Relatively cloud-free Landsat scenes within ±3 days of coincident in situ measures were retained for validation across the CONUS. The Landsat surface temperature data were interpolated to 30 m to ensure alignment with the solar reflective bands in the Level-1 products, which were in turn used as inputs to the surface temperature retrieval algorithms. For lake and reservoir validation, two sets of temperature data were evaluated: land-water boundary (180 m from shore) data. Estuary data were retained and evaluated only if they were >180 m from shore. The 59 m near shore threshold was chosen to be smaller than the finer of the two thermal band spatial resolutions for both the Landsat 7 ETM+ (60 m) and Landsat 5 TM (90 m) sensors.

This dataset is associated with the following publication: Schaeffer, B., J. Iiames, J. Dwyer, E. Urquhart, W. Salls, J. Rover, and B. Seegers. An initial validation of Landsat 5 and 7 derived surface water temperature for U.S. lakes, reservoirs, and estuaries. INTERNATIONAL JOURNAL OF REMOTE SENSING. Taylor & Francis, Inc., Philadelphia, PA, USA, 39(22): 7789-7805, (2018).