Final Report for LS10-232
Project Information
This research was designed to evaluate tillage, conventional tillage and no-till, and crop management on organic grain production in a rotation of wheat-soybean and rye/crimson clover-sunflower. Despite numerous difficulties, including inability to control weeds in no-till crops, planting sunflower in heavy residues of cover crop, and inability to control major insect pests, we were able to produce crops. However, yields for sunflower and wheat under both tillage regimes were lower than expected for the cultivars used in the study. Yields for both conventional-till and no-till soybean, though, were within the range reported for the cultivar grown in conventional production systems.
Objectives
- Develop systems of organic grain production that integrate sunflower into traditional grain rotations using either conservation or conventional tillage
- Determine effects of different tillage systems for organic grain production on soil quality and soil physical properties.
- Transfer organic grain production technology to producers, technical service providers, and students.
The purpose of this project is to expand organic grain production in the Southeast US. The growing demand for organic food in the US (20% annually over the past 10 years, OTA, 2008) has resulted in price premiums for organic grains (corn, soybean, wheat) two to three times greater than prices for conventionally grown grains (USDA-ERS, 2006). From 2006 to 2008 organic bakers experienced shortages of US organic grain which resulted in nearly doubling wholesale grain prices (Inland Empire Business Journal, 2008). Increased regional demand has been noted in North Carolina for organic corn and wheat by two mills (300,000 bushels or 8,000 Mg) and for organic feed by dairies (2000 tons or 1,815 Mg of grain and 2500 tons or 2,268 Mg of forage; NCSU, 2009). Limited organic grain production in the region also limits development of the organic poultry industry in the Southeast.
Increasing organic grain production in the Southeast will require efforts to overcome the challenges of potentially lower yields (Drinkwater et al., 1998; Delate and Cambardella, 2004; Cavigelli, 2008), greater weed pressures, and more complex nutrient management and crop rotation decisions. These challenges were acknowledged by participants at the Georgia Sustainable Agriculture Summit (UGA, 2008) who identified research on organic grain production as a high priority place-based research need.
There has been little research on problems facing organic grain producers in the Southeast. A survey of Southern SARE funded projects from the past 10 years identified seven Research and Education grants focused on organic commodity crops. Three of those were organic from a group working on peanut production, one was directed at adapting no-till systems for small-scale organic producers and the other three had emphases on grains, weed control, and sustainability. Common among these research programs was efforts on weed management, soil fertility, use of cover crops, and application of conservation tillage in organic production systems. Predominantly these projects have focused on soybean, corn, wheat, and peanut. Additional research (non-SARE funded) on organic grain production is being conducted at the Universities of Kentucky and Tennessee but results so far are limited. Continued expansion of research efforts on organic systems under different climatic conditions and with other crops is needed to create a critical mass of knowledge and expertise to help organic grain producers be successful.
Our proposal will strengthen regional efforts on organic grain production through our emphasis on conservation tillage practices and expansion of crops with sunflower in rotation with wheat and soybean. When included in the traditional organic grains rotation, sunflower will increase rotation length and add diversity. New crushing facilities for sunflower established at two locations in Georgia have expanded the production of these crops in the region. Managers of these facilities have expressed an interest in producing organic oils which could help shift traditional growers to organic production. The growing demand for organic oil by consumers and demand for organic meal in animal feed will provide additional economic incentives for producer conversion to organic grain production over time.
Sunflower is a summer annual crop that can be grown as an alternative to soybean or corn and may be particularly useful in organic production systems by creating longer rotation cycles. It is mostly grown in the Northern and Central Plains states (NSA, 2009) but can be grown in many other parts of the US. Organic sunflower seed is used primarily in bird feed and for sunflower oil with the meal used for livestock feed. Sunflower oil has a favorable combination of monounsaturated and polyunsaturated fats with low saturated fat levels. Three types of sunflower have been developed for oil production that differ in oleic levels and offer unique properties. In addition, sunflower produces nectar which provides food for bees and other insect pollinators as well as natural enemies of pest insects. Previous research on intercropping sunflowers in organic production of vegetables indicated that these flowers attracted numerous natural enemies of pests (Jones and Gillett 2005). This can be a major benefit for controlling insects like stink bug adults because there are no effective bio-pesticides for their control in organic cropping systems (Tillman et al. 2009).
Many organic producers are interested in using conservation tillage practices for organic crop production. However, a major limitation is the need for economical and effective measures of weed control (without herbicides). Traditionally, weed control on larger scale acreage has relied on intensive tillage practices. Conservation tillage systems that rely on large quantities of crop residues to enhance weed control in conventional agricultural systems should be effective in organic systems but will require additional weed control strategies (i.e. shallow tillage instead of herbicides).
Developing organic grain production systems will provide an opportunity for regional producers to meet the growing demand from food companies and animal producers. The project will offer producers additional enterprise options for increasing farm income. Handling and processing of organic grains will provide conventional grain handlers with an opportunity to expand and increase profits due to the greater value of organic grains. Organic grain production relies on limited use of chemical inputs for nutrients and pest control which will be a more environmentally friendly approach to grain production and more compatible with the growing mixed urban/agricultural landscape in the Southeast. Developing a regional network of organic grain production will reduce energy consumption for transporting grains from other regions and reduce the transfer of nutrients from one region to another. Potential social impacts of the project will depend on the success of the cropping systems to diversify farm operations and income. Conventional producers will be attracted to adopting organic practices due to the increased economic return. Expanding regional adoption of organic grain production will strengthen rural communities through positive impacts on their social, economic, and environmental health.
