The estimation of land surface emissivities at 24 GHz to 157 GHz using remotely sensed aircraft data

Rainfall Estimation fron passive microwave satellite data has been used wid ely over oceans but has been less successful over land. This is because ove r land surfaces, the high spatial and temporal variability in emissivity, c oupled with relatively low contrast between surface and rain cloud microwav e emissions, make the rainfall signal moo-e difficult to extract. The varia bility of emissivity is mainly due to variations in vegetation cover and so il moisture. Major improvements in reliability of rainfall estimates are po ssible if emissivities could be measured routinely at appropriate scales. T he possibility of estimating emissivity at the frequencies relevant to rain fall from vegetation and soil moisture measurements is explored in this pap er using data from airborne sensors over a semi-arid area in Spain. Results show a good correlation between vegetation cover (represented by Normalize d Difference Vegetation Index) and emissivity in dry conditions. This relat ionship is not significantly affected by vegetation type. Under wet conditi ons, the correlation is greatly reduced possibly due to the difficulty in a ccounting for cloud effects at higher frequencies. Attempts to quantity the effect of soil moisture using the Antecedent Precipitation Index were part ially successful but more accurate measurements would be needed for reliabl e retrieval of emissivities. The use of a soil-adjusted vegetation index pr oduced a higher correlation with emissivity than did the nonadjusted Normal ized Difference Vegetation Index. (C)Elsevier Science Inc., 2000.