Progress report for GW19-193
Regenerative agriculture has the potential to increase biodiversity and promote key biological processes while reducing farmer investment in mechanical and chemical inputs over conventional monoculture production systems. Almonds are California’s second highest grossing crop, and represent an excellent study system for comparing conventional and regenerative orchard management practices. This study provides an innovative systems-level comparison of best management practices in regenerative and conventional almond production in Central CA. This 2-yr study: 1) Characterized the soil quality and biodiversity in almond production systems, with a special focus on soil carbon and pest management services; 2) Measured the relative yield and profitability of regenerative and conventional almond systems; and 3) Will Disseminate results to producers using a variety of learning tools. A character matrix of practices is used to designate orchards as regenerative or conventional. In replicated plots, soil organic matter, total soil carbon and nitrogen, microbial biomass and diversity, water infiltration, and bulk density will be measured. Insect communities on the soil surface, as well as pest populations in the orchard canopy will be enumerated. Food borne pathogens in the soil will be quantified. Producer surveys will provide the basis for a cost/benefit analysis of each orchard. To date field work and data analysis is complete and we are working on completing the manuscript of the findings. Together with the Community Alliance with Family Farmers (CAFF) and Ecdysis Foundation/Blue Dasher Farm, findings will be shared via the organizations’ networks. By providing clear and transparent empirical assessment of these two systems, we will improve the profitability of farmers, improve the natural resource base on almond orchard, and increase the quality of life for farmers and their communities.
The overall goal of this research is to provide critical data and education that removes perceived barriers for the adoption of regenerative almond production systems. Specific objectives of the proposal include:
- Characterize the soil quality and biodiversity present on regenerative and conventional almond production systems, with a special focus on soil carbon and pest management services.
- Measure the relative yield and profitability of regenerative and conventional almond systems, identifying key cost and benefits of the two systems.
- Disseminate results to producers using a variety of learning tools, including web-based documents, presentations at grower meetings, in-person field days, interviews with local and national media outlets, and peer reviewed scientific articles.
- - Technical Advisor (Educator)
- - Technical Advisor (Educator)
Experimental site selection
Sixteen orchards were studied in 2018 and 2019 (n = 8 each year). The Web Soil Survey was used to select similar soil types for the conventional-regenerative orchard pairings (Soil Survey Staff (NRCS-USDA)). To further establish similar soil conditions between treatments in pairs, the ribbon test was performed in each study plot, and sand-silt-clay composition was determined via the hydrometer technique .
The average age of the conventional orchards was 13.6 ± 2.96 y and the average age of the regenerative orchards was 17.6 ± 3.27 y. all of the orchards in the study contained at least two varieties (Klonsky, Sumner, Stewart, & Gutierrez, 2016), and almond varieties varied among the orchards (table 1).
The treatments were defined by rankings derived from a character matrix of different practices that are considered as regenerative or conventional (LaCanne and Lundgren 2018). Use of organic amendments (compost, manure, mulch, compost teas), no-till, grazing, maintaining ground cover through planting cover crops or fostering resident vegetation, planting hedgerows were all considered regenerative and received a score of 1 in a binary system. Spraying synthetic insecticides, herbicides, fungicides (synthetic and metal based), use of chemical fertilizers, bare soil, and tillage are all considered conventional practices, and received a 0 score. Orchards that score greater than 4 in our matrix are declared regenerative orchards, and those that received a score of 4 or lower are categorized as conventional orchards. The regenerative orchards had an average score of 6.81 ± 0.21, while the conventional orchards had an average score of 1.56 ± 0.22.
Replicate plots (n = 4) were established on each orchard. The plots were 40 × 40 m and separated by at least 15 m, resulting in 64 total observation points for the study. Plots were established 20 m into the orchard to avoid field margin effects.
Invertebrate community assessments
The epigeal invertebrate communities were sampled using quadrats ( Jonathan G. Lundgren, Shaw, Zaborski, & Eastman, 2006), that were placed at two random locations in the inter-row areas of each plot. Sampling of the invertebrate communities occurred during the bloom, fruit development, and harvest periods. The invertebrate communities that could be collected from the soil surface and top 2 cm of the soil with aspirators in 15 min were stored in 70% ethanol. The biomass of the invertebrates per 0.5 m2 were weighed, and invertebrates were identified to the morphospecies level and sorted into functional groups for community analyses. Voucher specimens are all housed in the insect collection at Blue Dasher Farm, Estelline, SD.
We assessed the insect pest damage on 500 almonds per farm in 2018 and 600 almonds per farm in 2019, (< 20 from any one tree), from the orchard floor (Bentley et al., 2001; Doll, 2009; Legner & Gordh, 1992). The almonds were each categorized as having: no pest damage, navel orange worm damage (Amyelois transitella), ant damage (Formicidae), oriental fruit moth damage (Grapholita molesta), peach twig borer damage (Anarsia lineatella) ,leaf footed plant bug or stinkbug damage (Coreidae, Pentatomidae), and unknown pest damage.
Total Soil Carbon and Nitrogen (0-60cm)
Soil samples (60 cm deep) samples were taken at a random location in each plot (Geisseler & Horwath, 2016). The probe (2.54 cm × 91.44 cm Plated Replaceable Tip Probe w/ 61 cm Window and Hammer Head Handle) was inserted 60 cm deep and the resulting soil samples (2.28 cm diameter) were partitioned to 0-5, 5-10, 10-15, 15-30, 30-45, and 45-60 cm depths. Samples were weighed to the nearest 0.1 g, and then were air dried. before being prepared for elemental analysis of total soil carbon (TSC) and total soil nitrogen (TSN). The air-dried weight of the soil was recorded to 0.01 g. Ground samples were passed through a sieve with 0.180 mm openings. For each soil depth, three sub-samples underwent elemental analysis (ECS 8020, NC Technologies, Milan, Italy). To calibrate the analysis, each sample tray consisted of five bypass samples (12-15 mg of soil), two blanks, and four standards, 0.5-2.0 mg (Acetanilide, Costech, Valencia, CA), followed by the soil samples. Mass (Mg) of TSC per depth layer was calculated using the Equivalent Soil Mass (ESM) method (Wendt & Hauser, 2013). This resulted in the assessment of carbon as Mg of TSC/ha at the following ESM layers, with the average calculated depth to reference mass in parentheses: 500 Mg (6.1 cm), 1,000 Mg (11.0 cm), 1,500 Mg (15.8 cm), 3,000 Mg (30.4 cm), 4,500 Mg (45.2 cm), and 6,000 Mg (59.2 cm).
