Quantifying the impact of agriculture on tail-end lakes is key to the sustainable development of mountain-oasis-desert ecosystems.This study integrated remote sensing and machine learning to explore the linkages betwe...Quantifying the impact of agriculture on tail-end lakes is key to the sustainable development of mountain-oasis-desert ecosystems.This study integrated remote sensing and machine learning to explore the linkages between the cropland in the oasis and water dynamics of the tail-end lake in the Ebinur Lake Basin.A novel data-driven model is proposed to decouple cropland evapotranspiration(ET)into natural(ET_(n))and human-induced(ET_(h))components,which helps to quantify the contributions of climate change and human activities to cropland ET.The model performed well with R^(2) values ranging from 0.88 to 0.96 from 2003 to 2019.Ebinur Lake experienced a“shrinkage-expansion”change from 2003 to 2019.Between 2003 and 2015,the surface water area(SWA)decreased from 785.91 to 478.83 km^(2) with an accumulated water storage deficit of 0.32 km^(3).After 2015,SWA rebounded to over 700 km^(2) and the accumulated water storage deficit was almost eliminated due to increased rainfall and external water transfer.Concurrently,the oasis cropland expanded from 2,871.2 km^(2) in 2003 to 4,325.7 km^(2) in 2019.The expansion of the cropland and the increased cropland ETh after 2013 raised the total water consumption of the cropland from 1.37 km^(3) in 2003 to 2.21 km^(3) in 2019.The contribution of ETh increased from 67%in 2013 to 77%in 2019,but the share of ETn decreased from 33%to 23%.To restore the SWA of Ebinur Lake to its ideal 800 km^(2),an additional water of 0.29 km^(3) would be required.Alarmingly,Ebinur Lake has resumed rapid shrinkage since 2020,and urgent action is needed to prevent further degradation.展开更多
基金supported by the National Natural Science Foundation of China(42071271,41991232,and 42471319)the Youth Innovation Promotion Association of Chinese Academy of Sciences(2022125).
文摘Quantifying the impact of agriculture on tail-end lakes is key to the sustainable development of mountain-oasis-desert ecosystems.This study integrated remote sensing and machine learning to explore the linkages between the cropland in the oasis and water dynamics of the tail-end lake in the Ebinur Lake Basin.A novel data-driven model is proposed to decouple cropland evapotranspiration(ET)into natural(ET_(n))and human-induced(ET_(h))components,which helps to quantify the contributions of climate change and human activities to cropland ET.The model performed well with R^(2) values ranging from 0.88 to 0.96 from 2003 to 2019.Ebinur Lake experienced a“shrinkage-expansion”change from 2003 to 2019.Between 2003 and 2015,the surface water area(SWA)decreased from 785.91 to 478.83 km^(2) with an accumulated water storage deficit of 0.32 km^(3).After 2015,SWA rebounded to over 700 km^(2) and the accumulated water storage deficit was almost eliminated due to increased rainfall and external water transfer.Concurrently,the oasis cropland expanded from 2,871.2 km^(2) in 2003 to 4,325.7 km^(2) in 2019.The expansion of the cropland and the increased cropland ETh after 2013 raised the total water consumption of the cropland from 1.37 km^(3) in 2003 to 2.21 km^(3) in 2019.The contribution of ETh increased from 67%in 2013 to 77%in 2019,but the share of ETn decreased from 33%to 23%.To restore the SWA of Ebinur Lake to its ideal 800 km^(2),an additional water of 0.29 km^(3) would be required.Alarmingly,Ebinur Lake has resumed rapid shrinkage since 2020,and urgent action is needed to prevent further degradation.
