Project Results

2020 Crop Year

In 2020 (year 2), the project included 8 counties, 7 agents, 16 growers, and 16 cotton fields. Crisp County opted not to participate, the Decatur County agent left Extension and the new agent in Dooly County joined the project. The Mitchell County extension agent covered both Mitchell and Decatur counties. As with agents, some of the farmers that participated in 2020 were involved the previous year while several were new to the project. This may explain why the averaged survey responses (Tables 3 and 4) again indicated farmers and agents expressed limited knowledge of irrigation scheduling using soil moisture sensors or SmartIrrigation apps.

As in 2019, the participating farmers were again interested in trying both irrigation scheduling tools but still seemed to prefer the SMSS over the SmartIrrigation App. Again, like in the previous year, there were numerous technical issues with both the SMSS and the App, but the farmers were (mostly) willing to continue working with the technologies.

In Figures 13 and 14, data from the two soil moisture sensors on probes installed in two of the 8 counties in the project is presented. These two graphs illustrate the responsiveness of the SMSS, especially the shallow sensor, to rainfall and irrigation when telemetry connectivity was consistent and stable. The growers were also using the SmartIrrigation App in parallel with the SMSS.

Table 2 provides data from second year of an accompanying irrigation scheduling research project at the UGA Stripling Irrigation Research Park. The study again compared use of the SmartIrrigation App, an SMSS functionally equivalent to the Trellis SMSS, and the UGA Extension “checkbook” recommendation. Like in 2019, the SMSS scheduling tool led to significantly higher yields than the App or “checkbook” methods and had the highest IWUE as it required the least amount of irrigation during the season, by far. Note that the App method had the second highest yield with the “checkbook” method having the lowest yield. The App tool yield resulted in a higher IWUE than “checkbook” as the App method achieved that yield with nearly 2 inches less water applied.

Again, in 2020 only two counties were able to determine yield values from their two participating growers. In Terrell County, the two farmer yields were 1180 lb/ac and 1200 lb/ac. In Colquitt County, the two farmer yields were both 1200 lb/ac. Like most of the participating growers, these four did an exceptional job producing cotton as the Georgia state-wide average yield for 2020 was 887 lb/ac (includes both irrigated and dryland acres). The Stripling Park study did not produce yields quite as exceptional (as in 2019), but SMSS and App yields were above or at the state average.

Soil Moisture Sensor System

We opted to continue working with the Trellis company for soil moisture sensing gear as we had worked with that company during the 3 years of the Phase 1 – Pilot Project of AgWET. But during this Phase 2 project, Trellis hardware and telemetry had performance issues – with soil moisture values from sensors and with providing consistent, reliable, gap-free data from the farmer fields. These performance issues led to a loss or lack of confidence in agents and farmers in the data provided. Thus, few farmers or agents utilized the Trellis gear and data to actually schedule or trigger irrigation events.

SmartIrrigation Cotton App

As with most newly developed technologies, the SmartIrrigation Cotton App was updated and enhanced several times during this project timeframe. The foundational water balance algorithms were “tweaked” a few times to improve the performance of the App.

As noted earlier, to improve the performance of the App, programming changes were incorporated to link the App to data from the Trellis rain gauges that were installed in each farmer field. This linkage was between the cloud-based App and the server hosting the Trellis data. The App continued to use other weather data obtained from the nearest UGA Weather Network station.

During the course of the project, the Trellis rain gauge hardware physically had performance and accuracy issues plus there were problems with telemetry and data sharing. Thus the rainfall data from these gauges was often not particularly reliable nor consistent. Gaps of or inaccuracies in rainfall data caused the SmartIrrigation Cotton App to not perform optimally for many of the fields. This led to a loss of confidence in the App’s usefulness.

Where the App accessed data solely from a local UGA Weather Network weather station, such as at the Stripling Irrigation Research Park, the App performed very well.

Social Science Results

In late 2018 and early 2019, as part of a MARS Wrigley Confectionery funded project, a team of UGA social science collaborators, including an agricultural economist and agricultural communication specialists, interviewed 10 farmers and surveyed 86 additional “irrigators” in southern Georgia (all non-participants in this project). From these interviews and surveys, the social science team uncovered several major barriers to adoption of advanced irrigation scheduling tools. This barriers included, but were not limited to:

• Limited cell phone/broadband connectivity;
• Stress from time and money commitment;
• Installation/operation/maintenance costs;
• Sufficiency with current scheduling methods;
• Lack of knowledge about new methods; and
• See potential in technology, but are uncertain of dollars saved.