B43A-0541
A Harmonized Landsat-Sentinel-2 Surface Reflectance product: a resource for Agricultural Monitoring

Thursday, 17 December 2015
Poster Hall (Moscone South)
Martin Claverie1,2, Junchang Ju3, Jeffrey G Masek2, Eric Vermote2 and Christopher Owen Justice4, (1)University of Maryland College Park, College Park, MD, United States, (2)NASA Goddard Space Flight Center, Greenbelt, MD, United States, (3)Universities Space Research Association Columbia, Columbia, MD, United States, (4)University of Maryland College Park, Geographical Sciences, College Park, MD, United States
Abstract:
The combination of Landsat and Sentinel-2 data offers a unique opportunity to observe globally the land every 2-3 days at medium (<30m) spatial resolution. The Harmonized Landsat-Sentinel-2 (HLS) project is a NASA initiative aiming to produce surface reflectance data from Landsat and Sentinel-2 missions and to deliver them to the community in a combined, seamless form. The HLS will be beneficial for global agricultural monitoring applications that require medium spatial resolution and weekly or more frequent observations. In particular, the provided opportunity to track crop phenology at the scale of individual fields will support detailed mapping of crop type and type-specific vegetation conditions.

To create a compatible set of radiometric measurements, the HLS product relies on rigorous pre- and post-launch cross-calibration (Landsat-8 OLI and Sentinel-2 MSI) activities. The processing chain includes the following components: atmospheric correction, cloud/shadow masking, nadir BRDF-adjustment, spectral-adjustment, regridding, and temporal composite. The atmospheric correction and cloud masking is based on the OLI atmospheric correction developed at NASA-GSFC and has been adapted to the MSI data. The BRDF-adjustment is based on a disaggregation technique using MODIS-based BRDF coefficients. The technique has been evaluated using the multi-angular acquisition from the SPOT 4 and 5 (Take5) experiments. The spectral-adjustment relies on a linear regression that has been calibrated and evaluated using synthetic data and surface reflectance processed from a large number of hyperspectral EO-1 Hyperion scenes. Finally, significant effort is placed on product validation and evaluation.

The delivered data set will include surface reflectance products at different levels:

  • Using the native gridding, i.e. UTM, 30m for Landsat-8, and UTM, 10-20m for Sentinel-2
  • Using a common global gridding (Sinusoidal, 30m)
  • Temporal composite (Sinusoidal, 30m, 5-day)

During the first year of operation of Sentinel-2A, the HLS will be prototyped over a selection of 30 sites that includes some of the JECAM sites, Aeronet sites and Cal/Val sites. Then, the HLS spatial coverage will be increased as more Sentinel-2A data become available.

<< Previous Abstract | Next Abstract