Surface Bulk Density
Bulk density (BD) samples were taken in three plots per farm, following the protocol outlined by the NRCS (NRCS, 2017). A metal tube 7.62 cm wide and 12.7 cm tall was hammered to a depth of 8 cm. The volumes of soil in each core were estimated, and the samples were transferred to paper bags in the laboratory. Wet and dry weights of each soil core were recorded to the nearest 0.01 g and used to calculate BD and the soil’s gravimetric moisture percentage.
Preliminary soil texture was determined from a 15 cm core (2.54 cm diam). The ribbon test was performed on this soil core, and the texture class and soil moisture percentage were recorded (USDA-NRCS, 1998). Additionally, the soil samples (> 200 g) used to determine bulk density were analyzed for their sand, silt, and clay percentages using the hydrometer technique (Oregon State University-Central Analytical Laboratory- Standard Operating Procedures, 2017) .
Our analysis informs us that there is no difference in soil textures between treatments, with no correlation between matrix score and clay percentage. The mean, median, and standard for the farms’ sand silt, and clay percentages is found in Table 1.
Water Infiltration Rates.
Water infiltration rates were measured in one plot of each of the 2019 orchards (four regenerative and four conventional) during the bloom and fruit development stages of the crop. We followed the NRCS protocol, where 444 mL of water were poured into a sheet-metal ring (15.2 cm diam, 13.5 cm tall), which was hammered 6.5 cm into the soil (Doran, 1999). The time until all the water saturated into the soil was recorded to the nearest second.
Soil Macro and Micronutrients, Soil Organic Matter (SOM), Soil pH, the Haney Soil Health Score, and Soil Respiration.
Soil pH, soil macro and micronutrients, Haney soil health scores, and soil respiration were quantified in each orchard (Ward Laboratories, Kearney, NE). Soil cores (15 cm deep, 1.9 cm diam.; n = 16), were taken from the four replicate plots, totaling 16 samples per farm during the fruiting period of the orchard. To determine SOM, the Loss on Ignition (LOI) technique was used. Soil pH is quantified using the slurry method with a 1:1 ratio soil: water. Soil respiration, soil nutrients levels, and the Haney Soil Health Score were measured on samples that were dried at 50˚ C. The samples were ground to pass a 2 mm sieve and divided into three subsamples (two were 4 g each and one weighed 40 g. The 40 g soil sample is analyzed with a 24 h incubation test at 24o C. This sample is wetted through capillary action by adding 20 mL of deionized water to a 237 mL glass jar and then capped. After 24 h, the gas inside the jar was analyzed using an infrared gas analyzer (IRGA) (Li-Cor 840A, LI-COR Biosciences, Lincoln NE) for CO2-C. The two 4 g samples were extracted with 40 mL of deionized water and 40 mL of H3A, respectively. The samples were shaken for 10 min, centrifuged for 5 min, and filtered through Whatman 2V filter paper. The water and H3A extracts (Haney, Haney, Hossner, & Arnold, 2006) were analyzed on a flow injection analyzer (Lachat 8000, Hach Company, Loveland CO) for NO3-N, NH4-N, and PO4-P. The water extract was also analyzed on a Teledyne-Tekmar Torch C:N analyzer for water-extractable organic C and total N. The H3A extract was also analyzed on a Thermo Scientific ICP-OES instrument for P, K, Mg, Ca, Na, Zn, Fe, Mn, Cu, S and Al (Haney Test – Ward Laboratories Inc., 2020.).
Phospholipid fatty acid (PLFA) testing provides an index of a soil’s microbial biomass and composition (Frostegård, Tunlid, & Bååth, 2011). Soil cores (10 cm depth, 1.9 cm diam; n = 16), were taken from four replicates per site, totaling 16 samples per farm. This sampling occurred during the fruiting period. The samples were taken at random locations within each replicate, at least 5 m apart, using a transect that diagonally bisected the plot. The microbial biomass and community composition were recorded, and noting the Total microbial biomass, Undifferentiated microbial biomass, Total bacteria, Gram-positive bacteria, Actinomycetes, Gram-negative bacteria, Rhizobia bacteria, total fungi, Arbuscular mycorrhizal fungi, Saprophytic fungi, and Protozoa.
The ground cover height and composition in each of the replicates/plots were recorded during each of the three sampling periods. The percent ground cover was categorized as 0-25%, 25-50%, 50-75%, and 75-100%. Percent ground cover in the overall orchard was assessed using visual assessments of the percent ground cover in each invertebrate quadrat. The community composition and whether the ground cover was resident vegetation or planted was determined using information from the farmers and direct field observations.
A producer survey was used to determine management practices, costs and revenues that contribute to the direct net profitability of each operation (See supplemental materials for a copy of the survey). The factors used for determining profit were the production costs and material inputs, labor, and revenue. Under production operating costs, the study includes costs associated with winter sanitation, sampling for tree nutrient status and soil salinity, pH, and nutrient levels, irrigation and frost protection, fertilizers, insecticides, herbicides, fungicides, disease treatment sprays, trapping vertebrate pests, cover crop seed/bag, tillage, mowing, flamers, grazers, and harvest . Within the harvest category the study accounts for: people hours to conduct the harvest and/or the price paid to external contractors, kernel kg/ha, returns, and additional revenue streams such as almond hulls and co-products, as well as returns on harvesting the grazing livestock (Klonsky et al., 2016). No farm in the study reported additional revenue from grazers or additional revenue streams such as almond hulls. Two regenerative orchards reported revenues from selling value added products, such as almond butter. Under labor and operating costs, the study includes people hours worked (Klonsky et al., 2016). The 2016 labor rate for machine operators in almond orchards is $22.40/h and $15.40/h for non-machine operators. We used the mean, $18.90/h of these to incorporate people hours worked into overall orchard costs (Yaghmour et al., 2016). This is not a complete economic analysis as it does not account for cash overhead (insurance, taxes) and non-cash overhead (capitol recovery). Additionally, within the operating/cultural costs category the study does not account for the cost of pollination, custom pruning, and the cost of fuel for operating machinery (only propane for burning weeds).
