Savanna Crossman, Precision Agriculture Research Coordinator New York Corn and Soybean Growers Association
The 2016 field season marked the first year of testing for the variable rate planting model that is being developed by the Precision Ag Research Project. Growers across New York State know the challenges that the severe summer drought brought to our region. Crop yields were impacted across the state and the research was no exception. While unfortunate, it is advantageous to be able to test the model during a dry year and learn from how the crops reacts to the stress.
Across the board, the mid-to-lower seeding rates fared the best in the corn and soybean trials. The model was tested on five fields this year and only four made it to grain harvest due to severe drought stress. The results revealed that in three of the fields, there was not a significant difference in the profit produced by the model. While the average yield of the model was significantly less, the model was able to achieve similar profit per acre by using lower seeding rates. (Table 1)
Figure 1. 2016 Beach 2 model design. The left image displays the planting rate map and the right image displays the hybrid map.
A variation of the model design was planted on one field, Beach 2, in a split planter fashion with two contrasting hybrids. This varied design was used as it allowed for of multiple points of comparison, including hybrid comparison. Check strips were integrated every two passes to allows direct comparison of how the model performed to the typical grower practice rate. From there, the design becomes more complicated. The first pass would be planted at the model optimized rate for hybrid A, which meant hybrid B was also being planted at that same rate. Then the next pass would plant at the rate optimized for hybrid B while hybrid A was being planted at that rate as well. This allows us to examine the hybrid response to population in more depth. (Figure 1)
The hybrids P0216 and P0533 were selected due to their differences in plant architecture and responses to stress. In years of excellent growing conditions, the tight leaf structure and short stature of P0533 will produce aggressive yields. The hybrid P0216 will produce average yields in years of stress as well as in excellent conditions.
A 4,000 foot view of this field would show that there was not a significant yield difference between the model and the grower’s flat rate. The model yielded about 2 bu/ac more than the flat rate, but that difference was not statistically significant. When we separate the results out by hybrid, we see a much more telling story.
These hybrids resulted in a wonderful side-by-side comparison this year. When compared to the flat rate, P0533, regardless of optimization, yielded significantly more per acre and yielded an astounding $64/ac more. Conversely, P0216, regardless of optimization, yielded less than the flat rate and produced a profit $22/ac less than the flat rate.
Figure 2. P0216 optimized yield versus P0533 optimized yield.
A deeper look into the results showed that that when both hybrids were planted optimally, P0216 yielded almost 18 bu/acre higher than P0533 (Figure 2). It also demonstrated that P0533 exhibited a statistical significant response to model optimization. Meaning, when it was optimized the yield significantly improved over not being optimized (Figure 3). This is likely due to the fact that in a stressful year, P0216’s yields will not fall apart due to seeding rate while P0533 benefited from precise placement.
Figure 3. P0533 exhibited a hybrid response to population.
These same hybrids in Beach 2, however, exhibited the exact opposite hybrid response in 2014 which was a normal year in terms of weather conditions. Knowing this emphasizes the importance of multiple years of testing and data collection to create a robust algorithm. The biggest gain from the 2016 season has been the strong design and analysis process that has been developed. What the project has accomplished in these terms, is at the leading edge of the scientific community.
In order to build upon what the project has already accomplished, the project is still looking to get more producers involved and participating. The project aims to get fields in the research that have a large amount of variation and are fifty acres or greater. Any interested growers are highly encouraged to get in touch with the Project Coordinator, Savanna Crossman, at 802-393-0709 or savanna@nycornsoy.com.
By Bill Cox1, Eric Sandsted1, Jeff Stayton2, and Wes Baum2 1Soil and Crop Sciences Section – School of Integrated Plant Science, Cornell University; 2Cornell University Agricultural Experiment Station
The lower leaves of wheat were senescing in mid-June, despite more N being applied to high input conventional wheat (right), because of exceedingly dry conditions at Aurora and the droughty soil of the experimental area.
We initiated a 3-year study at the Aurora Research Farm in 2015 to compare the corn, soybean, and wheat/red clover rotation with different crop sequences in conventional and organic cropping systems during the 3-year transition period (2015-2017) to an organic cropping system. Three of the many objectives of the study are to determine 1) the best entry or 1st year crop (2015) to plant during the transition, 2) the best crop sequence during the 3-year transition (soybean-wheat/red clover-corn, corn-soybean, wheat/red clover, or plowed in red clover-corn-soybean) and 3) do corn, soybean, and wheat respond similarly to management inputs (high and recommended) in conventional and organic cropping systems? This article will compare the agronomic performance of organic wheat with conventional wheat following soybean in a soybean-wheat/red clover-corn sequence during the second year of the transition from conventional to an organic cropping system.
