Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Soil, crop yield, and cost trade-offs of organic nutrient management strategies across mixed vegetable… Norgaard, Amy 2020

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2020_november_norgaard_amy.pdf [ 2.1MB ]
Metadata
JSON: 24-1.0394816.json
JSON-LD: 24-1.0394816-ld.json
RDF/XML (Pretty): 24-1.0394816-rdf.xml
RDF/JSON: 24-1.0394816-rdf.json
Turtle: 24-1.0394816-turtle.txt
N-Triples: 24-1.0394816-rdf-ntriples.txt
Original Record: 24-1.0394816-source.json
Full Text
24-1.0394816-fulltext.txt
Citation
24-1.0394816.ris

Full Text

 SOIL, CROP YIELD, AND COST TRADE-OFFS OF ORGANIC NUTRIENT MANAGEMENT STRATEGIES ACROSS MIXED VEGETABLE FARMS IN SOUTHWEST BRITISH COLUMBIA   by   Amy Norgaard  B.Sc. The University of British Columbia, 2015    A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF  THE REQUIREMENTS FOR THE DEGREE OF    MASTER OF SCIENCE  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES  (Soil Science)     THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)    October 2020   © Amy Norgaard, 2020   ii The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, a thesis entitled:  Soil, crop yield, and cost trade-offs of organic nutrient management strategies across mixed vegetable farms in southwest British Columbia  submitted by Amy Norgaard in partial fulfillment of the requirements for the degree of Master of Science in Soil Science  Examining Committee: Dr. Sean Smukler, Faculty of Land and Food Systems Supervisor  Dr. Maja Krzic, Faculty of Land and Food Systems / Faculty of Forestry Supervisory Committee Member  Dr. Juli Carrillo, Faculty of Land and Food Systems Supervisory Committee Member Dr. Arthur Bomke, Faculty of Land and Food Systems Additional Examiner     iii Abstract Nitrogen (N) and phosphorus (P) are essential for crop growth but degrade the environment when lost from farming systems. While conventional farms have capacity to precisely calculate nutrient budgets based on the nutrient content of synthetic fertilizers, organic nutrient sources have inconsistent and difficult to predict nutrient supply. The objectives of this study were to: (1) inventory amendment and soil properties across three regions of southwest British Columbia (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)), and (2) evaluate effects of three nutrient management strategies on 20 farms across these regions on crop yields, economics (input costs), and selected soil properties (permanganate oxidizable carbon (POx-C), post-season available N and P). Nutrient strategies evaluated were: ‘high compost’ (HC): compost applied to meet crop N removal, ‘low compost + N’ (LC+N): compost applied to meet crop P removal plus an organic fertilizer to meet crop N removal, and ‘typical’ (TYP): the typical nutrient application used by the farmer (varying combinations of composts and organic fertilizers). While I found no differences in POx-C among nutrient management strategies, I did find HC had higher yields in the FV. However, principal components analysis (PCA) showed that HC was also associated with high post-season available N when high N composts and manures were used. Input costs tended to be least expensive in the lower Fraser Valley region, where TYP was less expensive than either HC or LC+N. The PCAs also showed that there was enhanced yield and POx-C values with LC+N when composts with high carbon to N ratios (C:N) were used. However, in regions where high nutrient composts are relatively inexpensive, productivity and   iv economic incentives encourage practices that contribute to high soil P and post-season available N.  The results of this study highlight the trade-offs between environmental and economic goals; even though organic farmers have land stewardship in mind, decisions are still largely influenced by economic principles, while in the bounds of organic regulations.   v Lay Summary  The goal of this research was to help organic farmers balance goals of environmental stewardship and crop production. Nutrients such as nitrogen and phosphorus are required for crop growth but are also environmental contaminants when lost from soil to either the atmosphere or hydrosphere. This study evaluated three organic nutrient management strategies across 20 mixed vegetable farms in southwest British Columbia. Strategies were evaluated in terms of crop yield, selected soil properties, and input costs. I found that nutrient management strategies were most successful when tailored to a farming system. Specifically, using specialty organic fertilizers (i.e. feather meal) in combination with lower nutrient composts are most likely to balance the trade-offs among crop production, potential environmental pollution and input costs. In farming systems with relatively inexpensive but high nutrient composts, maximizing yields and economics may come with an environmental trade-off of increased residual nutrients left in the soil after harvest.   vi Preface This thesis is original and unpublished work based on an experiment I, Amy Norgaard, carried out, with help from advisors, peers, and undergraduate students. As the lead investigator for this study, I was responsible for formulating specific research questions and completing data collection and analysis, and thesis composition. Assistance with collecting and analyzing samples was provided by undergraduate students Hannah Friesen, Katelyn Hengel, Conley Keyes, Jessica Mayes, and Carmen Wong. Dr. Sean Smukler designed the project, and was involved throughout, from research objectives to thesis editing, and is the main supervisory author on this study. Drs. Maja Krzic and Juli Carrillo provided input for methods and data analysis and helped with valuable edits to the thesis.     vii Table of Contents  Abstract ......................................................................................................................................... iii Lay Summary .................................................................................................................................v Preface ........................................................................................................................................... vi Table of Contents ........................................................................................................................ vii List of Tables ................................................................................................................................ ix List of Figures .................................................................................................................................x List of Equations ........................................................................................................................ xiii List of Abbreviations ................................................................................................................. xiv Acknowledgements ......................................................................................................................xv Dedication ................................................................................................................................... xvi Chapter 1: Introduction ................................................................................................................1 1.1 Nutrient management challenges on organic vegetable farms ....................................... 