Long Term Soil Quality Trial at Beech Grove Farm - Summer 2002
Special Supplement on On-Farm Research

By Anne and Eric Nordell
.pdf version with images (513 KB) and without (38 KB)

In 1993 we initiated a long term trial to see if twice-yearly soil testing could address a major concern about our weed control system: does the use of a bare fallow period midsummer deplete soil organic matter? Our hope was that monitoring four fields in May and October for a number of years would register the effect on soil quality of our controversial weed management strategy for vegetable production, which is based on a six week bare fallow sandwiched between spring and fall cover crops the year before production. We also thought that this simple procedure would be a good way to compare our usual fallow year sequence with a full year of cover crops, or a much abbreviated bare fallow, now that weed pressure was no longer a problem.

Fortunately for us, the newly-formed Pennsylvania Association for Sustainable Agriculture (PASA), with financial support from the American Farmland Trust, was actively engaging farmers in on-farm demonstrations. PASA decided to fund our long term soil quality trial. One of the advantages of working with this organization is we pretty much had a free rein on designing the experiment. On the other hand, PASA did not have the expertise to provide much oversight on conducting this long term on-farm demonstration.

For out part, we did not have the training – or the patience – to set up the trial on a scientific basis, using randomized, replicated test plots and all the sampling and number crunching that goes with it. Instead, we wanted the trial to be farmer-friendly and to reflect what actually happens in the field – with all of the uncontrollable variables and changes in management inherent to farming.

We did employ one measure of objectivity. That was to hire a Brookside Laboratory consultant to pull and process the soil samples before sending them in for analysis. As a check against lab error (and to try out several of the new soil quality tests available at the time) we sent split samples to four different labs over the course of the first six years of the trial.

After just a few years of testing, we realized we really could use some professional help! While the results for nutrient levels were pretty comparable between the different labs (and seemed to correspond with our management practices) the numbers for the other aspects of soil quality – such as organic matter, aggregate stability and respiration rate – were so variable from lab to lab, and between spring and fall testing periods, that we were not sure what to make of them. We simply could not see a rhyme or reason to some of these numbers, while other results seemed to correspond strongly with seasonal changes in soil moisture.

For instance, the Brookside Lab organic matter readings from all four fields followed more or less the same pattern regardless of whether the fields experienced a full year of clover, cultivated row crops, or our usual fallow year sequence of annual cover crops sandwiching a bare fallow mid-summer. Likewise, the dramatic changes in aggregate stability indicated by Woods End Lab followed pretty much the same curve for all of the fields in the trial. Although we could see pronounced differences in soil structure and moisture holding capacity between these three different treatments, the lab numbers suggested that the weather, not our cropping practices, was the key short term determinant of soil quality.

Whenever we had the chance we asked soil scientists what they thought of the results. Their conflicting conclusions made us ever more unsure of how to proceed with the trial. Some researchers suggested that the large seasonal fluctuations in organic matter were just what you would expect in a biologically active system. Others were adamant that these wide swings in organic matter were due to sampling or laboratory error.

To complicate things further, several soil scientists suggested that it is necessary to make sure that the same lab technician processes all of the samples over the entire course of the trial in order to get consistent results. Others recommended that testing of certain types of organic matter, or specific stages in its decomposition, such as particulate organic matter, would lead to more accurate and meaningful results.

There was also considerable disagreement about the validity and usefulness of the soil quality tests we had tried, like using the respiration rate of the soil as a measure of biological activity. But the researchers who claimed to have developed more accurate or meaningful soil quality analysis lacked the funding for on-farm research.

Needless to say, by 1999 we had mixed feelings about continuing the trial. On the one hand, we were not sure if our casual approach to conducting the experiment was producing results that were useful. But putting into place some of the researchers’ recommendations for improving the trial was way beyond our means, especially since PASA’s funding sources and priorities had changed over the years and we were paying for most of the soil analysis out of our pockets by the end of the ‘90s.

On the other hand, the connections between soil moisture and soil quality indicated by a couple of the tests encouraged us to change the emphasis of our cropping system from weed control to moisture conservation. With this new goal in mind, we experimented with reducing the depth of tillage in both the fallow and cash crops years of the rotation. The mulch effect created by working the cover crop residues into the surface of the soil helped the crops and the soil to survive the weather extremes we have experienced in recent years in much better condition.

We were also encouraged to continue the trial after talking with Klaas and Mary-Howell Martens, who raise organic field crops on 1300 acres in Penn Yan, NY. They had just completed the first year of their SARE-sponsored soil quality trial, tracking the seasonal influence of soil temperature and moisture on organic matter levels and nutrient availability. Testing the soil every five weeks from February through November in six of their fields indicated a substantial increase in both organic matter and phosphorus during the peak of the growing season, in synch with the nutrient needs of their high-yielding crops.

We replicated their trial on a couple of fields in 2000, pulling samples every month of the growing season and splitting the samples between two labs. Like the Martens, we saw a correlation between the weather and organic matter, although this had little effect on our already high phosphorus levels. If nothing else, the preliminary results from both farms suggest that the most appropriate, if hectic, time for soil sampling might be midseason — rather than testing in the spring or the fall, when the soil biology is naturally less active.

We have also been encouraged to continue the trial by the recent increase in funding and interest for on-farm soil quality research coming from public research institutions. For example, the Martens collaborated with researchers from the Geneva Experiment Station to design their soil quality trial and verify the results. Marianne Sarrantonio, who developed much of the cover crop and soil quality testing methodology at the Rodale Research Center in the ‘80s and early ‘90s, is now conducting a trial at the University of Maine, and on a nearby farm, comparing our fallow year approach to weed management with other cropping systems.

Closer to home, PASA is now working closely with Penn State to establish on-farm demonstrations of sustainable practices. Several new researchers at the university have expressed interest in developing various aspects of the trial. We may even get some definitive answers about the effect of our fallow year system on soil quality from Cornell scientists involved in SARE projects or part of the new NEON program.

For now, we are more comfortable and confident assessing soil quality based on crop health and visible changes in soil structure than relying on laboratory analysis. The past nine years of soil testing four fields twice a year, however, have indicated some surprising changes in nutrient levels that were not apparent simply by observing the crops and the soil. These findings have changed our approach to nutrient management.

For more details on this aspect of our long term trial, as well as the soil balancing philosophies of the labs involved in the testing, please see our Cultivating Questions column in The Small Farmers Journal. These articles are included in a booklet we put together this winter which also contains photo essays on cover cropping for weed control, moisture conservation and phosphorus availability as well as pieces on reducing purslane pressure, building portable hoophouses, and comparing intensive and extensive cropping systems. The cost of the 44-page booklet is $10 (postage included) from Ann and Eric Nordell, 3410 Rt. 184, Trout Tun, PA 17771.

Back to Summer 2002 TNF Page

The TNF is the quarterly publication of the Northeast Organic Farming Association. Click here to learn more about the TNF.

This page was last modified on March 06, 2004 at 9:08:07 PM.

Home | About Us | State Chapters | NOFA Interstate Calendar | NOFA Summer Conference | Search the Site
NOFA Video Project | Farmer to Farmer Exchange | NOFA Interstate Council | The Natural Farmer | Online Store