Global Land Surface Temperature Climatology

Land surface temperature (LST) is the temperature of the Earth's surface, measured from a satellite or other remote sensing platform. It represents the temperature of the topmost layer of the ground or vegetation, which can vary depending on factors such as time of day, season, and weather conditions. LST is an important parameter for understanding the impacts of climate change on the Earth's surface.

An LST anomaly is a measure of how much the current LS) deviates from the long-term average LST for a specific location. Unlike the LST itself, this value reveals the departure of the temperature from its "usual" state. As a critical tool, LST anomalies help detect and monitor changes in LST over time, which can be an indicator of climate change and potential environmental risks. They are particularly valuable for identifying areas susceptible to such risks.

A nine-year time series of JPSS LST data has been processed to create daily and monthly climatologies for daytime and nighttime, respectively. Notably, a major algorithm shift occurred in June 2019, transitioning from the previous surface-type based IDPS algorithm to the current emissivity-based enterprise algorithm. Accurate and consistent LST data are crucial for detecting and understanding changes in land surface temperatures over time. To address potential inconsistencies due to the algorithm change, the available datasets prior to June 2019 were reprocessed using the current algorithm, specifically focusing on data since April 2014. This reprocessing ensures the consistency and reliability of the entire dataset, enabling accurate analysis and interpretation of LST changes. The final product is a global LST composite for both daytime and nighttime with a spatial resolution of 0.05 degrees. This reprocessing effort now provides over nine years of consistent, algorithm-aligned time series data for comprehensive LST analysis.

Upon completion of reprocessing, the reprocessed time series was combined with operational dataset, forming a time series of more than eight years. A cloud screening procedure was applied to exclude possible cloud contamination. The data was divided to daytime and nighttime groups by the corresponding solar zenith angles of each pixel, i.e., pixels with solar zenith angles smaller than or equal to 85 K are categorized as daytime pixels and nighttime ones otherwise, consistent with their definition determined by the operational processing unit. All pixels falling in the same grid were grouped and one of them was selected based on certain priority. JPSS LST retrievals were associated with four-tier clear sky masks, including confidently clear, probably clear, probably cloudy, and confidently cloudy. This is used as the first priority during pixel selection, i.e., the pixel with best confidence of cloud free condition will be selected. The highest temperature was selected for daytime case and minimum for nighttime. This process aims to retain values close to the daily high/low temperature for the day.

For each month, the average temperature (mean) and its variability (standard deviation) are calculated using all available data. Additionally, the number of days with data for each grid cell is recorded to assess the reliability of the statistics due to potential cloud cover. To address cloud interference and increase flexibility in analysis, data from multiple satellites (currently SNPP and NOAA-20, with NOAA-21 coming soon) is incorporated. All data are treated as if they came from the same source. During selection, one pixel from any available satellite is chosen for each grid cell on a given day. For calculating the climatological LST average and variability, data from the same calendar month across different years are used. The number of days included in each calculation is also recorded. Finally, the monthly LST anomaly is simply obtained by subtracting the climatological LST average for the specific month from the LST for the same month and year.

Similarly, daily climatology and anomaly data were created and are routinely updated.