Attribute_Accuracy:
Attribute_Accuracy_Report:
In a previous study, we found remotely sensed surface temperature estimates to be accurate and highly correlated when compared to river water temperature data gathered by in situ instruments (r-square = 0.87, RMSE = 2.25 degrees C) within a large, well-mixed, Alaska river. Average daily water temperature measured by USGS stream gage over the multi-year dataset was 11.3 degrees C with a standard deviation of 5.7 degrees C. The average daily surface temperature estimate was 10.8 degrees C with a standard deviation of 6.1 degrees C. The bias (bias = [remotely sensed - in situ]) between mean daily water temperatures was -0.47 degrees C (Baughman and Conaway 2021).
Logical_Consistency_Report:
NoData indicates that, while data were recorded in the original Landsat scene, no data were available for a particular pixel due to masking of pixels that did not meet open-water, clear-sky conditions or other processing artifacts related to product derivation.
Completeness_Report:
Detection of water surface temperature was limited to open-water, clear-sky pixels within the bounds of a river's primary shoreline. Data were not consistently available over any one location due to variability of cloud conditions and river ice conditions. Additional linear data gaps within Landsat 7 scenes collected after May 31, 2003, are associated with the failure of the Scan Line Corrector (SLC). Within these scenes, open-water, clear-sky conditions existed at the time of acquisition but are still reported as NoData. The Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Emissivity Dataset (ASTER GED) provided by the NASA Land Processes Distributed Active Archive Center (LP DAAC) was used in the generation of Landsat Surface Temperature (ST) products. Where ASTER GED data is missing, there will be persistent data gaps in the Landsat U.S. ARD ST product, regardless of the scene. For the Kuskokwim River, no scenes were available or qualified for download for 1984, 1987, 1993, 1994, 1996, and 1998.
Positional_Accuracy:
Horizontal_Positional_Accuracy:
Horizontal_Positional_Accuracy_Report:
Landsat data have a 30-meter pixel resolution. Landsat Collection 2 U.S. Analysis Ready Data used in this study have a geometric accuracy of <12-m RMSE.
Lineage:
Source_Information:
Source_Citation:
Citation_Information:
Originator:
U.S. Geological Survey Earth Resources Observation and Science (EROS) Center
Publication_Date: 2023
Title: Landsat Collection 2 U.S. Landsat Analysis Ready Data (ARD)
Geospatial_Data_Presentation_Form: raster digital data
Publication_Information:
Publication_Place: Sioux Falls, South Dakota, USA
Publisher:
U.S. Geological Survey Earth Resources Observation and Science (EROS) Center
Other_Citation_Details:
Accessed in 2023 through EROS EarthExplorer.
Folder names in this data package are the ARD Tile Identifiers. The naming convention ([LXSS]_[US]_[HHHVVV]_[YYYYMMDD]_[yyyymmdd]_[CC]) can be parsed to determine the query parameters used to access the Landsat images in this data package and can be accessed from EROS EarthExplorer (via the USGS EROS image and data repository;
https://earthexplorer.usgs.gov).
[LX]: Landsat Sensor ("LT" = Thematic Mapper (TM), "LE" = Enhanced Thematic Mapper Plus (ETM+), "LC" = Operational Land Imager/Thermal Infrared Sensor (OLI/TIRS))
[SS]: Landsat satellite (04-09)
[US]: Regional grid of the U.S. ("AK" = Alaska)
[HHHVVV]: Landsat_ARD Horizontal and Vertical tile number
[YYYYMMDD]: Image acquisition date
[yyyymmdd]: Image processing date
[CC]: Landsat collection number ("02" = Collection 2)
Type_of_Source_Media: online database
Source_Time_Period_of_Content:
Time_Period_Information:
Range_of_Dates/Times:
Beginning_Date: 19840515
Ending_Date: 20221015
Source_Currentness_Reference: Image acquisition date
Source_Citation_Abbreviation: Landsat_ARD
Source_Contribution:
Processed Landsat satellite imagery and surface temperature products acquired between 1984 and 2022, used in the derivation of historical surface water temperature.
Source_Information:
Source_Citation:
Citation_Information:
Originator: Baughman, C.A.
Originator: Conaway, J.S.
