|
1000
|
Abstract/Summary
|
-
Crop yield is the result of the interactions between weather in the incoming season and how farmers decide to manage and protect their crops. According to Jones et al. (2000), uncertainties in the weather of the forthcoming season leads farmers to lose some productivity by taking management decisions based on their own experience of the climate or by adopting conservative strategies aimed at reducing the risks. Accordingly, predicting crop yield in advance, in response to different managements, environments and weathers would assist farm-management decisions(Lawless and Semenov, 2005). Following the approach described by Semenov and Doblas-Reyes (2007), this study aimed at assessing the utility of different seasonal forecasting methodologies in predicting durum wheat yield at 10 different sites across the Mediterranean Basin. The crop model, SiriusQuality (Martre et al., 2006), was used to compute wheat yield over a 10-years period. First, the model was run with a set of observed weather data to calculate the reference yield distributions. Then, starting from 1st January, yield predictions were produced at a monthly time-step using seasonal forecasts. The results were compared with the reference yields to assess the efficacy of the forecasting methodologies to estimate within-season yields. The results indicate that durum wheat phenology and yield can be accurately predicted under Mediterranean conditions well before crop maturity, although some differences between the sites and the forecasting methodologies were revealed. Useful information can be thus provided for helping farmers to reduce negative impacts or take advantage from favorable conditions.
|
