Long-Term (2013–2024) Reconstructed All-Sky Land Surface Temperature from MODIS Integrating Machine Learning and Cloud Conditions for South Korea

This dataset provides daily all-sky 1 km land surface temperature (LST) from MODIS over South Korea from 2013 to 2024. The dataset was generated using a Light Gradient Boosting Machine (LightGBM) model by integrating MODIS LST, in situ observations from Automated Surface Observing System (ASOS) stations, Local Data Assimilation and Prediction System (LDAPS) analysis data, and multiple auxiliary variables.The dataset includes LST at the four MODIS overpass times: Terra daytime, Terra nighttime, Aqua daytime, and Aqua nighttime, hereafter referred to as TD, TN, AD, and AN, respectively. Although the dataset was originally constructed on a daily basis from January to December, winter data (December–February) are excluded from this release due to relatively lower accuracy during this season. Winter data can be provided upon request by contacting the corresponding author.Each ZIP file corresponds to a single year and contains daily all-sky LST data at the four MODIS overpass times (TD, TN, AD, and AN) in NumPy NPZ format. All LST values are provided in degrees Celsius (°C).To provide geolocation information for each pixel, latitude and longitude grids are also included in the NPZ files.