1 1.1.1 Nitrogen management ............................................................................................. 5 1.1.2 Phosphorus management ........................................................................................ 7 1.2 Carbon management and impacts to soil health ........................................................... 10 1.3 Research objectives and hypotheses ............................................................................. 14 Chapter 2: Methods .....................................................................................................................18 2.1 Compost application rate calculations .......................................................................... 23 2.2 Compost analyses.......................................................................................................... 24 2.3 Soil analyses.................................................................................................................. 28 2.3.1 Soil bulk density ................................................................................................... 31 2.4 Crop yield sampling ...................................................................................................... 32 2.5 Economics: estimation of input costs ........................................................................... 33 2.6 Statistical analyses ........................................................................................................ 34 Chapter 3: Results & Discussion ................................................................................................37 3.1 Inventory of regional characteristics ............................................................................. 37 3.1.1 Soil properties ....................................................................................................... 37 3.1.2 Compost characteristics ........................................................................................ 40 3.1.3 Carbon and nutrient application ............................................................................ 44 3.1.4 Amendment costs by region.................................................................................. 47 3.2 Crop Yield ..................................................................................................................... 50 3.2.1 Crop yield and baseline soil and compost properties............................................ 54 3.3 Soil properties ............................................................................................................... 59 3.3.1 Permanganate oxidizable carbon .......................................................................... 59 3.3.2 Post-season available nitrogen .............................................................................. 61 3.3.3 Post-season available phosphorus ......................................................................... 68 3.4 Input costs ..................................................................................................................... 71 3.4.1 Cost per unit yield ................................................................................................. 73 3.5 Trade-offs of nutrient management strategies .............................................................. 75 Chapter 4: Conclusion .................................................................................................................78 4.1 General conclusions ...................................................................................................... 78 4.2 Strengths and limitations of the research ...................................................................... 79 4.3 Management implications for farmers .......................................................................... 80   viii 4.4 Directions for future research ....................................................................................... 82 Bibliography .................................................................................................................................85 Appendices ....................................................................................................................................95 Appendix A - Datasets .............................................................................................................. 95 Appendix B - Crop characteristics ............................................................................................ 97 Appendix C - Soil properties .................................................................................................... 99 Appendix D - Amendments .................................................................................................... 105 Appendix E - Yield sampling ................................................................................................. 118 Appendix F - Results .............................................................................................................. 119    ix List of Tables  Table 2.1 Characteristics of the three nutrient management strategies evaluated in my study, including the crop nutrient targeted for compost or feather meal (N or P) applications, and hypothetical outcomes in terms of N or P. .................................................................................... 23 Table 2.2 Estimated nitrogen (N) and phosphorus (P) crop removal rates (kg ha-1) based on estimated yields, averaged across major crop categories. Values shown are number in each category (n), mean   standard deviation (SD), minimum (min.) and maximum (max) values. .. 23 Table 2.3 Cover crop biomass dry weight (dry wt.) from five sites. ............................................ 27 Table 3.1 Chemical soil properties determined in the pre-season, averaged within regions, lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI), including total nitrogen (N), carbon (C), and inorganic carbon (IC), Mehlich-3 phosphorus (P), Mehlich-3 potassium (K), electrical conductivity (EC) and pH. Values are means  standard deviation and ranges with minimum – maximum. .................................................................................................................. 39 Table 3.2 Soil particle size distribution determined on the pre-season samples, and coarse fragment content (% volume) determined on the bulk density samples, at the 0-15 cm and 15-30 cm depths, averaged within regions, lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI). Values are means  standard deviation and ranges with minimum – maximum. ..................................................................................................................................... 39 Table 3.3 Descriptions of compost types used in the field trials. Composts that were produced by facilities whose main purpose is to generate and sell compost are indicated by ‘yes’ in the Commercial column; all other composts were sold by the farm or business generating the materials or were made on the farms that we worked with. ......................................................... 