1. Baseline Global Observation Scenario SDCG-9, Session 3 Gene Fosnight (USGS); Frank Martin Seifert (ESA) 25 Feb 2016, Frascati

2. Session Introduction Overview of 2016 Work Plan Outcomes Overview of Landsat, Sentinel, CBERS, and ISRO missions SDCG-9 Frascati, Italy February 22rd – 26th 2016 2

3. Outcome 1: Acquire Data Multiple annual Global Coverages by 2016 of the world’s forested areas – 2016: Multiple global annual coverages of the world’s forested areas from a suite of core mission sensors Optical core missions with free and open data policies – Landsat 7 and Landsat 8 – operational - global – Sentinel-2a – commissioning and operational - global – CBERS-4 – 2014 launch and 2015 commissioning - regional Radar core missions with free and open data policies – Sentinel-1a – operational - global – Sentinel-1b – 2016 launch and commissioning – global – ALOS and ALOS-2 – PALSAR annual Mosaics SDCG-9 Frascati, Italy February 22rd – 26th 2016 3

4. Outcome 2: Data Access Efficient and effective global flows of data – 2016: Complete and review global data flow study – 2016: Begin transition to Analysis Ready Data products Optical: Surface Reflectance products Radar: Orthorectified and slope corrected Optical and Radar – Tiled products – Composited products SDCG-9 Frascati, Italy February 22rd – 26th 2016 4

5. Outcome 3: Data to Information Global coverage with consistent information products – Identify Space Agency and expert partner information product initiatives relevant to GFOI and MDG. GA Data Cubes ESA Thematic Exploitation Platforms (TEP) USGS Land Change, Monitoring, Assessment and Prediction (LCMAP) FAO SEPAL – Intercalibration and interoperability studies SDCG-8 Bonn, Germany September 23rd – 25th 2015 5

6. Landsat 7 & 8 Status SDCG-9, Session 3 Gene Fosnight (USGS) 25 Feb 2016, Frascati

7. USGS Core Data Stream Report Landsat 7 and Landsat 8 Status Collection Management USGS Sentinel-2 hub SDCG-9 Frascati, Italy February 22rd – 26th 2016 7

8. Landsat 7 & 8 acquisitions SDCG-9 Frascati, Italy February 22rd – 26th 2016 8 Landsat 8 Lifetime acquisitions Landsat 7 Lifetime acquisitions Landsat 7 & 8 2015acquisitions

9. Landsat 7 status - 2015 Median: 443 images/day Max: 529 images/day SDCG-9 Frascati, Italy February 22rd – 26th 2016 9

10. Landsat 8 status - 2015 Mid latitude (57°) – 421 images/day – Reject 1.5 image/day – 99.6% High latitude – 226 images/day – Reject 16 images/day – 93.3% SDCG-9 Frascati, Italy February 22rd – 26th 2016 10

11. Landsat 7 & 8 combined Best Cloud Cover percent cloud cover <=10>10 & <=20>20 & <=30>30 & <=40>40 & <=50>50 Number of Scenes437512122400 SDCG-9 Frascati, Italy February 22rd – 26th 2016 11

12. Archive Distribution Products SDCG-9 Frascati, Italy February 22rd – 26th 2016 12

13. Collection Definition Progress The USGS defined three basic categories of products – NRT (Near-real time) – products that are processed using ancillary data such as predicted ephemeris or bumper mode parameters that may be improved by reprocessing – Tier 1 – products that meet the criteria for the collection definition (i.e. enable time-series stacking, <11.9m RMSEr) – Tier 2 – products that do not meet the criteria for the collection definition and have been processed using the best known ancillary data A single collection (i.e. “collection 1”) for all sensors (excluding MSS) as opposed to a separate collection per sensor

14. Collection Definition Study Findings Summary Radiometric variability is not a factor – Operational land imager (OLI) temporal uncertainty is better than 0.3% on average Based on on-board calibrator – ETM+ and TM are better than 2% Based on top of atmosphere reflectance measured over pseudo- invariant calibration sites Geodetic accuracy vary by sensor, source data type, the quality of PCD, and level to which the data have been processed (L1T, L1GT, L1G) – Some data (e.g. TMA and LGAC) are of lower quality due to poor quality or missing PCD MSS is highly variable 14

15. Near-real Time (NRT) Products Three Modes of NRT data collection that do not satisfy geometric accuracy 1.ETM+ products generated using predicted ephemeris data 2.ETM+ products generated that utilize predicted bumper mode calibration coefficients 3.OLI_TIRS products that utilize a preliminary line of sight (LOS) model based on “estimated” position of the scene select mirror. TIRS coefficients will initially be updated periodically, initially quarterly, increasing in frequency to twice per month after sufficient telemetry has been collected These products will be made available for download as soon as they are processed and will roll off the download cache once they become Tier-1 or Tier-2 collection products 15

16. Tier-1 Collection Definition Summary Need to be geometrically corrected to enable multi- spectral time-series stacking – Geodetic accuracy threshold of less than 11.9m radial root mean square error (RMSEr) based on a post fit analysis of data relative to the Global Land Survey (GLS) 2000 ground control – Results in about 57% of OLI_TIRS – A higher percentage of OLI science are collected over areas without ground control (Antarctic, Coastal Areas, Islands, higher cloud cover) – L1GTs can’t perform post-fit verification to GLS so they are part of the TIER-2 category 73% of all ETM+ 60% of all TM The intent is to make the full Tier-1 collection available for immediate download

17. Tier-2 Product  Products that do not meet the definition of the collection and have been processed using the best known ancillary data  Definitive ephemeris  Latest bumper mode parameters  Latest TIRS LOS coefficients Data will be made available at the highest processing level through the Tier-2 collection (some low-demand data may be available near-line) 17

18. Landsat Global Archive Consolidation (LGAC) Over 3.5 Million scenes collected so far Critical goals are data access and data preservation Ingest of Pakistan TM data nearing completion Received and ingested TM data from India Received additional Phase II TM and ETM+ data from ESA Received MSS and TM data from Thailand and Saudi Arabia Continued work with stations to finish delivering data L7 ETM+ image from CUB (Brazil) L5 TM image from KIS (Sweden)

19. Sentinel-2 from EROS Release awaiting final approval Need to switch from science hub to International hub – current throughput only allows us to pull 30% of new data Releasing as tiles SDCG-8 Bonn, Germany September 23rd – 25th 2015 19