We used a John Deere 1590 No-Till Grain Drill (7.5 inch spacing between drills) to plant the treated (insecticide/fungicide seed treatment) Pioneer soft red wheat variety, 25R46, in the conventional cropping system; and the untreated 25R46, in the organic cropping system at two seeding rates, ~1.2 million seeds/acre (recommended input) and ~1.6 million seeds/acre (high input treatment) on September 24, the day after soybean harvest. We applied about 200 lbs. /acre of 10-20-20 as a starter fertilizer to wheat in both conventional treatments. We also applied Harmony Extra (~0.75 oz. /acre) to the high input conventional treatment at the GS 2 stage (November 5) for control of winter perennials (dandelion in particular).
In both organic treatments, we applied the maximum amount of Kreher’s composted chicken manure (5-4-3 analysis), as a starter fertilizer, that would flow through the drill, or about 150 lbs. of material/acre. We also broadcast Kreher’s composted manure to provide ~60 lbs. of actual N /acre (assuming 50% available N from the composted manure) in the high input treatment in the organic cropping system immediately after planting. In addition, we also added Sabrex, an organic seed treatment with Tricoderma strains, to the seed hopper of 25R46 in the high input treatment in the organic cropping system.
We frost-seeded red clover into all the wheat treatments on March 9 to provide N to the subsequent corn crop in 2017. We applied ~60 lbs. of actual N/acre (33-0-0, ammonium nitrate) in the recommended input treatment in the conventional cropping system on March 21, about a week after green-up. In the high input conventional treatment, we applied ~45 lbs. of actual N/acre (33-0-0) on March 21 and then applied another 45 lbs. of actual N/acre on April 25 about a week before the jointing stage (GS 6). We also applied a fungicide (Prosaro) to the high input treatment on May 31.
We applied Kreher’s composted chicken manure to provide 75 lbs. of available N/acre in the recommended input treatment on March 21. Also, we applied an additional 55 lbs. of available N/acre to the high input treatment in the organic cropping system on March 21. All the plots were harvested with an Almaco plot combine on July 6. We collected a 500 gram from each plot to determine kernel moisture and test weight in the laboratory.
Nevertheless, the 10% greater plant density and lower weed density in organic compared with conventional wheat, especially in the recommended input treatment, did not translate into a yield advantage. In fact, organic wheat yielded ~7.5% lower than conventional wheat (Table 2) when averaged across input treatments (no response to high input treatments in either cropping system). We suspect that the use of an organic N source may have resulted in less available N to the organic wheat crop, although visual symptoms of N deficiency were not observed. We did sub-sample before harvest (two 1.52 m2 areas/plot) to determine yield components. Organic compared with conventional wheat did have higher spike densities (533 to 509/m2, respectively) probably because of its higher plant density. Organic wheat, however, had fewer kernels/spike (22.1 vs. 24.5, respectively) and lower kernel weight (311 vs.315 mg, respectively), which indicates that the organic wheat may have been short of N, similar to organic corn in 2015 (https://blogs.cornell.edu/whatscroppingup/2016/03/29/why-did-organic-compared-with-conventional-corn-yield-30-lower-during-the-first-transition-year/).
On the other hand, the recommended input (~75 lbs. of N/acre applied in late March) treatment yielded the same as the high input (~60 lbs. of N/acre in the fall followed by another ~55 lbs. /acre of N in late March) treatment in the organic cropping system. If available N were the limiting factor in organic wheat yields, then we would expect the high input treatment to yield higher because it received more total N (albeit at different timings). We will submit our wheat samples for total kernel N analysis. If total kernel N in organic and conventional wheat is similar, then total N availability may not have resulted in the 7.5% lower yields. Then we would have to explore the idea that perhaps the use of Kreher’s composted chicken manure as a starter fertilizer may not have provided adequate P or K to organic wheat.
