Application of Seasonal Climate Forecasts to Agriculture in the Southeastern United States

Researchers from the Florida Consortium (Florida State University, University of Florida, University of Miami) in the disciplines of meteorology, agricultural engineering, and economics have collaborated for four years on the application of climate information to support decision making in agriculture. The influence of ENSO on agricultural production and the new capacity to forecast climate suggest an unprecedented opportunity to tailor agricultural decisions to anticipated weather conditions, either to reduce risks of adverse conditions or to take advantage of favorable conditions.

In spite of the potential economic value for agricultural applications of climate forecasts, simply documenting the effects of climate variability and providing better climate forecasts are not sufficient. Our most important contribution has been the development of a framework for an effective climate information system. That framework has three components: (1) the generation of climate information; (2) the communication and comprehension of such information; and (3) the use of the information. We have explored extensively the functions of each of these components and how they fit together.

1. Generation of Climate Information

Associations between ENSO and climate in Florida. Significant associations exist between precipitation and temperature (maximum and minimum) and ENSO. These associations are strongest during winter (Dec-Feb) and are consistent throughout the state. One major impact is the 30% increase of winter rainfall during warm events. Cold events show a corresponding 10-30% decrease.

Associations between ENSO and crop yields in Florida. Several of Florida's highest valued crops are influenced by ENSO. These effects include decreased winter yields of tomato (28% of long-term average), bell pepper (31%), sweet corn (27%) and snap beans (18%) and increased prices of bell pepper (31%) and snap bean (31%) in warm events; increased sugarcane yields following cold events; and increased yields of grapefruit and tangerines but decreased lime yields in the harvest following warm events.

2. Communication of climate information

A central premise of our work has been close interaction with stakeholders in the agricultural sector to explore needs for climate information and perceptions about climate risk. Our principal activity in this regard has been the design and initiation of an operational system to disseminate agriculturally relevant climate information in Florida. That system is the result of a close collaboration between our Consortium and the Florida agricultural extension system and stresses program evaluation and feedback into research and project design.

3. Use of climate information

Decision-support tools we have developed:

- Stochastic weather generators to translate climate predictions into daily realizations of weather data for use in crop simulation models.

- Weather generators parameterized on ENSO phase so that numerous realizations of daily weather data can be generated for a location, given an ENSO categorical forecast.

- A method of using numerical climate model forecast output and generating daily weather data with monthly statistics matching climate model outputs.

- Crop management optimization procedures based on simulated annealing to evaluate management practices by ENSO phase.