Overall results and key findings on the use of UAV visible-color, multispectral, and thermal infrared imagery to map agricultural drainage pipes

Effective and efficient methods are needed to map agricultural subsurface drainage systems. Visible-color (VIS-C), multispectral (MS), and thermal infrared (TIR) imagery obtained by unmanned aerial vehicles (UAVs) may provide a means for determining drainage pipe locations. Aerial surveys using a UAV with VIS-C, MS, and TIR cameras were conducted at 29 agricultural field sites in the Midwest U.S.A. to evaluate the potential of this technology for mapping buried drainage pipes. Overall results show VIS-C imagery detected at least some drain lines at 48 % of the sites (14 out of 29), MS imagery detected drain lines at 59 % of the sites (17 out of 29), and TIR imagery detected drain lines at 69 % of the sites (20 out of 29). Three key findings, listed as follows and emphasized in this article by site examples, were extracted from the overall results. (1) Although TIR generally worked best, there were sites where either VIS-C or MS proved more effective than TIR for mapping subsurface drainage systems. Consequently, to ensure the greatest chance for successfully determining drainage pipe patterns in a field, UAV surveys need to be carried out with all three types of cameras, VIS-C, MS, and TIR. (2) Timing of UAV surveys relative to recent rainfall can sometimes have an important impact on drainage pipe detection results. (3) Linear features representing drain lines and farm field operations can be confused with one another and are often both depicted on site aerial imagery. Knowledge of subsurface drainage system installation and farm field operations can be employed to distinguish linear features representing drain lines from those representing farm field operations. The overall results and extracted key findings from this study clearly indicate that VIS-C, MS, and TIR imagery obtained with UAVs have significant potential for use in mapping agricultural drainage pipe systems.