Use of a thermal scanner image of a water stressed crop to study soil spatial variability

Abstract

The objective of this study is to evaluate the use of radiative surface temperature of a water-stressed crop to indicate sources of variability in soil characteristics affecting the availability of soil water. Use was made of a thermal infrared image collected by aircraft near solar noon of a 45-ha water-stressed wheat crop. The data was analyzed by the theory of regionalized variables to determine whether variability in observed radiative surface temperature was random or spatially related. Radiative surface temperatures varied between 24 and 29°C and semivariance analysis revealed that the variations were random but were structured. The analysis indicated that any two locations separated within the field by 400 or 700 m were spatially unrelated. Field observations found that, at about 700 m, the soil type changed from a more fertile Yooroobla clay soil to a less fertile Wilbriggie clay loam. At about 400 m, soil coring data indicated that the thickness of heavy clay at the surface declined from 3 m to less than 1 m. At a smaller distance, the semivariance analysis indicated microvariations, which field observations associated with either gilgais or the laser leveling of the land surface. It was concluded that these soil variations were likely to have caused variations in soil water availability and hence the observed spatial variability in radiative surface temperature. Semivariance analysis was successful in indicating the location of these sources of spatial soil variability.