Cooperators
Research
Study Location
The study site was located within the University of Georgia (UGA) Ponder Farm (31o 30’ 41” N, 83o 38’ 40” W) in Tift County Georgia, approximately 14 km NW of Tifton at an elevation of ~115 m (Fig. 1). A diagram of the experimental plots with a replicated crop rotation of wheat-soybean and rye/crimson clover-sunflower is shown in Fig. 2. The Natural Resources and Conservation Service’s Tifton County soil map shows Tifton loamy sand as the dominant soil in Rep 1, Rep 2, Rep 3, and the upper half of Rep 4, and Dothan loamy sand in the lower half of Rep 4. According to the descriptions given in the county soil map, the two soils are very similar in characteristics except that in the Tifton loamy sand, permeability is moderate throughout the profile whereas in the Dothan loamy sand it is moderate in the upper part of the subsoil and moderately slow in the lower. In both soils, typically the surface layer is dark grayish brown loamy sand about 25 cm thick (~>85% sand and <5% clay) underlain by dominantly sandy clay loam extending to a depth of 180 cm or more (<70% sand and 20-35% clay). In both soils, plinthite makes up of 5-10% of the lower part of the subsoil below a depth of ~75 cm. Many nodules of ironstone are present in the surface layer and upper part of the subsoil. Much spatial variability characterizes soils in farm lands in terms of these soil descriptions which gives rise to a similar typical spatial variability in other important soil properties such as soil water characteristics, physical and chemical properties, and fertility status, etc.
Experimental Design
This research was designed to evaluate tillage, conventional tillage and no-till, and crop management on organic grain production in a rotation of wheat-soybean and rye/crimson clover-sunflower. Crop rotation and different phases of crops over the 2-yr project are shown in Fig. 2. Each phase of the rotation was established initially so that crops were grown each year to capture year-to-year effects on production. The experiment had four replications arranged as a split-split plot design with tillage serving as whole plot and crop rotation phase serving as the split plot. To avoid possible weather-related failure of this budding research, the experiments were conducted under irrigated conditions. Plots were under two single-tower center-pivots each covering approximately 1.8 acre. Two of the 4 Reps fell under one pivot.
In treatments with crops grown with no tillage, cover crops were grown to suppress weeds. A cool-season rye/crimson clover cover crop was terminated with a front-mounted roller and sunflower simultaneously planted with a no-tillage vacuum planter. Cool-season wheat was harvested, if possible, and then soybean was planted. In treatments with crops grown under conventional tillage crop production, proven tillage practices for weed control were deployed at every phase of the cropping system. These included deep tillage once per season, shallow-tillage to control weeds and prepare a finished seedbed suitable for planting, and intense cultivation with a tine weeder.
Crops were selected based on traditional crops in the region and newly emerging markets for sunflower oil. Crop cultivars were selected based on their adaptation to the region and having attributes (i.e., pest resistance, agronomic qualities, and yield potential) that maximized their role in this cropping system. ‘Dixie’ crimson clover and ‘Wren’s Abruzzi’ rye were used. Group VII soybean (‘Woodruff’ soybean) was used because it tends to have greater vegetative growth than short-season cultivars which will be advantageous in weed suppression. ‘Georgia Gore’ wheat was used because it is adapted to the lower coastal plain regions of the southeastern US. ‘s668’ Sunflower was a cultivar with proven performance in the region and suitable for oil production. All crops were seeded at rates to maximize their yield potential and ability to suppress weeds as part of a cultural weed control strategy.
Weather
Daily weather data was periodically downloaded from the UGA Ty Ty Georgia Automated Environmental Monitoring Network Weather Station a few hundred feet from the research plots from Jan. 2013 through June 2014.
Soil Sampling
In mid-Oct. 2012, soil samples were collected from the 0-15 cm layer from six sites per plot and composited into one sample and sent to the UGA Soil and Water Analysis Lab to determine soil fertility.
In early April 2013 and 2014, soil samples were collected from the 0-15 cm layer from six sites per plot and composited into one sample and sent to the UGA Soil and Water Analysis Lab for nutrient analysis by the Mehlich 1 method.
In late Oct. 2012, core soil samples were collected (see Fig. 1 for location of samples) to get baseline information on soil C and N and mineralization rates. Soil cores were collected down to 1.2 m from 6 locations per plot. Cores were sectioned into 0-5, 5-15, 15-30, and 30-60 cm depths. Soil samples were then air-dried and composited by plot. Processed samples were sent to Dr. Alan Franzluebbers (USDA-ARS, Raleigh, NC) for NO3-N, NH4-N, total C and N, particulate organic matter (POM), soil microbial biomass C, flush of CO2-C, soil microbial biomass N and potentially mineralizable N (PMN), and aggregate stability analysis.
Soil Water
Soil sensors were installed (see Fig. 1 for location of sensors) to monitor soil water content. Sensors were installed (Fig. 3) in Reps 1, 2, and 3 in December 2012. In June 2013, three plots in Rep 4 were instrumented. Each sensor package from Spectrum Technologies came with a data logger with four ports to attach four soil water sensors into. We, therefore, installed sensors at 8, 22, 38, and 53 cm depths at each plot location to continuously monitor soil water content for the 0-15, 15-30, 30-45, and 45-60 cm depth soil profile. Soil water content sensing and logging occurred at ten minute intervals, and the data were downloaded to a laptop computer at one or two week intervals. Three unanticipated difficulties surfaced as we progressed with monitoring: calibration, cultivation activities, and critters.