Analyses were performed in Mini Tab 18 and R (R development Core Team, https://www.r-project.org/, version 4.0). The study used the lme4 package to create the Linear Mixed Models (Bates, Mächler, Bolker, & Walker, 2015). To calculate the biodiversity indices we utilized the vegan: Community Ecology package (Oksanen et al., 2019). For non-parametric statistical calculations and comparing models the study used the RVAideMemoire: Testing and Plotting Procedures for Biostatistics and the Modern Applied Statistics with S. Fourth Edition packages (Brian, Venables, Bates, Firth, & Ripley, 2002; Hervé, 2020) . Finally, for linear regression analysis and performing the Bonferroni Outlier Test the study utilized the Companion to Applied Regression (car) (John et al., 2019).
We tested the effect of orchards being defined as regenerative(Matrix Score>4) or conventional, as well as the orchards’ regenerative vs. conventional matrix scores (higher the score the more regenerative the orchard) on:: invertebrate biomass, invertebrate abundance, Arthropoda abundance, the Shannon-Wiener Index (H’ ) diversity indices, pest damage, TSC, TSN, bulk density, water infiltration rates, macro and micro nutrients, microbial communities, yield, revenue, costs, and profitability. As appropriate Linear Mixed Models, One-way ANOVA and Welch Two Sample t-tests were used to determine significance. In certain cases where the data were not normally distributed, and the samples size was less than N=15 the non-parametric Mood’s Median Test was utilized. Additionally, we quantified the effect of specific management practices (ground cover percentage, composition, and height , amendments, and various spraying regimes) on the response variables: invertebrate biomass, invertebrate abundance, Arthropoda abundance, the Shannon-Wiener (H’) diversity index, pest damage, TSC, TSN, bulk density, water infiltration rates, macro and micro nutrients, microbial communities, yield, revenue, costs, and profitability.
Finally, in order to elucidate potential connections within the agroecosystem we examined the effect of overall invertebrate biomass, overall invertebrate abundance, morphospecies and functional group Shannon-Wiener (H’) index biodiversity, and keystone species on the response variables: pest damage, soil organic matter, TSC, microbial biomass, and soil macro and micro nutrients. As four plots on each farm were sampled three times throughout the growing season the study utilized Linear Mixed Models (LMM) to determine significance with regards to the invertebrate community. For models encompassing the three sampling periods(bloom, fruiting, harvest) in 2018 and 2019 this results in 192 observations across the two groups, farm (n=16), and sampling trip (N=3)In the initial models Treatment (conventional or regenerative) was the fixed effect while farm (N=16) and sampling trip (N=3) were the random effects . In subsequent models the farm’s regenerative-conventional matrix score served as the fix effect (F), with farm and sampling trip continuing to serve as the random effects (R). When necessary the abundance and biomass data were log transformed or square rooted to meet the assumptions of the models and ensure a normal distribution of the residuals. The Shapiro-Wilks test was used to quantify the distribution of the model residuals. As is standard practice the vegan package log transformed the count of the invertebrate morphospecies to come up with the H’ index. To analyze the TSC and TSN soils data we built two LMM’s. The first LMM model designates treatment and clay percentage as fixed factors, with farm being the random factor (ESM Layer TSC/TSN/C:N x Treatment(F)+Clay%(F) + Farm(R), Number of obs: 49, groups: Farm, 16). The second LMM designates the orchards’ matrix scores and clay percentage as fixed factors, with farm being the random factor (ESM Layer TSC/TSN/C:N x Matrix Score(F)+Clay%(F) +Farm(R), Number of obs: 49, groups: Farm, 16). These two models are run for each ESM depth layer. As we only took three surface bulk density/soil texture samples per farm (the Atwater regenerative orchard was an exception; four samples were taken here), one soil sample per farm is randomly excluded from the LMM’s.
Finally, linear regression analysis was performed to quantify the key farm management practices contributing to the farms’ regenerative-conventional matrix scores, as well as a number response variables, such as TSC and biodiversity.
Unless otherwise noted data are reported as mean values followed by ±SE and significance was determine as P < 0.05.
Invertebrate biomass, abundance, and diversity (Shannon-Wiener (H’) and Simpson (DS=1−λ)) are significantly higher in regenerative orchards (P < 0.05). Additionally, earthworm abundance and arthropod abundance, H’ and DS are significantly higher in regenerative orchards. Further LMM’s assessing the relationships between biodiversity and pest damage finds a significant correlation between both, invertebrate and Arthropoda biodiversity, and reduced pest damage. A LMM shows that as invertebrate diversity (H) increases, the percentage of undamaged almonds increases significantly (χ2 (1)=5.20, p-value:0.023). Additionally, there is no significant difference in pest damage between regenerative and conventional orchards. The mean (± SEM) percent of non-pest damaged almonds in regenerative orchards is 92.7 ± 2.5%, while the mean percent of non-pest damaged almonds in conventional orchards is 91.1 ± 1.3% (N=16, welch p-value = 0.56).
To determine if there are particularly significant practices and/or landscape features contributing to higher levels of biodiversity in the orchards, we use a series of multiple linear regression and linear mixed models. We begin with two multiple linear regression models where Invertebrate H’ and Arthropoda H’ are the response variables and the regenerative-conventional matrix management categories are the explanatory variables (Grazers, Using Organic Amendments and/or Compost Tea , Not Using Synthetic Fertilizers, Herbicides and Insecticides (combined due to singularity), Not Using Synthetic Fungicides, Not Tilling, and Orchard Floor Ground Coverage 75-100%, and Hedgerows). Both of these models have significant p-values, with relatively high adjusted R squared values and significant intercepts ( Invertebrate H’: Adjusted R-squared: 0.61 F6,9: 4.97, p-value: 0.016, Arthropoda H’: Adjusted R-squared: 0.70, F7,8: 6.11 on 7 and 8 DF, p-value: 0.01). However, no single management factor is significant. These analyses in conjunction with the prior analyses examining the relationship between matrix score and H’, suggest that the more regenerative practices farmers implement on their farms the greater their farm’s biodiversity. Further, because we also see a significant relationship between farms meeting the regenerative threshold (regenerative practices>4) and biodiversity, it appears that scale emergent properties emerge when more than four of these practices are used on the farm, suggesting that farmers need to make systems level changes to their farms to increase biodiversity.
