Twenty years of change: Land and water resources in the Chindwin catchment, Myanmar between 1999 and 2019

被引：21

作者：

McGinn, Antonia J. ^{[1
]}

Wagner, Paul D. ^{[1
]}

Htike, Htike ^{[2
]}

Kyu, Khin Kyu ^{[2
]}

Fohrer, Nicola ^{[1
]}

机构：

[1] Univ Kiel, Inst Nat Resource Conservat, Dept Hydrol & Water Resources Management, Olshausenstr 75, D-24118 Kiel, Germany

[2] Myanmar Maritime Univ, Dept River & Coastal Engn, Thanlyin Township, Yangon, Myanmar

来源：

SCIENCE OF THE TOTAL ENVIRONMENT | 2021年 / 798卷

关键词：

SWAT model; Land use classification; CFSR; Data scarce region; Land use change; Interdisciplinary methods; USE CHANGE IMPACTS; CLIMATE-CHANGE; COVER CHANGES; HYDROLOGIC RESPONSE; VEGETATION CHANGES; SWAT; MODEL; CALIBRATION; SIMULATION; EVAPOTRANSPIRATION;

D O I：

10.1016/j.scitotenv.2021.148766

中图分类号：

X [环境科学、安全科学];

学科分类号：

08 ; 0830 ;

摘要：

Since 2011, Myanmar has undergone a more rapid socio-economic development, which may substantially have affected land use and land cover (LULC) and water resources. This study investigates the changes in land and water resources of the Chindwin River catchment (114,686.9 km(2)) in Myanmar over a twenty-year timespan from 1999 to 2019. The main aim of this study is to assess LULC change and evaluate its effects on the water balance and the people in the region. To this end, interviews were conducted, LULC classifications based on multitemporal multi-spectral satellite data and in-situ ground truth data were created, and a hydrologic model was built. The hydrologic model shows a reasonable performance for daily discharge simulation at the catchment outlet (percent bias between -2 and 13.2, Kling-Gupta Efficiency between 0.75 and 0.76, Nash-Sutcliffe Efficiency between 0.57 and 0.61, RMSE-observations standard deviation between 0.63 and 0.66). The LULC changes detected include a decrease in forest area of about 2%, an increase in shrubland area indicating increased degradation of the forest an increase in mining areas of 0.38%, an overall decrease in agricultural area (2.1%), but also the presence of new agricultural land pointing toward relocation of agricultural areas and an indication of an increase in settlement areas (1.5%). With the help of the hydrologic model, the most significant hydrologic impacts detected were a decrease in evapotranspiration and an increase in water yield which is correlated with the decrease of forest at the sub-catchment scale (R-2 = 0.72 and 0.46, respectively). Moreover, an increase of mining areas contributed to the increase in water yield (R-2 = 0.62). Interviews confirm that the identified LULC changes deforestation and increased mining activities contribute to major issues, e.g., water pollution, sedimentation, and changes in the river course. (C) 2021 Elsevier B.V. All rights reserved.

引用

页数：17

共 119 条

[41] Contextual land-cover classification: incorporating spatial dependence in land-cover classification models using random forests and the Getis statistic [J].