Document: RAM-3-90-3
Developing a biospheric forecast system.
NEMANI, R.* 1, S.RUNNING 1, J.ROADS 2, P.THORNTON 1, A.WEISS 1 and P.VOTAVA 1
NTSG/University of Montana, Missoula, MT 59812 USA 1
Scripps Institute of Oceanography, La Jolla, CA 92093 USA 2
Abstract:
Advanced warnings of the potential changes in key ecosystem variables such as soil moisture, snow pack, primary production and stream flow could enhance our ability to make better socio-economic decisions relating to natural resources management and food production. Accuracy of such warnings depends on how well the past/present /future conditions of the ecosystem are characterized. Forecasting skills of many current coupled Ocean-Atmosphere GCMs have steadily improved over the past decade. Given observed anomalies in SSTs from satellite data, GCMs are able to forecast climatic conditions 6-12 months into the future with reasonable accuracy. While such forecasts are useful for climatological purposes, analysis of their impacts on ecosystem response has been at best subjective. In this paper a conceptual framework for developing an Biospheric Forecast System (BFS) is provided. The proposed system assimilates NASA's Earth Observing System land products into an ecosystem simulation system, initialized/restored with observed weather data and forced with short to long-range weather/climate forecasts. The system will use the components of RHESSys (Regional HydroEcological Simulation System) as the land surface ecosystem model for the computation of distributed carbon, water and nutrient cycles, which have been used to estimate more basic resource information such as forest productivity, runoff production, snow dynamics and soil moisture stress. This model system is designed to be operated and forced by remotely sensed land surface state variables, daily meteorological information and widely available terrain data, and to allow the assimilation and updating of state variables by EOS/ MODIS land surface products. Successful implementation of such a system would bring together state-of-the art technologies in weather/climate forecasting, ecosystem modeling and satellite remote sensing, and would allow better management of floods, droughts, forest fires, irrigation requirements and crop/range/forest production and human health.
Keywords: ecosystem modeling, remote sensing, climate forecasts
This abstract is being presented at: 10:30 AM in session:
REMOTE SENSING