Probabilistic Yield Forecast Based on Aproduction Process Model

A method for probabilistic forecast of agricultural yield depending on meteorological variability, i.e. forecast of agrometeorological resources, is discussed. Forecast is based on the category of meteorologically possible yield (MPY)–the maximum possible yield for a given variety in the existing meteorological conditions. The forecasting process is realized by a potato production process model POMOD, which applies the principle of maximum plant productivity and method of reference yields. The yield diversity, granting probabilistic distribution was obtained from series of model calculations, whereby the weather realizations for post-forecast period were gained from a century-long meteorological data series. Three examples realized for extremely different years are discussed. The results of such forecast, presented as a cumulative distribution, allow user to adjust and plan activities to the sufficiently assured yield level. Forecast of agrometeorological resources can be transformed to the forecast of real commercial yield (CY) by incorporating the efficiency coefficient of using meteorological conditions (CY/MPY).