Final Report for GW15-046
Project Information
Simple on-site tests provide a possible avenue for farmers to improve their understanding of soil quality without the difficulty or cost associated with laboratory testing. Simple soil quality tests were compared with corresponding laboratory tests for their ability to distinguish between soils with known quality differences in an orchard setting. Despite overall weak correlations between simple tests and laboratory tests, several of the simple test results accurately differentiated the majority of orchard floor treatments based on soil quality. The best correlations between simple tests and laboratory findings were found between Solvita respiration and microbial biomass, the modified surface soil slake test and microbial biomass, Lamotte simple N test and laboratory measured N in the conventional orchard, and Mosser N and laboratory measured N. Modified slake tests and soil biodiversity/earthworm abundance counts consistently ranked as most preferred simple tests among growers in terms of user friendliness and cost.
Introduction
Maintaining soil quality is essential for the long-term prosperity of a farm or other land-based system (Wienhold et al., 2004). In the U.S., cropland loses an average of seven tons of soil per acre, per year (Sullivan, 2004). Maintaining soil quality can prevent loss in system productivity while also improving long-term financial outcome for farmers. For example, growers in Iowa were able to increase yield by 3-12% and reduce costs from inputs by 41-79% (Liebman et al., 2003). Despite attempts, little progress has been made in increasing grower involvement in maintaining soil quality (Herrick, 2000). Even when growers are interested in learning more about soil quality, soil quality tests are not always available, affordable, reliable, or feasible. (Friedman et al., 2001).
A number of simple soil health tests have been developed over the years, in particular, soil health cards and test kits such as the NRCS soil quality test kit. Soil health cards can be useful for soil health professionals in discussing soil health with growers. However, evaluating soil health based on comparisons to pictures or descriptions on cards alone can be subjective. (Friedman et al., 2001). The NRCS test kit is one of the most comprehensive test kits available, yet many of the tests are time consuming and confusing for someone new to soil testing (Friedman et al., 2001). Submitting soil samples to an analytical laboratory is the most straightforward testing method for growers. However, most laboratories do not offer biological and physical tests, and when they do, it is often cost prohibitive to a grower.
Aggregate stability is the ability of primary soil particles to remain attached under disruptive forces. There are generally three different categories of aggregate stability tests: 1) ease of dispersion by turbidimetric techniques (Emerson, 1967), 2) evaluation of aggregate strength related to raindrop impact (Bruce-Okine and Lal, 1775), and 3) aggregate stability by wet sieving (Yoder 1936). All three categories of soil aggregate tests have on-site versions. As rainfall simulators are often bulky and complicated to build, the most effective on-site aggregate testing options for growers are turbidimetric tests or wet sieving/slake tests. The NRCS incorporated a modified version of a slake test developed by Herrick et al. (2001) into their field test kit. Herrick et al. (2001) developed a stability test kit that could be made inexpensively with simple tools. Aggregate stability tests are useful in addressing a soil’s potential for erosion, in particular, for comparing the same soil type among multiple management systems (Kemper and Koch, 1966; Kemper and Rosenau, 1986).
Soil organisms are also important indicators of soil health as they may rapidly respond to shifts in management practices (Pankhurst et al., 1997). The most common simple biological test recommended is counting earthworms and or measuring soil respiration in a given volume of soil. However, earthworms are not native to all soils and soil respiration can be highly affected by weather (Friedman et al., 2001). Other tests to measure soil biological health include those for soil arthropods. The Berlese funnel test is commonly used to measure abundance of soil athropods in a laboratory (Macfadyen, 1953; Macfadyen, 1961; Sabu and Shiju, 2010). Foldable or collapsible Berlese funnels have been constructed for lightweight transportation (Saunders, 1959; Northon and Kethley, 1988). Hence, perhaps a Berlese funnel could be further modified as a convenient, affordable test for growers.
Chemical tests, such as nitrogen, potassium, phosphorus, and pH tests, are the one type of soil quality assessment easily available in most laboratories. But questions remain on the accuracy of do-it -yourself chemical test kits commonly available, such as Rapidtest kit, Lamotte test kit, and Mosser test kit. Accurate on site tests might increase adoption of soil testing by growers.