42 Table 3.4 Compost properties averaged by type within region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)), including total carbon (C), total nitrogen (N), the proportion of total N estimated to be plant available N (PAN), total phosphorus (P), carbon to nitrogen ratio (C:N), and nitrogen to phosphorus ratio (N:P). Values shown are number of samples per group (n), means  standard deviations, minimum values (min), and maximum values (max). ................................................................................................................................. 43 Table 3.5 Total carbon (C), total nitrogen (N), plant available N (PAN) and total phosphorus (P) input rates averaged by region for each nutrient management strategy for all farms and both years of the study. ......................................................................................................................... 46    x List of Figures  Figure 2.1 Map of farm locations in three regions of southwest British Columbia used in field trials in 2018-19. ........................................................................................................................... 18 Figure 2.2 Historical (1981-2010) and yearly data for total monthly precipitation and average temperature from the (A) Vancouver International airport, (B) Pemberton airport, (C) Nanaimo, and (D) Sidney weather stations (Government of Canada, 2019). ............................................... 21 Figure 3.1 Boxplot of fertilizer costs (cost of product + cost of transport) ($ kg-1) used in the LC+N and TYP plots by region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)). The center line indicates the median, means are shown as black dots and sample size indicated by n. ........................................................................................................... 48 Figure 3.2 Boxplot of compost costs ($ cubic yard-1) by region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)). The center line indicates the median, means are shown as black dots and sample size indicated by n. .................................................. 48 Figure 3.3 Boxplot of yield (kg m-1) by nutrient management strategy within each region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)). ANOVA F and p-values refer to main effect of nutrient strategy within each region over both years. Boxplots with different letters represent significant differences among treatments using Tukey’s post-hoc test at p<0.05. The center line indicates the median, means are shown as black dots and sample size indicated by n. ............................................................................................................................... 53 Figure 3.4 Boxplot of yield (kg m-1) in the lower Fraser Valley by nutrient management strategies within years. ANOVA F and p-values refer to main effect of nutrient strategy within year. Boxplots with different letters represent significant differences among treatments using Tukey’s post-hoc test at p<0.05. The center line indicates the median, means are shown as black dots and sample size indicated by n. ............................................................................................. 53 Figure 3.5 Principal Component Analysis (PCA) biplots for relative yields by nutrient management strategy (Typical:TYP, high compost: HC, and low compost + N: LC), plotted with pre-season compost and soil properties and grouped by region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)), where: (A) includes all nutrient management strategies, and (B) shows only the TYP strategy plotted with the quantity of applied nutrients (applied total phosphorus, carbon, nitrogen, and plant-available nitrogen (NO3- + NH4+) = appTP, apptTC, appN, and appPAN, respectively)). Pre-season compost and soil properties include: compost plant-available N (NO3- + NH4+) = compost_PAN, compost C to N ratio = compost_CN, soil C to N ratio = soil_CN, soil total organic C content = soil_C, soil mehlich-3 phosphorus = soil_P, and soil sand, clay, and pH. ........................................................................ 56 Figure 3.6 Principal Component Analysis (PCA) biplot of the relative POx-C measured in each nutrient management strategy (typical = TYP, low compost + N = LC, and high compost = HC), plotted with pre-season soil and compost properties, including compost plant-available N (NO3- + NH4+) = compost_PAN, compost C to N ratio = compost_CN, soil C to N ratio = soil_CN, soil total organic C content = soil_C, soil mehlich-3 phosphorus = soil_P, and soil sand, clay, and pH, and grouped by region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)).................................................................................................................................... 60 Figure 3.7 Boxplot of post-season NH4+-N (mg kg-1) (0-15 cm). (A) By nutrient management strategy, averaged over regions and years, and (B)  by nutrient management strategy within regions (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)) over   xi years. ANOVA F and p-value refer to main effect of nutrient strategy. Boxplots with different letters represent significant differences among nutrient strategies using Tukey’s post-hoc test at p<0.05. The center line indicates the median, means are shown as black dots and sample size is indicated by n. ............................................................................................................................... 64 Figure 3.8 Boxplots of post-season NO3--N (mg kg-1) (0-15 cm). (A) by nutrient management strategy, averaged over regions and years, and (B) by nutrient management strategy within regions (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)) over years. ANOVA F and p-value refer to main effect of nutrient strategy. Boxplots with different letters represent significant differences among nutrient strategies using Tukey’s post-hoc test at p<0.05. The center line indicates the median, means are shown as black dots and sample size is indicated by n. ............................................................................................................................... 65 Figure 3.9 Boxplot of post-season NO3--N (mg kg-1) (0-30 cm) by nutrient management strategy within regions averaged over years. ANOVA F and p-value refer to main effect of nutrient strategy within region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)). The center line indicates the median, means are shown as black dots and sample size is indicated by n. Red line indicates threshold  of 25 mg NO3- kg-1 for the 0-30 cm depth determined by the BC Ministry of Agriculture to trigger follow-up action. ................................ 66 Figure 3.10 Principal Component Analysis (PCA) biplot of the relative NO3- (0-30 cm) measured in each nutrient management strategy (typical = TYP, low compost + N = LC, and high compost = HC), plotted with pre-season soil and compost properties, including compost plant-available N (NO3- + NH4+) = compost_PAN, compost C to N ratio = compost_CN, soil C to N ratio = soil_CN, soil total organic C content = soil_C, soil mehlich-3 phosphorus = soil_P, and soil sand, clay, and pH, and grouped by region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)). ................................................................................................................ 