Adv. Space Res. Vol. 11. No. 3. pp.(3)143—(3)150, 1991 0273-1177/91 $0.00+.50 Printed inGreat Britain. All rights reserved. Copyright ©1991COSPAR AN OVERVIEWOF THE RESULTS OF THE FIFE-87 AND FIFE-89 CAMPAIGNS R. E. Murphy, P. J. Sellers, F. G. Hall, G. Asrar, B. L. Blad, E. T. Kanemasu,* R. D. Kelley, B. Markham, D. Strebel and J. R. Wang NASA Headquarters, Code EEL, Washington, DC 20546, U.S.A. ABSTRACT Results ofa major land surface-atmosphere exchange study, the First ISLSCP Field Experiment (FIFE)are summarized. Findings result from nearly 80 days of fieldmeasurements over a 15 kilometer square area involving 30 science teams, seven remote sensing aircraft, and 5 satellites which acquired over 1200lowresolution and35 high spatial resolution satellite images, maps of remotely sensed temperature, vegetation index and soil moisture, surface fluxmeasurements from more than 20 ground stations and airborne sensors, and transects ofvegetation, soil moisture and soil chemistry. Relationships between in-situ, airborne, and spaceborne remote measurements and the problem of scaling will bediscussed. Energy balance comparisons among sites were made. Diurnal measurements of latent heat fluxes indicated a strong correlation between the evaporative fraction at midday and the daytime average value. 1. INTRODUCTION The study of the Earth’s climate system has become an important focusfor Earth Science. Simulation models, such as general circulation models of the atmosphere (GCMs), are key tools forthe effort inunderstanding the mechanisms involved and potentially for predicting the Earth systems’ response to natural and human induced perturbations. More than half of theerror in a series of trial short-range numerical weather forecasts were associated withthe inaccurate specification of the initialstate of the system, and with the inadequacy of the physical parameterizations usedin the models(Anthes,1985). These parametenzations describethe essentially one-dimensional processes of surface-atmosphere interaction, convection, cloud-radiation feedback, etc., within GCMs. Research efforts have shown that land-surface-atmosphere interactions have a strong biophysical component; the Earth’s vegetation affects the interception of solar radiation, the exchange of momentum, mass,and energy between the lower atmosphere and the surface (turbulent transfer) and the partitioning of intercepted energy into sensible and latent heat fluxes (biophysical control ofevapotranspiration). Thefirst GCM-biosphere models incorporated crude descriptions of these processes, based on plantor evenleaf-scale models which havetheirorigins in small-scale biological studies. Additionally, these models are initialized,in terms of vegetation type, cover fraction, phenological stage and soil moisture, using climatologies based on geographic survey work or extrapolation. Inshort, the land surface parameterizations currently in use suffer from the (i) ‘scale gap’ existing between the biological andmeteorological disciplines and (ii) from the lack of a global observing system to initialize or update, andvalidate them. *Presenting Author. (3)143