Bio-fuel mandates are one of several policies, including crop insurance and research funding priorities, that currently maintain the profitability of corn production; however, these policies need not be the ones that define rotational landscapes, and increased funding for policies such as the Conservation Stewardship Program could better align farmers’ economic incentives with improved environmental health. When strong economic incentives encourage rotational simplification, our analysis suggests that it is more likely to occur on land with favorable biophysical conditions for corn growth. With our current policy structure, the highest quality lands in the Midwestern US therefore become the most prone to degradation through intensive management.Though there is some error in crop identification associated with the CDL, the overall accuracy is quite high across the states in our study area , suggesting a strong correlation between on-the-ground rotations and the indices in our analysis. Importantly, because of the way the RCI metric functions, the CDL does not need to correctly identify each crop species for RCI scores to be accurate, but rather distinguish if each crop is similar or different from crops grown in previous years. Because corn and soy account for over 86% of cropland in the study area, 4×4 grow table their correct identification is key to the metric’s success, and corn/soy show the highest accuracies of any crop type in the studied states.
We also find it relevant to compare our results to previous studies. First, we find that an RCI of zero, which corresponds to corn monoculture, accounts for 4.5% of cropland in the study area over the six year period from 2012-2017, while previous estimates in the region give the prevalence of four-year monoculture as 7% in 2010, and ten-year monoculture as 2% in 2019. Second, previous studies have examined the prevalence of corn-soy rotations in the Midwest, arriving at estimates of 53% of land planted in 3-year corn-soy rotations in 1997, and 72% of cropland in Iowa. Because these estimates come from sequences of shorter duration, they are likely to report higher estimated areas than our six-year sequences. Our analysis across the Midwest shows that 37% of the study area was in exactly a CSCSCS or SCSCSC sequence, while 57% of the study area had an RCI of 2.24, the value that corresponds to a corn-soy rotation for a 6-year sequence.As rainfall becomes more variable with changing climates, farmers around the world are contending with droughts that are increasing in both intensity and duration. For many farmers, restricted water use has become a constant and looming threat, forcing the agricultural sector to confront a key question: how can we adapt to water scarcity without jeopardizing farmer livelihoods? This question is particularly salient for California’s agricultural system, which has become increasingly fragile in recent decades due to its dependence on a shifting and shrinking water supply5,6.
Changing climates have caused droughts that not only result in massive financial losses , but also raise major concerns for farmers’ ability to maintain continuity in their farming operations. Because California’s waters are over-allocated even in years of typical rainfall, the Sustainable Groundwater Management Act, which requires sustainable groundwater use by 2040, implies that irrigation will need to be discontinued on hundreds of thousands of cropland acres. In this backdrop, dry farming, a practice in which farmers grow crops with little to no irrigation, has quickly garnered interest from farmers and policy-makers around the state. While dry farming is an ancient practice with rich histories in many regions, perhaps most notably the Hopi people in Northeast Arizona, dry farming emerged more recently in California, with growers first marketing dry farm tomatoes as such in the Central Coast region in the early 1980’s. In a lineage that likely traces back to Italian and Spanish growers, dry farming on the Central Coast relies on winter rains to store water in soils that plants can then access throughout California’s rain-free summers, allowing farmers to grow produce with little to no external water inputs. As water-awareness gains public attention, dry farming has been increasingly mentioned as an important piece in California’s water resiliency puzzle, however, while some extension articles exist, no peer-reviewed research has been published to date on vegetable dry farming in California. We therefore assembled a group of six dry farming operations on the Central Coast to collaboratively identify and answer key management questions in the dry farm community.
Growers identified three main management questions that would benefit from further research: 1. Which depths of nutrients are most influential in determining fruit yield and quality? 2. Are AMF inoculants effective in this low-water system, and more broadly, 3. How can farmers best support high-functioning soil fungal communities to improve harvest outcomes? Growers were primarily concerned with fruit yield and quality, with blossom end rot prevention and overall fruit quality being of particular interest due to the water stress and high market value inherent to this system. Managing for yields and quality present a unique challenge in dry farm systems, as the surface soils that farmers typically target for fertility management in irrigated systems dry down quickly to a point where roots will likely have difficulty accessing nutrients and water. Because plants are likely to invest heavily in deeper roots as compared to irrigated crops, we hypothesized that nutrients deeper in the soil profile would be more instrumental in determining fruit yields and quality. As deficit irrigation and drought change microbial community composition in other agricultural and natural systems, we hypothesized that dry farm management would cause shifts in fungal communities in response to dry farm management, which could in turn improve tomato harvest outcomes. Beyond general shifts in fungal communities, farmers were specifically interested in arbuscular mycorrhizal fungi inoculants, which are increasingly available from commercial sellers. Recent research has shown that AMF can help plants tolerate water stress, and we therefore hypothesized that commercial AMF inoculants might be beneficial. We organized a season-long field experiment from early spring to late fall of 2021 to answer these questions, sampling soils and collecting harvest data from plots on seven dry farm tomato fields on the Central Coast. Each farmer managed the fields exactly as they normally would, with AMF inoculation being the only experimental manipulation. We sampled soils for nutrients and water content at four depths down to one meter throughout the growing season to determine which nutrient depths influenced harvest outcomes. We also took DNA samples from soils and roots in surface and subsurface dry farm soils, as well as nearby irrigated and non-cultivated soils, cannabis drying system sequencing the ITS2 region to analyze the fungal community to verify inoculation establishment and more broadly characterize soil fungal communities to see how fungal communities changed under dry farm management and determine whether these changes or the introduction of an inoculant influenced harvest outcomes. We then used Bayesian generalized linear mixed models to estimate the effects of nutrient depths and fungal community metrics on yield and fruit quality data from 10-20 weekly harvests on each field. Our results highlight a tension between managing nutrients for fruit yield and quality, while fungal community metrics show promise for increasing fruit quality. The experiment was conducted on seven certified organic dry farm tomato fields in Santa Cruz and San Mateo counties in California during the 2021 growing season.