67 Figure 3.11 Boxplot of post-season Kelowna-extractable available P (mg kg-1) by nutrient management strategy, averaged over regions and years. ANOVA F and p-values refer to main effect of nutrient strategy. Boxplots with different letters represent significant differences among treatments using Tukey’s post-hoc test at p<0.05. The center line indicates the median, means are shown as black dots and sample size is indicated by n. ................................................................ 68 Figure 3.12 Principal Component Analysis (PCA) biplot of the relative post-season available P (0-15 cm) measured in each nutrient management strategy (typical = TYP, low compost + N = LC, and high compost = HC), plotted with pre-season soil and compost properties, including compost plant-available N (NO3- + NH4+) = compost_PAN, compost C to N ratio = compost_CN, soil C to N ratio = soil_CN, soil total organic C content = soil_C, soil mehlich-3 phosphorus = soil_P, and soil sand, clay, and pH, and grouped by region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)). ........................................................ 70 Figure 3.13 Boxplot of input costs ($ ha-1) by nutrient management strategy averaged over years within regions. ANOVA F and p-values refer to main effect of nutrient strategy within each region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)). Boxplots with different letters represent significant differences between nutrient strategies using Tukey’s post-hoc test at p<0.05. The center line indicates the median, means are shown as black dots and sample size is indicated by n. ........................................................................................................ 72 Figure 3.14 Principal Component Analysis (PCA) biplot of the relative cost per unit yield by nutrient management strategy (typical = TYP, low compost + N = LC, and high compost = HC), plotted with pre-season soil and compost properties, including compost plant-available N (NO3-   xii + NH4+) = compost_PAN, compost C to N ratio = compost_CN, soil C to N ratio = soil_CN, soil total organic C content = soil_C, soil mehlich-3 phosphorus = soil_P, and soil sand, clay, and pH, and grouped by region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)).................................................................................................................................... 74 Figure 3.15 Measured outcomes scaled to the maximum value observed among treatments, expressed as the mean within each treatment. Measured outcomes include: yield, input costs, post-season NO3- (0-30 cm depth) (NO3), post-season available P (0-15 cm depth) (soil_P), and permanganate oxidizable carbon (POXC). Measured outcomes with significant differences at alpha <0.05 between treatments are indicated with an asterisk (*). ............................................. 77    xiii List of Equations Equation 1:  ................................................................................................................................. 25 Equation 2: .................................................................................................................................. 26     xiv List of Abbreviations BC – British Columbia C – Carbon C:N – Carbon to nitrogen ratio EC – Electrical conductivity FV – lower Fraser Valley HC – the ‘high compost’ nutrient strategy kg – kilogram  lb – pound  LC+N – the ‘low compost + N’ nutrient strategy MOE – Ministry of Environment N – Nitrogen NH4+ – Ammonium  NO3- – Nitrate N:P – Nitrogen to phosphorus ratio PV – Pemberton Valley P – Phosphorus POx-C – Permanganate oxidizable carbon PCA – Principle component analysis PAN – Plant available nitrogen PHN – Post-harvest nitrate SOC – Soil organic carbon SOM – Soil organic matter TYP – the ‘Typical’ nutrient strategy  VI – Vancouver Island         xv Acknowledgements I could not have completed the work for this thesis without support from many people.   Frist, thank you to my supervisor, Dr. Sean Smukler, for the never-ending guidance and motivation. His passion for sustainable agriculture and creating change is a constant source of inspiration. I would also like to thank my committee members, Dr. Maja Krzic and Dr. Juli Carrillo, for sharing their time and knowledge to provide valuable insights for fine-tuning my project.  I am extremely grateful for the contributions of many research assistants, especially Carmen Wong, and also Conley Keyes, Katelyn Hengel, Hannah Friesen, and Dylan Peluso – field days with them were a constant source of joy.  I would also like to thank Carine Bening, from Agriculture and Agri-Food Canada in Agassiz, and Dr. Les Lavkulich, for resources and guidance for my phosphorus lab work.  Many, many thanks to my mentor and friend, Emma Holmes.  Finally, I have complete gratitude for my fellow soil science students and lab mates, who are too many to name, but provided countless hours of support, inspiration, enthusiasm, laughs, and ongoing friendship. Similarly, I could not have done this work without my community of friends here in Vancouver. This research was made possible with funding from Mitacs Career Connect and Organic Science Cluster 3 under the Agri-Science program of Agriculture and Agri-Food Canada.     xvi Dedication This thesis is dedicated to all of the farmers we collaborated with. They shared their time, knowledge, curiosity, and passion more generously than we could have hoped for.                    1 Chapter 1: Introduction  1.1 Nutrient management challenges on organic vegetable farms Organic agriculture is often promoted as an environmentally sound alternative to conventional production (Gomiero et al., 2011). This has contributed to the growing consumer demand for organic production, which has resulted in a market totalling $5.4 billion in Canada in 2017 (Canada Organic Trade Association, 2017). However, organic yields are still typically lower than conventional (Gomiero et al., 2011), largely due to challenges with nutrient availability in organic farming systems (de Ponti et al., 2012), which is primarily linked to the mismanagement of nitrogen (N) (Berry et al., 2002; Seufert, et al., 2012). Improving nutrient management has been ranked a top priority by organic farmers in North America (Jerkins & Ory, 2016). While conventional systems work with flexible nutrient budgets based on concentrated and often single-nutrient synthetic fertilizers, organic agriculture promotes the use of, and relies on, carbon-based nutrient sources. These nutrient sources, such as composts, cover-crops, and specialty fertilizers (i.e. feather meal, blood meal, guano, fish emulsion, etc.) present a number of challenges, including an inconsistent ratio and mobility of nutrients, which are difficult to predict and often do not match plant nutrient demand (Gale et al., 2006; Maltais-Landry et al., 2016).  Farm-scale nutrient budgets can be used to assess and prevent nutrient imbalances and inform targeted nutrient application rates (Watson et al., 2002). However, balancing nutrients in mixed vegetable farms can be particularly challenging given their high and variable nutrient demands within and across fields. A meta-analysis by Seufert et al. (2012) found that the organic vegetable production, compared to conventional, performs exceptionally poor for yields (-33%), relative to other sectors, such as fruit and oilseed production (-3% and -11%, respectively).   