There was no statistical difference in soil pH between regenerative and conventional orchards (Welch p-value:0.06), with the mean pH of conventional orchards being 7.3 ± 0.07 and the mean pH of regenerative orchards being 7.5 ± 0.10.
LMM’s show that through all of the Equivalent soil mass (ESM) soil layers (500 Mg, 1000 Mg, 1500 Mg, 3000 Mg, 4500 Mg, 6000 Mg), regenerative orchards contain significantly higher levels of TSC and TSN. Through the 6000 Mg ESM layer regenerative orchards contain 54.52 Mg TSC/ha±4.76 Mg TSC/ha, while conventional orchards contain 41.37 Mg TSC/ha± 7.27 Mg TSC/ha. Through the 6000 Mg ESM layer regenerative orchards contain 54.52 Mg TSC/ha± 4.76 Mg TSC/ha, while conventional orchards contain 41.37 Mg TSC/ha± 7.27 Mg TSC/ha. At the 6000 Mg ESM layer regenerative orchards contain 5.96 Mg/TSN/ha ±- 0.45, while conventional orchards contain 5.00 Mg/TSN/ha ± 0.53. At the 6000 Mg ESM layer being a regenerative orchards affects Mg TSC/ha (χ2 (1)=5.38 treatment p-value: 0.02, χ2 (1)=35.53 clay p =0.0), increasing it approximately 11.58 Mg TSC/ha ± 4.99. Each percent increase in clay percentage increases TSC by 1.52 Mg TSC/ha ± 0.25. At the 6000 Mg ESM layer there is a significant correlation between the orchards’ matrix scores and TSC (χ2 (1)=4.98 Matrix Score p-value:0.026, χ2 (1)=36.01 clay p =0.0), with each 1 point increase in matrix score, increasing TSC by 2.10Mg TSC/ha ±0.94. Each percent increase in clay percentage increases TSC by 1.54 Mg TSC/ha ±0.26. Through the 1500 Mg ESM layer an orchard’s regenerative conventional matrix score is the dominant factor in determining its TSC and TSN. Clay becomes a significant factor in determining TSC at the 1500 Mg ESM layer, but with TSN clay is not a significant factor until the 3000 Mg ESM layer. At the 3000 Mg ESM layer matrix score and clay percentage become equally significant TSC determinants. Proceeding down through the 4500 and 6000 Mg ESM layers matrix score is still highly significant in determining TSC and TSN, but soil clay percentage becomes the more significant of the two factors.
The soils in regenerative orchards have lower bulk densities and more effectively infiltrate water. The mean surface bulk density in regenerative orchards is 1.16 ± 0.02 g/cm3, and in conventional orchards it is 1.33 ± 0.05 g/cm3 (n=16, t10.14 = 3.41, df = Welch p-value: 0.007). Higher regenerative-conventional matrix scores correlate to lower bulk densities. Each percentage increase in matrix score is associated with a 0.034 ± 0.008 g/cm3 reduction in bulk density. In regenerative orchards, water infiltrates 3,846 ± 954 seconds faster than that in conventional orchards (Adjusted R-squared: 0.70 F2, 5: 9.21, model P = 0.02 matrix score P = 0.01, clay P = 0.06, n=8).
Soil macronutrients and micronutrients, and soil respiration are calculated on a per farm basis (N=16, depth:15cm). The nutrients we examined are (bolded=significant differences between treatments): 1:1 Soluble Salts(Electric Conductivity), Water Extracted Total N, Water Extracted Organic N (WEON), Water Extracted Total Organic C (WEOC), H3A extracted (H3A) Nitrate, H3AAmmonium, H3AInorganic Nitrogen, H3ATotal Phosphorus, H3A Inorganic Phosphorus, H3AOrganic Phosphorus, H3AICAP Potassium, H3AICAP extracted Calcium , H3AICAP Aluminum , H3AICAP Iron, H3AICAP Sulfur, H3AICAP Zinc, H3AICAP Manganese, H3AICAP Copper, H3AICAP Magnesium, H3AICAP Sodium, % Microbially Active Carbon(MAC), Organic C:N (Conventional: 13.42 ± 1.19 , Regenerative: 12.21 ± 0.80) , Organic N:Inorganic N (Conventional: 1.39 ± 0.27, Regenerative: 1.90 ± 0.40), Organic N Release, Organic N Reserve, Organic P Release, Organic P Reserve, Available N, Available P, and Available K. Additionally, we examine differences in Haney Soil Health Scores, which is a soil health index( H3A is a soil extractant that extracts NH4,NO3, and P from soil. It removes the need for separate phosphorus (P) extractants for acid and calcareous soils, while maintaining the extract pH, on average, within one unit of the soil pH. The extractant, H3A , is made up of organic root exudates, lithium citrate, and two synthetic chelators (DTPA, EDTA) (Haney et al., 2006).
In this section we present the nutrients that are found in significantly different levels between the regenerative and conventional orchards. We determine significance via Welch t-tests, Mood’s Median tests, and simple linear models, with the regenerative-conventional matrix score/treatment and the soil clay percentage (when it improves the model) as fixed factors (Nutrient × Matrix Score/Treatment + Clay %).
Regenerative orchard soils had higher levels of Water Extractable Organic Matter (WEOM), which is comprised of WEON and WEOC. WEOC and WEON are the fraction of soil organic carbon and nitrogen present that is mobile and bioavailable to the microbial community (Grebliunas, Armstrong, & Perry, 2016; Zhang et al., 2011; Zsolnay, 1996). Regenerative orchards have 15.32 ± 1.9 ppm of WEON, while conventional orchards have 9.79 ± 1.2 ppm (N=16, Welch p-value: 0.03). Regenerative orchards contain higher levels of WEOC, with 178.9 ppm ±15, while conventional orchards have 122.6 ppm ±10 (Welch p-value: 0.009).
