USING MEASURED DATA AND EXPERT OPINION IN A MULTIPLE-OBJECTIVE DECISION-SUPPORT SYSTEM FOR SEMIARID RANGELANDS

A Decision Support System (DSS) can be used to structure information i n a way that leads to improved decision making for natural resources. The decisions will only be as good as the information on which they ar e based. As the applications of a DSS are outpacing the available data bases and simulation models, there is an increasing reliance on expert opinion for information on resource management systems. As a result, the effect of information source on the outcome from the DSS is an imp ortant issue. This article compares the outcomes from a prototype DSS (P-DSS) developed by the USDA-ARS Southwest Watershed Research Center in Tucson, Arizona, when measured data and expert opinion are used to quantify eight decision criteria in the evaluation of four management systems (yearlong and rotation grazing, each with mesquite trees (Pros opis velutina Woot.) retained or removed) for semiarid rangelands. The decision criteria are sediment yield, channel erosion, runoff rate an d quantity, rangeland condition, aboveground net production, and wildl ife habitat for quail and javelina, although the analysis is not restr icted to these criteria. When measured data are used to quantify the d ecision criteria, rotation grazing with mesquite removed is the prefer red management system, whereas yearlong grazing is the preferred syste m when expert opinion is used. The experts also directly ranked the fo ur management systems. The difference between the experts' ranking and the P-DSS results based on expert inputs is a concern for future use of decision support system technology, particularly when information s ources are blended.