Soil water sensors typically come with default calibrations for mineral soils. Manufacturers often assure clients that default calibrations would be adequate for general purposes but recommend on-site calibration from much improved accuracy. In our case, it was clear from the initial set of data we obtained that the latter option had to be pursued. Even following rainfall and wet conditions, sensors were giving soil water contents that would be expected under dry or very dry conditions. Therefore, we ran a series of calibrations in the field (undisturbed) and greenhouse (disturbed). The greenhouse calibration gave better curves. We settled on two curves, one for the 0-30 cm depth (loamy sand) and the second for the 30-60 cm depth (sandy clay loam) (Fig. 4).
The frequency of secondary cultivation in conventional tillage fields to keep weeds in check proved to be much more than expected, due perhaps to the above normal rainfall conditions in the growing season in 2013. For soil water monitoring, the result was that sensors had to be taken out of the fields frequently to avoid interference with implements and damage to the sensors. The end result was that data from conventional tillage fields were obtained less frequently than desired. When no-till sunflower plots had to be replanted, sensors had to be removed until better plant emergence in the replanted plots. The result was that not only did we lose soil monitoring data, but it also became impossible to compare soil water in no-till against conventional tillage sunflower due to the differential age of the crop in the two tillage treatments at any one time because of the replanting.
Towards the end of August 2013, we observed an increased infestation of fire ants and mice. The latter managed to create nesting within the 25-cm diameter PVC tubing covering the data loggers and to chew through the plastic sheath protecting above ground sensor wire cables in some fields. The sensors come with a standard 6-m cable length, much of which had to be above ground (although partially protected).
Soil Infiltration Measurements
We conducted soil infiltration measurements using double ring infilrometers in early April 2013 in eight wheat plots. The tall height of growing rye prevented us from collecting this data in this cover crop.
Plant Nutrients
We collected sunflower and soybean plant samples (eight plants per row for four random rows per plot) at approximately 21 day intervals starting early June through August 2013, and separated them into roots, stem, petiole, leaves, head, seed, and determined dry weight for quantification on a per hectare basis. At the same time, we measured plant height. We also measured leaf area before drying.
Dried and weighted plant samples were milled (Thomas-Wiley Model 4) as a precursor for nutrient analysis. To throw some light into the association of temporal and spatial distribution of insect infestation density with plant nutrient status (minerals, carbohydrate, protein, fiber, digestible matter, fat, starch, etc.), we have gotten the cooperation of Dr. Uttam Saha, Program Coordinator, Feed and Environmental Water Laboratory, Agricultural and Environmental Services Labs., University of Georgia, Athens, to run NIR (Near Infrared Reflectance Spectroscopy) and Wet Chemistry analysis on leaf samples collected from sunflower and soybeans. The NIR is a fast, and when properly calibrated, accurate procedure and less costly than Wet Chemistry analysis. Results from Wet Chemistry are used to calibrate NIR readings. We have two objectives to this collaborative study. One is that we get accurate nutrient content of leaves that can be used to interpret/assess insect infestation patterns. The second is that there currently are no NIR calibration curves for soybeans and sunflowers in the southeast. The effort will create a database that can be used by the larger SARE community in the southeast for a fast and relatively accurate assessment of plant nutrient status at reasonable cost. The 120 samples that we have recently sent to Dr. Saha are still being analyzed.
Planting Activities, Crop Plant Density, Weeds, Insects, Yields
Soil water conditions delayed soil cultivation to early October 2012. Field plots were first tilled to bury weed and crop debris and then deep turned and bedded to make similar starting conditions for all plots (Fig. 5). In late Oct. 2012, rye/crimson clover plots were planted. The rye/crimson clover cover crop was swept (24-ft of row per sample; one sample per plot) for insect predators and pests during the growing season.
No-till sunflower was planted on 4/24/2013. Rye/crimson clover was rolled and crimped and sunflower seeds were planted into the senescing rye/crimson clover cover crop (Fig. 6). It was very difficult to plant sunflower seed into the heavy residue of the rye/crimson clover cover crop, especially since the crimson clover was still living, and thus the resulting plant stand was very poor (Fig. 7). Due to lack of effective organically-certified herbicides, we were unable to kill the crimson clover prior to planting. Sunflower was re-planted in no-till plots on 5/31/2013. By adding row cleaners (like opposing circular saw blades angled into a chevron) to the Monosem no-till planter, a 6” wide strip was tilled in the seed row. Modification of the planter along with delay in planting until crimson clover was dead resulted in a better stand of sunflowers (Fig. 8).
In early-April, blind cultivation with a tine weeder was used for weed management in conventional-till sunflower. These plots were planted on 4/29/2013. Sunflower plant density was determined by counting the number of plants per 10 ft of row. One sample was obtained for each of 20 rows per plot. Sunflower heads within 6 ft. of row were sampled for insect predators and pests during the growing season; two samples were obtained for each of 6 random rows per plot. Eight (one for each of 8 random rows) mature sunflower seeds heads were collected per plot to determine head diameter and percentage of seeds damaged by larvae of the sunflower moth. Mature seed heads were collected from conventional-till plots in mid-June and from no-till plots in mid-August. Head diameter for each head was measured and then the number of undamaged seeds and the number of seeds damaged by sunflower larvae were determined per head. Because we did not have access to a sunflower combine to harvest this crop, sunflower yield was estimated using the model developed by retired USDA-ARS sunflower geneticist Jerry Miller (The Sunflower, August/September 2008).
