Submitted to: International Conference on Precision Agriculture Abstracts & Proceedings
Publication Type: Abstract only
Publication Acceptance Date: 7/12/2002
Publication Date: 7/12/2002
Citation: GRAFF, C.D., KOSKINEN, W.C., ANDERSON, J.A., HALBACH, T.R., DOWDY, R.H. USING GEOSTATISTICS TO UNDERSTAND THE SPATIAL DISTRIBUTION OF SOIL PROPERTIES AND THE FIELD DISSIPATION OF HERBICIDES. INTERNATIONAL CONFERENCE ON PRECISION AGRICULTURE. 2002. ABSTRACT P. 149. Interpretive Summary:
Technical Abstract: Current research in precision agriculture has focused on solving problems dealing with soil fertility and crop yield. Specifically, geostatistical methods are being used to link the spatial distribution in soil physical and chemical properties to spatial patterns in crop yield. Because soil properties such as OC and clay contents, and pH, which affect field dissipation of herbicides, can significantly vary spatially across the landscape, it is possible that herbicide dissipation rates have spatial structure as well. The spatial variability of surface soil properties was characterized for a watershed cultivated in corn at the Rosemount Experiment Station in Dakota County, MN and displayed on a digital elevation model (DEM). Utilizing Arc/Info, terrain attributes such as slope, aspect, plan and profile curvature were derived from the finely sampled DEM in order to determine whether such attributes enhanced prediction of soil characteristics and/or herbicide residues. Acetochlor was applied at the uniform recommended rate of 2.2 kg ha-1 for both the 2000 and 2001 field season. At 2 days, 2, 6 and 11 weeks, soil samples were taken to a depth of 70 cm at numerous georeferenced locations having a range in soil properties representative of the site. Soils were analyzed at 0-10, 10-20, 20-40, 40-60 and 60+ cm for herbicide residues. Field dissipation rates (k) at the surface were calculated using a first order rate equation at each georeferenced location. The spatial distribution of k and herbicide concentrations at each sampling time were modeled in Vesper and displayed with Surfer. Various regression and spatial techniques were applied to try and correlate those patterns to the distribution of soil properties and terrain attributes across the field. Understanding how herbicide dissipation rates vary with soil characteristics could lead to reducing off-site transport via precision management.