Jornada Experimental Range: Albedo from AVHRR BRDFs
It has been suggested that many LiSK BRDF model inversions fail when inverted
against "noisy" AVHRR data, with unreliable and negative retrieved model
parameters leading to estimates of albedo which are out of range (i.e.
not in the range 0..1). While the preprocessing (calibration, registration,
atmospheric correction and contamination-screening) of AVHRR data are not
trivial, this is not borne out by experiments with multiscene AVHRR over
New Mexico (Chopping, Large-scale BRDF retrieval over New Mexico with a
multiangular NOAA AVHRR dataset, Remote Sensing of Environment 74(1):163-191).
When an iso-LiSparseMODIS-RossThin model is used, the vast majority of
retrieved parameters are positive in both visible and near-infared channels
and the corresponding white sky and black sky albedos are also reasonable:
Visible (VIS) and Near-Infrared (NIR) White Sky Albedos from AVHRR BRDFs
and AMBRALS v3.0
(U. of Colorado PROVE data set, reprocessed by MJC as per Remote Sensing
of Environment 74(1):163-191; AMBRALS iso-LiSparseMODIS-RossThin model
(reciprocal kernels), displayed via a 2-SD stretch, with histograms):
Some results with AMBRALS v2.4
Use Back button | Mark's BRDF page
| Disclaimer