In late Nov. 2012, wheat plots were planted. Wheat was swept (24-ft of row per sample; one sample per plot) for insect predators and pests during the growing season. Using a wheat combine, wheat was harvested by plots in late June 2013. Grain crops are not a research priority at this location, and there are few combines available for harvest, with the few available having uses of higher priority. Thus, wheat harvest was delayed causing a delay in planting subsequent soybean (see below).
Blind cultivation with a tine weeder was used for weed management in conventional-till soybean in early June. Both soybean tillage treatments were planted in late June 2013. Soybean plant density was determined by counting the number of plants per 6 ft of row. One sample was obtained for each of 20 rows per plot. Soybean plants within 6 ft. of row were sampled for insect predators and pests during the growing season; two samples were obtained for each of 6 random rows per plot. Using a soybean combine, soybeans were harvested by plots in mid-Nov 2013.
In early Dec. 2013, rye/crimson clover plots were planted. Very little crimson clover survived this extremely cold winter. Wheat was planted in mid-Dec. 2013. Wheat was not harvested in 2014 due to extreme cold-weather damage and subsequent potential low yield (potential maxed out at 10-12 bu/A). Also, wheat harvest was delayed due to the unavailability of a combine, and while waiting for a combine wheat was lodged due to heavy rain and wind-storms. Conventional-till and no-till sunflower were planted in late May 2014. Conventional-till and no-till soybean were planted on 7/25/2014. Wheat and weeds were flail mowed prior to planting.
Weather
January and May of 2013 had below average rainfall (mean -66 mm) but February proved very wet (+328 mm) (Table 1). Rainfall in April and May was slightly under normal. The summer 2013 rainfall was above normal (+188, +29, +72 for June, July, and August, respectively). Except for November (+22 mm), rainfall for the rest of 2013 was below normal (-30, -49, -13 for September, October, and December, respectively). Because of the February rains, annual rainfall in 2013 was above average (+398 mm). This also made cumulative potential evapotranspiration greater than rainfall by 200-500 mm (Fig. 9). In 2013, average monthly temperature stayed around 11-12oC in winter sharply rising to 18 and 21oC in April and May, respectively, and holding steady at 25oC through the summer and early fall, before declining to 19oC in October and 13oC in November and December.
For the first six months in 2014, except in January (-40 mm), rainfall was 20-40 mm above normal in all months but April (+172 mm) (Table 2). Average temperature steadily rose from 6oC in January to 23oC in May and 27oC in June.
Soil Sampling
Based on the results of 2012 soil tests, Nature Safe fertilizer was applied to wheat plots at 1000 lbs/A 10-2-8 at preplant and to rye/clover plots at 800 lbs/A 10-2-8 at preplant.
Nutrient levels were comparable across treatments, and values in 2014 were slightly smaller than those in 2013 (Table 3). Generally, nutrient levels were in the medium range. These values were used for determining fertilization rates per UGA recommendations.
Based on the results from core soil samples collected in 2012 alone (Table 4), we can conclude that the plots were relatively uniform to initiate this 2-year field study, as few differences occurred prior to treatment implementation. Also, the soil was relatively low in soil biological activity (e.g. flush of CO2 values less than100 mg/kg, cumulative C mineralization less than 300 mg/kg/24 d, and net N mineralization less than15 mg/kg/24 d at all depths). The soil was obviously very sandy and that contributed to the low concentrations of total and particulate organic C and N fractions, as well as to the low soil biological activity and microbial biomass concentration. No evidence of depth stratification of any of the responses suggested that the soil had been extensively disturbed with tillage in previous years. Continuing the study into the future and assuming future soil samples collected should result in significant depth stratification of all C and N fractions. These results from 2012 provide a baseline condition for changes that might occur into the future with continued management of the experiment.
Temporal and Spatial Soil Water Content Data
Figs. 10, 11 and 12 show several features of the soil water dynamics in the surface 8 cm depth. This 8 cm soil profile depth dries out fast and so might better show the soil water differential by tillage.
Replicate Variability: No-till Rye-Sunflower and No-till Wheat-Soybeans
Fig. 10 shows the variability by rep for no-till winter rye and summer sunflower and no-till winter wheat and summer soybean. In January and early February 2013 in no-till rye-sunflower, the soil was dry but responded quickly to saturation to the February rains (Fig. 10A). In March and April soil water was again relatively low, but the graph shows quick responses to the small rain events. Soil water content in Rep 1 appears to be consistently in the lower end while that in plot Rep 2 seems to be in the upper end. Sensors were removed in late April to allow field preparations for planting of the summer crop. In June through August, the graphs show a fairly close similarity in soil water content in Reps 1, 3 and 4 (for part of the time). Soil water in Rep 2, however, was consistently greater by up to10%. This might have been due to soil variability (where the sensor is installed). Soil water content increased substantially in early July in Rep 3 to closely match that in Rep 2. Again soil water content responded fairly fast to rainfall events in all plots. Soil water content dynamics for no-till wheat-soybean was fairly similar to what was described above (Fig. 10 B). In winter, soil water content in Rep 1 was at the upper end and that in Rep 2 at the lower end. In the first half of summer soil water content was fairly similar in Reps 1, 2, 3, and 4 (sensor installed late). There was a little more separation in the second half of summer so that soil water content in Reps 3 and 4 was at the upper end and that in Rep 2 at the lower end. We conclude that there was significant variability in soil water content among reps.
