Final Report for FNE11-710
This project confirmed previous field observations and identified that wood ash amendments of four, six, and eight tons per acre significantly affected the biological competitiveness of no-till hairy vetch when planted into perennial grass dominated hay sod typical of degraded hayfield field conditions. Vetch in the wood ash treatments averaged 30% of the biomass at an average of 1100 lbs/acre and would provide over 30lbs/nitrogen for the following crop. This compared with no vetch biomass in the control. The average biomass for no treatment was 3834lbs/acre and the average for the wood ash treatments was 5079, or about 1200 lbs more than the control. This shows that the vetch growth in the wood ash treatments accounted for almost all the additional biomass. No other treatments yielded a significant difference from the no-treatment control.
These results confirm a third year of observations and a second year of replicated trials. These results represent fourteen treatments with four replications in a randomized complete block experiment, three wood ash treatments yielded the only significant improvement in organic no-tilled vetch biomass. Other treatments included carbon treatments of biochar and sawdust, Liming treatments with fast acting lime and conventional lime, and Potassium treatments, and combinations of these treatments.
This project follows three years of work in organic no-till systems and followed exploration of techniques to not only organically no-till a row crop, into a roller crimped cover crop but to analyze some of the practices and amendments that might make organic no-till establishment of the cover crop itself possible. Success would enable greatly reducing tillage and energy inputs to establish new crops directly into hay sod. This project also relied on equipment and techniques developed with a 2009 NH Conservation Innovation Grant that demonstrated no-and low-till combined front and rear mounted equipment that established single pass planting of no-till vetch into hay sod. Those field trials showed that wood ash applications created conditions that favored the leguminous hairy vetch cover crop over the existing perennial sod (See attachments later in this report for photos of specialized equipment and of pilot projects and associated Cornell soil health test).
This project was based on previous year's field observations and replicated trial was designed to isolate the wood ash effects and more clearly identify those being most important in increasing vetch competitiveness. The primary effects being tested were pH, Potassium, and Carbon and combinations of these. The range of woodash treatments also tested various levels to identify a level of diminishing returns. The carbon effect was tested with biochar and hardwood shaving treatments. The lime and Potassium were tested with both lime and fast acting lime as well as mineral KCL.
A no-till drill on 7” rows was used to establish a vetch cover crop directly into the hayfield test plots. This is done after first removing hay, and then scalping sod with a Kuhn eight foot flail mower two weeks later. The two mowing closely following each other is believed to be important to keep the grasses in their non-photosynthetic stages which depletes stored energy reserves in the roots. Following the second mowing the Vetch was seeded at 30 lbs/acre with a nurse crop of oats also at 30 lbs/acre using a Great Plains eight foot no-till drill.
Site Description and Methodology. Field experiments were conducted in 2012 at a hay meadow located in Lee, New Hampshire. The soil at the site is a Buxton silt loam (fine, illitic, frigid Aquic Dystric Eutrudepts, USDA) The site had been hayed once in July each of the previous five years prior to the study. (please see attached baseline soil test results) The baseline Cornell Soil Health report for the site follows this section.
This trial had two primary objectives. The first was to test the reliability of previous field observations which showed a dramatic effect on no-till vetch when woodash was applied and to identify the level of diminishing returns for wood ash application. The second objective was to further isolate the reasons for the effectiveness of woodash on organic no-till hairy vetch establishment. The fourteen treatments of the randomized complete block design (RCBD) were focused on these two objectives. The treatment and plot design is attached. The primary performance target was to identify the treatment or treatment combinations which most benefited the competitiveness of vetch biomass over the established grass sod.
The experiment was established on September 3, 2012 on a 1 acre site. The design was a randomized complete block, with hairy vetch was planted throughout, and thirteen soil amendment treatments included. Treatments were established by first mowing all plots close to ground level with a 8 foot Kuhn flail mower (Kuhn North America, Inc., Brodhead, WI). Following mowing, amendments were applied to each plot with a drop spreader (Gandy Co., Owatonna, MN) pulled by hand. The tilled treatment was established with a single pass of a 8 foot Howard rotovator (Howard Australia Pty. Ltd.) set to 6 inches. The vetch and oats were planted with a Great Plains no-till drill (Great Plains Manufacturing) after all amendments were applied. Vetch was seeded at a rate of 30 lbs.acre and oats were seeded at 30 lbs/acre in a second pass. All treatments were replicated four times and individual plots measured 16’x20’ with 4’ mowed ally-ways on all four sides .
