LNext SNR Note.pdf
PDF 514 KB Posted
- Attached to
- Landsat Next Federal contract opportunity
- Solicitation number
- RFI2020LandsatNext
About this file
This document contains a request for information from the National Aeronautics and Space Administration regarding a Landsat Next mission concept. The Landsat Next mission would collect superspectral land observations with higher spatial resolution than previous Landsat missions, including increased resolution for applications such as agriculture, ecology, urban analysis, and water resources. Additional targeted spectral bands would improve the suitability of data for applications involving surface water quality, cryospheric science, geology, and agriculture including crop water consumption. The mission would also acquire additional thermal infrared bands to allow improved estimation of surface temperature by constraining emissivity. Spatial, geometric, and radiometric requirements would reflect Landsat 8/9 heritage with signal-to-noise ratio comparable to Landsat 8 OLI performance once aggregated to 30 meters. The Landsat Next operations are planned to begin in the late 2020s.
View the file
Other files for this federal contract opportunity
| File | Type | Posted |
|---|---|---|
| Landsat Next Draft SMRD.pdf | ||
| Attachment A - RFI Tables.pdf |
On GovTribe
Work with this file on GovTribe
- Download the original file
- Contacts named in this file
- Similar government files
- Ask GovTribe AI about this file
Text version
Notes on Landsat Next Signal-to-Noise (SNR) and Correction to SMRD Table 3 (November 9, 2021)
It has been pointed out that Landsat Next SNR requirements for the Landsat 8 heritage bands appear to represent an increase in performance over the L8/OLI as-built performance once aggregated to heritage 30m resolution.
The increased SNR in the new specification arises from the increase in the reference radiance associated with using the COPERNICUS S2/MSI reference radiance (L.ref) instead of the LDCM/L8/OLI typical radiance (L.typ). In most cases, the reference radiance increased by a factor of 3 to 5.
Using the OLI radiometric response and noise model, which is very close to a shot-noise limited performance (SNR ~ sqrt(signal) for higher radiance levels), we computed how OLI would have responded to the L.ref at 30m. From there we took signal and noise and scaled it to what we would expect at the new specification spatial sampling/GSD.
For example, the CA band, which is still at 30m GSD, is specified with a reference radiance level (Lref) of 129 W/m2/sr rather than the OLI Ltyp value of 40 W/m2/sr. This increased the required SNR from the OLI SNR of 235:1 to 495:1.
For the Blue band, the 30m SNR went from 262:1 to 880:1. However, the new 10m GSD requirement required a lower SNR in the native 10m pixel in order to maintain the OLI SNR when aggregated to 30m. Assuming uncorrelated random noise and an aggregate of 9 10m pixel to obtain a 30m pixel, this gives an SNR of 293:1 at 10m GSD:
30m Signal = 9 * 10m signal 30m noise = SQRT (9 * (10m noise)^2) = 3* 10m noise 30m SNR = 3 * 10m SNR
Finally, in reviewing the SNR values (SMRD Table 3) we did discover some minor errors in the scaling to the new Lref values. Corrected values are shown below (bold text, green background). The effect is a small reduction in SNR for affected bands, although a slightly bigger SNR reduction for the Cirrus band.
Table 3. Landsat Next Image Data Radiometric Performance Requirements
File details come from the government source that posted it. Updated .