Treatment Differences by Reps
Before the February rains, the soil was dry (Fig. 11). Thereafter during winter in Rep 1 plots, no-till wheat (plot 104) consistently had greater soil water content than no-till rye/clover (plot 101) perhaps suggesting greater soil water usage by the rye/clover mix (Fig. 11A). During the summer in Rep 1 plots, no-till soybean (plot 104) had slightly greater soil water content than conventional-till sunflower (plot 102) or no-till sunflower (plot 101). For the first half of August soil water data was limited to one plot meaning sensors had been removed as explained earlier.
For Rep 2, soil water content in winter was fairly similar between no-till rye/clover (plot 204) and no-till wheat (plot 201) (Fig. 11B). In the summer, no-till sunflower (plot 204) had greater soil water content than conventional-till sunflower (plot 203). This likely was a result of replanting of the no-till sunflower making it less physiologically advanced than that in conventional tillage and hence demanding less soil water than the more mature conventional-till sunflower.
For Rep 3, soil water content in winter was fairly similar between no-till under wheat (plot 302) and no-till rye/clover (plot 303) (Fig. 12A). Similarly in the summer, soil water content was fairly similar between no-till soybean (plot 302), conventional-till sunflower (plot 304), and the early part of the summer for the no-till sunflower (plot 303).
There were not enough sensors to install in Rep 4 during the 2013 winter, but they were installed during the summer in this rep. In the first half of July, responding to a series of rainfall events, conventional-till sunflower (plot 401) had greater soil water content than no-till sunflower (plot 402) (Fig. 12B). This is contrary to expectations and might have been related to soil differences or cultivation variables. The soil dried out in late July through mid August because of plants transpiring and reduced rainfall. There was a spike in soil water content in the latter half of August due to increased rainfall, but conventional-till sunflower, no-till sunflower, and no-till soybean (plot 401, 402 and 403) responded similarly.
Soil Infiltration Measurements
As expected there was much variability in soil infiltration (Fig. 13). In three conventional-till wheat plots (plots 103, 301, 404), the steady state infiltration rate was approximately 15, 10, and 28 cm/hr, respectively. The experiment had to be abandoned in the fourth conventional-till plot (202) because infiltration was negligible after an hour of run, likely due to a compacted layer close to the surface at the location the rings were setup. In the no-till wheat plots, the steady state rate was approximately 23, 16, 14, and 35 cm/hr. Infiltration is typically spatially very variable. In our case, the limiting variable likely was the sandy clay loam subsoil and how close it was to the surface because it has a much lower infiltration rate than the surface loamy sand. In addition, compaction due to machinery (which can also be very variable spatially) could have played a role as suspected in the case of the abandoned site. The measured values were within the range reported in the literature for these soils.
Planting Activities, Crop Plant Density, Weeds, Insects, Yields
There was an excellent stand of rye (Fig. 14) with approximately 15-20% crimson clover (Fig. 15); only a few weeds, mainly mustard and vetch, occurred in these plots. Rye grew well; mature height was approximately 7 ft. (Fig. 16). Beneficial insects, including big-eyed bugs, pirate bugs, spiders, and lady beetles, were abundant on rye/crimson clover. There were no major insect pests on this cover crop combination.
Sunflower heads were present from early June through late July in conventional-till plots (Fig. 17) and from early July through mid-August in no-till plots (Fig. 18). Light infestations of crabgrass, morning glory, and sicklepod occurred in all sunflower plots for both tillage treatments. Some plots in both tillage treatments had moderate levels of yellow nutsedge (Fig. 19). Sunflower plant density was significantly higher in conventional-till plots (10.4 ± 0.8 plants per 10 ft. of row; Fig. 20) compared to no-till plots (6.2 ± 0.5 plants per 10 ft. of row; Fig. 21). Heavy populations of honey bees and native bees visited flowering sunflowers (Fig. 22). The main insect predators on sunflower seed heads were big-eyed bugs, spiders, lady beetles, and fire ants. The main insect pest was the sunflower moth. Birds also sometimes fed on sunflower seed. Feeding by larvae of the sunflower moth damaged sunflower seeds. Percentage of damaged seeds was significantly lower for seed heads in conventional-till plots (4% damaged seeds per head) compared to those in no-till plots (38% damaged seeds per head). Head diameter was significantly higher for heads in conventional-till plots (5.1 ± 0.2 in. per head; Fig. 23) compared to those in no-till plots (4.1 ± 0.1 in. per head). Estimated sunflower seed yield was significantly higher for conventional-till plots (1213.4 ± 68.9 lbs/acre) compared to no-till ones (325.7 ± 88.9 lbs/acre). These yields were somewhat lower than the average yield for ‘s668’ sunflower with plenty of rainfall. Heavy rainfall resulted in standing water in Reps 3 and 4 during maturation of sunflowers (presumably these two reps were at a lower elevation than the other two reps). Estimated seed yield for conventional-till sunflower was numerically lower for these two reps (1318.2 ± 57.4 lbs/acre) compared to the other two reps (1108.6 ± 56.7 lbs/acre) indicating that flooding over a period of time during the period of sunflower emergence and/or reproduction may reduce yield as has been reported in the literature on sunflower production.
