Fig. 4

Image of cotton plants with the infrared thermometer (IRT) field of view (FOV) outlined in red (a), where FOV1 is centered on the cotton crop and FOV2 is no longer centered on the cotton crop due to a row centerline deviation (wobble). The “wobble” visualized in QGIS software within the experimental plot boundaries (b). Changes in the sensor FOV due to sensor boom roll and pitch alterations because of holes and bumps in the tractor drive path (c). An example of pre-outlier (d) removed data where the distribution is highly variable, and outliers caused by edge effects are visible in the north end of the plot. An example of post-outlier (e) removed data from an infrared thermometer after a single iteration where the edge effect outliers are removed, and the data better resemble a normal distribution (blue line)