Estimation of spatially distributed latent heat flux over complex terrain from a Raman lidar

A method is presented in which estimates of evaporation may be made over an area approaching three quarters of a square kilometer, with relatively fine (25 m) spatial resolution, using three-dimensional measurements of water vapor concentration from a scanning Raman lidar. The method is based upon Monin-Obukhov similarity theory applied to spatially and temporally averaged data, Data from the lidar is used to sense the location and orientation of the surface and the location of the water vapor measurements with respect to that surface. Maps of the spatial distribution of evaporation have been produced showing the evaporation rates at regular intervals throughout the day. The method was applied to the SALSA experimental site during the 1997 summer field campaign. The estimates of evaporation rates made during the campaign compare favorably with estimates made using sap flux methods with RMS differences of 18W/m(2). While the method has certain limitations, the three-dimensional character of the data allows for the detection of anomalous situations so that analysts may alter the analysis technique or reject the estimates from the affected regions. This information can be used in a wide variety of ways to study the spatial variations in evaporation caused by changes in soil type and moisture content, canopy type and topography. (C) 2000 Elsevier Science B.V. All rights reserved.