Assessing User Needs and Model Accuracy of Seasonal Climate Forecasts for Winter Wheat Producers in the Southern Great Plains

Successful agriculture is essential to global food security, but productivity can be severely impacted by drought, flood, or heat. Seasonal climate forecasts can warn agricultural producers of unseasonal conditions and help them reduce crop losses by providing critical decision-support in farm management. Seasonal climate forecasts have been used by producers for decades, but they are criticized for being inaccurate and for lacking relevant information for specific farm decisions. Our research is intended to address these criticisms. We assessed seasonal forecast needs of winter wheat producers and tested the capability of a state-of-the-art, high-resolution seasonal forecast model whether it provides decision-relevant information with sufficient accuracy.

At AMS, we present results from two studies that (1) improve our understanding of the forecast needs of winter wheat producers in the Southern Great Plains (Texas, Oklahoma, Kansas, Colorado), and (2) quantify the error of a state-of-the-art seasonal forecast model for forecast elements most preferred by winter wheat producers. Through an online survey of 119 agricultural advisors we determined which forecast elements are most important to winter wheat farmers and in which months of the year, and how much lead-time is necessary to inform farmers in time. We then compared hindcasts from the Geophysical Fluid Dynamics Laboratory FLOR-B01 forecast-oriented seasonal forecast model with observations for the most requested forecast elements (monthly average temperature and total precipitation, frequency of extreme precipitation, and number of days without precipitation) to quantify the error of the model across the study region and to compare the error to that of a persistence forecast, namely the average of the month of interest for the five years preceding the forecast month.