Farmers stated that soil tests often confirmed what they already knew about their soil and did not add new information. For this reason, some farmers used results from a soil test as a guide, while other farmers found results to be redundant and therefore less useful to their farm operation. Because issues with soil fertility were sometimes linked to inherent soil characteristics within a particular field, such as poor drainage or heavily sandy soil, farmers found that soil tests were not able to provide new insight to overcome these environmental limitations. “I’m not able to correct that environmental limitation [ie, poor drainage] by adding more nitrogen,” one farmer emphasized. A different farmer echoed this sentiment, saying that “I’m not going to magically get rid of issues that soil tests show… I can only slightly move the needle, no matter what I do.” Most farmers recognized that soil tests produced inconsistent results because of differences in timing and location of sampling. As one farmer noted, “You can take the same sample a couple months apart from the same field and get very different results.” Likewise, another farmer shared that, “I still struggle with the fact that I can send in two different soil tests and get two very different results. To me that seems like the science is not there.” Farmers also emphasized that each of their “fields are all so different” with “a lot of irregularity in [their] soil.” According to several farmers, soil tests did not account for variations in soil texture and soil structure, despite their observations of the influence of both edaphic characteristics on soil test results. For example, one farmer pointed out that fields that were plowed or were previously furrow irrigated created marked differences in soil test results. Similarly, plant pot with drainage another farmer shared that if a sample for soil testing was taken from an irregular patch in a field with heavier clay, differences in soil texture across samples skewed soil test results.
If a systematic sampling approach was not considered, several farmers emphasized that results of soil tests might be “misleading.” Another source of inconsistency that farmers voiced stemmed from variation in protocols used across different labs that processed soil samples. One farmer stated that in their experience, “soil tests are not really accurate, because if I use a different lab, a different person [ie, consultant] doing the soil test, it’s all different.” For example, one farmer pointed out that they do not use soluble forms of nitrogen, and instead relied on their animal rotations and cover crops to supply nutrients as part of their fertility program; this farmer emphasized that, “I think we need to get to a place with soil testing where it would be more applicable or be more accurately useful for a farm like mine. For example, with soil testing, if the standards you’re setting, and the markers you’re setting are based on farms that are putting fertilizer on the soil, I don’t think my numbers are going match up.This farmer questioned if available soil tests were calibrated to their type of farm, given that soil tests were designed for conventional agriculture . Several additional farmers interviewed also raised similar concerns. Relatedly, farmers expressed that soil tests often did not match up with their own observations of their soil and fields. One farmer plainly stated, “I’ve had soil tests that I felt were wrong; they often do not match up with what I’ve observed and gathered.” So instead, this farmer created a work around, “I usually just rent a backhoe every year and dig up one of my fields.” Another farmer also discussed this gap in soil tests, and stated the reason for this misalignment in farmer knowledge of soil and soil test results occurred because soil tests only provided “snapshots” and that observation was “just more practical in the end” because of the historical, iterative knowledge-making farmers engage in.
To this farmer, these snapshots were a “another tool” but not as powerful as direct observation; as a result, soil test results did not inform decision-making on this farm. These sentiments were often directly related to the issue of sampling discussed above. By far, the largest limitation of soil tests that nearly all farmers discussed related to the lack of analysis and interpretation of results provided by most commonly available tests. Farmers used a variety of metaphors to get at this general point. For example, one farmer likened using soil tests as a fuel gauge. This farmer stated that “the soil test tells me my tank is half empty, but it doesn’t tell me how far you’re going to be able to go… I think what’s lacking from soil tests, if someone with experience [could] help me interpret the results.” Another farmer wished they could ask “someone who has a lot of experience with doing soil tests—what do the results mean to you? Then I would incorporate my thoughts into the results… but there is not expertise and no dialogue.” This lack of dialogue was echoed by several farmers that saw the usefulness of soil tests in the collaborative interpretation of the results. Farmers emphasized that this dialogue needed to occur not with a farm consultant, but a neutral, third party expert who could “interpret relationships.” PCA indicated strong relationships among several key management variables; the results of the PCA also provided strong differentiation among farms along the first two principal components, which together accounted for 77.4% of the variability across farms . The first principal component explained 55.1% of the variation, and the second component explained 22.3% of the variation observed across all farms. Both components had eigenvalues greater than 1.0. Additional N-based fertilizer represented the management variable most associated with PC 1—followed by tillage, and inversely ICLS. While crop diversity, cover crop frequency, and crop rotation patterns also contributed to the overall variation explained by PC 1, these management variables were weaker in comparison to N-based fertilizer additions, ICLS, and tillage. On the other hand, variables with the strongest contribution to PC 2 were crop diversity, cover crop frequency, and crop rotation patterns.
