Landsat Science Team (LST) members meeting in early February in Phoenix, AZ, were briefed or provided input on a number of important issues involving the Landsat program—everything from updates on Landsat 9 development to the implications of Landsat 7’s orbit change.
LST February Meeting
Comprised of 20 Landsat experts from U.S. universities, private industry, and Federal and international agencies, the team visited the Northrop Grumman satellite manufacturing facility in Gilbert, AZ, to check progress on the assembly of Landsat 9.
The group also was briefed on Sustainable Land Imaging Technology (SLI-T) and discussed future Landsat mission priorities. It reviewed the status of Landsat Collection 2 processing, and learned about Landsat’s commercial cloud architecture, as well as Landsat data products and access.
Christopher “Chris” Crawford, the U.S. Geological Survey (USGS) Landsat Project Scientist and USGS co-chair of the LST, provided these highlights from the meeting.
Status of Landsat 9
Project Managers Del Jenstrom (NASA) and Brian Sauer (USGS) updated the team on the status of Landsat 9 development. The two primary instruments, Operational Land Imager-2 (OLI-2) and Thermal Infrared Sensor-2 (TIRS-2) have been delivered to Northrop Grumman for integration on the spacecraft. Work is progressing but technical issues and contention for testing facilities have resulted in the launch readiness date being moved to March 2021.
“It was a good tour,” Crawford said. “It was neat to see Landsat 9. Once the instruments complete integration, I think the next phase is getting some of the test data back.”
Landsat 7 End of Life Plans
After a review of the decaying Landsat 7 orbit, the LST recommended that once the satellite’s Mean Local Time (MLT) crossing reaches 9:15 a.m., users need to be warned explicitly that solar illumination levels may not be consistent with the historical record, and data should be used with caution.
Crawford said Landsat 7 is expected to be at 9:23 a.m. MLT by April 1, 2021, and 8:52 a.m. MLT by April 1, 2022.
“The really prevailing wisdom was, ‘Let’s just continue to acquire the data like we do today until the MLT crossing is 9:15 a.m.,’ ” he said. “I think the consensus was, ‘Let’s not worry too much about Landsat 7 until we know a firmer idea of when Landsat 9 is going up.’ ”
Global Analysis Ready Data (ARD)
Though the LST has been a strong advocate of the ARD concept, providing pathfinding input to the development of U.S. ARD products and pushing for a global ARD product, decisions on technical specifications for global ARD have still been challenging. That’s particularly true as far a recommendation on a map projection.
With USGS staff having gone through several rounds of analyzing the options for potential global map projections, the team concluded at Phoenix that the sinusoidal projection, tailored to five continents, is the best technical solution that the USGS should consider for further testing, evaluation, and prototyping.
Prototype products, or a method to generate prototype products in the recommended projections, will be made available for LST member review and evaluation. At the meeting’s conclusion, the LST mentioned that releasing a global ARD product is its top priority, and it strongly encouraged it to be part of Collection 2 if possible.
Making Use of Early Landsat Multispectral Scanner System (MSS) Data
LST members have long been encouraging the USGS to carry out investigations needed to improve the quality of the MSS collection so that the data can be more easily used with the full Landsat record. The first 12 years of Landsat—the Landsat 1-3 MSS record—cover a period in which global environmental changes were accelerating, thus making the scientific importance of the record invaluable.
Because MSS is based on earlier imaging technology, the image quality is not consistent with the Landsat 4-9 archive. Crawford said the geometry calibration/validation team has a plan to update the density of ground control points to improve the georegistration accuracy of older MSS images. Additionally, work is needed on cloud detection and atmospheric correction.
The team discussed the strict standards used for Landsat 4-8 tiered data production and whether they can be loosened to accommodate MSS data. But they also acknowledged that MSS data need to be improved and added to ARD products.
Crawford said USGS hopes to begin rolling scene-based products out of Collection 2 in late March or early April of 2020. Collection 2 will provide an upgrade to the Landsat data inventory that will make data easier to use alongside other satellite data sources, be available in a cloud-optimized format, and offer global access to some of the ARD products that Collection 1 made possible.
Collection 2 reprocessing will be implemented in the USGS’s Cloud Hosting Solutions (CHS) commercial cloud, Crawford said, with U.S. ARD Collection 2 expected to arrive later in 2020.
“We’re going to try to get the LST involved in some of the test evaluations once we actually generate cloud-optimized products,” he said. “I thought they were very warm to that idea.”
Interoperability with Europe’s Sentinel-2 program
The LST urged the USGS to consider the interoperability of Sentinel-2 with Landsat products. In the future, the integrated use of data from the two programs will become more commonplace, and so compatibility of products will have significant benefits to data users, LST members said.
In addition to the above highlighted items, many of the LST members gave presentations on their Landsat-related research. The meeting presentations are available for viewing on the USGS Landsat Mission Website under the Landsat Science Team tab at this link.