Regenerative orchard soils had higher levels of total Phosphorus, inorganic Phosphorus, and organic Phosphorus, and available Phosphorus. In regenerative orchards the median amount of H3A Total Phosphorus is 157.5 ppm, while in conventional orchards it is 55.0 (Mood’s p-value: 0.01). In regenerative orchards the median amount of Inorganic Phosphorus is 140.5 ppm, while in conventional orchards it is 47.5 (Mood’s p-value: 0.01). The median amount of available Phosphorus in regenerative orchards is 350.15 ppm and in conventional orchards it is 114.74 (p-value: 0.01).
Soils in regenerative orchards contain higher levels of Calcium, 897 ppm ±73 and 616 ppm ±61 in conventional orchards (Welch p-value: 0.01). Conventional orchards contain higher levels of Aluminum with 175.7 ppm ±19, while regenerative orchards have 112.5 ppm 12 (Welch p-value: 0.02). Regenerative orchards average 18.27 ppm ± 3.7 of Sulfur, while conventional orchards average 8.07 ppm ± 2.2 (Welch p-value: 0.037). The soil quality index, Haney Soil Scores, are higher in regenerative orchards. The mean Haney Score in regenerative orchards is 8.16 ± 0.98 and in conventional orchards it is 5.47 ± 0.64 (Welch p-value: 0.04).
The soil microbial community is quantified on a per farm basis (N=16, depth: 10cm). Via PLFA testing we examine (bolded=significant differences between treatments): Total Biomass, Diversity Index, Bacteria %, Total Bacteria Biomass, Actinomycetes %, Actinomycetes Biomass, Gram (-) %, Gram (-) Biomass, Rhizobia %, Rhizobia Biomass, Total Fungi %, Total Fungi Biomass, Arbusular Mycorrhizal (AMF) %, Arbuscular Mycorrhizal Biomass, Saprophytic %, Saprophytes Biomass, Protozoan %, Protozoa Biomass, Gram (+) Biomass, Gram (+) %, Undifferentiated %, Undifferentiated Biomass, Fungi:Bacteria, Predator:Prey, Gram(+):Gram(-), Sat:Unsat, Mono:Poly.
Regenerative orchards contain higher levels of total microbial biomass (ng/g). The mean microbial biomass in conventional orchards is 3642.2 ± 388.16 ng/g. Regenerative practices increase the microbial biomass by 1947.5 ng/g ± 896.7 to 5589.6 ng/g (Adjusted R-squared: 0.20 F-statistic: 4.72 on 1 and 14 DF, p-value: 0.048). Total bacterial biomass is greater in regenerative orchards. The mean bacterial biomass in conventional orchards is 1409.46 ± 216.07 ng/g. Regenerative practices increase the bacterial biomass by 1151.0 ± 471.4 ng/g to 2624.29 ng/g (Adjusted R-squared: 0.25 F1,14: 5.96 p-value: 0.029). Being regenerative increases an orchard’s Gram (+) biomass by 761.2 ng/g ± 261. to 1558.52 ng/g.
Specific management practices and Total Soil Carbon
For TSC as the response variable we run multiple regression analyses for each of the ESM layers (N=16). The explanatory variables are matrix categories consisting of : Grazers, Using Organic Amendments and/or Compost Tea , Not Using Synthetic Fertilizers, Herbicides and Insecticides (combined due to singularity), Not Using Synthetic Fungicides, Not Tilling, and Orchard Floor Ground Coverage 75-100%, and soil clay percentage. At the 500 and 1000 Mg ESM layers the sole significant practice is maintaining Orchard Floor Ground Coverage at the 75-100% level. At the 500 Mg ESM layer maintaining ground cover at the 75-100% level increases a soil’s TSC by 9.93 Mg TSC/ha ± 4.16 (Adjusted R-squared: 0.779 F6,9: 8.572 Model p-value: 0.004, Ground Coverage p-value: 0.044). At the 1000 Mg ESM layer maintaining ground cover at the 75-100% level increases a soil’s TSC by 12.04 Mg TSC/ha ± 5.06 (Adjusted R-squared: 0.791 F6,9: 11.43, Model p-value: 0.003, Ground Coverage p-value: 0.044).At the 1500 Mg ESM layer no one response variable is significant , however the model is still highly significant (Adjusted R-squared: 0.7445, F6,9: 7.251, p-value: 0.006). At the 3000(Adjusted R-squared: 0.61, Model p-value:0.03 , clay p-value:0.02 , clay coefficient:0.65 ± 0.23 ), 4500(Adjusted R-squared:0.56 , Model p-value: 0.04 , clay p-value: 0.01 , clay coefficient: 1.11 ± 0.33), and 6000 Mg (Adjusted R-squared: 0.54, Model p-value: 0.049, clay p-value:0.007 , clay coefficient: 1.50 ±0.42) ESM layers the only significant response variable is soil clay percentage, with the models maintaining significance. However, we must remember the results from the models in discussed previously which show that the orchards’ matrix scores significantly contribute to Mg TSC/ha through the 6000 Mg ESM layer, suggesting that it is not any one management practice but rather the combination of multiple management practices that makes management a significant factor in determining Mg TSC/ha.
