Daniel L. Thomas

Evaluation of the EPIC model for simulating crop yield and irrigation demand

An understanding of water needs in agriculture is a critical input in resolving the water resource issues that confront many southeastern states. Unfortunately, how much water is required and how much water is actually being used for irrigation in Georgia is primarily estimated and largely unknown. The objective of this study was to evaluate the performance of the Environmental Policy Integrated Climate (EPIC) model in simulating crop yield and irrigation demand for three major crops in Georgia. Model evaluation is necessary to provide credibility in applying the model for simulating water use by agriculture. Seasonal yield and irrigation data for the 1990 through 2001 crop variety trials conducted at five agricultural experiment stations were used to evaluate simulation of yield and irrigation amount. The root mean squared deviation (RMSD) for yield was 0.29 t/ha for cotton, 0.39 t/ha for soybean, and 1.02 t/ha for peanut. The RMSD for peanut was large because the model tended to underestimate high yields and was not as sensitive to the factors responsible for the year-to-year variability of peanut yield. The RMSD for total amount of irrigation was 75 mm for cotton, 83 mm for soybean, and 87 mm for peanut. The model simulated the mean irrigation amount and the magnitude of annual variability very well. The component of mean squared deviation (MSD = RMSD2) related to the pattern of annual variability in irrigation amount contributed most to MSD. Overall, the results showed that the EPIC model can be a useful tool for simulating crop yield and irrigation demand at a field level. Future efforts will focus on using the model for regional estimation of water use for irrigation in Georgia and other southeastern states.

MEASUREMENT OF WATER TABLE FLUCTUATIONS IN A SANDY SOIL USING GROUND PENETRATING RADAR

A study was conducted to assess the impacts of ground-penetrating radar (GPR) calibration techniques on the water table depths calculated from the GPR graphical output. A Geophysical Survey Systems, Inc., SIR-8 impulse radar with a 120 Mhz antenna operating at a scanning rate of 100 ns was used to produce GPR images along seven, 130 m transects of a field site. The study site is located near Tifton, Georgia, in the Coastal Plain physiographic region of southeastern U.S. The soil on the study site is classified as a Lakeland sand with profile depths from 1.9 to 4.4 m. Underlying the sand is a restricting layer consisting of tight clays.

THE PEANUT YIELD MONITORING SYSTEM

The most essential component of precision farming is the yield monitor, a sensor or group of sensors installed on harvesting equipment that dynamically measure spatial yield variability. Yield maps, which are produced using data from yield monitors, are extremely useful in providing the farmer a color–coded visual image clearly showing the variability of yield across a field. University of Georgia scientists recently completed development work on PYMS, the Peanut Yield Monitoring System. PYMS uses load cells for instantaneous load measurements of harvested peanuts and has proven to be accurate to between 2% and 3% on a trailer–load basis and to approximately 1% on a field basis when using data collected during combine operation. PYMS data are accurate to around 1% on a basket–load basis when using data collected under static conditions. The instantaneous accuracy of PYMS was calculated to be 700 kg/ha. Basing management decisions on the yield of individual pixels of PYMS yield maps is not realistic. The strength of PYMS is in differentiating yield trends and evaluating management practices. The system was extensively and successfully field–tested over a 3–year period and evaluated by 11 users during 1999, all of whom were able to use the resulting yield maps to evaluate current management practices or to develop future management plans. The University of Georgia has submitted a patent application for PYMS, and the technology has been licensed.

DECONVOLUTION OF SITE-SPECIFIC YIELD MEASUREMENTS TO ADDRESS PEANUT COMBINE DYNAMICS

During the development of a peanut yield monitoring system, experiments were conducted on a two-row peanut combine to determine the duration of time lag between pickup and yield measurement, and to characterize the convolution of peanut flow within the combine. The research indicates that the two-row peanut combine used in the experiment subjects harvested product to significant convolution. A simple time lag correction will not recover the site specific (short term accuracy) of yield measurements. The distance and time period required to achieve a yield estimate error less than 20% (95% confidence) is greater than 19.7 m (17 s) for simple time lag correction while it is 5.8 m (5 s) for deconvoluted data. The net result is that smaller regions of yield variability may be recognized with greater confidence using the deconvolution method than with the simple time delay method.

Drainage-subirrigation design for Pelham loamy sand

ABSTRACT This project was conducted on the Bell farm located in Pierce County, GA on Pelham loamy sand soil. The study area included 40 ha of land under controlled drainage-subirrigation (CD-SI) system of which 38 ha were in blueberries. The system installation included two drain spacings of 15.3 and 20 m, and two types of control structures for the drainage system outlet water level, which were an open ditch and a closed conduit system network. The blueberry section of the field was in a closed conduit system network with lateral drains spaced at 15.3 m. Seventeen punch-tape recorders were used to measure the water table elevations in the soil profile within the field over the drain tiles, and midway between drainlines, at the open ditch and closed system control structures (one for each), and in an undrained-nonirrigated section of the farm. A punch-tape rainfall recorder was also used to measure rainfall at the site. Surface runoff, drainage effluent, or subirrigation volume was not measured. Experimental results showed that the water table and soil-water conditions could be adequately managed in the blueberry field, thus excellent crop growth and yield resulted. DRAJNMOD, a water management model for shallow water table conditions, was used to simulate the system performance for the study site. Simulation results indicated that a subsurface drain spacing of 20 m is satisfactory for controlled drainage-subirrigation (CD-SI) systems for Pelham loamy sand soils in Pierce County, GA. Simulations were done using 20 years of climatological data from Augusta, GA. Additional research is needed to develop specific design guides for other soil types in the Georgia Flatwoods region.

