Machine Learning for High-Resolution Downscaling in the Hawaiian Islands
Active
By Climate Adaptation Science Centers
December 31, 2023
Project Overview
Climate change is expected to change rainfall patterns on pacific islands like Hawaiʻi, but current global models lack the resolution to accurately predict local rainfall. Researchers supported by this Pacific Islands CASC project will use machine learning models and historical data to improve rainfall predictions and create detailed rainfall maps for Hawaiʻi that can be used to better understand how climate change will impact the region’s water resources.
Project Summary
Climate change will likely shift precipitation patterns on tropical islands, including Hawaiʻi, with significant consequences for water resources. Currently, models of global climate change lack the resolution needed to model the processes that create most of Hawaiʻi’s rainfall. Specialized models, including statistical models trained on historical data, are needed to generate better predictions of how global changes will affect local climate and precipitation in the region.
This project will develop machine learning methods to predict precipitation at locations where no measurement data is available, using rainfall measurements (or downscaled projections of future rainfall) from nearby locations. Unlike traditional methods that interpolate between measurement stations, this project will use machine learning to incorporate information about the physical features of the landscape to build better predictions.
Using these improved spatial interpolation models, this project will create high-resolution, accurate historical rainfall maps. The project will also test the method for projecting future rainfall and compare predictions to existing statistical downscaling models. These rainfall maps, both historical and future, will be shared through the Hawaiʻi Climate Data Portal to help resource managers and decision makers better understand climate change effects on water resources in the state of Hawaiʻi.
Climate change is expected to change rainfall patterns on pacific islands like Hawaiʻi, but current global models lack the resolution to accurately predict local rainfall. Researchers supported by this Pacific Islands CASC project will use machine learning models and historical data to improve rainfall predictions and create detailed rainfall maps for Hawaiʻi that can be used to better understand how climate change will impact the region’s water resources.
Project Summary
Climate change will likely shift precipitation patterns on tropical islands, including Hawaiʻi, with significant consequences for water resources. Currently, models of global climate change lack the resolution needed to model the processes that create most of Hawaiʻi’s rainfall. Specialized models, including statistical models trained on historical data, are needed to generate better predictions of how global changes will affect local climate and precipitation in the region.
This project will develop machine learning methods to predict precipitation at locations where no measurement data is available, using rainfall measurements (or downscaled projections of future rainfall) from nearby locations. Unlike traditional methods that interpolate between measurement stations, this project will use machine learning to incorporate information about the physical features of the landscape to build better predictions.
Using these improved spatial interpolation models, this project will create high-resolution, accurate historical rainfall maps. The project will also test the method for projecting future rainfall and compare predictions to existing statistical downscaling models. These rainfall maps, both historical and future, will be shared through the Hawaiʻi Climate Data Portal to help resource managers and decision makers better understand climate change effects on water resources in the state of Hawaiʻi.
- Source: USGS Sciencebase (id: 667da0ded34e67892486505c)
Project Overview
Climate change is expected to change rainfall patterns on pacific islands like Hawaiʻi, but current global models lack the resolution to accurately predict local rainfall. Researchers supported by this Pacific Islands CASC project will use machine learning models and historical data to improve rainfall predictions and create detailed rainfall maps for Hawaiʻi that can be used to better understand how climate change will impact the region’s water resources.
Project Summary
Climate change will likely shift precipitation patterns on tropical islands, including Hawaiʻi, with significant consequences for water resources. Currently, models of global climate change lack the resolution needed to model the processes that create most of Hawaiʻi’s rainfall. Specialized models, including statistical models trained on historical data, are needed to generate better predictions of how global changes will affect local climate and precipitation in the region.
This project will develop machine learning methods to predict precipitation at locations where no measurement data is available, using rainfall measurements (or downscaled projections of future rainfall) from nearby locations. Unlike traditional methods that interpolate between measurement stations, this project will use machine learning to incorporate information about the physical features of the landscape to build better predictions.
Using these improved spatial interpolation models, this project will create high-resolution, accurate historical rainfall maps. The project will also test the method for projecting future rainfall and compare predictions to existing statistical downscaling models. These rainfall maps, both historical and future, will be shared through the Hawaiʻi Climate Data Portal to help resource managers and decision makers better understand climate change effects on water resources in the state of Hawaiʻi.
Climate change is expected to change rainfall patterns on pacific islands like Hawaiʻi, but current global models lack the resolution to accurately predict local rainfall. Researchers supported by this Pacific Islands CASC project will use machine learning models and historical data to improve rainfall predictions and create detailed rainfall maps for Hawaiʻi that can be used to better understand how climate change will impact the region’s water resources.
Project Summary
Climate change will likely shift precipitation patterns on tropical islands, including Hawaiʻi, with significant consequences for water resources. Currently, models of global climate change lack the resolution needed to model the processes that create most of Hawaiʻi’s rainfall. Specialized models, including statistical models trained on historical data, are needed to generate better predictions of how global changes will affect local climate and precipitation in the region.
This project will develop machine learning methods to predict precipitation at locations where no measurement data is available, using rainfall measurements (or downscaled projections of future rainfall) from nearby locations. Unlike traditional methods that interpolate between measurement stations, this project will use machine learning to incorporate information about the physical features of the landscape to build better predictions.
Using these improved spatial interpolation models, this project will create high-resolution, accurate historical rainfall maps. The project will also test the method for projecting future rainfall and compare predictions to existing statistical downscaling models. These rainfall maps, both historical and future, will be shared through the Hawaiʻi Climate Data Portal to help resource managers and decision makers better understand climate change effects on water resources in the state of Hawaiʻi.
- Source: USGS Sciencebase (id: 667da0ded34e67892486505c)