The goal of this study was to increase adoption of soil quality testing by growers through assessing the effectiveness and use of a number of simple on-site chemical, biological, and physical soil quality indicator tests. Simple tests for measuring soil physical, biological, and chemical properties were correlated to comparable lab analysis for their ability to distinguish between soils of known soil quality characteristics. Physical simple tests measured aggregate stability and included the NRCS slake test and other modified slaking tests. The biological simple tests included in this study were the Solvita respiration test which measures CO2 evolved in a given mass of soil over 24 hours, simplified Berlese funnel tests, earthworm abundance tests, and soil biodiversity tests measuring arthropods, earthworms, and organism diversity in soils respectively. The chemical tests included, LaMotte, Mosser, and Rapid soil test kits to measure macronutrients and pH, and the Hana pH meter to measure only pH. Tests that compared favorably with corresponding lab analysis were taught to orchardists through demonstrations, and survey results were collected on their perceptions of these tests. Surveys on soil quality were also administered to Utah orchardists, to gain a better understanding of their current level of interest in and knowledge of soil quality.
1) Compare results of simple soil testing strategies to comparable standard lab tests.
2) Determine the most predictive soil quality tests by comparing results to an existing database collected from the Kaysville systems orchards.
3) Conduct onsite training sessions with growers to determine usefulness of the tests in a range of field settings.
4) Collect feedback from growers on ease of use and applicability.
Cooperators
Research
Objective 1. Compare results of simple soil testing strategies to comparable standard lab tests. Simple soil tests designed to be easy for a grower to conduct themselves were selected based on the accessibility of the test or test components in terms of cost, availability, and reasonable time commitment. The most expensive test kit purchased was the NRCS test kit ~$800, the chemical test kits were all purchased for under $100, the Solvita test kit was purchased for ~$60 for a set of six tests, and as much focus as possible was placed on tests that could easily be made from materials for under $20. Emphasis was also placed on tests that could be completed in less than an hour. Many different types of test kits are available online, and one of the most comprehensive soil quality test kits available is the NRCS test kit (Friedman et al. 2001). The slake test and earthworm abundance test were selected from the NRCS test kit based on simplicity and necessary time commitment. In efforts to further simplify the NRCS slake test, a modified slake test using a kitchen sieve was also developed. Additional biological tests chosen were Berlese funnel tests as a measurement of soil biodiversity modified for in-field use, and litterbag tests as a measure of decomposition rates. The chemical tests chosen (Rapidtest kit; Lamotte: Hanna pH meter, and Mosser test kit) were either available locally or readily available online.
Simple biological tests: The earthworm and biodiversity tests were conducted two to three days after an irrigation event during August. To determine earthworm/biodiversity counts, a 30 x 30 x 30 cm hole was dug in each designated test plot. The soil from the hole was placed in a bucket and visually inspected one handful at a time for earthworms and other macroscopic soil organisms. The number of earthworms and any other visible organisms were recorded.
The Berlese funnel tests were conducted two to three days after an irrigation event during August. The soil collected was mixed and one large handful was placed on top of a piece of cheesecloth placed in a funnel. The spout of the funnel was placed into a glass jar, and the space between the funnel and the jar was sealed with aluminum foil.
The funnels were left in the sun for three hours. The temperature was about 28.9ºC during the time of the test. The funnels were carefully removed from the jars. The contents of the jars were poured onto a piece of paper in order to count the organisms. The number and type of organisms were recorded.
The Solvita test was conducted two days after an irrigation event in late June. The Solvita test kit included plastic jars, lids, and CO2 reactive probes. Each jar was marked with the required soil volume, about 64 grams from each designated plot. The CO2A probe was removed from its metallic pouch and placed into the soil within the jar with the color indicator side facing upward. The jars were sealed with lids, placed in a cool dark place for 24 hours after which the probe color was matched to test kit indicator sheet. The corresponding soil respiration number was recorded.
Laboratory biological tests: Simple biological tests were correlated to results from laboratory respiration and dehydrogenase enzyme activity. Samples for the laboratory analyses were taken in the end of June and were analyzed in the first two weeks of July. Mineralizable carbon (RMC), basal respiration (BR), and microbial biomass (Cmic) determined by substrate induced respiration (SIR) were measured with an infrared CO2 analyzer (Model 6251, LICOR Biosciences) on day 12, 13, and 14 of an incubation at 25°C and 22% moisture as described by Anderson and Domsch (1978) and Davidson et al. (1987). Dehydrogenase enzyme activity (DHA) the reduction of triphenyl tetrazolium chloride of 2.5g soil dried weight equivalent at 22% moisture was measured as described by Tabatai (1994).
Simple physical tests: Physical simple tests were conducted on soil collected in August both years, two to three days after an irrigation event.
The NRCS slake test was completed as described in NRCS (2001). We also developed and tested two modified version of this simple slake test.