The Invertebrate Community and Total Soil Carbon
The study examines the relationship between the invertebrate community and the critical soil quality metric, TSC, on the per farm level (n=16). Lumbricina abundance, invertebrate biomass, and H’ are positively correlated to Mg of TSC through the 1500 Mg ESM layer (Adjusted R2=0.242, F1,14: 5.79, p-value:0.031) (Adjusted R2=0.368, F1,14: 9.72, p-value:0.008) (Adjusted R2=0.367 F1,14: 9.68, p-value:0.008). Further, in examining the relationship between the farms’ H’ and the Mg of TSC at all the ESM layers sampled in the study, we find the strongest correlations at the shallowest ESM layers. The strength of the correlations declines as the ESM layers increase, with significant correlations ending after the ESM 1500 Mg layer . The diversity of a farms’ invertebrate community is largely dependent on the complexity and connectedness of the landscape (Bianchi, Booij, & Tscharntke, 2006; Dainese et al., 2019; Eilers & Klein, 2009; J. G. Lundgren & Fausti, 2015). The orchards with the higher regenerative-conventional scores have more complex landscapes as a result of higher rates of ground coverage percentage, number of plant species found on the orchard floor, and hedgerows. Additionally, maintaining a robust ground cover, using compost, integrating grazers, not tilling, avoiding synthetic pesticides, and a soil’s clay percentage are key factors for maintaining and building carbon in the soil (LaCanne & Lundgren, 2018; Ryals & Silver, 2013; Weil & Brady, 2017). Therefore, in order to try and isolate the effect of H’ on TSC we run multiple regression analyses, with TSC of the six different ESM layers as the scalar response variable and H’, ground cover, the use of compost or compost teas, not using synthetic herbicides, insecticides, and fertilizers (these three combined due to collinearity), not using fungicides, grazers, and not tilling, as the explanatory variables. Originally, clay was also a response variable, however ANOVA analysis of the two models returns p-values>0.05, so soil clay percentage is not included at the 500,1000, and 1500 Mg ESM layers. At the 3000, 4500, and 6000 Mg ESM layers we use the same response variables as the lower layers, with the addition of soil clay percentage (p<0.05). These 500 Mg ESM layer models show that none of the explanatory variables are significant (p>0.05). Though epigeal invertebrate H’ has the lowest p-value among the explanatory variables (p: value: 0.053) However, in the previous section we learn that the one management practice that has a significant influence on TSC at the 500 and 1000 Mg ESM layers is maintaining orchard floor coverage. Therefore, as a next step we run models with the 500 and 1000 Mg ESM layers as the response variables, while invertebrate H’ and maintaining 75-100% ground coverage are the explanatory variables. At the 500 Mg ESM layer we see that epigeal invertebrate H’ and maintaining 75-100% ground coverage play equally significant roles in building TSC. Increasing H’ increases TSC by 6.94 Mg TSC/ha ± 3.04 (p-value: 0.040), while having ground cover increases TSC by 6.41 Mg TSC/ha ± 2.66 (p-value: 0.032) (Adjusted R-squared: 0.771 F2,13: 29.62 , Model p-value:0.000). As we move to the 1000 Mg ESM layer epigeal invertebrate H’ no longer has a significant effect on TSC (p-value: 0.45), with ground coverage increasing TSC by 12.35 Mg TSC/ha ± 3.48 (Adjusted R-squared: 0.750 F2,13: 23.53, Model p-value:0.000, Ground Coverage p-value:0.004 ). In the top layer of soil the inputs of carbon from plant litter and rhizodeposition play critical roles in building soil carbon, but equally important is having a robust invertebrate community to break down the plant litter(Weil & Brady, 2017).
Yield and Profitability
There is no difference in yield between conventional and regenerative orchards. The mean yield in conventional orchards is 1713.08 lbs./acre ± 281.31, while the yield in regenerative orchards is 1193.34 lbs./acre ± 221.34 (Welch p-value: 0.17, N=14) . The yields in conventional orchards range from 480 lbs./acre to 2700 lbs./acre, with the yields in regenerative orchards ranging from 629 lbs./acre to 2200 lbs./acre. Additionally, there is no correlation between yield and matrix score (Adjusted R-squared: 0.006, F1,12: 1.074 , p-value: 0.32). We also examine the relationship between yield and: orchard age, clay %, SOM, TSC, soil nutrients, H’, and the regenerative-conventional matrix score categories. The only model with a significant correlation is the one examining, yield x orchard age + clay%. In this model increasing the soil clay percentage reduces yield by 38.49 lbs./acre ±16.28 (Adjusted R-squared: 0.39 F2,11: 5.23 , Model p-value: 0.025, clay % p-value: 0.038, orchard age p-value: 0.054).
Regenerative orchards are approximately twice as profitable as conventional orchards. The profitability of regenerative orchards, accounting for value added products (almond butter, etc.) is $6092.83/acre ± 1154.79. The profitability of regenerative orchards not accounting for value added products is $5928.67/acre ± 1090.15, while the profitability of conventional orchards is $ 2876.63/acre ± 732.97 (Welch p-value: 0.045). Additionally, each one-point increase in matrix score correlates to a $660.40/acre ± $241/50.00 increase in profitability (Adjusted R-squared: 0.351 F1,11: 7.476 p-value: 0.019) .
Regenerative orchards are twice as profitable as conventional orchards because they generate approximately twice the per acre revenue, with no difference in operating costs between the two management systems . The operating costs of regenerative orchards are $ 1374.86/acre ± 425.28 and for conventional orchards they are $ 1009.29/acre ± 36.55 (Welch p-value: 0.42).
There is no correlation between operating costs and matrix score (p-value:0.34). However, there is a correlation between revenue and matrix score, with each one-point increase in matrix score correlating to a $757.30/acre ±259.8 increase in revenue. Since, there is no difference between farming systems with regards to operating costs and yields, the greater per acre revenue of the regenerative orchards is due to the premium paid for almonds grown in regenerative systems. All the regenerative producers in this study are certified organic, allowing them to get the organic premium in the wholesale market. Additionally, some of the regenerative producers in the study sell the product directly to consumers, allowing them to receive an even greater premium for their product. Finally, as we mention in the methods section this is not a complete economic analysis. We solely focus on operating/cultural costs and do not account for the cost of pollination, custom pruning, and the cost of fuel for operating machinery (only propane for burning weeds).
