Estimation of Ca²⁺ wet deposition in the Northern Hemisphere by use of CNN deep-learning model

Wet deposition, the process where substances from the atmosphere are deposited through precipitation, involves components like calcium (Ca²⁺), which play a critical role in influencing ecosystem health by affecting nutrient cycling and maintaining the acid-base balance in soils and waters. Despite its ecological significance, the lack of a comprehensive dataset on Ca²⁺ wet deposition has limited our understanding of its impacts. This study addresses this gap by developing a convolutional neural network (CNN) framework to estimate Ca²⁺ wet deposition across the Northern Hemisphere from 2000 to 2022, with a high spatial resolution of 0.5° x 0.5°. The model has been rigorously validated, showing high accuracy with correlation coefficients (R²) of 0.84 for annual and 0.67 for monthly estimations. Our results identify deposition hotspots in southwestern China, Southeast Asia, and regions near the Mediterranean. Furthermore, this study quantified the neutralization factor of Ca²⁺ (NF[Ca²⁺]), a crucial indicator for evaluating how well precipitation neutralizes acidity. Notably, NF[Ca²⁺] decreased by 47.2 %, 39.6 %, and 32.2 % in 2014 compared to 2006 in Japan, China, and South Korea, respectively. Across land cover types, NF[Ca²⁺] in grasslands, wetlands, and settlements experienced reductions of 24.6 %, 31.4 %, and 12.6 %. A more rapid decrease in Ca²⁺ wet deposition compared to NO₃^– and SO₄^2- (attributed to emission control) has weakened the ability to counteract acid rain in East Asia. Our findings provide insights into Ca²⁺ wet deposition patterns and underline ecological importance, supporting future environmental management.