Sunflower seed production was higher in conventional-till plots likely because stand density was higher, percentage of damaged seed was lower, and head diameter was higher. Presumably, the later planting date in no-till plots had a significant negative impact on sunflower seed production. However, plant density was still significantly lower for no-till versus conventional-till sunflower in 2014 even though both tillage treatments were planted at the same time. We have not yet been able to analyze the 2014 data, but head diameter is visually higher for conventional-till sunflower compared to no-till sunflower. Nevertheless, percentage damage of seed heads by larvae of the sunflower moth appears to be similar. These results/observations indicate that tillage treatment affected plant density and head diameter while planting date influenced sunflower moth infestations. Neem 7-Way (azadiractin), an organically-certified insecticide, can effectively kill lepidopteran larval pests, and thus will likely kill exposed sunflower moth larvae. The problem, though, with using this product is that these larvae feed inside sunflower seed. Therefore, the management tactic most likely to result in low infestation of the sunflower moth is planting sunflower early. Unfortunately, this may be difficult to accomplish in no-till sunflower.
The wheat stand was excellent (Figs. 24 and 25). In the spring, only a few weeds, mainly mustard and vetch were present in this crop. Beneficial insects, including big-eyed bugs, pirate bugs, spiders, and lady beetles, were abundant on wheat. There were no major insect pests on this crop. There was no statistical difference in wheat yield between conventional-till (19.8 ± 2.2 bushels/acre) and no-till (16.2 ± 2.9 bushels/acre) treatments. These yields were lower than the average yield for ‘Georgia Gore’ wheat. Perhaps, the unavoidable delay in harvesting reduced potential yield.
Soybean plant density was significantly higher in no-till plots (47.4 ± 4.4 plants per 6 ft. of row; Fig. 26) compared to conventional-till plots (31.6 ± 6.2 plants per 6 ft. of row). However, there was no significant difference in soybean plant height between no-till plots (41.0 ± 4.0 cm high) and conventional-till plots (39.9 ± 6. plants cm high). There were moderate to heavy infestations of weeds, including crabgrass, crowfoot grass, Florida pusley, yellow nutsedge, sicklepod, fennel, and morning glory in no-till soybean plots. Levels of weeds, including yellow nutsedge, crabgrass, and sicklepod, were relatively low in conventional-till soybean plots (Fig. 27). Insect predators and pests were present on soybean plants from late June through early September. Again, the main predators were big-eyed bugs, spiders, lady beetles, and fire ants. The main pest was the newly invasive kudzu bug. In a side study, we determined that Neem 7-Way, Pyganic (pyrethrins), and Azera (azadiractin + pyrethrins) did not effectively control this pest. Thus, there currently are no effective biopesticides for managing this pest in organic production. Mean number of kudzu bug adults per sample in no-till plots (8.3 per sample) was significantly lower than that for conservational tillage plots (13.7 per sample). In addition, the mean number of kudzu bug eggs per sample in no-till plots (2.2 per sample) was significantly lower than that for conservational tillage plots (4.1 per sample). A new parasitoid, Paratelenomus saccharalis, parasitized kudzu bugs eggs. Parasitism of kudzu bug eggs was significantly higher in no-till plots (52.8%) compared to that in conservation tillage plots (40.4%). Thus, tillage had a significant influence on kudzu bug density and parasitism of eggs in soybean. Why this occurred is a mystery, but perhaps wheat stubble deterred the dispersal kudzu bugs into soybean. There was no statistical difference in soybean yield between conventional-till (57.5 ± 6.3 bushels/acre) and no-till (52.6 ± 5.1 bushels/acre) treatments. These yields were within the normal range for ‘Woodruff’ soybean. Total dry weight of fruit also was not significantly different between conventional-till (23.6 ± 1.7 g) and no-till (22.6 ± 1.3 g) treatments. Overall, plant density was higher, kudzu bug adult and egg density was lower, and kudzu bug egg parasitism rate was higher for no-till soybean compared to conventional-till soybean. Inexplicably, soybean yields nevertheless were similar for both tillage treatments. Perhaps, the higher infestation of weeds in no-till soybean relative to conventional-till soybean resulted in some reduction in yield in no-till plots.
In the 2014 plots, moderate infestations of weeds, including crabgrass and yellow nutsedge are present in conventional-till sunflower plots. Heavy infestations of weeds, including crabgrass, Florida pusley, yellow nutsedge, and morning glory occurred in no-till sunflower plots. A light infestation of yellow nutsedge grew in conventional-till soybean. Even though wheat and weeds were flail mowed prior to planting, moderate infestations of weeds, including crab grass and yellow nutsedge occur in no-till soybean. This is unfortunately all the data we have to present for this season due to June 30, 2014 end of the SARE grant. However, we will continue the research project throughout the remainder of the second year of the study.