In conclusion, organic wheat, despite not receiving an insecticide/fungicide seed treatment, had better stands than conventional wheat and fewer weeds in both the fall and spring. Organic wheat, however, yielded 7.5% lower than conventional wheat in the second year of the transition from conventional to an organic cropping system. We expect that net returns will also be ~7.5% lower for organic compared with the recommended input conventional treatment because the lower seed costs, associated with no insecticide/fungicide seed treatment, will be offset by the higher costs for N, associated with the cost of Kreher’s composted chicken manure vs. ammonium nitrate. Many growers, however, practice high input wheat (high seeding rates, fall herbicide application, split N application, and a fungicide application), which provided no additional yield response to conventional wheat in the dry 2016 growing season. Consequently, organic wheat with recommended inputs will provide a greater return to conventional wheat with high inputs in this study in 2016.
Authored by: Savanna Crossman, Research Coordinator, CCA
Statistical Analysis by: Margaret Krause, Cornell University PhD Student
*This article is part of an ongoing series. Previous articles can be found at www.nycornsoy.org/research*
New York State has always presented a unique challenge to grain growers due to the large amount of in field variability. In recent years, growers have also added adverse weather conditions to that list. From the project’s perspective, two of the past three growing seasons have fallen far outside the conditions of a normal year. The 2015 season brought early precipitation amounts far above than the historical average while the 2016 season is setting up to be one of the driest in decades. These conditions have resulted in significantly lower, less uniform yields than a typical year such as 2014 (Figure 1). Variable rate seeding technology is one of the many tools that NYS growers can use to help overcome these challenging conditions. However, mainstream companies have yet to design a prescription writing software that is developed to meet the unique conditions of New York State and the Northeast. This project seeks to address this void by developing a software that will do just that.
The project has been collecting data on a large scale since 2014 in order to create a model that will select hybrids and population rates given certain soil properties and characteristics. To do this, six major data types are being examined; seeding rate, hybrid, topographical information, NRCS soil survey maps, Veris soil sampling data, and grid soil sampling data. Each data type consists of many variables which are analyzed individually and as interacting networks.
Figure 2. This example random forest regression analysis demonstrates that phosphorus is the variable with the largest effect on yield in this field.
To examine the effect that each variable has on yield, a statistical approach called random forest regression is being used. This method essentially ranks each variable based on its importance to yield. The greater the importance number that is assigned to a variable, the larger effect that variable has on yield (Figure 2).
The project has seen that the variables can rank very differently given the field, crop type, or year. Each field location is unique and thus has a unique combination of variables influencing yield. Some fields exhibit a very strong yield response to seeding rate, while others exhibit a strong yield response to fertility factors or topography.
Figure 3. 2014 and 2015 resulted in similar population curves on this corn field in Clyde, NY.
Though each field may be different, it is important to see stability within a field across years. For example, this 80 acre corn field in Clyde, NY produced similar population curves in two drastically different seasons. The first year, 2014, resulted in high and uniform yields across the field. The second year, 2015, yielded dramatically lower with a large variance in yield uniformity. Though the two seasons were very different, both demonstrated a negative yield response to increased seeding rate (Figure 3). The lowest rate of 27,000 sds/ac yielded the highest across the two years and which was 5,000 sds/ac lower than the grower’s typical rate. The random forest regression confirmed that seeding rate was the most important variable influencing yield across both years.
This year to year stability in yield response to seeding rate has been seen between crop types as well. This 60 acre field in Pavilion, NY is managed as a conventional till field in a corn-soybean rotation. In 2014, its soybean crop exhibited a strong positive yield response to increased seeding rate. The random forest regression confirmed that seeding rate held a dramatically greater importance than any of the other variables. The next year, 2015, the field was planted with corn and again exhibited a positive yield response to seeding rate. This time, the analysis showed that while seeding rate was still the most important variable, many other factors were also important. This difference could be due physiological preferences between the two crops or the different weather conditions between the two years. (Figure 4)
Figure 4. Random forest regression analysis of 2014 soybean and 2015 corn of a sixty acre field in Pavilion, NY.
To explore the idea of physiological differences between corn and soybean, some further analysis was conducted. In this same Pavilion field, soybean exhibited positive relationships with calcium and pH, while corn exhibited negative relationships with the same variables. These observations are likely related to the differences in crop preference for pH. Soybeans grow best in more neutral soils where the rhizobia bacteria that provide the soybean plant nitrogen are most active. Whereas the corn plant is known to prefer a slightly acidic soil where some key micronutrients, such as zinc and manganese, are more available. It is understanding relationships such as these from an agronomic and a statistical perspective that will result in a reliable model for NYS growers.
