e–JUST – Environmental Justice using Urban Scalable Toolkit: Development and Enhancements

e-JUST - Environmental Justice using Urban Scalable Toolkit is a sophisticated web-based system designed to support decision-making related to air quality and heat stressors in the Chicago Metropolitan Region. It offers users the ability to explore a wide range of spatiotemporal data, including land surface temperature measurements from NASA satellites like ECOSTRESS (70-m resolution), and MODIS (1-km resolution). The system also integrates hourly temperature data from meteorological stations and climate forecast data for variables such as temperature, NO2, PM2.5, O3, and CO, produced using urban climate models. To enhance data accuracy, we refine environmental data down to the street level through advanced machine-learning algorithms for spatiotemporal interpolation. This allows us to deliver predictions that are not only more precise but also more relevant to the needs of local communities. e-JUST further empowers users by enabling the exploration of socio-demographic data from the 2020 US Census, facilitating the study of correlations between climatic and socio-demographic factors, and supporting decision-making for populations at risk from extreme heat, poor air quality, and environmental disadvantages. Over the past year, e-JUST has undergone significant advancements. It now supports various data formats, including CSV, NetCDF, and JSON, an essential capability given the prevalence of NetCDF in climate models and satellite data. The system also offers more spatial integration methods, such as average, maximum, minimum, and weighted average. Additionally, the user interface has been refined to be more intuitive, allowing for the simultaneous exploration of multiple spatial layers simplifying visual comparisons. The tool also includes the capability to calculate forecast verification metrics, such as RMSE and ME, helping users assess how closely forecasts align with reality. Users can examine relationships between variables over time through customized graphs like scatter plots and gain insights into variable behavior with heat matrices. For more in-depth correlation analysis, we have introduced the calculation of Bivariate Moran’s I over space. We have also enhanced the user interface to offer more flexibility, allowing flexible configuration of interaction types tailored to each data type which is helpful to a diverse audience, including experts, researchers, and the general public. Such customized data visualizations and functionalities within weather forecasts can inform decision-making and environmental justice efforts in the region.