The first modified slake test, the surface structure test, was conducted by taking a 20 cm diameter kitchen sieve filled to the rim with un-sieved soil from the designated plot, with rocks and large pieces of organic material removed. Pictures were taken of the sieve and notes were taken on the general appearance of the structure of the soil. The sieve was soaked in a bucket of water for five minutes, after which the sieve was raised and submerged four times, allowing water to drain (about five seconds) in between. The sieve was removed and a second picture was taken of the soil surface structure. The structure of the soil was again noted, and estimation was made of the percent soil structure remaining intact.
The second modified slake test, the hose test, was conducted on the same sieve of soil directly after completing the first modified slake test (the surface structure test). The sieve filled with soil was brought to a hose. The hose was turned on, using one and three quarters turn to the knob, to maintain the same water pressure on all of the tests. The sieve was held about half a meter from the hose and then sprayed for one minute in a circular motion, while maintaining an equal distribution of water flow over all surface points of the soil in the sieve. The amount of soil remaining by the end of one minute was recorded.
Laboratory physical tests: The laboratory procedure correlated to the physical simple tests was the machine aggregate stability test as described by Kemper and Rosenau (1986). Four grams of sieved and air dried soil was pre-moistened with steam to 4.75g soil wet weight (19.5% water content) and placed in sieves in a mechanical sieving device (make, model number, and place of manufacture). The instrument submerged the soil into water and raised and lowered it at regular intervals for three minutes. The soil that was lost during the sieving process was oven dried at 40 ºC and weighed. The process was repeated in a 0.2% sodium hexametaphosphate solution (NaPo3)6. The soil removed from the sieves by the (NaPo3)6 solution represented the stable aggregates.
Simple chemical tests: Soil samples were taken the last week of July each year for both laboratory and simple test kit chemical analyses. Instructions were followed according to the respective manuals for the testing of nitrogen, potassium, phosphorus, and pH by the Rapidtest kit, Lamotte test kit, and Mosser test kit. Instructions were also followed according to the manual for the testing of pH by the Hanna pH meter.
Laboratory chemical tests: For the laboratory chemical analysis, soils were sieved and stored at -15 ºC and processed within 10 days for measuring available nitrogen. Laboratory measured nitrogen was measured by nitrate and ammonium extraction using 1M Potassium Chloride and analyzed by Lachat (Quickchem 8500, Hach Company, Loveland, CO) using the X and Y methods respectively. Olsen’s (1954) sodium bicarbonate extraction method was used for measuring phosphorus and potassium and were measured after sieving soils at 4mm and air-dried for two weeks.
Statistical analysis: Each simple test was compared to a relevant lab-based test using Pearson’s correlation. The estimated percentages of stable soil aggregates from the simple slake tests were also correlated with biological laboratory procedures as the physical qualities of the soil are often directly linked to biological activity in the soil. For the statistical analysis, only Pearson’s correlations were measured and not P values. This was because the analyses did not meet P value assumptions; individual observations were not independent of treatment and or replicated blocks.
Objective 2. Determine the most predictive soil quality tests by comparing results to an existing database collected from the Kaysville systems orchards.
Soil samples were collected from two experimental peach orchards - conventional and organic - located on the Utah State University Research farm in Kaysville, Utah. The orchards consisted of 12 replicated orchard floor treatments with documented differences in soil quality. The conventional orchard consisted of five tree-row treatments, all with grass alleyways: 1) NPK fertilizers and herbicides (HN); 2) NPK fertilizers and herbicides and converted to organic practices after tree establishment (HNC); 3) Herbicides plus compost for N (HC); 4) Paper mulch with reduced herbicide in addition to NPK fertilizers (PR); 5) Paper mulch, organic herbicide, and compost for N (PC). In the second year a second conventional orchard was used which had noticeably more clover than grass in the alleyway. The organic orchard comprised six different treatments: 1) Straw mulch in the tree row with a grass alleyway (SG) 2) Straw mulch in the tree row with a legume (birdsfoot trefoil, Lotus corniculatus) alleyway (ST); 3) Living mulch (low-growing shallow rooted alyssum, Lobularia maritima) in the tree row with a grass alleyway (LG); 4) Living mulch in the tree row with a legume alleyway (LT); 5) Woven plastic mulch in the tree row with a grass alleyway (WG); 6) Tilled tree rows with a grass alleyway (TG). All treatments were used to compare the simple chemical tests; however, only four treatment types were used for biological and physical tests: SG, ST, TG, and HN. Each treatment consisted of four replicates in a random incomplete block design (RIBD).
