SDSU researchers explore automated controlled drainage systems

South Dakota State University's Department of Agricultural and Biosystems Engineering has recently been exploring the viability of automated controlled drainage systems in eastern South Dakota. The systems would allow farmers to maximize the available water and potentially improve their yearly crop yields. 

Wet springs, dry summers. That's been the trend in eastern South Dakota, and farmers aren't particularly thrilled. But what if early rains could be saved, and utilized, as relief for a later dry spell?

It may no longer be just wishful thinking. Researchers in South Dakota State University's Department of Agricultural and Biosystems Engineering are investigating if an emerging technology — automated controlled drainage systems — could hold the answer.

"Controlled drainage has been around for a while and has proven benefits for farmers with tile drainage systems installed in their fields," said John McMaine, an SDSU associate professor and the Griffith Chair in Agriculture and Water Resources. "Automated controlled drainage takes this technology one step further."
 
What are controlled drainage systems?
Controlled drainage structures were developed to add more flexible management for tile drainage (also known as subsurface drainage) systems on cropland in the Midwest, where the systems are widely used. According to a study conducted by Iowa State University, more than 40% of Midwest farm fields utilize tile drainage systems. Although tile drainage is a long-used agricultural practice with many advantages, it is recognized there are also potential detrimental effects the systems can have on the surrounding environment.

Maryam Sahraei is a graduate research assistant in SDSU's Department of Agricultural and Biosystems Engineering. Her research seeks to understand how farming practices affect downstream water quality.

"My research looks to improve water quality," Sahraei said. "One of the biggest potential issues is that if outflow from tile drainage contains high levels of nitrate, there is no buffer before this water enters a river or stream."

While tile drainage systems effectively move water away from the root zone of growing crops, if excess nitrate or soluble phosphorus is present, then they move these excess nutrients into waterways as well.  Scientists have determined that excess nutrients cause algae blooms in downstream water bodies, disrupting the ecosystem's natural balance and creating unfavorable conditions for water recreation enthusiasts. In addition, algal blooms can cause potential human health issues, as was seen in the 2014 Toledo Water Crisis. In this case, microcystin from a harmful algal bloom contaminated the drinking water for approximately a half million people.

In the Midwest, most waterways eventually find their way to the Mississippi River and eventually, the Gulf of Mexico. There, the detrimental effects of nutrient-laden water are even more bleak.

According to the U.S. Environmental Protection Agency, nitrate from tile-drained fields in the Midwest and other sources — such as leaking septic systems — are a major contributor to hypoxia zones in the Gulf of Mexico. These zones, often referred to as "dead zones," decreases the available oxygen for aquatic life, contributing to a reduction in the number of harvestable fish and shellfish in the Gulf. Many states (not including South Dakota) along the Mississippi River have been tasked with reducing nutrient loading in the river's many tributaries.

"How we manage will have an effect on downstream water quality," Sahraei said. "In this case, if we lose excess nutrients in eastern South Dakota, it may have an effect on water over 1,000 miles away."

According to previous research, controlled drainage has resulted in a 40-50% nitrate reduction in water being discharged into nearby lakes or rivers. However, Sahraei's research has also found that nitrate loss varies across drainage outlets, from around 2 mg/L to over 60 mg/L.
 
Searching for solutions
Although tile drainage continues to provide benefits, researchers began investigating to see what solutions there may be to address these issues. In response, farmers began implementing controlled drainage systems, which function in tangent with tile drainage systems.

Controlled drainage technology ("flow-control") is quite simple: a structure is placed within the drainage pipe to hold back water or let it flow. Boards allow the farmer to choose how much water can be discharged at a given time. The more boards in place, the more water that is held back. When the boards are removed, the water flows conventionally. These boards can be best described as "mini-dams."

Simply, controlled drainage increases water retention in the soils to raise the depth of drainage outlets, operating more or less like spillways. The system helps to hold back more moisture in a field during periods when drainage is not needed, according to McMaine.

One of the challenges with these types of systems is knowing exactly when to hold more water and when to let water go. For example, boards are often added to hold back water during the growing season. However, extra attention must be taken to prevent a high-water table from limiting crop growth. Most farmers go off past experiences or feel — which can sometimes cause the system to be inefficient or even ineffective.

The research being conducted by McMaine and his graduate students on automated controlled drainage will take the technology one step forward by creating algorithms that will automatically let water go when drainage is needed, eliminating the guesswork and improving the system's effectiveness. By utilizing real-time data, the system will utilize precision water management principles to use resources as efficiently as possible. Early studies have shown that this precision allows for higher crop yields and improved downstream water quality. 
 
