Santiago Mendoza Paz’s Post

View profile for Santiago Mendoza Paz, graphic

Research scientist in hidrology-hydraulics and climate change

We invite you to review our recent publication in MDPI Water, titled "Adapting to Climate Change with Machine Learning: The Robustness of Downscaled Precipitation in Local Impact Analysis." In this study, we examined the efficacy of machine learning algorithms, specifically support vector machines and random forest models, to establish non-linear relationships that facilitate the downscaling of precipitation data from a global to a local scale. This process, known as statistical downscaling, is critical for enhancing localized climate impact assessments. Given the relative novelty of these techniques in the field of downscaling, our investigation rigorously assessed their strengths, limitations, and underlying assumptions. Our findings reveal the significant potential of these methodologies, particularly in their capacity to rectify large-scale precipitation data. We identified what we term "robust changes" across multiple locations in Bolivia—defined as alterations that are consistently supported by the majority of global climate models and substantial enough to exceed the thresholds of natural variability. It is important to note that the local climatic changes observed in Bolivia are heterogeneous, reflecting the region's diverse environmental conditions. We encourage you to explore our results in detail. I am pleased to share this publication, which culminated from my final year of doctoral research at KU Leuven, in collaboration with my colleague Mauricio F. Villazón from UMSS, under the guidance of Prof. Patrick Willems. This research was made possible through funding from KU Leuven. https://lnkd.in/eEXCz_4s

Adapting to Climate Change with Machine Learning: The Robustness of Downscaled Precipitation in Local Impact Analysis

Adapting to Climate Change with Machine Learning: The Robustness of Downscaled Precipitation in Local Impact Analysis

mdpi.com

To view or add a comment, sign in

Explore topics