Objective 3. Conduct onsite training sessions with growers to determine usefulness of the tests in a range of field settings. Soil quality training opportunities were presented to local farmers. Seven orchardists volunteered to be trained in soil quality and on-site soil quality tests. Tests demonstrated were those determined most accurate under objective 1 and 2. They included modified slake tests, NRCS slake test, Solvita soil respiration, and earthworm abundance/biodiversity test. A demonstration of the same simple on-site soil quality tests taught to the volunteers was also given at the summer field tour organized by the Utah State Horticultural Association (USHA).
Objective 4. Questionnaires were given to each of the seven farmers involved in the one-on-one simple test trainings as well as the participants of the field demo hosted by USHA in order to obtain feedback. A questionnaire was also distributed through a USU orchardist listserv to obtain general feedback from Utah orchardists on their knowledge and interest in soil quality and testing methods. The results from the seven growers and those who attended the field demo were combined, and the results from the online survey remained separate.
Biological test results:
Results from the Solvita soil respiration test kit correlated most highly with laboratory tests. Solvita soil respiration was able to differentiate between plots with mulch and plots without mulch. It is possible that precision could be improved by lessening the amount of time that the soil probes were incubated as many of the organically managed soils maxed out the range of the test within a few hours of the specified 24 hour incubation.
The earthworm abundance test, although often recommended by the NRCS as well as others, proved to have little relationship with laboratory soil biological testing measures. In previous trials the earthworm abundance test correlated with soil microbial activity and differentiated between conventional and organic treatments fairly well. However, in 2015, the earthworm test was not correlated with microbial activity and only weakly correlated with laboratory measured soil respiration (R=0.33). The correlation between the number of different organisms found in the 30 cm3 pit to soil microbial biomass was higher (R =0.68), and could potentially be improved by more repetitions. The results for the on-site Berlese funnel tests hardly compared to laboratory tests (R=0.48 correlation with microbial biomass) and was not able to readily differentiate between treatments. The sieves used may have been too deep, allowing the organisms to remain in a comfortable environment for the duration of the test. A longer test period may also have improved the results. For this kind of test, it is important to choose a sunny day with temperatures over 25ºC.
Physical test results:
There were no correlations between the machine aggregate tests and any of the simple slake testing methods, although several of the simple tests correlated quite well with the biological tests. It was beneficial to look at the physical soil properties on a larger scale. Using our modified slake test, the surface soil test, it was fairly easy to observe soil structure upon wetting and working the soil. Using smaller on-site slake tests such as with the NRCS test, these visual cues were absent.
The best physical test correlation was between the surface soil test, and microbial biomass (R = 0. 64). The hose test also clearly distinguished between most treatments, even moderately distinguishing between the soil quality of the tree row with a trefoil alleyway and the tree row with grass alleyway (R = 0.42).
Chemical test results:
The best correlations among simple and lab based chemical tests were with available soil nitrogen (N). The exception was the Lamotte simple nitrogen test in the organic orchard. The next best tests were the Lamotte simple tests correlated to potassium, with much better results when testing soils from the organic field over the conventional one. Overall, the correlations with soil P and pH were poor, regardless of the test used. The test kits often attained results in the relative range of the laboratory based tests, but sometimes could not even accurately accomplish that. These three test kits came in packages of N,P,K and pH. To purchase a kit but to only use particular tests is not the most efficient use of a product.
The Lamotte test kit had somewhat more sensitivity to nutrients and pH than the Rapidtest kit. Value 1 on the Lamotte scale correlates to 0-8 ppm. Although the precision was not exact, the results roughly fit the laboratory measured soil nitrogen. The simple Lamotte potassium test was the next best test (and the best potassium test out of all of the other simple tests used) and was considerably better in detecting amounts of potassium in the organic orchard over the conventional orchard. Since both the Lamotte potassium and nitrogen simple tests faired differently in the organic and conventional orchard, it is possible that there could have been an effect of humic acids on some of the chemical solutions from the higher soil organic matter in the organic orchard.
The Mosser nitrogen test had the best correlation out of all of the chemical simple tests. The potassium test was not correlated. The test correctly identified the treatments with the greatest levels of potassium;however overall, the identification of potassium was often undervalued and not very precise.
Grower feedback on simple soil quality tests:
The growers who worked with researchers one-on-one or attended the soil testing demo indicated the first modified slake test/the surface structure test as the most likely test they would use on their own farm, followed by the earthworm abundance test. The main concern cited (mentioned by two growers), was that the water flow rate used in the hose test might be challenging to keep consistent. The other concern mentioned 2-3 times by different growers was the cost of the Solvita test and how that would limit their ability to use that test. The main positive feedback mentioned by more than two growers on the simple tests demonstrated was how hands-on the tests were. One farmer in particular had assumed the tests would only involve vials and chemicals. Another grower said it gave her a new way to think about testing the soil.