Field Measurements. Biomass was measured on June 12, 2012 when the majority of individuals were flowering. Total plant community biomass was measured by harvesting all individual stems rooted within two randomly placed 20”x20” quadrats within each treatment replicate. Harvested biomass was sorted to species (except grasses) and dried to constant weight at 60?C and weighed to the nearest 0.01g. Data from the two subsamples were averaged to obtain a plot-level measure of plant community abundance. Soil chemical and biological properties were measured in each plot on June 14, 2012 (approximately 40 weeks after application of treatments). Soils were collected by taking four soil cores to a depth of 6 inches from within the portion of each plot that was harvested for above ground biomass. Soil, wood ash, and biochar samples were analyzed for pH, organic matter, and nutrients by Agro-One Soils Laboratory (Ithaca, NY). (see results attached)
Biomass, grass biomass, and vetch biomass and soil data, we used partial least squares regression (PLSR) to investigate how variation in Vetch biomass in might be related to any changes in the above ground plant community or soil biophysical properties resulting from the soil amendment treatments. Partial least squares regression (PLSR) is a multivariate technique that uses latent variables (factors) to determine the relationship between predictor and response variables and has advantages over multiple regression in that it performs well when predictors are highly correlated.
In this study, we used PLSR to determine whether the variation in vetch biomass we observed in response to the soil amendment treatments was most associated with changes in above ground or below ground variables, both of which would likely be affected by our soil amendment treatments. Above ground predictor variables used in the PLSR were total plant community biomass, total biomass of all broadleaf species, total biomass of all grass species, and hairy vetch biomass. Below ground predictor variables used in the analysis were soil pH, soil organic matter, Al, Ca, Fe, K, Mg, Mn, P, and Zn.
Outcome showed significant effect for woodash treatments but not for any other treatments. The woodash effect has now been observed on four sites over three years and has been tested at the field scale as well as in the replicated trial and has been successful. Although the mechanism is not clear the systems effect is clear.
The enclosed overhead images clearly show the plots treated with the woodash which also corresponds with dramatic field observations and the collected biomass data. The 4,6, and 8 ton/acre woodash clearly provided a competitive advantage for organic no-tilled vetch growth.
There was not a definitively measured carbon, pH or potassium effect, but potassium shows the greatest potential, and would warrant greater investigation. This are of investigation seems promising since the baseline soil data shows significant K deficiency and since the larger quantities of woodash applied more K than the mineral KCL treatments.
At a systems level analysis, it is clear that woodash amendments at levels greater than 2 tons to the acre is effective at increasing the growth of no-tilled hairy vetch in the hay field. No other amendments were significantly different than the control of no treatment or the no-treatment and tillage. See statistical analysis and charts enclosed.
At the 30 lb seeding rate the average vetch biomass yield for the four six and eight ton woodash treatments was 1100 lbs /acre which would yield about 33 lbs/acre of additional nitrogen for the following crop. The highest recorded vetch biomass was over 5400lbs /acre which would have yielded about 160 lbs of nitrogen for the following crop. A complete table of results is attached.
The results also highlight that the soil chemistry and amendments proved to be a greater factor on vetch growth than tillage practices. This observation alone, provided an important insight for future farm practices beyond this trial.
This project successfully completed replicated field trials and related presentations, workshops, and field days and also reproduced and verified for a third year the woodash effect on no-till hairy vetch at the systems level. The results also helped narrow down the level of inquiry to the biochemical mechanisms at work that result in the woodash effect.
The trials also had several other significant accomplishments. In the course of conducting the trials, several new methods for data collection were tested as well as new methods for disseminating and distributing the analysis and communication of agricultural field trials . In addition to collecting on-the ground biomass soil data using traditional labor intensive methods, the plots were also documented using very low cost digital cameras and weather balloons to provide high resolution digital imagery of the plots. The digital imaging analysis borrows from lower land scape level remote sensing technology and used Public laboratory’s (www.publiclaboratry.org) open source image stitching technology as well as hyper 3d’s open source technology to generate three dimensional models, and NDVI near infrared imaging technology also developed by Public Labs members. For this project we also collaborated with Farm Hack www.farmhack.org to coordinate the field day and open source documentation approach. The imaging clearly shows the woodash treatments in the visual spectrum showing up as darker green. Darker green coloring is also an indicator of greater amounts of chlorophyll and nitrogen levels. Future projects using a SPAD chlorophyll meter to calibrate the results and would add greater accuracy to what is already evident on casual observation. The airial imagery was also fed into a 3D-engine provided by Hyper 3D (http://www.hypr3d.com/)which also provided a representation of topographic volume which could also be used to calculate the total biomass within the plot. The NDVI images which indicate photosynthetic activity, and visual spectrum images also provide data which could be used for both species identification and generation of percent cover calculations. All of these calculations could be performed multiple times a year and with far less labor than conventional methods of harvesting representative samples from each plot, and separating, drying and weighing each sample. A poster of this research approach was created by one of Public Labs members. http://publiclaboratory.org/notes/cfastie/10-29-2012/agricultural-mapping Some of these images are attached for reference.
