Thermal Characterization and Spatial Analysis of Water Stress in Cotton (Gossypium Hirsutum) and Phytochemical Composition Related to Water Stress in Soybean (Glycine Max)

Authors: Thomson, Steven; Defauw, Sherri; ENGLISH, PATRICK - MSU-DREC; Hanks, James; Fisher, Daniel; Foster, Phelesia; Zimba, Paul

Submitted to: International Conference on Precision Agriculture Abstracts & Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 5/7/2008
Publication Date: 7/23/2008
Citation: Thomson, S.J., Defauw, S.L., English, P.J., Hanks, J.E., Fisher, D.K., Foster, P.N., Zimba, P.V. 2008. Thermal Characterization and Spatial Analysis of Water Stress in Cotton (Gossypium Hirsutum) and Phytochemical Composition Related to Water Stress in Soybean (Glycine Max). International Conference on Precision Agriculture Abstracts & Proceedings.

Interpretive Summary: Crops planted in highly heterogeneous soils prevalent in the Midsouth US may exhibit varying levels of stress depending on the interactions of many factors including soil texture, meteorological conditions, and pest pressures. A series of studies were designed to explore the spatial relationships of water and/or heat stress in cotton and soybeans and to assess factors that may influence yield potential. One experiment used an Electrophysics thermal imaging camera mounted on agricultural aircraft, VERIS 3100 soil EC mapping system, and yield monitor to obtain preliminary data on the suitability of thermal imaging for soil texture mapping. Thermal images matched VERIS-derived soil maps quite well visually, indicating good potential for thermal mapping of soil texture. Use of a calibrated thermal imaging camera to map soil texture could save time for characterization over large field areas. Spatial yield relationships for cotton over a five year period were also compared, and these yield patterns were highly influenced by soil texture combined with precipitation patterns. A study on soybeans indicated good results using pigment analysis to detect differences in water stress between irrigated and non-irrigated soybeans. Preliminary results from thermal imaging and hyperspectral sensing to detect differences were inconclusive for the soybean fields, but data will be analyzed further to indicate possible trends. An investigation that focused on detection of spatial relationships between irrigated cotton canopy cover change and canopy temperature differences in field zones demonstrated that this particular combination of datasets could serve as a useful alternative to vegetative indices (VIs) for in-season prediction of yield potential and site-specific application of insecticides and defoliants/harvest aids.

Technical Abstract: Studies were designed to explore spatial relationships of water and/or heat stress in cotton and soybeans and to assess factors that may influence yield potential. Investigations focused on detecting the onset of water/heat stress in row crops using thermal and multispectral imagery with ancillary physicochemical data such as soil moisture status and photosynthetic pigment concentrations. One cotton field with gradations in soil texture showed distinct patterns in thermal imagery, matching patterns measured by the VERIS 3100 soil EC mapping system. Yield of cotton obtained from 2003 to 2007 also followed mapped soil textures, and this depended to an extent on timing and amounts of precipitation. Thermal images were obtained of a soybean canopy in irrigated and non-irrigated fields, and samples were taken from three leaves at four locations of both fields for pigment analysis using high performance liquid chromatography (HPLC). Although leaf sample size for this preliminary study was rather small, significant differences between leaves sampled from irrigated vs. non-irrigated soybean plants were seen for three pigments at the 0.10 level. For another cotton field, several thermal images acquired over a five-week time interval (July-August 2006) were composited to produce a cumulative thermal map. Changes in canopy cover (derived from intensity-normalized color infrared imagery) were also mapped. Composited thermal imagery combined with tracking canopy cover change at key phenological stages could serve as a useful alternative to vegetative indices (VIs) for the in-season prediction of production potential as well as early senescence promoted by heat/water stress in highly heterogeneous fields, and foster the development of site-specific applications of insecticides to protect high-yielding areas and cost-effective application of defoliants/harvest aids for cotton. Thermal mapping was seen as a useful tool for characterizing soils to predict yield depending on temporal precipitation patterns and thus could be useful for determining supplemental irrigation needs.