Submitted to: Agronomy Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: October 28, 1997
Publication Date: N/A
Interpretive Summary: Precision farming has enjoyed widespread, growing interest as a result of better equipment, better computers, global positioning systems, and an increasing awareness of the economic and environmental benefits it offers. As with any new technology, precision farming creates almost as many questions as answers. One question particularly important to users is how much data is needed to make intelligent decisions about how much fertilizer, pesticide, etc. to put where. Two-dimensional grid soil sampling, for example, costs 4 times as much when you cut the spacing in half. Obviously, the farmer does not want to oversample, but if too few samples are taken, the information may be nearly meaningless. We summarized literature findings and compared them to research in the SE USA to develop practical guidelines for precision farmers. Where researchers in the Midwest have generally recommended grid spacing from 100 to 200 feet, and commercial firms have used from 330 to 440 feet, local soil variability suggests that important differences may be missed unless the spacing is reduced, possibly to less than 100 feet. The high cost of doing so emphasizes the need for sensor development and analysis tools to provide detailed spatial information at lower cost.
Technical Abstract: Precision farming has created a critical need for spatial data, but because costs of data increase faster than information content, users need practical guidelines for spatial resolution. Our objectives were to 1) describe observed variation in the SE region, 2) compare it to other regions, 3) describe relationships among soil and crop characteristics that taffect resolution needs, and 4) offer suggestions for precision farmers in the SE USA. Corn (Zea mays), wheat (Triticum aestivum), soybean (Glycine max), and grain sorghum (Sorghum bicolor) yields were mapped from 1985 to 1995 at Florence, SC. Also available, were topography, depth to clay, and in 1993, plant height on one date, canopy temperature on 4 dates, and detailed crop and soil information at selected sites. Yield of all crops in all years was significantly (p<0.0007), correlated to soil map unit. In 1993, canopy minus air temperature was also correlated to soil map unit, even after a 46-mm rain. Spherical semivariograms fitted to yields had ranges of 59-248 m (median=169). Nugget:sill ratios were 0.0-0.5 (median= 0.26. Ranges for canopy temperature and plant height were from 45 to 75m. Captioning spatial variation in Coastal Plain soils may require more intensive sampling than commonly used in precision farming.