2 When considering performance by crop types, they found that perennial and legume crops have the lowest yield gap due to a greater N use efficiency by these crops. Additionally, phosphorus (P), another macronutrient needed in large quantities, is especially challenging to balance in relation to N budgets in farming systems that rely on composts and manures as primary nutrient sources (Nelson & Janke, 2007). This is mainly because of the mismatch of N:P ratios between crops and composts/manures. Because these amendments “couple” N and P, their sole use rarely balances both N and P supply with crop demand; instead, one of these nutrients are more commonly over- or under-applied.  While insufficient nutrients lead to yield reduction, their excess can have extensive environmental impacts beyond the farm gate. Nitrogen losses (as ammonia (NH3) or nitrous oxide (N2O) gases or nitrate ions (NO3-)) from agricultural fields occur through several pathways, depending on the source of N and environmental conditions in the field. These losses contribute to degraded air quality, climate change, and disrupted waterway ecology and contaminated drinking water. Although P is lost through one primary pathway (erosion), its contribution to eutrophic waterbodies is substantial (Kleinman et al., 2011). When composts and manures are applied at rates to match and not exceed crop P requirements, crop N requirements are typically not met (Nelson & Janke, 2007; Maltais-Landry et al., 2016). Options to top-up N requirements without additional P include high N specialty organic fertilizers (i.e. blood and feather meal) and legume cover crops through biological N fixation (Maltais-Landry et al., 2016; Ackroyd et al., 2019; Maltais-Landry & Crews, 2020). However, specialty fertilizers can be more expensive per unit N than composts and manures, especially in areas where concentrated livestock industries or urban populations generate these materials. Similarly, cover crops can be costly in terms of inputs, labour and field space   3 (Daryanto et al., 2019) and knowledge intensive to implement appropriately, including choosing species, seeding rates, and planting time, and then incorporating them into complex crop rotations (Moore et al., 2016). While organic farms may have overarching goals of environmental protection and an emphasis on ecological processes such as on-farm nutrient cycling, they face specific nutrient management challenges that hinder not only production, but also the farm’s ability to meet economic and environmental goals simultaneously. Additionally, both N and P are uniquely difficult to provide in their plant-available forms in the right rate and timing to match crop demands within organic farming systems. The research I present in this thesis aims to help address this challenge for organic vegetable production in British Columbia (BC), Canada.  In the following sections I introduce how these challenges relate to organic farming systems for the climates and soils found in BC. British Columbia ranks third in Canada in field vegetable production, behind Quebec and Ontario, with 5,518 hectares planted to field vegetables on 1,135 farms, generating a total farm gate value of $103 million in BC in 2019 (Agriculture and Agri-Food Canada, 2020). Important field vegetables grown here include potatoes, sweet corn, cole crops (i.e. broccoli, Brussel sprouts, cabbage), beans, cucurbits (i.e. pumpkins, squash, zuchinni), lettuce, carrots and peas (Government of British Columbia, n.d.-b).  Fruit and vegetable production lead the organic farming sector in BC. Of the 569 certified organic farms in BC in 2011, 70% reported growing fruit, vegetables, and/or greenhouse products, while only 25% reported producing hay, field crops or animals (Government of British Columbia, 2012). The farm gate value of certified organic vegetables in BC is currently   4 estimated at $9.6 million (Statistics Canada, 2020), ranking just second in organic vegetable sales in Canada, behind Quebec. Most field vegetable production in BC takes place in the lower Fraser Valley region, while the Vancouver Island – Coast region has the second largest production area. In 2016, 68% and 12% of the total hectares of field vegetable production in BC were in the lower Fraser Valley and in the Vancouver Island – Coast region, respectively; the remaining ~20% of production area is dispersed throughout other regions of the province (Government of British Columbia, 2017).  The soils, climate, and proximity to markets make the lower Fraser Valley one of the most intensive agricultural regions of the province. However, rising land prices, coupled with developing markets elsewhere in the province, will presumably push production outside of the Fraser Valley. Already, organic farms are more predominant in agricultural regions other than the lower Fraser Valley. While 29.8% of all BC farms are located in the lower Fraser Valley, just 26.4% of organic farms in BC are found here. The opposite trend is found in other regions of the province. The Vancouver Island – Coast, Thompson – Okanagan, and Kootenay regions have 15.9, 27.2, and 6.6%, respectively (total = 49.7%) of all farms, but 16.2, 40.4, and 9.8% (total = 66.4%) of BC’s organic farms are found in these regions, respectively (Government of British Columbia, 2017). While the lower Fraser Valley is an important agricultural region, less is known about these other, increasingly important production regions of BC. Therefore, the Vancouver Island and Pemberton Valley regions were included in this study to represent different combinations of soils, climates, and amendment characteristics that exist elsewhere in the province.   5 1.1.1 Nitrogen management There are two main pools of N in the soil: organic and inorganic. Over 95% of soil N is stored in the organic form and held in soil organic matter (SOM) (Berry et al., 2002). Organic farms tend to have greater SOM than their conventional counterparts, and therefore greater organic N, due to their emphasis on carbon (C) inputs over time to build long-term soil fertility (Gomiero et al., 2011). Organic N must undergo mineralization by soil microbes to be available for crop uptake. Specifically, the two predominant forms of inorganic and plant-available N (PAN) in the soil are ammonium (NH4+) and nitrate (NO3-) ions, which are taken up by crops from the soil solution. Therefore in these systems that emphasize and largely rely on organic amendments (composts, manures, specialty fertilizers) and cover crops, producers are also relying on the process of mineralization to fulfill crop N demands. Mineralization is a microbially-mediated process involving two steps: first, organic N undergoes ammonification, releasing NH4+, which can then be transformed to NO3- through the second step referred to as nitrification. The timing and rate of these steps is difficult to predict (varying with any parameter that impacts soil microbial activity, such as soil conditions like temperature and moisture, and substrate qualities like C to N ratio (C:N)), and is a main reason that N management in organic systems is more challenging than in conventional farming systems. Even though composts, manures, and cover crops are used as sources of N and C in both conventional and organic systems, conventional farming systems