 |
 |
The bidirectional reflectance distribution function (BRDF) characterizes the angular distribution of reflected solar radiation as a function of viewing direction. Different surfaces reflect solar radiation in very different ways. Snow and sea ice surfaces, for example, reflect radiation in a nearly isotropic (Lambertian) manner; oceans are characterized by bright reflection in the forward (specular) direction, known as sunglint; many vegetation surfaces are characterized with enhanced reflection in the antisolar (backscattered) direction, known as the 'hot spot;' and water clouds are characterized by enhanced reflection in the forward scattering direction but with noticeable glory (backscatter) and rainbow reflectance patterns.
The Cloud Absorption Radiometer (CAR) has been used for over 10 years to obtain the spectral and angular distribution of reflected solar radiation from a wide variety of surfaces, including: (i) Atlantic Ocean and Persian Gulf water with sunglint, (ii) dense forest and forested wetland, (iii) dense smoke layers, (iv) snow-covered and melt-season sea ice, (v) snow-covered and summer tundra, (vi) Saudi Arabian Desert, (vii) Arctic stratus clouds and Namibian stratus clouds with glory and rainbow, (viii) salt pans of Namibia and Botswana, Africa, (ix) and savanna and woody savanna of Brazil, South Africa, Botswana, and Zambia. From these measurements it is possible to correct for atmospheric scattering from molecules and aerosols above and below the aircraft, and to calculate the atmospherically corrected reflectance of the underlying surface. From these measurements, it is possible to compute the spectral albedo of these natural surfaces. The MODIS satellite sensor has also been used to derive the spectral albedo of all land surfaces of the globe every 16 days at a 1' resolution (approximately 2 km at the equator). This has been accomplished by using the MODIS spectral albedo measurements that are already atmospherically corrected, and filling in for all missing data due to cloud cover, especially year-round in the tropics and seasonally over India and South-East Asia (monsoon season), and due to persistent snow cover during the long winters in Siberia and North America. This filling in process utilizes the temporal fit of nearby cloud-free pixels of the same ecosystem classification (e.g., savanna, woody savanna, broadleaf forest) to replace nearly half of the yearly data and generate global maps that are spatially, spectrally, and temporally complete. This value-added global surface reflectance product is of value not only for cloud retrieval work in which the underlying surface reflectance must be estimated, but also for analysis of AERONET inversions of aerosol optical properties, assessment of the impact on urban environments on local climate, and for land-surface climate models. Contact: Eric Moody
|
||||||||||||
|
|