Machine learning in spatial predictive modelling: challenges and opportunities
Spatial modelling involves predicting and mapping the distribution of environmental variables across geographic areas, such as precipitation, and soil organic carbon. Machine learning can be a powerful tool in helping to analyze large and complex datasets, identify patterns in spatial data, and make accurate predictions based on those patterns. However, machine learning is not inherently spatial but there are several ways how to incorporate spatial aspect into machine learning. In this presentation, I will give an overview of how machine learning can be used in spatial modelling of environmental variables (e.g. soil organic carbon) based on our experiences in several practical use cases. I will also cover the main pitfalls (e.g. overfitting, interpretability) of using machine learning in spatial predictive modelling.