do not exclusively rely on these slow, unpredictable releases of PAN, and can directly apply inorganic N via synthetic fertilizers to closely match crop needs in both rate and timing. For example, on the nine organic farms that Berry et al. (2003) observed, even though PAN supply from just SOM was enough to   6 fulfill crop demands (i.e. in excess of 300 kg N ha-1 yr-1), crops could still be N limited due to issues with timing, rather than the total amount mineralized over the year. This mismatch between the timing of N mineralization and crop demand not only impacts crop yields but can also have environmental consequences (i.e. Maltais-Landry et al., 2019). While high levels of inorganic soil N during the growing season are necessary for crop growth, if large quantities of N continue to mineralize after the crop is harvested, this inorganic N can be lost to the surrounding environment. Although the cation, NH4+, can be protected from leaching through adsorption to clays, the anion, NO3-, is highly mobile and susceptible to leaching with percolating water. For example, high precipitation from November to March (Figure 2.2) in the lower Fraser Valley of BC means any NO3- in agricultural fields at the end of the season will be leached and unavailable to crops the following season (Zhang et al., 2019). A substantial portion of the NO3- leached from agricultural fields into the Abbotsford-Sumas Aquifer in the lower Fraser Valley has been traced to high rates of poultry litter application (Wassenaar, 1995). This highlights the possibility for undesirable environmental outcomes from the use of organic inputs. Accordingly, the BC Ministry of Environment is currently phasing in the Agricultural Environmental Management Code of Practice, making soil testing for post-harvest NO3- (PHN) mandatory for many growers (depending on the size, location, and type of the operation) (Government of British Columbia, n.d.-a). This will mean that PHN tests above 100 kg NO3--N ha-1 in the 0-30 cm depth may require producers to complete nutrient management plans and adjust nutrient management practices (Government of BC, n.d.-a). While conventional farms can top-up N applications to meet crop N demands with predictable and concentrated N fertilizers, organic farmers will need organic options that balance their need to meet productivity goals.   7 1.1.2 Phosphorus management Similar to N, a very small portion of total soil P occurs in plant-available forms (phosphate ions, H2PO4- and HPO42-) in the soil solution. In contrast to N, only a very small portion of the soil P is present in organic compounds as part of SOM, while most of soil P is found in secondary minerals through fixation reactions with ions such as calcium (Ca), magnesium (Mg), iron (Fe), aluminum (Al), in primary minerals, or tightly bound to clays in adsorption reactions. The total and relative sizes of these pools depend on soil mineralogy, pH, and quantity and type of organic and inorganic inputs of P and is therefore highly region and farming-system dependent. Given that most P fertilizers come from mined phosphate rock, a finite resource, the global trajectory is towards a widespread P shortage. Therefore, developing management practices for more efficient P use is necessary (Schneider et al., 2019). One way to accomplish this is through enhanced soil P cycling. This is important because only a small proportion of total soil P is in plant available forms, even in soils with excessively high total P. For example, phosphorus-solubilizing microbes (bacteria and fungi) are being studied to help cycle P from soil reserves to meet crop demands both soils low in P (Sharma et al., 2018), as well as in soils high in P. Such studies have potential to help farmers reduce (or even eliminate) starter fertilizers required to overcome low P availability in cold soils (Bittman & Kowalenko, 2004).  In North America, soil P status is region-specific and highly variable (Carr et al., 2019). In regions where additions of P to soil are not economically feasible, crop uptake of available P will inevitably lead to P deficiency (i.e. Berry et al., 2003). For example, in Eastern Canada, soil P is often lower on organic than conventional dairy farms because organic farms rely less on   8 expensive grain rations (which would bring P into the farm system) and more on forages produced on-farm (which simply cycle P in the system) (Martin et al., 2007). Similarly, organic grain farms in the Canadian Prairies experience yearly exports of P through grain harvest, without replacements of this important nutrient (Martin et al., 2007). In contrast, regions with a dense livestock industry (especially livestock reliant on grain imports for rations) often experience excessively high soil P due to application of various types of manures (i.e. poultry, swine) used as soil amendments with a high nutrient content (Ackroyd et al., 2019; Zhang et al., 2019; Carr et al., 2019).  In regions with intensive livestock production, applying composts or manures at N-based rates (i.e., to meet crop N demand) is common. However, due to a high P to N ratio in these amendments relative to crop requirements (Watson et al., 2002; Maltais-Landry et al., 2016; Nelson & Janke, 2007; Carr et al., 2019), this practice leads to soil P build-up. This is exacerbated by the low first-year availability of N from these sources (i.e., -10 to 54% of total N; Gale et al., 2006), in contrast to a greater P availability (i.e., 70 to 100% of total P; Nelson & Janke, 2007). Additionally, while some N that is applied in excess can be stored in SOM, much of the applied N can be lost from agricultural fields. In contrast, P is less mobile and therefore continually accumulates in soils unless removed through crop harvest or lost with soil erosion. An 11-year study on three farms in Ohio with high soil P content found no yield differences between plots receiving either no P fertilizer, or P fertilizer at 1x or 2x crop P removal (Wade et al., 2019). The lack of yield reduction in the plots receiving no P fertilizer (compared to 1x or 2x crop P removal) highlights that it can take years (or decades) to drawdown a ‘legacy’ high soil P content through small annual removals in crop harvests.   9 Accumulation of soil P beyond optimum for crop growth is not only an agronomically inefficient use of nutrients but is also an environmental concern. Loss of P through erosion to local water systems is one of the most substantial environmental impacts of agriculture, degrading water quality and causing eutrophication (Correll, 1998). The potential for agricultural fields to cause P pollution of local waterways depends on a combination of soil, crop, and landscape features. In terms of soil, managing soil P reserves is an important factor in preventing negative outcomes.  