Bates, D., Mächler, M., Bolker, B., & Walker, S. (2015). Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software, 67(1), 201–210. https://doi.org/10.18637/jss.v067.i01
Bentley, W. J., Hendricks, L., Duncan, R. A., Silvers, C., Martin, L., Gibbs, M., & Stevenson, M. (2001). BIOS and conventional almond orchard management compared. Calif. Agric., 55(5)(5), 12–19. https://doi.org/doi:10.3733/ca.v055n05p12
Bianchi, F. J. J. A., Booij, C. J. H., & Tscharntke, T. (2006). Sustainable pest regulation in agricultural landscapes: A review on landscape composition, biodiversity and natural pest control. Proceedings of the Royal Society B: Biological Sciences, Vol. 273, pp. 1715–1727. https://doi.org/10.1098/rspb.2006.3530
Brian, A., Venables, B., Bates, D. M., Firth, D., & Ripley, M. B. (2002). Modern Applied Statistics with S. Fourth Edition. https://doi.org/ISBN 0-387-95457-0
Dainese, M., Martin, E. A., Aizen, M. A., Albrecht, M., Bartomeus, I., Bommarco, R., … Steffan-Dewenter, I. (2019). A global synthesis reveals biodiversity-mediated benefits for crop production. Science Advances, 5(10), eaax0121. https://doi.org/10.1126/sciadv.aax0121
Doll, D. (2009). Taking a harvest sample for orchard IPM check-up – The Almond Doctor. Retrieved September 13, 2017, from UCANR website: http://thealmonddoctor.com/2009/09/08/taking-a-harvest-sample-for-orchard-ipm-check-up/
Doran, J. (Soil Q. I. (1999). Soil Quality Test Kit Guide. Retrieved from United States Department of Agriculture website: https://www.nrcs.usda.gov/Internet/FSE_DOCUMENTS/stelprdb1044790.pdf
Eilers, E. J., & Klein, A.-M. (2009). Landscape context and management effects on an important insect pest and its natural enemies in almond. Biological Control, 51, 388–394. https://doi.org/10.1016/j.biocontrol.2009.07.010
Frostegård, Å., Tunlid, A., & Bååth, E. (2011). Use and misuse of PLFA measurements in soils. Soil Biology and Biochemistry, 43(8), 1621–1625. https://doi.org/10.1016/J.SOILBIO.2010.11.021
Geisseler, D., & Horwath, W. R. (2016). Soil Sampling in Orchards. Davis.
Grebliunas, B. D., Armstrong, S. D., & Perry, W. L. (2016). Changes in water-extractable organic carbon with cover crop planting under continuous corn Silage production. Air, Soil and Water Research, 9, 45–54. https://doi.org/10.4137/ASWR.S30708
Haney, R. L., Haney, E. B., Hossner, L. R., & Arnold, J. G. (2006). Development of a New Soil Extractant for Simultaneous Phosphorus, Ammonium, and Nitrate Analysis. Communications in Soil Science and Plant Analysis, 37(11–12), 1511–1523. https://doi.org/10.1080/00103620600709977
Haney Test – Ward Laboratories Inc. (n.d.). Retrieved February 20, 2020, from https://www.wardlab.com/haney-test/
Hervé, M. (2020). RVAideMemoire: Testing and Plotting Procedures for Biostatistics. Retrieved from https://cran.r-project.org/package=RVAideMemoire
John, A., Weisberg, S., Price, B., Adler, D., Bates, D., Baud-bovy, G., … Winsemius, D. (2019). Companion to Applied Regression, Third Edition. Retrieved from https://socialsciences.mcmaster.ca/jfox/Books/Companion/
Klonsky, K., Sumner, D. A., Stewart, D., & Gutierrez, C. (2016). 2016 SAMPLE COSTS TO PRODUCE ORGANIC ALMONDS. Retrieved from https://coststudyfiles.ucdavis.edu/uploads/cs_public/24/7c/247c069c-8290-4060-aa96-3e556b0f339b/16almondorganicnsjvfinaldraft42716.pdf
LaCanne, C. E., & Lundgren, J. G. (2018a). Regenerative agriculture: merging farming and natural resource conservation profitably. PeerJ, 6, e4428. https://doi.org/10.7717/peerj.4428
LaCanne, C. E., & Lundgren, J. G. (2018b). Regenerative agriculture: Merging farming and natural resource conservation profitably. PeerJ, 2018(2), e4428. https://doi.org/10.7717/peerj.4428
Legner, E. F., & Gordh, G. (1992). Lower Navel Orange worm (Lepidoptera: Phycitidae) Population Densities Following Establishment of Goniozus legneri (Hymenoptera: Bethylidae) in California. Journal of Economic Entomology, 85(6), 2153–2160. https://doi.org/10.1093/jee/85.6.2153
Lundgren, J. G., & Fausti, S. W. (2015). Trading biodiversity for pest problems. Science Advances, 1(6), e1500558–e1500558. https://doi.org/10.1126/sciadv.1500558
Lundgren, Jonathan G., Shaw, J. T., Zaborski, E. R., & Eastman, C. E. (2006). The influence of organic transition systems on beneficial ground-dwelling arthropods and predation of insects and weed seeds. Renewable Agriculture and Food Systems, 21(04), 227–237. https://doi.org/10.1079/RAF2006152
NRCS. (2017). Soil Quality: Indicators: Bulk Density. Retrieved November 27, 2017, from Soil Quality for Environmental Health website: http://soilquality.org/indicators/bulk_density.html
Oksanen, J., Guillaume Blanchet, F., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., … Wagner, H. (2019). Vegan: Community Ecology Package. R Package Version 2.5-2.6. Retrieved from https://cran.r-project.org/package=vegan
Oregon State University-Central Analytical Laboratory- Standard Operating Procedures. (2017). Retrieved from http://oregonstate.edu/ehs/sites/default/files/pdf/lab_safety_requirements.pdf
Ryals, R., & Silver, W. L. (2013). Effects of organic matter amendments on net primary productivity and greenhouse gas emissions in annual grasslands. Ecological Applications, 23(1), 46–59. Retrieved from https://www.marincarbonproject.org/file/2018-documents/2_Effects-of-organic-matter-amendments-on-net-primary-productivity-and-greenhouse-gas-emissions-in-a
Soil Survey Staff (Natural Resources Conservation Service United States Department of Agriculture). (n.d.). Web Soil Survey. Retrieved February 13, 2020, from https://websoilsurvey.sc.egov.usda.gov/App/HomePage.htm
USDA-NRCS. (1998). Estimating Soil Moisture by Feel and Appearance. Retrieved from ftp://ftp-fc.sc.egov.usda.gov/MT/www/technical/soilmoist.pdf
Weil, R., & Brady, N. (2017). The Nature and Properties of Soils. 15th edition.
