Drones map agricultural drainage pipes
Locating "lost" drainage pipes in the field has for many years been both time-consuming and expensive, but research from Aarhus University, US universities and research organizations has investigated the use of drones for locating drain pipes. Drone imagery is a solution that will save farmers time and efforts when the drains need to be retrofitted or repaired.
Subsurface drainage has for a century been the common practice of agriculture across the world to remove excess water in agricultural fields. However, drainage pipe installations are often poorly documented, which is problematic when they need to be retrofitted or repaired. In addition, there is a risk of nitrate and phosphate leakage to surrounding aquatic systems as the farmer drains his field. This can have major negative consequences for the aquatic environment. Knowledge about the location of the drainage systems and the distance between the pipes play an important role in assessing the risk of nutrient leakage.
Up to now it has been both expensive and time-consuming to locate the subsurface drainage pipes in the field, but researchers from Department of Agroecology have, together with a group of American colleagues, investigated the possibility of using drone imagery in the search for the hidden drainage system.
“Until now the preferred methods have been tile-probing and trenching to locate the drainage pipes. But this is very time-consuming, and the risk of destroying the drainage pipes is huge. We have looked into using ground penetrating radar (GPR), but it is a very expensive and impractical solution to use for large areas,” says PhD student Triven Koganti from Department of Agroecology at Aarhus University.
As part of his PhD study, he has worked with scientists from the U.S. Department of Agriculture and other American organizations. Together they have investigated the possibility of using drone imagery to locate drainage pipes in the fields.
Different methods
Over the years, several methods have been tested for locating the drainage pipes, and some of the most effective have been GPR or images taken from airplanes. Both are impractical and expensive solutions and might not be successful every time.
“With the advent of drone technology and increase in affordability of cameras, we have new opportunities to explore their applications in agriculture. Mapping the drainage system of the field is one such application, which is neither time-consuming nor expensive,” says Triven Koganti.
In trying to map drainage systems with drones, the researchers have worked with three different camera systems to find the most reliant one.
- Visible-color (VIS-C)
- Multispectral (MS)
- Thermal Infrared (TIR)
“Directly above the drainage pipes, the soil is often drier than the soil between the drainage pipes, which is especially evident just after it has rained. Sometimes after it has rained you will be able to see the difference between the dry soil and the wet with the visible-color and multispectral cameras. And you will also be able to capture it with the thermal infrared camera, because the soil warms up at different speeds depending on whether it is dry or wet. The soil directly on top of the drainage pipes will have a different temperature and would appear as a linear feature on the thermal infrared images,” says Triven Koganti.
Drones have great potential
"Our studies show that there is great potential for using drones to locate drainage pipes in the field, but it is not a silver bullet," says Triven Koganti.
The success rate for the three different cameras and measurement methods is not 100%. For the three types of cameras, the researchers found the following success rates:
- Visible-color (VIS-C) - 48%
- Multispectral (MS) - 59%
- Thermal Infrared (TIR) ??- 69%
Thus, even when using drones, it is not certain that you can find the exact location of the drainage pipes.
"That is why we suggest using all three cameras if at all possible when measuring in the field," says Triven Koganti. “We have had some examples where we, for example are able to find the pipes by using multispectral and not the other two, and vice versa. So, if you want to safeguard yourself, you can use all three. Moreover, with the high-resolution imagery from VIS-C camera you can generate a highly accurate digital terrain model which is an extremely useful information for drainage contractors. We also suggest combining with GPR technology, where you can pinpoint the location and depth of the pipes down to a few centimeters, while the positional accuracy of, for example, our thermal infrared measurements can be 4 meters without accurate georeferencing.”
According to the researchers, a mix of methods will be the best solution for finding the agricultural drainage pipes.
“Drone technology is an optimal and cost-effective solution for mapping field drains, but it requires a combination of imaging methods before the result is certain, as sometimes linear features caused by field operations (e.g. harvest or tillage tracks) also show up in the imagery and can be misinterpreted as drain pipes. Nevertheless, we find that there is great potential in drone technology and endeavor to research its application in greater detail,” says Triven Koganti.
Behind the research
Collaborators: Department of Agroecology at Aarhus University, U.S. Department of Agriculture, Ross County Soil and Water Conservation District, U.S. Geological Survey - Ohio-Kentucky-Indiana Water Science Center, Central State University - C.J. International Center for Water Resources Management, University of Tennessee Institute of Agriculture, Green Aero Tech USA, Attica - United States. |
Funding: U.S. Department of Agriculture and Aarhus University. |
Conflict of Interest: None |
Read more: You can read the publication “Overall results and key findings on the use of UAV visible-color, multispectral, and thermal infrared imagery to map agricultural drainage pipes” here. It is written by Barry Allred, Luis Martinez, Melake K. Fessehazion, Greg Rouse, Tanja N. Williamson, DeBonne Wishart, Triven Koganti, Robert Freeland, Neal Eash, Adam Batshelet, and Robert Featheringill. |
Contact information: PhD student Triven Koganti, Department of Agroecology, Aarhus University. Email: triven.koganti@agro.au.dk |