Note: These datasets are part of our research, which was published in Environmental Science & Technology. Please cite this paper when you use these datasets: Yuanhui Zhu, Soe W. Myint, Danica Schaffer-Smith, Daoqin Tong, Yuyu Zhou, Yubin Li, and Rebecca L. Muenich. Rising Environmental Inequalities and Their Relationship to Racial and Socioeconomic Disparities in the US Southwest, 2025, Environmental Science & Technology. https://doi.org/10.1021/acs.est.4c14369 For any questions or concerns, contact Yuanhui Zhu at jta90@txstate.edu This study aims to provide insight into the US Southwest social and environmental inequality problems by combining high-resolution ECOSTRESS data, including Land Surface Temperature (LST), Evaporative Stress Index (ESI), and actual Evapotranspiration (ETa) with sociodemographic data at the block group level acquired from US Census. ESI and ETa represent drought and consumptive water use, respectively. Further, disparities of environmental changes over the past two decades in connection with races/ethnicities are explored using Landsat-based LST and ET from 2000 to 2020 across major US Southwest cities in light of global climate changes. We narrow our investigations to the summer months, including June, July, and August, when environmental issues are more pronounced during the day, such as heat-related mortality and morbidity and higher water consumption. This dataset covers major US Southwest cities. The dataset includes social-environmental data at the block level and remotely sensed environmental data. The description for each individual dataset is listed as follows, T01(T01_dataset_meta_race_ethnicity_economy_env_southwest.csv): This dataset provides statistics on race/ethnicity, social economy, and arithmetic mean of environmental variables at the block group level. T02: These datasets provide processed environmental data, including Daytime Land Surface Temperature (LST), Actual Evapotranspiration (ETa), and Evaporative Stress Index (ESI) images at 70m resolution. T03: These datasets include annual Landsat-based summer LST, and ETa means at 30m resolution from 2000 to 2020. The images of LST and ETa in 2012 are missing because Landsat datasets in 2012 are unavailable. We collected all available Landsat images with cloud cover lower than 75% in the summer from 2000 to 2020 to produce the dataset. The high-resolution (30m) LST and ETa are processed on Google Earth Engine (GEE). There are a total of 8860 scenes (including Landsat 4, 5, 7, and 8) from 2000 to 2020 to cover all major US Southwest urban areas. We also include two sample images of 2020 LST and ETa for demonstration. T04: These datasets contain LST and ETa Sen's slope results, which represent the median rate of change of LST and ETa from 2000 to 2020 on an annual basis. T05: The boundary of the major US Southwest urban areas