High soil P has been well documented (Sullivan & Poon, 2012; Zhang et al., 2019) in the lower Fraser Valley. In a 2012 survey of soil nutrient status by Sullivan & Poon (2012) in the Fraser Valley, 94% of fields surveyed (n=177) had (Kelowna-extractable) available P levels in the ‘High’ (>50 mg P kg-1) or ‘Very High’ (>100 mg P kg-1) class ratings. Alternatives to the coupled application of N and P in composts and manures need to be developed and assessed for their ability to provide appropriate levels of available N and P to crops, sustain soil health, and be economically feasible. This requires that N budgets can be achieved without building up soil P. One option is the use of leguminous cover crops which bring N into the system through biological N fixation. For example, yields were maintained when poultry litter applications were reduced to P-based rates by using cover crops for N on organic farms in Maryland with high soil P (Ackroyd et al., 2019). While cover crops can be used to enhance nutrient cycling by fixing, trapping, and releasing nutrients (Maltais-Landry & Frossard, 2015), their use is limited by various management challenges (Daryanto et al., 2019).  Common barriers include management complexity (lack of knowledge and technical assistance), increased costs (seeds, labor) (Daryanto et al., 2018), and residue management (Brennan, 2017). In some regions, a barrier to cover crop adoption is also related to high land prices (Silva & Moore, 2017), which can be even further exacerbated by a short growing season (Carr et al.,   10 2020). In lower Fraser Valley of BC, cover crop use is further challenged by additional pressures brought by wildlife (i.e. waterfowl) that may completely consume the aboveground biomass of cover crops (lower Fraser Valley farmer, personal communication, April 18, 2018). Another option to decouple N and P applications through organic amendments is through application of specialty, organic certified fertilizers such as feather meal, blood meal, alfalfa meal/pellets, and fish meal, which are all high in N but low in P. These specialty, organic certified fertilizers require substantially less prior knowledge about management, and labor than cover crops, but do not provide the same benefits to soil health that below- and aboveground biomass of cover crops do. In addition, specialty fertilizers could be relatively expensive, especially in regions with intensive livestock industries, where manures and composts can be cheaper and/or easier sources of N and P. Overall, this option to reduce compost rates by using high N organic fertilizers has yet to be adequately assessed within these important parameters (i.e. N, P, and C cycling, economics), and is therefore the focus of this study.  1.2 Carbon management and impacts to soil health Building and sustaining soil health is a key principle of organic farming. The soil health is viewed as its ability to “contribute to environmental quality and to promote plant, animal and human health” (Bünemann et al., 2018, p.106). It is assessed through comparison against baselines or reference values within minimum datasets (Gregorich et al., 1994), which can include both static and dynamic properties, but are largely focused on measures of the latter as a gauge of the sustainability of the management of a soil (Bünemann et al., 2018; Gregorich et al., 1994). Soil organic matter, and its primary constituent, soil organic carbon (SOC), are among the most widely used measures of soil health (Bünemann et al., 2018).    11 Different nutrient management strategies, based on varying combinations of composts, manures, cover crops, and/or specialty organic fertilizers, not only impact nutrient cycling and nutrient provision to plants but will also have implications for whether SOM is being built-up, maintained, or depleted due to differences in C inputs, and differences in residence time in the soil system between C compounds from these different sources. While the use of composts, manures, and cover crops often represent large C inputs (depending on application rate for amendments or biomass production for cover crops), the specialty organic fertilizers, in contrast, are small sources of C at typical rates of their application. For example, during an 8-year study on organic vegetable farms in California, White et al. (2020) measured cumulative carbon inputs from a pelleted poultry manure / feather meal product applied at 55-65 kg N ha-1 year-1, a food scrap / yard trimming compost applied at 14.25 Mg ha-1 year-1, and various annual winter cover crops. They found much lower cumulative C inputs from the specialty fertilizer (6.5 Mg C ha-1)  than from the food scrap / yard trimming compost (36.1 Mg C ha-1) or the cover crops (23.5 – 34.7 Mg C ha-1). Other studies have also illustrated that a combination of compost and feather meal at rates to not exceed crop P removal, result in much lower C applications than just compost/manure applications to meet crop N requirements (Eghball, 2002; Maltais-Landry et al. 2019a). Even though SOC and SOM are commonly used as indicators of soil health they are slow to change due to management practices; hence, cannot be used to make short-term management decisions (Gregorich et al., 1994). Given that it can take years for different soil management practices to result in detectable changes in SOM, it is important to measure these impacts with more sensitive indicators (Gregorich et al., 1994; Weil, 2003).    12  The labile carbon pool of soil organic matter is more responsive to management changes and more closely linked to soil processes. There are various methods to measure this C fraction, such as particulate organic matter, hot water-extractable carbon, water-soluble (dissolved organic) carbon, and POx-C (Bunemann et al., 2018). Additionally, a full soil health assessment or monitoring program would also ideally encompass measures of chemical, biological, and physical properties. Common chemical indicators (besides SOC) include pH, available P and K, total N, electrical conductivity and cation exchange capacity, while indicators such as water storage, bulk density, texture, structural stability, penetration resistance, porosity, aggregation, and infiltration are common soil physical properties used in soil health assessments (Bunemann et al., 2018). Finally, while the shift in language from soil quality to soil health reflects an understanding of the importance of soil biology, these indicators are the least common to include in assessments (Bunemann et al., 2018). Common biological indicators include soil respiration, microbial biomass, N mineralization and simple counts of earthworm (or other biota) populations.  In collaboration with three organic farms in Southwestern Ontario, Hargraves et al. (2019) assessed the sensitivity, repeatability, and interpretation of six soil health assessment indicators / tools, spanning chemical, biological, and physical soil properties. These included POx-C, SOM, wet aggregate stability, phospholipid fatty acids, Haney soil health test and the Haney nutrient test. The latter two tests consisted of measures of soil respiration, water-extractable organic C and N, SOM, and total and weak-acid extractable nutrients. Surprisingly, Hargraves et al. (2019) found that pH, wet aggregate stability, soil respiration and water-extractable organic C did not differentiate between (farmer-assessed) fields of high and low productivity. In contrast, POx-C, nitrate, and total P and K differentiated between the high   13 productivity, low productivity and references fields at each of the organic farm sites. Among all indicators evaluated in that study by Hargraves et al. (2019), POx-C was highly correlated with other biological, physical, and chemical indicators, and that it is consistently interpretable as an indicator (i.e. more is better). Other studies have also found Pox-C