- Agronomic: corn, cotton
- Crop Production: application rate management, fertilizers, nutrient management
I’m about to become a graduate student in Mississippi State University’s (MSU) agronomy program. Last year, I became an FAA certified remote pilot and am now working with a team of researchers who use sUAS technology to create variable rate prescription N maps. Previous research in Arkansas showed a reduced N requirement in cotton (from 110 lbs/acre to 89 lbs/acre) with a 3 lbs/acre increase in lint yield for every 1 lb/acre reduction in total N. This algorithm needs to be tested in both cotton and corn.
My father, Tap Parker CEO of Parker Farms, agrees that our Lake Providence, LA, farm site is ideally suited to test these tools. Our farm operation aims to be environmental stewards and cooperates with environmental programs aimed at increasing N-use efficiency. However, we are still applying N at rates that are, most likely, higher than optimal. I would also like to test a new means of sensing early crop N status using emerging unmanned aerial technologies.
My graduate program director supports my research effort in giving me access to an unmanned flight system that carries a state-of-the-art vegetation sensor; valued at $10,000. I train under an unmanned flight specialist in autonomous quad-copter technology and I fly fixed-wing unmanned aircraft, but have not yet gained full command of unmanned, autonomous fixed-wing flight systems. The quad system is available to me, as a student, at no charge from the University. I must hire the fixed-wing technology from UAS Solutions (Clinton, MS) because the platform is highly specialized and unproven. Our program currently does not own this technology but my research may validate a need to obtain the system.
Because there is no research to validate quad- vs. fixed-wing technology for VRN mapping in early crops, I believe my research will be ground-breaking and may elucidate a means to sense much larger areas than can currently be achieved using common quad-copter technology.
The purpose of this grant request is to secure funds to: 1) test novel algorithm for improving N-use efficiency in corn and cotton, and 2) test two different flight systems for best use practices. It is unknown which unmanned system will provide the most accurate data for this project. Previously, the research used quad-copters but their battery life is limited to about 30 acres or less. I would like to test the difference between quad-copter technology and fixed-wing technology, which we believe can fly five times longer than the quad-copters. The results of this research will demonstrate how accessible and affordable this technology is to producers, and this research will also demonstrate sustainable, N-reducing techniques.
Project objectives from proposal:
This study will be conducted for two years on corn and cotton in Lake Providence, LA, between 2020-2021. The fields are five miles west of the Mississippi River and managed under the Mississippi River Basin Initiative (MRBI) Environmental Quality Incentives Program (EQIP), administered by the National Resources Conservation Service.
Prior to planting, soil sample cores will be taken at 0-15 and 15-30 cm, and sent to Waypoint Analytical (Jackson, MS) for total N percent (NO3-N and NH4-N) composition. The resulting soil N map will be derived through IDW interpolation. This will aid establish the difference between optimal crop performance occurring from the variable rate N application and/or residual soil N resources.
Both corn and cotton will be planted according to the timetable provided. Approximately half the total required crop N [URAN®-32 (32-0-0)] will be knifed in at planting. At peak early corn and cotton growth stages (V4-V6 and pinhead square, respectively) sUAS missions will be flown at 400 ft above ground level with MicaSense RedEdge® (Seattle, WA) sensor technology. Both the quad-copter (flown at 20 mph) and fixed-wing (flown at 40 mph) platforms will be used to compare efficiency in each method. Time trials on acres flown will be documented to establish sensing efficacy of each flight system.
The VRN treatments will be established on 24-row plots, Randomized Complete Block Design, with a full fixed rate as a control and two different VRN applications (from two different aircraft imaging missions) as treatments. Four reps of the three treatments will be established on 60 acres for each crop.
The imagery data returns estimate early crop N status, which supports the calculation of variable rate N sidedress recommendations. Immediately after flying these missions, leaf tissue samples will be taken to relate leaf N% to the VRN map and validate imaging efficiency. The leaf N% samples will aid in validating the research method of assessing early crop N status.
Once the sensing missions are completed, the data is process and the maps are created, the VRN maps will be delivered to the equipment operator to be applied. Throughout the growing season, crops will be managed in accordance to recommendations for irrigation scheduling, pest and weed control.
Harvest data will be collected in digital format for relating yield to VRN prescription. This will aid in determining the efficacy of the VRN mapping operations from unmanned systems.. Also, the yield data will be compared to total N calculation (residual soil N + VRN prescription) to find the full N amount used throughout the growing season. Linear statistics will be used to model the optimal yield rates against total N fertilizer applied. Data outliers will be determined using Cook’s D statistics. Yield points that exceed or fall below the expected range will be examined in a separate study in order to not skew data results. N-use efficiency will be reported as Yield per unit N applied. Comprehensive N-use efficiency will be reported as Yield per unit N applied including residual soil N observed prior to planting.