In summary, we faced serious challenges at various phases of this research project:
1) inability to control weeds in organic no-till sunflower and no-till soybean if weeds escape suppression by cover crops
2) inability to kill a rye/crimson clover cover crop before planting the subsequent crop
3) inability to control sunflower moth larvae in sunflower and kudzu bugs in soybean
4) difficulty in planting no-till sunflower into heavy cover crop residue
5) delay in obtaining a wheat combine caused additional delays in planting subsequent crops, which led to difficulties in managing weeds and planting subsequent crops
6) unavailability of a sunflower combine
7) variability in soil conditions among plots
8) unfavorable weather conditions – extremely wet or cold at times
Nonetheless, we were able to successfully plant a rotation of wheat-soybean and rye/crimson clover-sunflower under both tillage regimes, conventional-till and no-till. Yield for conventional-till sunflower was somewhat low for this cultivar under good rainfall, but still within the range of yields reported under dryland conditions. Wheat yields were lower than expected, but there was no significant difference between tillage treatments. Yields for conventional-till and no-till soybean were within the range reported for the soybean cultivar.
Educational & Outreach Activities
Participation Summary:
An Advisory Committee, including Donn Cooper with Georgia Organics, Ray Hicks, the County Agent for Screven County, and Al Clark, an organic grower, was assembled in September of 2102. In early October 2012, an overview of the objectives of the project was presented to this committee by the researchers during a conference call. One of our challenges was identifying and finding varieties of organic wheat suitable to this area. Ray suggested Pioneer and Southern States or contacting breeders at the University of Georgia (UGA) Griffin campus and Auburn University. Harry Schomberg suggested that David Marshall with USDA-ARS, Raleigh, NC might have some adapted organic seed. The group decided to look at relay cropping soybean into wheat because of concerns about weeds especially as wheat senesces and weeds begin to grow. We decided to no-till soybean in living wheat late-April early-May for June 1 wheat harvest. There was discussion about how we can harvest wheat profitably after soybeans are planted. Ray thought we would lose 60 to 70% of wheat yield. We decided to have another discussion with farmers and the North Carolina organic grain researchers later this winter to discuss details of how this might be best done. Ray recommended monitoring for the newly-invasive kudzu bug on soybean. He also mentioned that sunflowers can become infested with seed maggots. We continued to communicate with this Advisory Committee throughout the study via the two workshops, phone conversations, and e-mails. The committee chose to plant ‘Georgia Gore’ wheat and ‘Woodruff’ soybean. We tried relay cropping soybean into wheat. However, this did not work well because too much wheat was flattened with press wheels and tractor tires. We estimated that approximately 50-60% of the wheat yield potential was going to be reduced (as predicted by Ray) simply by planting the soybean as a relay-crop. Therefore, by consensus, we decided to harvest wheat and then plant soybean. Also, as cautioned by Ray and mentioned in results, we did have a significant kudzu bug infestation in soybean and a sunflower moth infestation in sunflower.
An “Organic Grain and Oilseed Workshop” for growers and extension county agents was held on June 27, 2013 in Tifton, GA. There were 40 participants, 20 farmers and 6 county extension agents. The morning of the workshop featured presentations on organic grain production by researchers in the southeastern US, including Dr. Chris Reberg-Horton, North Carolina State University (NCSU), Raleigh, NC and Drs. Harry Schomberg and Steven Mirsky, USDA-ARS, Beltsville, MD, and by Mr. Robert Davis with AgStrong Oilseed press. Then a panel of organic grain farmers, including, Mr. Al Clark with Clark Farms and Ms. Charlotte Swancy with Riverview Farms, presented their on-farm experiences and answered questions from the workshop participants. Dr. Glynn Tillman, USDA-ARS, Tifton, GA, informally presented information on the biology and distribution of the kudzu bug in Georgia. The afternoon field session featured equipment to plant into heavy residue and a demonstration of a newly developed subsurface banding applicator for poultry litter. Evaluations indicated that most of the farmers were using organic practices but were not certified organic producers. Seventy-five percent of the participants indicated they would try a few things differently or planned to make major changes. The corn production presentation and farmer panel were the most highly ranked topics. Evaluation comments indicated there was confusion around the terms GMOs and hybrids, and several participants indicated they wanted more information for organic grains on a small scale. That evening, 13 researchers from UGA, NCSU, and USDA-ARS, Tifton, GA, Auburn, Al, and Beltsville, MD, met to discuss their projects, potential collaborations, and needs for research on organic grains and oilseeds. An outgrowth of this meeting was a project to develop three videos on specific aspects of organic grain production.
An “Organic Grain Production In-Service Training” was held for extension county agents in Tifton, GA on March 11, 2014. There were 8 participants: Donn Cooper with Georgia Organics and 7 county extension agents. In the morning, presentations on critical aspects of organic grain production were given by researchers with expertise on various subjects. Dr. Chris Reberg-Horton, NCSU, Raleigh, NC, spoke on best practices for corn, wheat, soybean and canola production. Dr. Steven Mirsky, USDA-ARS, Beltsville, MD, spoke on using cover crops for weed management and the equipment needed to plant through heavy residue cover crops. Dr. Glynn Tillman, USDA-ARS, Tifton, GA, discussed insect issues in organic grain production including kudzu bugs. Dr. Dinku Endale, USDA-ARS, Tifton, GA discussed the interaction between cover crops and soil moisture and the challenges this can present to growers. In the afternoon, the group went out the research site at the UGA Ponder Farm to look at equipment for organic grain production and to watch demonstrations of the equipment in use. The emphasis was on cultivation equipment for weed management including the use of blind cultivation with tine weeders and equipment for planting into cover crop residues emphasizing the importance of cutting coulters, planter set up, and use of different closing wheels. We looked at rolling and crimper rye at an early growth stage and determined how efficiently the equipment planted through the cover crop. Evaluations indicated that most of the agents work with clients that use organic practices. The participants indicated they planned to use or might use the information provided in their programming. Sixty-seven percent of the participants said that the information was new to them. They also said that the training session was better than expected, presentation and delivery were good, and the training was well organized. Their knowledge of the topics presented was improved, they were interested in the topics covered, and they left with confidence in using the skills discussed and demonstrated during the training session. The presentations on overview of organic grain production, recommended practices, insect issues and demonstrations of cultivation for weed management and planting into cover crop residue in the field were the most highly ranked topics. Agents indicated they wanted more information on cover cropping, crop varieties, insect management, especially information on best places for beneficial insect refuges, crop diseases, and nutrient management for crops.
