Socio-environmental risk in Peru: Identification, characterization, and categorization of 1874 districts in 2019, using machine learning and spatial econometrics

Abstract

The environmental crisis due to climate change has forced many States to direct efforts towards environmental transition to reduce the probability of occurrence of a situation with a negative impact on their population or environment. Peru is no exception. In this sense, the need arises to identify and categorize its districts according to a certain socio-environmental risk. Faced with this challenge, a multistage quantitative methodology was developed and implemented, which made use of both machine learning (supervised and unsupervised) and spatial econometrics. The results of this methodology, visualized through emerging risk indixes, evidenced the existence of 165 districts considered socio-environmental risk zones (SERZ, in Spanish known as ZRS), mostly located in the coastal strip. Finally, it is concluded that the pattern and replicability of urban development model in Peru is currently not coherent with efforts towards environmental conservation and preservation.