This year has marked the first infield testing of the model which will provide side by side comparison of grower practice to the model’s prescriptions. Each year of additional data collected will serve to further the development of the model into a robust and reliable resource to growers of the State.
The project is currently looking to bring on additional participants for the 2017 season and encourages any interested growers to contact Savanna Crossman at (802) 393-0709 or savanna@nycornsoy.com .
Brian Caldwell, Matthew Ryan, and Charles Mohler
Soil and Crop Sciences Section, School of Integrative Plant Science, Cornell University
In 2015, over 1000 certified organic farms were operating in New York State (NYS Dept. of Agriculture and Markets). Nationwide, New York ranks third in number of organic farms and organic cropland harvested (USDA 2011). Of the approximately 5000 dairy farmers in NYS, about 430 are currently certified organic. This number is expected to rise to 500 within two years (Fay Benson, Cornell Cooperative Extension, personal communication). Thus 10% of NYS dairy farms will then be organic. However, organic grain production has not kept up with demand, and well over half of feed grains sold to New York livestock farmers are from out of state (Mary-Howell Martens, Lakeview Organic Grain, personal communication). Consequently, land-grant university research is needed to support more organic feed and forage production in NYS.
The Cornell Organic Cropping Systems Grain Experiment (OCS) was initiated in 2005 at the Musgrave Research Farm in Aurora, NY. The purpose of this long-term experiment is to compare four approaches to organic production. Results from 2005-2010, including the 3-year transition period, were documented previously (Caldwell et al. 2014). This article discusses recent findings from the experiment and its future prospects.
Experimental Design
The OCS compares organic cropping systems: high fertility (HF), low fertility (LF), enhanced weed management (EWM) and reduced tillage (RT) organic cropping systems. We consider them systems because they are different in multiple ways. They have evolved over time to address production challenges with help from our organic farmer advisory board. Currently, HF employs higher nutrient additions during each rotation than the others, and uses both belly-mounted and rear-mounted cultivators. LF receives only corn starter fertilizer once during every 3-year rotation and only rear-mounted cultivators are used. EWM has an intermediate nutrient regimen and employs both types of cultivators, short tilled fallows, and extra cultivation to reduce weeds. In contrast to the moldboard plow-based tillage program of the other systems, RT uses a mixture of deep zone tillage, ridge tillage, and chisel plowing depending on the crop. It has an intermediate soil nutrient regimen.
The experiment includes four replications and two rotation entry points of each system. Plots are 30 x 100 feet and are managed with farm scale equipment. Soils are in the Lima series, relatively flat calcareous silt loams with fair internal drainage. All systems started with a
rotation for the first six years (RT used other legumes instead of red clover in the spelt year). A group of local organic farmers and extension educators advise on the management of this experiment.
Weed biomass in HF and RT systems was much higher than in LF and EWM by 2010 (Figure 1), and was reducing yields significantly. It was decided by the OCS researchers based on advisory group input to change the rotation for HF and RT to address this issue. The rotation for HF and RT was lengthened to six years:
In essence, a double crop of winter barley and buckwheat was substituted instead of corn at year 4 (2013 for EP A and 2014 for EP B). This enabled extra mid-season tillage to reduce weeds, particularly perennials. It also meant that no red clover was grown that year. In the other years of the rotation, crops were similar to those in LF and EWM.
Figure 1. Weed biomass in soybeans before and after 2013-2014 crop years, average of entry points.
Results 2005-2010
Results from 2005 to 2010 were reported in Caldwell et al. (2014). Briefly, applied organic chicken manure compost increased spelt yields but not corn yields. The LF system had the best overall financial returns. Corn was a poor choice during the transition period to certified organic production, but soybeans performed relatively well and spelt was intermediate. After the transition period, corn yields increased and were similar to Cayuga County averages. Organic crops with an arbitrary 30% price premium (chosen to reflect a conservative value) were more profitable than analogous conventional crops with County average yields. In recent years, the organic premium for corn and soybeans has often been higher than 30%. Currently (7/8/16) it is over 100% for corn and about 50% for soybeans (USDA, Chicago Board of Trade).