Publication_Date: 2021
Title:
Comparison of Historical Water Temperature Measurements with Landsat Analysis Ready Data Provisional Surface Temperature Estimates for the Yukon River in Alaska
Geospatial_Data_Presentation_Form: journal article
Series_Information:
Series_Name: Remote Sensing
Issue_Identification: 13(12):2394
Publication_Information:
Publication_Place: online
Publisher: MDPI
Other_Citation_Details:
Baughman, C.A., Conaway, J.S., 2021. Comparison of historical water temperature measurements with landsat analysis ready data provisional surface temperature estimates for the Yukon River in Alaska. Remote Sensing 13(12):2394
https://doi.org/10.3390/rs13122394
Type_of_Source_Media: Publication
Source_Time_Period_of_Content:
Time_Period_Information:
Range_of_Dates/Times:
Beginning_Date: 19840515
Ending_Date: 20221015
Source_Currentness_Reference: Publication date
Source_Citation_Abbreviation: Baughman and Conaway 2021
Source_Contribution:
A feasibility study supporting the utilization of Landsat Analysis Ready Data surface temperature products in Alaska.
Process_Step:
Process_Description:
STEP_1: We iteratively batch-downloaded scenes from the Landsat_ARD Collection 2 dataset in EarthExplorer (
https://earthexplorer.usgs.gov/) for each tile intercepted by rivers featured in this dataset. Scenes were filtered by cloud cover (<80%), month (May-October), and only downloaded if open water could be seen in the rivers. Downloaded scenes were delivered from EarthExplorer as compressed folders (.tar format) containing multiple Landsat_ARD processed data products derived from each scene.
Process_Step:
Process_Description:
STEP_2: The Landsat_ARD product "Pixel Quality Assessment" (QA_PIXEL) includes information about Cloud, Cloud Confidence, Cloud Shadow, and Snow/Ice flags. We reclassified the QA_PIXEL raster associated with each Landsat_ARD scene so that all pixels previously interpreted as clear-sky/open-water were assigned a value of 1 and all others were assigned a 'NoData' value. For Landsat 4-7, the QA_PIXEL value was 5504. For Landsat 8-9, the QA_PIXEL value was 21952. This produced a raster for masking cloud-free, open-water surface temperature data.
Source_Used_Citation_Abbreviation: Landsat_ARD
Process_Date: Unknown
Process_Step:
Process_Description:
STEP_3: Landsat_ARD includes several Surface Temperature (ST) products generated from the thermal infrared bands and designated by "_ST_" in the filename. Using the reclassified QA_PIXEL mask from STEP_2, we masked the Landsat_ARD ST raster (B6 for Landsat 4-7, B10 for Landsat 8-9) so that only clear-sky/open-water pixels retained surface temperature values.
Source_Used_Citation_Abbreviation: Landsat_ARD
Process_Date: Unknown
Process_Step:
Process_Description:
STEP_4: We made shapefiles to capture the full historical extent of the river channels, including migration of channels and sloughs. We used the shapefiles to clip the masked Landsat_ARD ST raster from STEP_3, retaining only pixels that fell within the bounds of the river of interest. In this step, we also decoded the original Landsat_ARD surface temperature values from integer scaled-Kelvin to floating point degrees Celsius using the equation: ((ARD_SURFACE_TEMPERATURE_VALUE * 0.00341802) + 149) - 273.15).
Source_Used_Citation_Abbreviation: Landsat_ARD
Process_Date: Unknown
Process_Step:
Process_Description:
STEP_5: The Landsat_ARD product "Surface Temperature Quality Assessment" (ST_QA) provides the surface temperature product uncertainty (in Kelvin) using a combination of uncertainty values and distance to cloud values. Using the reclassified QA_PIXEL mask from STEP_2, we masked Landsat_ARD ST_QA raster so that only clear-sky, open-water pixels retained surface temperature uncertainty values.
Process_Date: Unknown
Process_Step:
Process_Description:
STEP_6: We used the shapefiles from STEP_4 to clip the masked Landsat_ARD ST_QA raster from STEP_5, retaining only pixels that fell within the bounds of the river of interest. In this step, we also scaled the original Landsat_ARD surface temperature uncertainty values back to units of degrees using the equation: ((ARD_SURFACE_TEMPERATURE_UNCERTAINTY_VALUE * 0.01).
Process_Date: Unknown