Education & Outreach Activities and Participation Summary
Results and practices developed for this trial were presented at the 2013 NESAWG plenary presentation, In a 2012 Quivira Resilience Journal article and at the 2011 Quivira coalition conference in SantaFe New Mexico. The results and on-farm research approaches were also presented at the 2013 Vermont NOFA winter conference. Public Lab members also presented posters and published the results on the public laboratory web site (www.publiclaboratory.org). In addition, a paper has been submitted by Dr. Richard Smith, UNH agroecologist, that was based on the data collected for this trail which also showed a significant woodash effect of reducing yellow rattle biomass. During the fall of 2012, NE SARE also did a video interview of the project, which documented the field corn results and the no-till drill planting hairy vetch after a four ton woodash amendment at a field scale. The plots were also toured more than a dozen times including a Spring 2012 NOFANH promoted covercrop walk and in conjunction with UNH research faculty and extension programs.
Immediate Agronomic Contribution
An immediate outcome of the trial is that our farm is putting the results directly into its land management practices. This year we spread woodash on ten acres of hayfield adjacent to the trials and no-till seeded hairy vetch using the methods demonstrated by this trial. Last year we also trailed two acres of corn using the same method with no additional nitrogen amendments and had excellent results. (see attached images) The current practice does not crimp to kill the vetch but instead skim-tills (3 inches) the vetch to lightly incorporate prior to planting the row crop – either corn or sunflower. This practice achieves a complete vetch kill and, we believe, provides for more nitrogen release for the following crop. This method saves a significant amount of time and energy during the fall planting season and on nitrogen requirements from other sources. It is especially useful for fields that are not close to our manure storage facility.
Agricultural systems understanding
This trial also contributed to a greater systems level understanding of organic no-till, and the sensitivity organic no-till systems to soil chemistry as well as species composition and operation timing. This trial has provided another key piece of the puzzle to enable more consistent and favorable results in organic no-till, and highlighted the importance of soil health in determining the species composition and competitiveness.
In addition to the field trial data itself, this project also made significant contributions to expand the potential for on-farm research by potentially drastically lowering the labor costs for field trial data collection. The potential for digital imaging as a method for collecting field trial data is not only that it is low cost and extremely fast, but it provides other benefits to the farmer as well. It has a “fun” factor as well as generating almost immediate results in the form of overhead images which can be used for purposes other than just the current field trial. Digital aerial imagery is also non-destructive and can be replicated many times throughout the growing season which could enable much greater insights to the biochemical mechanisms at play. This would have been especially true for this trial if the wood ash effect has to do with the solubility/availability of the nutrients at a particular growth stage of the vetch. Had the technology been available at the beginning of the trial a much more detailed analysis would have been possible. In addition to the visual spectrum imagery that was immediately available, additional and more sophisticated NDVI and spectral analysis for high resolution field trial analysis was also an outcome of this trial. Although not fully developed yet, there is potential for species recognition, biomass, photosynthesis level measurements, percent cover calculations. The following link hosts a poster on the these methods and demonstrates the contribution that this trial has had for lowering the cost and accessibility of on-farm research. http://publiclaboratory.org/notes/cfastie/10-29-2012/agricultural-mapping
The results have been successful enough that our farm has incorporated 4 tons of wood ash as a pre-planting treatment prior to no-tilling hairy vetch as the first step in renovating older hay fields and bringing them into corn and sunflower production. The nitrogen from the vetch has provided sufficient for excellent stands in field scale tests, as documented by the NE SARE film project. Ten acres of vetch were seeded with this technique in the fall of 2012 and In spring of 2013 the vetch will be skim tilled and planted into sunflower and field corn.
Another area for future trials to improve would additional potassium treatments in form and quantity. Based on nutrient analysis, the wood ash treatments provided more K per acre than the lime and K treatment and the release of the K may have been slower with the mineral K than from the wood ash. However, as a practical matter on the farm the effect is clear and replicable and uses an organic product that is readily available in the New England that is also a byproduct of timber management.
Future tests to investigate the “wood ash” effect and expanded K and pH variables could be carried out at both the laboratory and field scale. The laboratory controlled experiments could also potentially isolate the rate of nutrient availability and the effect due to the timing and availability of the nutrients. This could also be done at the field scale in future trials by also testing the soil in each plot in the fall and spring during other stages of growth. In addition to greater frequency of soil testing, greater resolution could also be achieved by testing at varying soil depths.
Field scale trials using digital aerial imagery could also identify early differences in growth patterns which could further indicate the sensitivity of the system to the solubility and availability of the nutrients available in the wood ash or other amendments.
Both field and laboratory tests would focus on how to manage and expand the mismatch in timing of nutrient release and uptake between plant species, and improve the physical treatment of the soils and crops to facilitate an organic no-till approach that will most radically and consistently change the species composition of the field.
Both the laboratory and field scale trials would focus on the available potassium and liming effect during the vetch early growth stages. This approach would also test the theory that the grasses, in a later stage of their perennial growth cycle, were less able to absorb the nutrients before becoming dormant for the winter. The trial would also benefit from an overgrazing treatment to substitute for the haying and flail mowing approach to stress the sod growth.
There is also significant work in better understanding the overgrazing/scalping effect. A trial that looked at both mechanical and animal effects to this process would add additional layers of biological complexity, and would also offer some exciting possibilities for an even lower energy input and higher output system.