• Alireza Tahai - Professor of Quantitative Methods, Mississippi State University
• Benjamin F. Blair - Assistant Professor of Economics, Mississippi State University
• Keith Coble - Assistant Professor of Agricultural Economics, Mississippi State University
• J. Alex Thomasson - Assistant Professor of Agricultural and Biological Engineering, Mississippi State University

• Scott A Shearer - Associate Professor of Biosystems and Agricultural Engineering, University of Kentucky

Effective farm management requires the farm manager to make two types of decisions. The first category of decisions -- crop management decisions -- include what, when and where to plant, when and where to fertilize, irrigate, etc. Current research in remote sensing is exploring how the technology can be used to aid the farmer in these types of decisions. However, perhaps just as important, is the second category of decisions -- risk management decisions. Risk management decisions are used by farmers to limit their exposure to uncertain future outcomes. E.g., before planting, a farmer may purchase crop insurance which will limit the farmer=s financial losses in the case of a poor harvest. While the decision to purchase crop insurance must be made before planting, there are other risk management strategies that can be employed throughout the growing season. These intra-season strategies include futures hedging, cash forward contracting, and purchasing options contracts. As opposed to a "static" risk management plan in which the farmer makes all decisions prior to planting, this study will analyze the use of a more "dynamic" plan in which the farmer can adjust and readjust his exposure to risk throughout the growing season.

The choice of risk management strategy and the payoff to the farmer of adopting a particular intra-season risk management strategy depends upon the timing of the decision and the accuracy of the information the farmer has when the decision is made. An important piece of information the farmer needs is some prediction about the future yield of the crop. For example, pre-harvest hedging strategies are conditioned on both price and yield expectations. Failure to recognize yield uncertainty can lead to the situation where more of the crop has been priced in the futures market than is produced at harvest. In such a situation, the risk averse producer is placed in a speculative position, rather than achieving the desired risk reduction. This study will examine the usefulness of reducing intra-season yield uncertainty in shielding the farmer from risk throughout the growing season.

In the past, future yield predictions have been based upon data gathered using conventional technology such as temperature and rainfall patterns. With the advent of remote sensing technology, a much richer and more timely data set can be compiled with which it may be possible to make more accurate and timely predictions about future crop yield. More accurate and timely crop yield forecasts allow the farmer to make better risk management decisions which, in turn, may increase the profitability of the farming enterprise. Whether or not profitability is increased will depend upon the relative improvement in accuracy of forecasts based on remote sensing data over forecasts based on conventional data and, in addition, any cost differences associated with achieving the improved accuracy.

To address these issues, our project involves three main steps:

1. Development of a theoretical economic model of risk management that can be used to assess the value of reduced yield uncertainty.
2. Empirically investigate the applicability and accuracy of current remote sensing data on the prediction of crop yield.
3. Investigate the comparative value of the remote sensing forecast estimates in the development of a farmer's risk management plan.