MD 20771, USA.
² Hydrological Sciences Laboratory, NASA Goddard Space Flight Center, Mail Code 617, Greenbelt, MD 20771, USA.
³ Spatial Sciences Laboratory, Department of Ecosystem Science and Management, Texas A&M University, College Station, TX 77843, USA.
⁴ School of Earth Ocean and Environment, University of South Carolina, Columbia, SC 29208, USA.

Abstract

In this work, we have used the Soil & Water Assessment Tool (SWAT) to examine streamflow variability of the Lower Mekong River Basin (LMRB) associated with changes in the Upper Mekong River Basin (UMRB) inflows. Two hypothetical experiments were formulated and evaluated for the LMRB, where we conducted runoff simulations with multiple inflow changes that include upstream runoff yield increase and decrease scenarios. Streamflow variability of the LMRB was quantified by two streamflow metrics that explain flow variability and predictability, and high flow disturbance. The model experiments were performed for the Lower Mekong River Basin with identical climate, soil, and other watershed characteristics data. Remote sensing precipitation (Tropical Rainfall Measurement Mission, TRMM, and Global Precipitation Measurement mission, GPM), meteorological data as well as spatial data that include a digital elevation model, newly developed soil information (Harmonized World Soil Database, HWSD), and land use and land cover were processed as input to the LMRB model simulations. Observed daily streamflow data along the Lower Mekong River from Chiang Sean, Thailand to Kratie, Cambodia were used for calibration and validation. Our work results suggest that the Lower Mekong River streamflow is highly variable and has a low predictability (Colwell index of about 32%). We found that releasing more water from upstream Mekong during rainfall months by 30% would result in a reduction in the Lower Mekong streamflow predictability by about 21%. This reduction in predictability is mainly attributed to a decrease in the Contingency index. Our work shows that the ability to predict floods/droughts at the Lower Mekong River would be reduced if there is any anticipated change (i.e., increase/decrease) from UMRB releases. Our results also show that releasing more flows from the upstream Mekong would also affect flood duration and the frequency of flood occurrences downstream. The results of this work thus help to quantify the sensitivity of streamflow variability at the Lower Mekong River Basin to upstream anthropogenic changes.

Keywords: Flooding; Mekong River; Remote sensing; SWAT; Streamflow predictability; Streamflow variability.

Abstract

Grants and funding