Five blocks were established on each field over the course of a full growing season , for a total of 70 experimental plots. These fields are managed by six farms; one farm contributed two fields at two separate sites. Each farmer continued to manage their field for the duration of the experiment according to their typical practices. Each dry farm crop was preceded by a crop in the winter prior to the experiment, either in the form of a cover crop , or continuous winter production . All fields were disked prior to planting, and two fields additionally ripped down to 60-90cm. Each field’s plant and bed spacing, plant date, and tomato variety are listed in Table 1, along with amendments added to the soil. Fields also varied in their rotational history . The mapped soil series, measured texture, and soil pH are listed in Table 3. From March 2 to October 27 there were 15 rain events greater than 1 mm recorded at the De Laveaga CIMIS weather station , none of which occurred between the months of May and October . Monthly weather data is summarized in Table 4. A nested experimental design was used to account for management and biophysical differences across fields. Each plot contained 12 plants, and plots were divided across two beds with a buffer row between . Plots were randomly selected to be inoculated in the first experimental row and then paired with a counterpart in the second experimental row that received the opposite inoculation condition to achieve a randomized complete block design with five blocks per field. Here we refer to a pair of inoculated and control plots as a block. There were three non-inoculated buffer plants between each plot and at least twenty buffer plants at the start and end of each experimental row. A commercial AMF inoculant was used to inoculate transplants. This inoculum has been shown to impact crop physiology and improve plant water status in various field and greenhouse applications. Each of the 12 plants in plots in theinoculation condition received 0.2 g of inoculum, which was mixed with 40 mL of water and then poured at the base of the plant within three days of transplanting, as per manufacturer instructions. Harvests began when farmers indicated that they were beginning to harvest the portion of their field that included the experimental plots. Each field was harvested once per week from its start date to its end date, with the exception of Farm 5, which was harvested twice per week, in accordance with farmer desires. All red tomatoes were harvested from each plot and sorted into marketable, blossom end rot, sunburnt, or “other unmarketable” fruits and then weighed. Harvests stopped when there were no remaining tomatoes in the field or when farmers decided to terminate the field. Fruit size and quality were assessed on the third, sixth, and ninth week of harvest at a given field. Ten representative marketable tomatoes were taken from each plot, weighed, dried at 70 degrees C and then weighed again to establish the percent dry weight . PDW was used as a proxy for fruit quality, with fruits with a lower water content increasing fruit quality up to a certain point. Extension research has linked dry farm fruit quality with lower fruit water content, as opposed to specific compounds that are elevated in dry farm tomatoes, and we expect PDW to correlate highly with the concentration of flavors previously found to create dry farm fruits’ superior quality. After eliciting quality categorization from farmers in the study, we determined that fruit quality increases up to a PDW of 8%, peaks between 8 and 12%, and falls above 12%.Soil samples were taken three times over the course of the field season: once at transplant , once mid-season , and once during harvest . Each time samples were taken from four depths at each plot. Samples were homogenized and a sub-sample was immediately put on ice for transport to the lab. Each sample was then divided into fresh soil , dried at 60 degrees C , and dried at 105 degrees C . Ammonium and nitrate levels were measured after using 2M KCl to extract samples from transplant , midseason , and harvest samples using colorimetry. As soil pH was close to neutral, Olsen P37 was used to measure plant-available phosphate on samples from transplant and midseason . Gravimetric water content was assessed for all samples. Samples from transplant were composited by depth at each field, and texture was assessed using a modified pipette method38. At transplant, a soil core was taken with a bucket auger down to one meter from a central plot in each field and used to calculate bulk density at each depth increment. We then took a weighted average of GWC at each plot to calculate available water using bulk density and a pedotransfer function based on soil texture.