Deep learning: To better understand how human activities affect the
value of ecosystem services
Abstract
The increase in human activities is one of the important factors
affecting the value of ecosystem services. However, understanding of the
driving mechanisms of human activities is limited. We established a deep
learning model to approximate the ecosystem service value (ESV) of
Nanjing City using 23 socioeconomic factors. A multi-view analysis was
then conducted on feasible impact mechanisms using model disassembly.
The results indicated that factors such as the proportion of ecological
waters in the land-use structure and secondary industry output value had
their own independent effects on ESV. Other intrinsically related
factors, for instance, industrial water consumption and industrial
electricity consumption, were likely to be composited together to affect
ESV.