988 
parameters as vegetation cover and biomass and soil moisture. Data from an airborne laser system were related 
to vegetation height and cover, and surface aerodynamic roughness. Active microwave (ERS-1 SAR) data proved 
useful for mapping soil temperature and moisture; and there was evidence the passive microwave could be used 
to map soil moisture and infer the distribution and amount of rainfall. These results addressed many of the 
challenges in the use of remotely-sensed data in regions with sparse vegetation. 
Substantial progress was also made in the refinement of interstorm models to estimate energy and water 
balance (box 2). An important aspect of this research was the investigation of the relation between remotely- 
sensed surface temperature and sensible heat flux. Previously-published studies have resulted in a mixed 
message; for example, Sellers et al. (1992) reported large discrepancies between measured and calculated surface 
fluxes when using surface temperature, yet others (e.g., Brutsaert and Sugita (1991) and Kohsiek et al. (1993)) 
have found a good correlation between surface temperature and H. This discrepancy may be linked to differences 
in the computation of aerodynamic resistances to heat and momentum transfer. Results from Monsoon’90 and 
WG’92 indicated that if this difference was accounted for (resulting in an additional resistance term), reliable 
estimates of H could be obtained from remotely-sensed surface temperature measurements for both sparsely- and 
densely-vegetated sites. 
The Monsoon’90 Experiment resulted in one of the first attempts to merge remotely-sensed data with a storm 
model to produce distributed maps of runoff (boxes 1 and 5). Though this work was limited to the catchment 
scale, it provided direction for further work at the regional scale. In particular, it emphasized the importance of 
accurate estimates of input rainfall to achieve reasonable runoff simulations. 
With the progress made in storm and interstorm modeling, the next logical step is to exploit the 
interdisciplinary expertise of Monsoon’90 and WG’92 participants to interface the storm and interstorm models 
and develop an integrated hydrologic model. Work is currently underway to georegister the satellite data with 
digital elevation data, and utilize a grid of meteorological stations to produce corresponding maps of air 
temperature, wind speed, humidity and incoming solar radiation. With these regional data layers, it will be 
possible to use a Geographic Information System (GIS) to analyze hydrologic processes at the regional scale. 
Procedures are being implemented for coupling an existing mesoscale model (Avissar and Pielke, 1989) with 
hydrologic and land surface characterization data layers derived from WG’92 satellite-based spectral images. 
This coupled model will allow investigation of the model sensitivity for application to a topographically diverse 
regions and allow definition of such model input requirements as temporal/spatial resolution and accuracy. Once 
the model sensitivity and input requirements have been defined, the approach will be ready for application to 
other sites for calibration and validation, and subsequent investigation of regional hydrologic processes. 
The preliminary results of Monsoon’90 and WG’92, coupled with the extensive experimental data sets, have 
shown great potential for confronting the hydrologic modeling challenges in arid and semiarid rangelands. 
However, the value of these experiments lies not only in investigation of arid rangelands, but also in providing 
the foundation for design of future large-scale experiments at sites with different climates, biomes, and 
topography. For example, results from Monsoon’90 and WG’92 underscored the need for a future study to 
investigate how spatial scales of heterogeneities affect their importance towards certain regional-scale processes. 
SALSA-MEX is an experiment designed to explore the transferability of Monsoon’90 and WG’92 results to 
different biomes and different scales, and to investigate the effects of mountain topography on hydrologic 
processes. 
4 - SALSA-MEX: CONCEPTUALIZATION OF A FUTURE REGIONAL-SCALE LAND-SURFACE 
EXPERIMENT 
Mountainous areas are often the primary source of available water in arid and semiarid regions of the world. 
Increased understanding of the processes significantly influenced by topography, such as enhanced precipitation, 
snowmelt, and groundwater recharge, are essential for population sustainability in these areas. Nonetheless, many 
experiments, including Monsoon’90 and Walnut Gulch’92, were designed to address the effects of local and 
regional scale advection on land surface-atmosphere and hydrologic processes over relatively flat terrain. In 
response to this, a regional-scale, semiarid mountain experiment [termed SALSA-MEX, proposed for execution 
in the late 1990s] has been designed to specifically address the profound effects of topography on hydrological 
and meteorological processes at virtually all scales. 
The experimental site will be the San Pedro basin, encompassing an area of roughly 12,000 km 2 in 
southeastern Arizona and northern Sonora, Mexico. Elevation ranges from roughly 900 to 2800 m in the upper 
basin, providing exceptional ecological and climatic diversity in a relatively compact area. The major biome 
types represented in the basin are: Chihuahuan Desert, Semiarid Grasslands, Oak Savannah Chaparral, Pinyon-