Results 2011-2016
Weather extremes
The current six-year cycle, starting in 2011, will finish at the end of this season for both entry points. HF and RT will complete one 6-year rotation and LF and EWM will complete two, 3-year rotations. The period 2011-15 was marked with a dry July (2011) and August (2012) and two very wet Junes (2013 and 2015). OCS stands were poor and areas of crops were severely stunted in 2013 and 2015 due to insufficient drainage, but crops tolerated the dry spells. Figure 2 shows yields of OCS crops over the period. Yields were normalized as a percentage of Cayuga County (if available) or NYS average yields for each year and crop, then placed in two groups based on wet or “normal” years. Buckwheat yields were not included due to lack of State or County averages. In 2011, 2012, and 2014, yields were close to County averages for the conventionally tilled systems, whereas in 2013 and 2015 they were quite reduced. The RT system had about 60% of County yields in all years, regardless of June precipitation. In the wet years, the low fertility system was affected most severely.
Figure 2. OCS yields, 2011-2015. Buckwheat not included due to lack of published County or State yields. “Normal” years were 2011, 2012, and 2014. Wet June years were 2013 and 2015.
In 2015, 8 inches of rain fell in June, whereas in 2016, the June total was only 0.74 inch, the lowest growing season monthly precipitation during this experiment. It appears likely that such extreme weather periods will be common in the future. Our results indicate that under organic management on this soil type, higher nutrients can ameliorate some of the negative effects of excess rainfall. The extremely dry June of 2016 was preceded by a dry May, and drought continued into July. Whereas the winter spelt crop looks excellent in HF, EWM, and RT as of this writing, corn growth has been slowed dramatically. Corn harvest this fall will give us insight into whether any of these systems are better able to withstand severe dry spells.
Figure 3. Spelt prior to harvest in the High Fertility (HF) system on July 5, 2016.
New crop rotation
Weed biomass was reduced in HF and RT after the barley/buckwheat year in their expanded rotations (Figure 1). Whether weed biomass will remain lower in these systems is not yet clear. However, this strategy under our constraints was likely unprofitable. The introduction of new crops such as winter barley and buckwheat into the crop mix often requires new equipment and knowledge. Our buckwheat yields in particular were low because of equipment limitations and unfamiliarity with harvesting this crop. Local organic buckwheat farmers often use a swather and combine pickup head to harvest buckwheat. The swather mows and gently windrows the buckwheat, allowing it to remain in the field to fully mature and dry. The windrows are then gathered into the combine using the pickup head. Instead, we direct-harvested the crop, a method that can result in field losses (Bjorkman 2010). Similarly, our inexperience with barley also resulted in some harvest losses. Although we have not yet put together financial budgets for these crops, net returns for the barley and buckwheat with our yields would likely have been much lower than those from corn achieved in LF and EWM in corresponding years. Our experiences mirrored those of many farmers when starting out with new crops.
Future plans
The OCS grain experiment begins a new phase in 2017.The first twelve years have yielded valuable insights into nutrient regimens, crop yields, and weed dynamics, but farmers are now facing additional challenges and attractive opportunities. For example, climate change seems to make “normal” seasons rarer and rarer. Extremes of rainfall and drought are encountered more frequently. On the plus side, markets for organic dairy feed including balage and other forages are strengthening. Buyers for crops such as sunflowers (Bob Gelser, personal communication) are looking for local producers. Over the next year, we will work with our organic farmer advisory group to plan out the next 12 years of the experiment. It will start in 2017 with a uniformity trial in which the same crop (sorghum sudangrass) will be grown over all plots. This will allow us to assess cumulative effects on soil nutrients and weeds from 12 years of management using four different organic management systems. In addition to updating management practices and data collection protocols, we will also work to improve the research site by installing new tile drainage in the alleyways between plots
This new phase of the Organic Cropping Systems Grain Experiment will explore scenarios and issues that we and our advisors anticipate will impact farmers in our region in coming years.
Caldwell, B., C. L. Mohler, Q. M. Ketterings, and A. DiTommaso. 2014. Yields and Profitability during and after Transition in Organic Grain Cropping Systems. Agron. J. 106:871-880.
Schipanski, M.E. & Drinkwater, L.E. 2011. Nitrogen fixation of red clover interseeded with winter cereals across a management-induced fertility gradient.NutrCyclAgroecosyst 90: 105-119.
