关于HLS项目的光谱响应函数是否公布，若没公布则是不是可以用OLI的光谱响应函数近似替代
Translation:
Regarding whether the spectral response function of the HLS project has been published, if not, can it be approximated by the spectral response function of OLI?
Spectral response function of the HLS project
-
LP DAAC - dgolon
- User Services
- Posts: 181
- Joined: Tue Dec 03, 2024 2:37 pm America/New_York
- Endorsed: 2 times
Re: Spectral response function of the HLS project
Hello @gaosen We are looking into your question and will report back when we have an answer. Thanks -- Danielle
Subscribe to the LP DAAC listserv by sending a blank email to lpdaac-join@lists.nasa.gov.
Sign up for the Landsat listserv to receive the most up to date information about Landsat data: https://public.govdelivery.com/accounts/USDOIGS/subscriber/new#tab1.
Sign up for the Landsat listserv to receive the most up to date information about Landsat data: https://public.govdelivery.com/accounts/USDOIGS/subscriber/new#tab1.
-
LP DAAC - jwilson
- User Services
- Posts: 385
- Joined: Mon Sep 30, 2019 12:39 pm America/New_York
Re: Spectral response function of the HLS project
@gaosen The pdf is too large to attach. See section 4.4 comments on how the MSI (Sentinel-2) and OLI (Landsat) relative spectral response functions compare and the discusses how that relates to HLS.
Below are the section 4.4 comments:
HLS V2.0.
4.4.Bandpass adjustment
The spectrally corresponding OLI and MSI bands (Table 1) have similar but not the same relative spectral responses (RSR). Sensor cross calibration puts the TOA measurements from all sensors on the same scale by comparing measurements at a small number of calibration sites and adjusting for many factors including the different RSR, i.e., the bandpass effects (Teillet et al., 2007). OLI and MSI are cross calibrated well (Barsi et al., 2018; Lamquin et al., 2019). But away from the cali-bration sites, the surface reflectance spectra can be quite different, and when they are convolved with the slightly different RSR of OLI and MSI, slightly different TOA measurements are expected even under the same observing conditions. And the retrieved surface reflectances are affected in turn.
The HLS processing adjusts the Sentinel-2 MSI surface reflectance obtained after the atmospheric correction and the BRDF correction, with the OLI bandpass as the reference for the seven spectrally corresponding bands. The HLS V2.0 adjustment applies the same algorithm used for HLS V1.4 (Claverie et al., 2018); the algorithm details were described in Claverie (2023). In short, simulated Landsat and Sentinel-2 surface re-flectances were generated by convolving the 10-nm Hyperion surface reflectances with the respective RSR of Landsat 8 OLI and Sentinel-2 MSI, and were used to train a group of bandpass adjustment tech-niques, and finally a linear regression model was selected for its modest accuracy and its simplicity (Claverie, 2023). The linear regression model coefficients are summarized in Table 5.
Landsat 8 and 9 are very similar in spectral response and are cross calibrated (Kaewmanee et al., 2023), but subtle difference exists be-tween them, like for the two MSI sensors. Adjustment of Landsat 9 is not implemented in this version of HLS but will be considered in the next version.
Here is the link to the paper: https://www.sciencedirect.com/science/article/pii/S0034425725001270
The-Harmonized-Landsat-and-Sentinel-2-version-2-0-s_2025_Remote-Sensing-of-E.pdf
Let me know if you are able to locate the paper referenced.
Thank you,
LP DAAC
Below are the section 4.4 comments:
HLS V2.0.
4.4.Bandpass adjustment
The spectrally corresponding OLI and MSI bands (Table 1) have similar but not the same relative spectral responses (RSR). Sensor cross calibration puts the TOA measurements from all sensors on the same scale by comparing measurements at a small number of calibration sites and adjusting for many factors including the different RSR, i.e., the bandpass effects (Teillet et al., 2007). OLI and MSI are cross calibrated well (Barsi et al., 2018; Lamquin et al., 2019). But away from the cali-bration sites, the surface reflectance spectra can be quite different, and when they are convolved with the slightly different RSR of OLI and MSI, slightly different TOA measurements are expected even under the same observing conditions. And the retrieved surface reflectances are affected in turn.
The HLS processing adjusts the Sentinel-2 MSI surface reflectance obtained after the atmospheric correction and the BRDF correction, with the OLI bandpass as the reference for the seven spectrally corresponding bands. The HLS V2.0 adjustment applies the same algorithm used for HLS V1.4 (Claverie et al., 2018); the algorithm details were described in Claverie (2023). In short, simulated Landsat and Sentinel-2 surface re-flectances were generated by convolving the 10-nm Hyperion surface reflectances with the respective RSR of Landsat 8 OLI and Sentinel-2 MSI, and were used to train a group of bandpass adjustment tech-niques, and finally a linear regression model was selected for its modest accuracy and its simplicity (Claverie, 2023). The linear regression model coefficients are summarized in Table 5.
Landsat 8 and 9 are very similar in spectral response and are cross calibrated (Kaewmanee et al., 2023), but subtle difference exists be-tween them, like for the two MSI sensors. Adjustment of Landsat 9 is not implemented in this version of HLS but will be considered in the next version.
The-Harmonized-Landsat-and-Sentinel-2-version-2-0-s_2025_Remote-Sensing-of-E.pdf
Let me know if you are able to locate the paper referenced.
Thank you,
LP DAAC