Three organic grain education videos were produced in partnership with North Carolina State University Extension and USDA-ARS. The topics were selected based on feedback from participants in the organic grain workshops. The first video, entitled “Success with Organic Grains – Seedbed Preparation”, was filmed with Drs. Carroll Johnson in Tifton, GA and Steven Mirsky in Beltsville, MD. The brief video focuses on proper seedbed preparation and how timely cultivation is an integral part of a weed control program. The second video, entitled “Success with Organic Grains – No-Till Soybeans”, was filmed with Dr. Chris Reberg-Horton in Raleigh, NC. The video explains key aspects of producing organic soybeans in a no-till situation and using heavy residue rye cover crop as weed control. The third video, entitled “Success with Organic Grains – Nitrogen Management in Corn”, was filmed with Dr. Steven Mirsky in Beltsville, MD. This video highlights the different management strategies a farmer may want to use to supply nitrogen to corn when needed as based on the soil quality at their farm. The videos will be posted at the SARE Learning Center, the University of Georgia Sustainable Agriculture webpage and facebook page, and the North Carolina Organic Grain project website.
We presented a poster entitled “Using Sunflowers in an Organic Grain Rotation” at the Georgia Organics Annual Conference in Feb. 2013. This was a current overview of the research project. Conference had 1200 attendees.
Glynn Tillman presented a talk entitled “Methods for Managing Stink Bugs” at the Southeastern Vegetable Conference Organics Seminar in Jan. 2013. This presentation covered information on natural enemies of stink bugs and the importance of providing nectar on-farm to enhance predation and parasitism of these pests. This seminar had 100 attendees.
Glynn Tillman presented talk entitled “New Parasitoid on Kudzu Bugs – Breaking Research” at the Southeastern Vegetable Conference Organics Seminar on Jan. 12, 2014. This presentation provided information on identification of newly-invasive pest insects in the southeast and a parasitoid of the kudzu bug. This seminar had a 100 attendees.
The NIR calibration curves for soybeans and sunflowers in the southeast will be posted at the SARE Learning Center.
Project Outcomes
This research project has led to new knowledge of management effects on organic grain systems and a better understanding of organic grain production methods in the southeast through presentations and informal conversations at scientific meetings attended by researchers, extension, industry, and producers. Also, in the near future, three videos on organic grain production will be posted at the SARE Learning Center, the University of Georgia Sustainable Agriculture webpage and facebook page, and the North Carolina Organic Grain project website which will increase new knowledge on organic grain production to researchers, extension, industry, and producers. Scientific papers will be published on results and outcomes of the research project.
This project has increased extension agent’s knowledge and skills of organic grain production systems and ability to provide technical support for sustainable agriculture practices. These were assessed via evaluations of student knowledge from classroom training and field learning sessions with extension agents.
This project has increased producer’s knowledge and skills of organic grain production systems.This was assessed through evaluations of producer knowledge at workshops and meetings.
This project will provide NIR calibration curves for soybeans and sunflowers in the southeast to researchers and producers for the first time.
Farmer Adoption
At least 500 farmers have been reached by this research project so far through field days, workshops, training sessions, and scientific meetings which farmers regularly attend. A large number of farmers will be reached by the research project through the organic grain videos that will be readily available on-line in the near future. There has been a lot of interest by farmers in conventional-till production and no-till equipment and production. Farmers are becoming increasing aware of the benefits of using cover crops and no-till production practices. We expect significant adoption of organic grain production over time. Conventional-till soybean production was successful during the first year of this project, and thus this production system would be a good starting place for farmers considering organic grain production. We encourage farmers to plant and harvest crops within recommended times even though difficulties may arise. We recommend that farmers use a GPS auto-steer system, especially for no-till grain production. Keep fields need to be well drained, especially for sunflower production.
Areas needing additional study
More research needs to be conducted on planting no-till sunflower into heavy residues of rye. Lower plant densities and reduced growth in no-till sunflower relative to conventional-till sunflower indicate that sunflower seeds were not placed into moist soil and/or growth of emerging seedlings was negatively impacted by planting into heavy residues of cover crop. We also need to find a method to successfully produce no-till versus strip-till sunflower under heavy residues. Research needs to be conducted to determine if cover crop residue reduces leaching of nutrients in the soil in these production systems. Because soybean is adaptable to a wide range of planting dates, we need to conduct studies to determine how planting date may influence kudzu bug density in soybean. Since kudzu bugs move from kudzu into soybean early in the summer, delaying the planting of soybean in these systems may reduced kudzu bug infestations. Research needs to be conducted on reducing weed pressure in no-till soybean; rolling down wheat may work, or another cover crop or a cover crop mixture may need to be utilized. We need to conduct research on using continuous ground “residue” to reduce the weed seed bank in organic grain production.