Nimbus-7 SMMR brightness temperatures have been reprocessed in a form that simplifies the data handling associated with estimation and analysis of global land parameter fields. Spatial and temporal analyses of the brightness temperatures indicate their sensitivities to land surface and atmospheric conditions, and show trends linked to changing surface moisture, temperature, and vegetation patterns. An estimation algorithm is used, calibrated to sites in arid regions and forests, to obtain preliminary estimates of the land parameters for a given geographic region focusing on semiarid areas were the potential for observing moisture variability using the SMMR channels is greatest, the estimates are compared with model output data. Results of the study are helpful in evaluating the potential of future L-band spaceborne radiometer measurements, which would have greatly increased capability for surface moisture estimation if available.
Water and energy exchanges at the land-atmosphere interface play a key role in determining patterns of regional and global climate. However, accurate estimation of surface fluxes of sensible and latent heat over arid and semiarid regions is a challenging task. In this study, a scenario for assimilating satellite data in the visible-infrared (AVHRR) and in the microwave (SSM/I) spectral ranges in a hydrological flux model is presented. The aim of this investigation over the HAPEX-Sahel area in West Africa is to show that the use of multispectral remotely sensed data, in conjunction with radiative transfer models and a hydrological flux model, can provide reasonable estimates of the surface fluxes. A discussion of the potentials and limitations of the approach is presented.< >
The current global change research emphasis on understanding water and energy fluxes at the land-atmosphere interface provided renewed interest in using passive microwave satellite data for land studies. Radiative transfer models of microwave emission and scattering in the soil-vegetation-atmosphere column, though still in the process of development, can be used to estimate the accuracies with which land surface parameters can be derived from satellite data. These parameters include surface soil moisture, surface temperature, vegetation water content, and atmospheric water content. A microwave radiative transfer model is used to develop linear and nonlinear versions of a multichannel retrieval algorithm to simulate retrievals of the surface and atmospheric parameters. These simulations include the effect of additive noise, and examine the effects of nonlinearities in the radiative transfer models. However, effects of spatial heterogeneity are not considered. Satellite data to which the retrieval algorithms may be applied include the SSMR and SSM/I, and in future will include the Multifrequency Imaging Microwave Radiometer (MIMR) to be launched as part of the Earth Observing System (EOS). The Scanning Multichannel Microwave Radiometer (SMMR) on the Nimbus-7 satellite provided data from 1978 to 1987 at five microwave frequencies between 6.6 and 37 GHz. The series of Special Sensor Microwave/Imager (SSM/I) instruments launched on the DMSP satellites provide data from 1987 to the present at four frequencies between 19.35 and 8.5 GHz. Spatial resolutions of the data range from approximately 12 to 120 km depending on the frequency. Examples using data from these sensors are shown to indicate the results of applying retrieval algorithms based on model simulations to real data.
The scanning multichannel microwave radiometer (SMMR) launched aboard the Nimbus-7 satellite provided global data at five frequencies (6.6, 10.7, 18, and 37 GHz) and two polarizations. The global data over land surfaces from this instrument are available for the period of January 1979 to December of 1985. The 6.6 GHz data set, being the lowest frequency of SMMR is especially important for studying the vegetation canopies and the moisture variations of the underlying soil surface. In this study, a microwave emission model for vegetation canopies has been developed to simulate the 6.6 GHz channel of SMMR. The canopy model consists of three layers of crown, trunk and underlying soil. The emissivity from the canopy is obtained by first computing the bistatic radar cross section of the three layer canopy using the distorted Born approximation and then integrating the radar cross sections over the hemispherical scattering angles according to the conservation of energy. In this formulation, each layer of the canopy is modeled as a random distribution of canonical shape dielectric scatterers (discs and cylinders). The dielectric constant of the scatterers are determined according to the available moisture in various components of the canopy. The model is then verified over homogeneous agricultural canopies using a ground stationed radiometer system. In particular, the properties of the polarization ratio with respect to the available vegetation biomass and soil moisture have been analyzed and it is found that this ratio becomes less sensitive to soil moisture as the vegetation biomass increases. The model has then been modified to take into account the large spatial resolution of the SMMR data by introducing a distribution of gaps in each resolution cell. The model simulations are then used in conjunction with monthly averaged 6.6 GHz SMMR data over two areas of Amazon forest and North-West US to study the effect of the moisture and vegetation changes.< >
