Rapid Indo-Malaysian Land Cover and Land Use Change over the last 35 Years

The Indo-Malaysian region, which covers much of East, South, and Southeast Asia, is widely considered a hotspot of rapid land cover and land use change (LCLUC). Yet there is little consensus about the rates and magnitudes of such change, which creates uncertainties in regional carbon fluxes, long-term environmental change, and therefore the future land use policies that should be pursued. Here we use temporal Convolutional Neural Networks (TempCNNs) based on NOAA AVHRR GIMMS Normalized Difference Vegetation Index third generation (NDVI3g) to generate a spatiotemporally consistent LCLU dataset for nearly 35 years (1982-2015), validated against Google Earth and Landsat imagery with 80% accuracy. Our results both confirm and complicate estimates from earlier work that relied on decadal, rather than interannual, changes in regional land cover. Consistent with previous work, we find forest decreases in mainland and maritime Southeast Asia along with increases in South China and South Asia. We find cropland expansion was a driving force for deforestation in mainland Southeast Asia with increasing shrublands playing a secondary role, suggesting widespread forest degradation in this region. In contrast to earlier work, we find that South China's increasing forest cover came from shrubland, rather than grassland, conversion. The explicit interannual LCLUC patterns, rates, and transitions identified in this study provide a valuable data source for studies of environmental change and climate in this region.