All growers (7) that worked one-on-one with researchers said that they learned something from the demonstrations. And 18 out of 19 from the demo survey stated that they had learned something from the simple testing methods. The simple soil tests developed here, particularly the modified slake test surface structure test, show real potential in terms of involving growers more intensely in soil quality testing.
Grower survey:
Out of the 400 surveys sent via email through the USU extension grower listserv, 101 growers completed the survey. The questions asked the growers whether they tested their soil, why or why not, what kind of tests they used, and what soil quality meant to them. The respondents were primarily men between the ages of 55-64, although women did represent 43% of the respondents. The greatest number of respondents owned acreage between 1-5 acres.
When growers were asked how they rated their knowledge on soil testing, 46% of the respondents mentioned that they had some knowledge, while 25% mentioned having minimal to no knowledge on soil testing. Most growers affiliated healthy soil with healthy plants and good yields, followed by healthy populations of soil microorganisms, and good soil structure.
Sixty-nine percent of respondents said they test their soil. With the majority of those respondents indicating they complete chemical tests (69%), only 11% of respondents indicated having done biological or physical soil tests. This is likely due to unavailability of such tests in most laboratories, costs, and issues associated with sending soils to laboratories with tests available, and the lack of opportunities to learn about biological or physical soil tests on-site.
Respondents most often indicated that the reason they tested their soil was to determine soil fertility so they could apply the proper amendments. The next most common response was in order to track the soil properties. The most common reason for not testing was that growers had not seen any issues with their plants, followed by expense. One grower in particular mentioned that to test his soils at a lab it costs a minimum of $150 because of the distance he lives from a soil testing laboratory and the need to send his soil through the mail.
In order to understand the current perceptions on soil testing strategies and what growers thought about them, they were asked to what extent they agreed that usefulness, affordability, and ease were common traits among current testing strategies. The largest percentage of respondents indicated that these were common traits among testing strategies, while the next largest percentage of respondents indicated that they were not particularly common traits but also not missing as traits. Less than 15% of respondents for each trait indicated that they were not common traits.
When asked what could be improved on standard tests known to or available to the public, most growers actually felt that the methods available to them were sufficient and nothing needed improvements. The next most common response was for more information on organic systems, in particular recommendations for organic amendments and making the tests more affordable. Mentions were also made about making tests more convenient and comprehensible, as well as an emphasis on biological tests. The last question asked on the e-mail questionnaire was is if they would be interested in learning more from researchers on soils quality testing, and the majority of respondents, 87%, said yes.
Research Outcomes
Education and Outreach
Participation Summary:
Posters and Presentations:
Thomsen, E., J.R. Reeve, C.M. Culumber. 2015. On-farm soil quality testing in organic, and conventional peach orchard systems. Soil Science Society of America. November 15-18, Minneapolis, MN. Poster presentation.
Thomsen, E., J.R. Reeve, C.M. Culumber. 2015. On-farm soil quality testing in organic, and conventional peach orchard systems. Soil and Water Conservation Society. July 26-29, Greensboro, NC. Oral presentation. 30 Participants.
Thomsen, E. and J.R. Reeve. 2015. Simple on-site soil quality testing, Urban and Small Farms conference, February 19th, West Jordan, Utah. Oral presentation. 17 Participants.
Thomsen, E. Simple Soil Health Testing Strategies. Utah State Horticultural Association, June 30, 2015, Santaquin, Utah. 20 Participants.
Graduate Thesis – Utah State University
Thomsen, E., 2016. Simple Soil Quality Tests and Organic Management Practices in Orchards in the Intermountain West. M.S. Thesis. Utah State University.
Research paper:
Thomsen, E, J. Reeve, C.M. Culumber, R. Newhall, D. Alston, and G. Cardon. Simple soil tests for onsite evaluation of soil health in orchards. In preparation.
Fact sheet:
Thomsen, E, J. Reeve, R. Newhall, D. Alston, and G. Cardon. Simple soil tests for onsite evaluation of soil health in orchards. In preparation.
Short demonstration video:
Thomsen, E. Simple soil tests for onsite evaluation of soil health in orchards: physical and biological tests. Footage complete, editing in progress.
Brochure:
Thomsen, E, J. Reeve, D. Alston, and G. Cardon. Simple Soil Quality Testing. In progress.