Going deep

Soil moisture sensors are a valuable tool that can be utilized to schedule irrigation. Sometimes it can be difficult to determine how to make decisions from sensor data on when and how much irrigation a crop requires. These factors are determined by soil type, crop growth stage, irrigation capabilities and current environmental conditions. In a study from 2022, a corn trial was planted in a Lucy loamy sand soil and sensors were integrated into one probe at 8, 16 and 24 inches deep. The main focus of this trial was to utilize soil moisture sensors to aid in determining both yield and economic penalties for over- and under-irrigating corn during the production season. While yield data were not available for this article, the main focus here is on in-season soil water tension and the implications of varying soil water tension levels. This study is being conducted at the University of Georgia’s Stripling Irrigation Research Park, near Camilla, Georgia. Nine methods were implemented under a variable rate lateral irrigation system. They included 20, 30, 40, 50, 60 and 70 kilopascals of soil water tension (SWT), the SmartIrrigation Corn App, UGA checkbook method and nonirrigated.

During the middle of the corn production season in Georgia in 2022, it became very hot and dry. Corn is typically planted during the middle to end of May in Georgia and reaches peak water use during May to June. Thus, hot and dry conditions during this time period show up very rapidly in soil moisture data and irrigation requirements. It should be noted that the data shared here ranges from May 2 to Aug. 3.

The corn was planted on March 29, and between that date and June 21, there were only 9.5 inches of rainfall received at SIRP. Three significant rainfall events affected the SWT on May 24, June 1 and 14. It is important to note these dates, as these events affected the SWT in some treatments. For irrigation treatments, the soil water tension data was weighted by rooting depth, meaning in early season the deeper soil moisture sensors were not utilized. As the season progressed, the data from the deeper sensors were integrated into irrigation decisions.

As shown in figure 1, overall, deep moisture was retained until the end of June. However, at that time, it can be seen by the drop in the 16-inch sensor depth (orange line) that water use was occurring at a more rapid rate. In this treatment, drastic changes only occurred in the 8-inch sensors (blue line), but soil moisture was recovered with irrigation relatively easily. Past studies have shown that this trigger level, while not yield-limiting, causes profit loss due to the cost of overirrigating without associated yield returns. Figure 2 represents more drastic changes in all three depths, showing that not maintaining soil moisture at a higher level caused the crop to utilize moisture from deeper levels faster than the 20 kPa treatment. Except for the 24-inch sensor (gray line), soil moisture was able to be recovered at the 8- and 16-inch depths with irrigation.

As the SWT was allowed to increase to a higher average level in the 40 kPa treatment (figure 3), the changes in the 8-inch sensor were more drastic, and we could not recover moisture at the 16-inch depth as easily with irrigation. However, moisture was not utilized at the 24-inch depth providing the assumption that there was still adequate shallow moisture in the profile.

Similarly, figure 4 represents the 50 kPa treatment, where more drastic changes were observed in all three depths and there was no recovery of water use at the 16-inch sensor without a significant rainfall event. While past studies have shown this SWT treatment was not yield-limiting and did not maximize profit, any mistakes at this soil moisture level will lead to yield reductions. Figures 4 and 5 are very similar, except the scale in figure 5 reached a higher level and it was more difficult to recover deep (16-and 24-inch) moisture in the 60 kPa treatment (figure 5). The rainfall events were able to lower the SWT levels, but it should be noted that there was a delay as the moisture took time to infiltrate to the deeper levels.

Figure 6 shows a very rapid utilization of soil moisture at all levels very early in the season. This is proof that if there is not adequate soil moisture the plants will work to access soil moisture where it is available. It can be assumed that the extra energy required to access moisture at these levels is most likely yield-reducing. It is also important to note that deep soil moisture was not recovered except during the three significant rainfall events mentioned earlier, but with a significant delay in recovery. The lagging graph lines for the deeper depths show this delay.

The SmartIrrigation Corn App (figure 7) keeps the soil moisture levels at an adequate range during the entire season and follows a similar pattern to the 40/50 kPa treatments. Past studies have shown that the app performs well. This is very good news showing that there is a tool that can perform even in “extreme” weather conditions without sensors in the field.

Overall, the checkbook method was designed to, and tends to, overirrigate. This can be seen up until mid-May. After this point, a general decline in overall soil moisture compared to some of the lower SWT treatments is present, especially the deep moisture, reinforcing the fact that the checkbook method was developed on a historical average and will miss the target in years that are not “average.” In this case, 2022 was hotter and drier than average and the checkbook method was not able to keep up with the total water requirements later in the season, since evapotranspiration
was higher than average.

Figure 9 or the nonirrigated treatment rapidly depleted soil moisture by mid-May. As can be seen, all sensor depths reached a SWT of 150 (or higher) by mid-May, it should be noted that the sensors do not read over 200 kPa, showing that there is basically no moisture left at any of these depths. Even though there is some recovery of the soil moisture after the rainfall events, there was very little to no recovery in the deepest sensors even with higher rainfall events. This treatment will have a very low yield due to the lack of any moisture and heat during tassel.

As stated above, figure 10 shows the 9.5 inches of rainfall that has been received over the past 80 days. These intense hot and dry conditions during the testing time period allowed us to learn interesting things about soil moisture management using sensors in-season.

Wesley Porter, PhD, is an associate professor of crop and soil sciences at the University of Georgia.