EXPERT SYSTEM-BASED COUPLING OF SOYGRO AND DRAINMOD

ABSTRACT A combined crop growth (SOYGRO 5.4) and water management (DRAINMOD 3.4) modeling system was developed using an expert system shell to coordinate input and output between the models. An iterative simulation procedure was used to keep the models intact and modular for inclusion of additional crop growth, pest, economic, and environmental impact models in the future. DRAINMOD 3.4 was modified to accept evapotranspiration and effective rooting depths from SOYGRO and allow feedback control. The feedback control system performed satisfactorily. SOYGRO 5.4 was enhanced to accept from DRAINMOD water movement due to upward flux from a water table. In addition, the model was modified to account for the crop stress due to oxygen deprivation caused by a water table extending into the rooting zone. Simulated soybean yields and root length densities were greatly affected by a raised water table and by climatic variations for the tested conditions. However, if the water table was maintained within 90 cm of the soil surface during the entire growing season, weather patterns had only a minor influence on root length densities and yield.

Estimating Water Demand For Irrigation Using A Crop Simulation Model

Crop yield and water demand for irrigation under rainfed and irrigated conditions for four major crops in Georgia were estimated using the Environmental Policy Integrated Climate (EPIC) model. Seasonal yield and irrigation data during 1990-2001 for Tifton, Plains and Midville in the Coastal Plain region, Griffin and Athens in the Piedmont region, and Calhoun in North Georgia were used for evaluating simulated yield and irrigation. Under rainfed conditions, the model performs fairly well for different crops, weather and soil conditions across Georgia. In general, the model tends to overpredict for low yielding conditions and underpredict for high yielding conditions. Under irrigated conditions, the model overpredicted to a greater extent for low yielding conditions and underpredicted to a greater extent for high yielding conditions. Only for cotton, the model simulated the year-to -year variability in measured irrigation fairly well.

Pecan mechanical harvesting parameters for yield monitoring

A study was conducted to examine how pecan yield from individual trees can be correlated to pecan yield measurements from a mechanical harvester equipped with a yield monitor that weighs pecans as they are dropped into a wagon. For purposes of determining individual tree yields, a method was devised for taking the yield of pecans and foreign material collected by a mechanical harvester close to the trunk of the tree and extrapolating to determine the entire yield for that tree. Data were collected for 2 years on 12 pecan trees. Total pecan yield for a tree was closely correlated (r 2 = 0.84) to the gross weight of pecans (pecan yield plus foreign material) in a 4 m section of harvested material. It is expected that comparison of tree yields could help improve pecan grower’s selection process when removing unproductive trees and help researchers assess yield differences for test treatments.

Simultaneous Assessment of Cotton Yield Monitors

The most essential component of precision farming is the yield monitor . a sensor . or group of sensors . installed on harvesting equipment that dynamically measure spatial yield variability. Yield maps, which are produced using data from yield monitors, are extremely useful in providing a visual image to clearly show the variability of yield across a field. In response to the demand for a reliable and accurate cotton yield monitor, several monitors have recently become commercially available. We assessed the AgLeader, AgriPlan, FarmScan, and Micro-Trak cotton yield monitors in southern Georgia for five harvest seasons between 1997 and 2001. During 2001 we also assessed a prototype yield monitor. Each year, three or four yield monitors were mounted on a cotton harvester and used during harvest of several farmer-owned and managed fields. The accuracy of each sensor was tested by comparing the weight of each harvested load to data produced by the yield monitors. Yield maps from each yield monitor were also produced with the respective software packages and compared. Feature comparisons of each monitor were included. Each of the cotton yield monitoring systems we assessed have something to offer a user interested in creating yield maps. All are capable of producing an adequate yield map provided the system is properly calibrated, operated, and maintained.

Noise reduction in a load cell based peanut yield monitor using digital signal processing techniques

Precision farming describes the process of measuring and mapping land crop characteristics and then using these measurements to develop precise and intelligent application strategies that improve overall farm production. Yield monitoring is the phase of precision farming in which the crop yield variation within a field is measured and mapped. Yield maps from previous seasons can be used to determine the needed inputs in the field, whereas post harvest yield maps can be used to evaluate the implemented methods and make adjustments for the next season. Grain yield monitors are available, however there are few if any monitors for other crops. This paper examines the use of strain gauge load cells in a yield monitor for peanut combines and the development of methods for minimizing measurement noises thereby increasing reliability.