Tools for Optimizing Management of a Spatially Variable Organic Field

Abstract

Geostatistical tools were used to estimate spatial relations between wheat yield and soil parameters under organic farming field conditions. Thematic maps of each factor were created as raster images in R software using kriging. The Geographic Resources Analysis Support System (GRASS) calculated the principal component analysis raster images for soil parameters and yield. The correlation between the raster arising from the PC1 of soil and yield parameters showed high linear correlation (r = 0.75) and explained 48.50% of the data variance. The data show that durum wheat yield is strongly affected by soil parameter variability, and thus, the average production can be substantially lower than its potential. Soil water content was the limiting factor to grain yield and not nitrate as in other similar studies. The use of precision agriculture tools helped reduce the level of complexity between the measured parameters by the grouping of several parameters and demonstrating that precision agriculture tools can be applied in small organic fields, reducing costs and increasing wheat yield. Consequently, site-specific applications could be expected to improve the yield without increasing excessively the cost for farmers and enhance environmental and economic benefits.

Keywords

• GRASS
• raster images
• principal component analysis
• organic farming
• precision agriculture
• geostatistics