Satellite remote-sensing is used to collect important atmospheric and geophysical data at various spatial resolutions, providing insight into spatiotemporal surface and climate variability globally. These observations are often plagued with missing spatial and temporal information of Earth’s surface due to (1) cloud cover at the time of a satellite passing and (2) infrequent passing of polar-orbiting satellites. While many methods are available to model missing data in space and time, in the case of land surface temperature (LST) from thermal infrared remote sensing, these approaches generally ignore the temporal pattern called the ‘diurnal cycle’ which physically constrains temperatures to peak in the early afternoon and reach a minimum at sunrise. In order to infill an LST dataset, we parameterize the diurnal cycle into a functional form with unknown spatiotemporal parameters. Using multiresolution spatial basis functions, we estimate these parameters from sparse satellite observations to reconstruct an LST field with continuous spatial and temporal distributions. These estimations may then be used to better inform scientists of spatiotemporal thermal patterns over relatively complex domains. The methodology is demonstrated using data collected by MODIS on NASA’s Aqua and Terra satellites over both Houston, TX and Phoenix, AZ USA.

These land surface temperature data were collected by MODIS onboard NASA's Aqua and Terra remote-sensing satellites, over Houston, TX (28 June-4 July 2014) and Phoenix, AZ (11-17 Sept 2003). Raw data are available at https://lpdaac.usgs.gov/products/mod11a1v006/ and https://lpdaac.usgs.gov/products/myd11a1v006/, for the Terra and Aqua satellite, respectively. The data were collected twice daily for each satellite, yeilding a total of four satellite overpasses each day. The Aqua satellite passes over the regions at about 02:00 and 13:00 local time, while Terra overpasses at about 11:00 and 22:00 local time. These raw data were trimmed and prepared for use by Leiqiu Hu, coauthor on the corresponding manuscripts. Data for the Houston area are in a 151x151 grid of 1km^2 pixels, while data for Phoenix are in a 100x150 grid of 1km^2 pixels. Because we were interested in land surfaces only, we removed the water pixels -- lakes and inlets -- in the Houston area, so these appear as missing for all satellite overpasses. There were no lakes or inlets in the Phoenix area. In addition, the data quality was rated as poor for some pixels on certain days, mostly due to cloud cover. These poor-quality data were also removed, leaving additional missing data. Land surface topology is included for all pixels in both regions, in two separate data files. These land cover data were originally obtained from https://lpdaac.usgs.gov/products/mcd12q1v006/, then modified to fit the grid structure of the satellite readings of land surface temperature.

The zip file contains two folders, along with a README.txt file. The folder "Houston_Data_Dryad" contains all of the data for the Houston region, while the "Phoenix_Data_Drayd" folder contains all of the data for Phoenix. Each folder in "Houston_Data_Dryad" contains three types of .csv files. First, "Houston_Land_Cover.csv" contains the land cover type for each of the 1x1km pixels in the 151x151km area. The nine types of land cover include (i) barren, (ii) cropland, (iii) forest, (iv) grassland, (v) open shrubland, (vi) savanna, (vii) urban, (viii) water, and (ix) other. "Water" is the only land cover type used in the main reference paper, but we believe that the entire land cover information may be helpful in future use of this data. Second, the files "Houston_lat.csv" and "Houston_long.csv" include the latitude and longitude, respectively, of each of the 151x151 pixels. Third, the files "Houston_Temp_yyyy_mm_dd_tt/tt.csv" contain land surface temperature data (measured in degrees Kelvin) for each of the 151x151 pixels, at various times in the collection period, where "yyyy" is the four-digit calendar year, "mm" indicates the month, "dd" indicates the day of the month, and "tt/tt" indicates the time of day, measured in local standard time. For example, "Houston_Temp_2014-07-03_21/54.csv" contains land surface temperature data for the Houston region, collected at 21:54 local standard time on July 3, 2014. Data collected around 02:00 and 13:00 local time were collected by the Aqua satellite, while data collected around 11:00 and 22:00 local time were collected by the Terra satellite. Missing data (typically due to cloud cover) in these files are indicated with the value "NA". Analogous land cover, lat/long, and land surface temperature files exist in the "Phoenix_Data_Dryad" folder, with the prefix "Phoenix_" rather than "Houston_". The Phoenix .csv files are 100x150 tables representing a 100x150km grid of 1x1km pixels, rather than the 151x151km grid for Houston. One method of data use is demonstrated in the reference papers.