Experimentation at Southeast Research Farm
In fall 2023, McMaine and the graduate students under his tutelage ventured to SDSU's Southeast Research Farm, outside of Beresford, to install the automated controlled drainage systems in the tile-drainage system fields. Installing the systems in the fall ensured it is ready to collect data as soon as planting season begins this spring.

Josh Becker, an agricultural and biosystems engineering major, is a graduate student working under McMaine's tutelage. Becker was on-site this fall helping to install the automated drainage systems and gained valuable insight into this technology's potential.

"My experience with this project has been eye-opening," Becker said. "Growing up in an agricultural background, I had gained a fair understanding of tile drainage used for crop production. However, controlled drainage was a new topic to me, and automated drainage dives further into that technique. Using precision agriculture practices, farmers can now autonomously control structures in either real-time or through pre-programming."

Preliminary data will be gathered first after the spring growing season and then again in the fall, after the harvest season is complete in October 2024. 
 
Algorithm development
Murad Ellafi is a postdoctoral research associate working under the guidance of McMaine. He had previously earned his doctorate from Cranfield University in the United Kingdom, where he focused on using artificial neural networks to design drainage systems in regions with scarce data.

For this project, Ellafi is charged with creating the algorithms needed for the automated aspect of the controlled drainage system, which will occur once the data collection period is complete.

The algorithm creation process will center around running hundreds or thousands of simulations with DRAINMOD, a drainage simulation designed to replicate the hydrology and water quality of poorly trained soils with high water tables.

"DRAINMOD is widely applied, especially in the Midwest, to predict the effect of drainage on water quality, soil-water regime and yield," Ellafi explained. "Since we have a calibrated model, we will be able to incorporate historical weather data to simulate the effects of automated controlled drainage systems on nutrient load reduction and potential crop yield improvements. These simulations aim to understand the systems' performance under varying weather conditions, focusing on their efficacy in managing nutrient loads and enhancing yields."

The simulation will help inform the machine learning-derived algorithms, which aim to predict the most effective drainage strategies for a given field or farm.

"The goal is to finalize the algorithms by summer 2025," Ellafi said. "This timeline allows for the thorough collection, analysis and the incorporation of the necessary data into DRAINMOD, followed by the development and calibration of the algorithms based on the model's simulations." 
 
Advantages to automated controlled drainage systems
Controlled drainage systems are intended to improve downstream water quality by reducing nitrate discharge. The system is also intended to improve yearly crop yields by utilizing the available water as efficiently as possible. Automated controlled drainage systems intend to make the systems even more effective by eliminating the guesswork and supporting the necessary decision making with data.

"Automated controlled drainage elevates water conservation efforts to a new level — a crucial advancement given the agricultural impacts of more extreme weather patterns in South Dakota," Ellafi said. "The state has experienced a notable increase in temperature, compared to the average of the last century, alongside shifts in rainfall patterns. These climate changes pose a direct threat to agricultural productivity."

Past research from SDSU examined crop insurance data from four eastern South Dakota counties. The data showed that soil moisture extremes — both excess and drought — were present in 90% of the years between 1990 and 2020. Managing water efficiently and effectively has never been more crucial for agricultural productivity and sustainability.

"This technology seems applicable in many situations in eastern South Dakota and western Minnesota," Becker said. "This technology serves to grow a more consistent crop yield, while significantly reducing the amount of nutrient contribution to nearby water bodies. A farming practice which simultaneously improves yield and contributes to protecting the environment is certainly one to get excited about." 

An added bonus is the economical nature of the controlled drainage structures. As McMaine notes, the structures themselves are relatively cheap, especially in comparison to other types of drainage strategies.

But how much water could the system actually save?

According to McMaine's and others' research, controlled systems save as much one inch of water in a single growing season, which would save somewhere between 16-24% of a crop’s yield in ideal circumstances.

The team’s data collection process will provide more specific data on automated systems for eastern South Dakota, but considering the changing climate conditions, any water saved could have major economic implications to farms in the region.

"Ultimately, these systems support the dual goals of environmental responsibility and agricultural efficiently, making them a valuable tool for South Dakota's farming community," Ellafi said.

"The biggest takeaway at this point in the project is that automated controlled drainage structures have a bright future," Becker said. "As the demand for food grows in the changing environment, new technologies must be developed to meet the need. Automated controlled drainage has shown promising signs as being a key contributor to a sustainable future, harmonizing productivity and sustainability."