Wendt, J. W., & Hauser, S. (2013). An equivalent soil mass procedure for monitoring soil organic carbon in multiple soil layers. European Journal of Soil Science, 64(1), 58–65. https://doi.org/10.1111/ejss.12002
Yaghmour, M., Haviland, D., Fichtner, E., Sanden, B., Viveros, M., Sumner, D., … Gutierrez, C. (2016). Sample Costs to Establish an Orchard and Produce Almonds. In University of California Agricultre and Natural Resources Cooperative Extension Agricultural Issues Center UC Davis Department of Agricultural and Resource Economics. Retrieved from https://coststudyfiles.ucdavis.edu/uploads/cs_public/27/59/27599258-3e53-4e5f-aefb-08d12f5b0cb9/16almondssjvnorthfinaldraft81116.pdf
Zhang, M., He, Z., Zhao, A., Zhang, H., Endale, D. M., & Schomberg, H. H. (2011). Water-extractable soil organic carbon and nitrogen affected by tillage and manure application. Soil Science, 176(6), 307–312. https://doi.org/10.1097/SS.0b013e31821d6d63
Zsolnay, A. (1996). Dissolved Humus in Soil Waters. In Humic Substance in Terrestrial Ecosystems (pp. 171–223). https://doi.org/10.1016/B978-044481516-3/50005-0
Educational & Outreach Activities
To date we have presented the preliminary results of the project at a variety of conferences and through webinars. At the EcoFarm conference the preliminary results of this project were presented as part of the keynote address by Dr. Lundgren (1/9/2020) and a pre-event full day workshop (1/8/2020), while Tommy Fenster presented a poster on this project (1/9/2020). Before the California Small Farms Conference Dr. Lundgren presented the preliminary results of this project at the Cal Poly San Lois Obispo Sustainable Agriculture Lecture Series (2/27/2020). At the California Small Farms Conference Tommy Fenster, Dr. Lundgren, and Ward & Rosie Burroughs (two of the farmers in the study) conducted a conference workshop based around the preliminary results of this study (2/28/2020). On 2/29/2020 a large beekeeper we work with who works with a number of Almond growers in the Central Valley, took us out to consult with several farmers he supplies bees to about the study’s findings. On 7-16-2020 Dr. Lundgren and Ward and Rosie Burroughs (two of the farmers in the study) lead a webinar hosted by Chico State Regenerative Ag and Science program discussing the preliminary findings of this study.
Originally, our plan was to host two Farmer field days (one in Chico and one in the San Joaquin Valley) with the Community Alliance of Family Famers on regenerative farming in almond orchards. However, due to Covid-19 we are transitioning to a webinar field day format. CAFF has successfully executed several of these webinars to date, so we are confident in this being an effective outreach event. Our plan is to host this event in the first two weeks of October and have famers from both Chico and the San Joaquin Valley participate, combining the two field days into one webinar. We think this will be very informative for farmers looking to transition as our San Joaquin Valley participants will be Ward and Rosie Burroughs who have been operating a regenerative system a number of years, as well as the neighbor Wes Sperry whose conventional orchard was a part of this study. Further as a result of sharing the preliminary findings of this study with Wes in 2019 he began to transition one of his blocks from a conventional to a regenerative system. Thereby, providing workshop participants with key farmer resources who can speak to both what an established system looks like as well as the transition process. The farmers in Chico indicated that they would like to do a field day, but we still need to get verbal confirmation before locking them in to the workshop.
As we are now doing this field day in a webinar format we will provide the participants with electronic pre and post event surveys. When they sign up for the webinar we will require that they provide us with an email address. On the day before the webinar we will send the participants a survey to quantify their management practices and their familiarity with regenerative agriculture. Six months later we will send that out a survey to determine if the workshop resulted in the adoption of any new management practices.
Originally, we stated that “results from the study will be shared via the CAFF and Ecdysis Foundation: Blue Dasher Farm websites, social media accounts, and e-newsletters from the beginning of 2020 through the spring of 2020.” However, due to Covid-19 the lab at Cal State East Bay was shut down until August of 2020. This delayed the completion of some of the results. Instead of submitting the papers from the study in the Spring of 2020 we will now be submitting the results to journals this Fall. Therefore, the plan is to create fact sheets, etc. and begin sharing the finalized results via websites, e-newsletters, and social media starting in the late Fall of 2020, continuing through the Spring of 2021.
To date we have presented the preliminary findings of this project at several conferences, via a couple webinars, and the network we have built with famers and beekeepers. However, as we just finished all of the data analysis a few weeks ago, we have not yet engaged in extensive outreach. The goal is to begin submitting the paper(s) from this study this Fall. Further, we have a webinar on regenerative almond systems planned for the first week in October. Once we formally publish the results we hope that this project will have a large positive impact on agricultural sustainability. The results from this study indicate that when farmers take a regenerative systems based approach they create a more diverse and ecologically farming system with the biology of the system providing the same effect as synthetic inputs in conventional systems. Further, these regenerative systems provide key ecosystem services while reducing the environmental health impact to the farm community. For example, this study finds that regenerative almond systems have equivalent yields to conventional systems, no differences in pest damage (positive correlation of invertebrate community to reduced pest damage). This is all while being twice as profitable, holding more carbon, having more robust and diverse invertebrate communities, and infiltrating water more effectively. We hope that these results and the associated outreach can demonstrate to farmers that a transition to sustainable agriculture not only improves the environmental health of their farms, but also their farm’s economic outlook, while making their farms more resilient to abiotic and biotic pressures.
This project has been an incredible learning experience for me. Since I am a Masters student this is my first research project, and this project taught me how to plan and execute a research project involving sustainable agriculture. I have learned that sustainable agricultural systems may vary by practices, but it is key for sustainable systems to take a systems level approach where multiple sustainable practices are working hand in hand to create a system where the whole is greater than the sum of the practices. Farmers cannot just add one sustainable practice and expect to have a sustainable system. Further, going into this project I hypothesized that the sustainable systems would see significantly lower yields than the conventional systems. However, seeing the results from this project and then extensively reviewing the literature on regenerative agriculture and talking to farmers practicing regenerative systems I learned that regenerative systems can have equivalent yields, while supplying key ecosystem services.
We will provide in the next report.