Comprehensive assessment of land use / land cover change and associated carbon emissions and uptake in the Mekong Drainage Basin by time series analysis of MODIS and Landsat data

Abstract:

Land use land cover (LULC) change caused by human activities is one of the main sources of anthropogenic carbon emission and drivers of climate change. The Mekong Region is the most dynamic but relatively understudied in terms of LULC change in the past decades. This study presented a comprehensive assessment of LULC change for the Mekong Drainage Basin over 2001-2019 using time series analysis (CCDC) of MODIS data. The overall accuracy of the LULC data product is 74.4 ± 1.9% (82.1 ± 1.7% if merging tree dominated classes). A novel M-CCDC enhancement process which fuses time series analysis with existing MODIS data products was able to improve the overall accuracy by 5.6%. Two of the largest components of LULC change in this region is conversion to plantation, estimated at 33617 ± 7342 km2 and agricultural expansion estimated at 14915 ± 4682 km2 over 2001-2016. Carbon emissions and uptake associated with the LULC change activities were estimated using a spatiotemporal carbon bookkeeping model. Total carbon emissions over 2001-2019 is estimated at 12.8 ± 2.3 Tg C yr-1 with an additional 81.5 ± 15.2 Tg C to be released over time after 2019. Total carbon uptake for the same period is estimated at 6.3 ± 0.7 Tg C yr-1. This study provided the first spatially and temporally continuous estimation of LULC conversions and associated carbon fluxes for the Mekong Drainage Basin. The capability of mapping LULC and LULC change using MODIS data is largely limited by the coarse spatial resolution of MODIS. Comparison of the LULC data product with a prototype results from time series analysis using Landsat data on Google Earth Engine suggested that using input data with higher spatial resolution have the potential to capture the sub-MODIS-pixel scale LULC dynamics.