Statistical and Neural Methods for Site–specific Yield Prediction

Abstract

Understanding the relationships between yield and soil properties and topographic characteristics is of critical importance in precision agriculture. A necessary first step is to identify techniques to reliably quantify the relationships between soil and topographic characteristics and crop yield. Stepwise multiple linear regression (SMLR), projection pursuit regression (PPR), and several types of supervised feed-forward neural networks were investigated in an attempt to identify methods able to relate soil properties and grain yields on a point-by-point basis within ten individual site-years. To avoid overfitting, evaluations were based on predictive ability using a 5-fold cross-validation technique. The neural techniques consistently outperformed both SMLR and PPR and provided minimal prediction errors in every site-year. However, in site-years with relatively fewer observations and in site-years where a single, overriding factor was not apparent, the improvements achieved by neural networks over both SMLR and PPR were small. A second phase of the experiment involved estimation of crop yield across multiple site-years by including climatological data. The ten site-years of data were appended with climatological variables, and prediction errors were computed. The results showed that significant overfitting had occurred and indicated that a much larger number of climatologically unique site-years would be required in this type of analysis.

Crop Models as Decision Support Systems in Crop Production

“…Based on these considerations, another used measure of model accuracy is the root mean square error (RMSE). A major advantage of using RMSE over R² for model evaluation is that RMSE provides information about both on the calibration data and on new data not used in developing the model to estimate the true predictive ability of the model www.intechopen.com (Drummond et al, 2003). Cross-validation is another more robust, reliable method of measuring prediction accuracy (Stone, 1973) of crop models.…”

Overview of the Mark Twain Lake/Salt River Basin Conservation Effects Assessment Project

“…These data showed considerable spatial variation in crop yield and soil properties. In general, soil texture and topsoil depth (as inferred by apparent soil electrical conductivity ]), along with topography, had the most persistent relationships with yield because of their effect on soil water holding capacity and within field water redistribution (Drummond et al 2003). The shape of the relationship was dependent on the climate during the particular growing season, specifically the amount of rainfall received in July and August .…”

Production Potential Prediction for Wheat, Barley and Maize Based on Soil Characteristics Using Artificial Neural Networks in Varamin Region, Iran

“…It is also true about wheat compared to maize. Drummond et al (2003) mentioned the more capability of ANN for yield prediction of crops with more resistance and environmental comparability.…”