to be more sensitive than SOM to changes in management (Culman et al., 2012; Hurisso et al., 2016). The fraction of soil C measured by POx-C is estimated to constitute about 1 to 4% of SOC (Hurisso et al., 2016; Culman et al., 2012), and POx-C and SOC have been found to be closely related (r=0.93; Morrow et al., 2016). Initially, POx-C was proposed as a measure of “active C” (Weil et al., 2003), but it is now considered as a relatively processed fraction of labile soil C. Specifically, the pool of C measured by POx-C is more closely related to heavier than lighter particulate organic C (Culman et al., 2012). Hence, POx-C is a better indicator of organic matter stabilization than mineralizable C (CO2 evolution upon re-wetting method), which better indicates short-term nutrient availability and SOM turn-over (Hurriso et al., 2016). Therefore, short-term changes of POx-C are reflective of long-term changes of SOM (Weil et al., 2003). Hurriso et al. (2018) found that POx-C predicted crop productivity better than other soil C fractions (including SOC). Studies have also shown that POx-C can detect changes in management, such as tillage or inputs, after 2 to 4 years (Culman et al., 2012) or as short as one year when these changes were substantial (Wilson et al., 2018). When several soil health indicators were tested across three organic farms in Southwestern Ontario, POx-C was found to be the only indicator that was sensitive and repeatable, with consistent interpretation, and well-correlated with other biological, physical, and chemical indicators (Hargreaves et al., 2019). Finally, determination of POx-C involves a relatively easy and low-cost analytical method. A   14 field-kit (and its protocol) has been developed too. When 59 soil samples were analyzed with both laboratory Pox-C method and the field kit, very similar results (R2 = 0.98 between the two analyses) were obtained (Weil et al., 2003). Given the increasing interest by land managers in tracking soil health on their farms (Hargreaves et al., 2019; Idowu et al., 2008), meaningful field tests for soil health can provide opportunities for greater involvement in the process. 1.3 Research objectives and hypotheses In 2015 and 2016, a field experiment at the Centre for Sustainable Food Systems at the University of British Columbia (UBC) was established to assess the outcomes of several different organic nutrient management strategies on crop productivity, soil properties, and greenhouse gas (GHG) emissions (Maltais-Landry et al., 2019a). This research site is located in the lower Fraser Valley of BC. The study found both increased yields and residual soil N from manure applications with a high N demand crop. They also found both lower yields and lower residual soil N when compost was used for a low N demand crop. In contrast, the use of an organic high N fertilizer in combination with compost targeting P demand was found to reduce nitrous oxide emissions and P surpluses compared to manure, but with a yield reduction for crops with a low N:P ratio.  The study by Maltais-Landry et al. (2019a) demonstrated trade-offs of nutrient management strategies and highlighted variability by crop type, but since it has been done on just one study site it remained unclear how outcomes of these nutrient management strategies would differ under other climate and soil conditions found in other regions of BC or elsewhere.  Building on the findings of the study by Maltais-Landry et al. (2019a), additional studies are needed to enhance understanding of effects of other combinations of amendments, nutrient targets (i.e. N or P), and their trade-offs across a range of soil types and climatic conditions. This   15 would be of great relevance for farmers’ decision-making process to achieve favourable outcomes that would be more broadly applicable. The goal of my research is to evaluate the trade-offs of three organic nutrient management practices on working, organic mixed vegetable farms in three important agricultural regions of southwest BC – Fraser Valley, Pemberton Valley, and Vancouver Island. These regions differ by climate, soil properties, and access to and use of amendment types. The following three nutrient management strategies evaluated in this study represent common but contrasting nutrient management approaches: High compost (HC): compost applied at a rate to target crop N removal; Low compost plus organic N fertilizer (LC+N): compost applied at a rate to target crop P removal plus an organic fertilizer (feather meal) at a rate to meet crop N removal; and Typical (TYP): the nutrient application that the farmer would typically use for the specific crop (varying combinations of organic fertilizers, composts, and manures, or no amendments applied). The overarching objective of this study is to identify nutrient management strategies that are most likely to result in outcomes that have multiple benefits and minimize trade-offs among them across a range of conditions. The three specific objectives of my study were to: 1. Inventory the current use of organic nutrient management strategies, amendment characteristics, and soil properties across three diverse regions of south west BC; 2. Evaluate the effect of the three nutrient management strategies on farms across three important agricultural regions of southwest BC on crop yield, selected soil properties (POx-C, and post-season available N and P) as well as indicators of farm economics (determined as input costs).   16 3. Identify the edaphic, environmental, and input quality factors that best explain the variation in outcomes among nutrient management strategies and compare their potential trade-offs  Assumptions for objective 1 (regional characteristics) Lower Fraser Valley: This region has both a substantial urban population and livestock industry, so I predict that most farms in the lower Fraser Valley will be using a compost or manure product as part of their typical nutrient management strategy. Amendments (compost, manure, fertilizer) are likely the least expensive in this region due to greater availability than the other regions. This region is also known to have high soil P (Sullivan & Poon, 2012). Pemberton Valley: I anticipate that most farms in this region will use a commercially produced food-scrap / yard waste compost because this is available from a local compost facility. Based on past personal experience with farms in this region, I anticipate finding low soil P on farms in my study in this region. Vancouver Island: Farms in this region are likely the most diverse because they span the largest geographic area. Overall, inputs are likely the most expensive here. Fertilizers are shipped (across a ferry) from the mainland, and urban and livestock populations are substantially smaller than in the lower Fraser Valley, so fewer wastes are generated for use as compost feedstocks. HObj.2: Hypotheses for objective 2 (nutrient management strategy outcomes) The following hypotheses are organized by the three outcome areas used to evaluate trade-offs of the nutrient management strategies and compare the results across regions. H Obj.2 A. Crop yield: There will be no differences in crop yields among the three management treatments and crop yield outcomes will not depend on region because all treatments are expected to meet crop nutrient demands.   17 H Obj.2 B. