Spatial patterns of China's carbon sinks estimated from the fusion of remote sensing and field-observed net primary productivity and heterotrophic respiration

Abstract

Accurately assessing the carbon sink and spatial distribution pattern of China's terrestrial ecosystems is of great significance to the implementation of climate change and carbon neutrality strategy. However, the views of various studies are still very controversial due to the differences in carbon sink estimation methods and data sources. In this study, vegetation net primary productivity (NPP) and ecosystem heterotrophic respiration (Rh) estimation models were constructed based on machine learning methods by fusing multisource data, such as remote sensing and ground observation data. The magnitude and spatial pattern of carbon sink in China from 2000 to 2018 were then revealed, and the carbon sink capacity of various ecosystems was quantitatively assessed. The main conclusions include the following: (1) The use of scale-matched carbon input and output data can help reduce the system error in carbon sink estimation. (2) China's terrestrial ecosystem carbon sink since the twenty-first century is approximately 0.458 Pg C/yr, which is equivalent to 22.72% of China's anthropogenic carbon emissions. (3) Deciduous forest has a higher carbon sink capacity than evergreen forest, while coniferous forest has a more stable carbon sink capacity than broad-leaved forest. The magnitude and spatial distribution of carbon sink in China reported in this study provides a scientific reference for achieving carbon neutrality and sustainable development.