MODIS
Emissivity (Sobrino et al., 2003)
Emissivities are estimated for day-time acquisitions from FVC and MODIS band 1 information (ρ1), following the methodology presented by Sobrino et al. (2003). These emissivities correspond to MODIS thermal bands 31 and 32, and are estimated differently depending on the vegetation proportion within a given pixel.
These emissivities are expressed as average emissivity ε (for bands 31 and 32) and spectral difference of emissivities Δε for different pixel characteristics:
- Vegetation (NDVI > 0.5)
ε=0.99 ; Δε=0
- Mixed pixel (0.2 ≤ NDVI ≤ 0.5)
ε=0.971+0.018(FVC) ; Δε=0.006(1-(FVC))
- Bare soil (NDVI < 02)
ε=0.9832-0.058ρ1 ; Δε=0.0018-0.060ρ1
In the case of night-time acquisition, such method cannot be implemented due to the lack of solar radiation, therefore the emissivity estimates during the previous day are reprojected to a lat/lon grid, averaged and reprojected back to the night-time acquisition configuration for further calculations.
FVC (Fraction of Vegetation Cover; Gutman and Ignatov, 1998)
FVC is estimated from NDVI parameter following Gutman and Ignatov (1998) for day-time acquisitions only, as a normalization of NDVI between standard bare soil and dense vegetation values. In the case of MODIS, these values are respectively 0.15 and 0.90 (Camacho et al., 2006). Therefore,
References
Gutman, G. & Ignatov, A. (1998). The derivation of the green vegetation fraction from NOAA/AVHRR data for use in numerical weather prediction models, International Journal of Remote Sensing, 19, 8, 1533-1543.
Camacho de Coca, F., Jiménez-Muñoz, J.-C., Martínez, B., Bicheron, P., Lacaze, R. & Leroy, M. (2006). Prototyping of fCover product over Africa based on existing CYCLOPES and JRC products for VGT4Africa, Proceedings of the 2nd RAQRS symposium, 722-727, Torrent, 22-25 September 2006
Sobrino, J. A., El Kharraz, J. & Li, Z. L. (2003). Surface temperature and water vapour retrieval from MODIS data, International Journal of Remote Sensing, 24, 5161-5182.