Figure 1 summarizes the spatial distribution of all farms based on PCA results with PC 1 as the x-axis and PC 2 as the y-axis. As shown in Figure 3, the results of the nearest neighbor analysis order each farm from 1 to 13, and provide a basis for visualization of the gradient in management. Therefore, this gradient in management, strongly driven by the amount of external N-based fertilizer applied on-farm, served as the basis for further visual comparison of Fields A and Fields B across all farms . As shown in Figure 2a, the difference in soil ammonium concentration between fields was low among farms on the low end of the gradient. At the middle and high end of the gradient, farms showed greater soil ammonium concentrations in Field B compared to Field A—with the exception of two farms. Farm by farm, net N mineralization rates followed trends identical to soil ammonium concentrations. Soil nitrate concentrations varied widely among farms and did not produce any consistent trends ; however, a majority of farms showed greater soil nitrate concentrations in Field B compared to Field A regardless of the management gradient. Like net N mineralization rates, net N nitrification rates followed trends analogous to nitrate concentrations farm by farm. For both mineralization and nitrification rates, a majority of farms showed greater rates in Field B compared to Field A, regardless of the gradient in management. Differences between Field A and Field B for total N, total C, and POXC followed identical trends farm by farm . Among farms on the high end of the gradient, the difference in total C between fields was consistently low . Similarly, the difference between fields in soil protein values were also consistently low at the high end of the gradient . Radar plots provided further comparison of Field A and Field B across all eight indicators for soil fertility along the gradient in management developed above . As mentioned, because the level of N-based fertilizer input was a strong driver of the management gradient, pot with drainage holes radar plots were divided to reflect low, medium, and high N-based fertilizer inputs. Shown in Figure 3L is the high overlap in soil indicators, with the exception of net N mineralization and nitrification rates, between Field A and B. However, among farms with medium N-based fertilizer input , the overlap of soil indicators between fields is minimal; Field B tended to show higher concentrations of soil ammonium and soil nitrate than Field A, while Field A tends to show higher values for total N, total C, POXC, and soil protein among these farms. Among high input farms , differences between fields were less evident in terms of soil ammonium concentration, total N, total C, POXC, and soil protein, though soil nitrate concentrations and net N mineralization and nitrification rates did show noticeable differences in values between the two fields.
The results presented above are reflective of the perspectives, observations, and experiences of a sample of organic farmers in Yolo County, California, USA, and offer an enhanced understanding of soil health and fertility from this particular node of the organic movement . Here, we focus less, as prior studies have commonly done, on a comparative analysis that quantitatively compares farmers perception of soil health to results of soil laboratory analyses ; instead, we lead the discussion with farmer knowledge of soil health and fertility, and explore emergent synergies with ongoing soil health research and soil indicator results. Establishing definitions of soil health among farmers in this study was important to gauge as a starting point to discuss soil fertility, and also for selecting fields used for soil testing. Among farmers in this case study, there was general consensus on defining soil health, with strong overlap in the particular language used by farmers. Because farmers who participated in this study were geographically located within a significant node of the organic movement in California and many of the farmers interviewed participated directly or indirectly in the growth of this movement , the similarity in responses to define soil health suggests that—on the one hand, these farmers continue to draw their understanding of soil health from the culture and guiding principles of the organic movement to this day . Indeed, maintaining healthy soils was a central component of the organic movement, as stewardship of soil represented a direct connection to the land and a form of environmental protection . At the same time, the aspects of soil health that farmers touched on here were also similar to findings by other previous studies , which suggests that—on the other hand, more recent codification of the five soil health principles by the US Department of Agriculture Natural Resources Conservation Service has led to widespread integration of a national soil health lexicon, as put forth by federal policy . This soil health lexicon, in combination with farmers’ deep cultural history with organic agriculture, likely unified definitions of soil health among farmers in this study. Interestingly, while nearly all farmers interviewed touched on the first four soil health principles in some capacity, even farmers who used integrated crop livestock systems did not explicitly mention the importance of livestock integration . This finding suggests that perhaps due to sensitivity around food safety concerns, farmers may not openly emphasize livestock integration in conversation, because although this practice may be considered beneficial to their soil, in reality, they face structural and policy limitations . Despite the emphasis on understanding nutrient cycling and nitrogen availability to crops in soil health research and fertility management , we found that for most farmers interviewed in this study, tracking nutrient levels was less important than other aspects of fertility management. Moreover, for these farmers, managing for soil fertility required a holistic approach that went beyond understanding nutrient levels. Farmers also underscored that measuring indicators for soil fertility was not particularly useful to maintaining soil fertility in practice, because assessment of soil indicators lacked integration with management practices. In most farmers’ experiences, assessing soil indicators was often associated with prescriptive rather than holistic solutions. In this sense, farmers stressed that the synergy of multiple management practices over space and time guided their approach to building and assessing soil fertility on-farm, rather than using soil nutrient levels as a guide—a key finding that is also emerging in recent literature .