Title:

Feasibility of satellite vicarious calibration using HYPERNETS surface reflectances from Gobabeb and Princess Elisabeth Antarctica sites

Publication Year:
2024
Abstract:

The HYPERNETS project developed a new hyperspectral radiometer (HYPSTAR®) integrated in automated networks of water (WATERHYPERNET) and land (LANDHYPERNET) bidirectional reflectance measurements for satellite validation. In this paper, the feasibility of using LANDHYPERNET surface reflectance data for vicarious calibration of multispectral (Sentinel-2 and Landsat 8/9) and hyperspectral (PRISMA) satellites is studied. The pipeline to process bottom of atmosphere (BOA) surface reflectanceHYPERNETS data to band-integrated top of atmosphere (TOA) reflectances and compare themto satellite observations is detailed. Two LANDHYPERNET sites are considered in this study: theGobabebHYPERNETS site inNamibia (GHNA) and Princess Elizabeth Base in Antarctica (PEAN). 36 near-simultaneous match-ups within 1 h are found where HYPERNETS and satellite data pass all quality checks. For the Gobabeb HYPERNETS site, agreement towithin 5%is foundwith Sentinel-2 and Landsat 8/9. The differenceswith PRISMA are smaller than 10%. For theHYPERNETS Antarctica site, there are alsoa numberofmatch-upswith good agreement towithin 5%for Landsat 8/9. The majority show notable disagreement, i.e., HYPERNETS being over 10% different compared to satellite. This is due to small-scale irregularities in the wind-blown snow surface, and their shadows cast by the low Sun. A study comparing the HYPERNETS measurements against a bidirectional reflectance distribution function (BRDF) model is recommended. Overall, we confirm data from radiometrically stable HYPERNETS sites with sufficient spatial and angular homogeneity can successfully be used for vicarious calibration purposes. Keywords: HYPERNETS, hyperspectral, validation, surface reflectance, uncertainty, fiducial reference measurements, earth observation, vicarious calibration.

Publication Title:
Frontiers in Remote Sensing
Volume:
5
Number:
1323998
Item Type:
Journal Article
Language:
en