UC Davis

Changing climate and the increase in agricultural demand has increased water scarcity and highlighted the importance of how irrigation is managed. Precision and deficit irrigation are two management strategies that have been used to save water. Precision irrigation can simply be described as applying the right amount of water to the right place, at the right time. Deficit irrigation can simply be described as providing the plant with less water than it could use. In both cases, evapotranspiration (ET) is an important tool for management decisions as it provides the “right amount” information for precision irrigation and can be used as an indicator of plant stress in deficit irrigation. Maps of ET are an even stronger tool as they can indicate spatial patterns throughout the entire field, answering the “right place” for precision irrigation and indicating potential troubled areas for deficit irrigation. With the advancement of remote sensing technologies, and specifically drones, maps of ET have become more available to growers and other managers. In this dissertation, I used remotely sensed ET for three different applications related to precision and deficit irrigation. The High-Resolution Mapping of Evapotranspiration (HRMET) model was used to create maps of ET for this dissertation. Chapter 2 evaluates the impact of capture elevation on drone-based ET estimates in a commercial Wisconsin potato field. Missions were conducted at an elevation of 30, 60, and 90 m. While there was a noticeable increase in resolution with the lower elevation missions, there were no observable differences in ET between the different elevations. This indicated that for Wisconsin potatoes it would be preferable to go with the higher capture elevation as it has smaller storage and data processing requirements while producing similar results. Additionally, I conducted a Monte Carlo based uncertainty analysis on the inputs HRMET model in Chapter 2. The spatial inputs for HRMET were the biggest source of uncertainty, stressing the importance of following best practices for collecting remotely sensed data and would help guide data collection for the other two chapters. Chapter 3 evaluates the impact of four deficit irrigation treatments on California processing tomato yield and marketable quality and identifies connections between above and below ground stress indicators. This chapter uses data from the 2020-2022 growing seasons in three different commercially managed processing tomato fields in Fresno County, CA. Each season used a different “late season” processing tomato cultivar with similar cultivars. The four deficit treatments were initiated after the fruit had set and ranged from 37-70% of crop ET. I found that there was no significant reduction in yield or marketable value with the increase in the amount of deficit irrigation. In some cases, deficit irrigation led to an improvement in marketable values like brix and hue. These findings, in addition to the 29% reduction in irrigation, imply that deficit irrigation is a commercially viable strategy to reduce water use in CA processing tomato. Remotely sensed surface temperature and evapotranspiration were the above ground stress indicators and soil moisture was the below ground stress indicator used in Chapter 3. Remote imagery was collected bi-weekly prior to the fruit set and weekly after the fruit has set for each season. Soil moisture and salinity data were collected continuously during the 2022 growing season. The above and below ground stress indicators were evaluated using a Bayesian regression model. The above ground stress indicators had a mixed response, where the 2020 season had a significant response while the 2022 season saw no significant difference between the deficit treatments. This could be attributed to cultivar differences, as some cultivars some show signs of water stress sooner than others. The below ground stress indicators had significant differences between treatments in soil moisture and salinity as the season progressed. However, differences in soil moisture and salinity were also seen within the treatments as the season progressed, indicating that soil moisture and salinity already had significant changes with the grower standard treatment. Chapter 4 compares the performance of self-preservation based functions for scaling hourly to daily drone ET estimates in multiple crops, climates, and biophysical conditions. Drone based ET is a powerful tool, but the nature of fluxes measured gives it an hourly timestep. To best be utilized, hourly ET estimates should be scaled to daily values. Numerous scaling methods have been developed for this purpose, based off the theory of self-preservation. The self-preservation theory assumes that throughout the daytime, the components of the energy balance maintain a constant ratio. This chapter used data from commercial potato, cherry, and processing tomato fields during the 2019-2022 growing seasons in Wisconsin and California. The sine, net radiation ratio, crop coefficient, and evaporative fraction methods were selected as the scaling functions for their ease of calculation and general performance in prior studies. Scaling functions were compared by crop and then the combined dataset was evaluated by groups based off differences in ET driver parameters. The evaporative fraction method performed the best when the scaling functions were applied by crop group, providing an easy recommendation for scaling. However, there was an improvement in performance when the missions were separated by vapor pressure deficit groups, providing a framework for future decision support. Overall, this dissertation addresses knowledge gaps in the application of remote sensing, and in particular the application of drones, in precision and deficit irrigation.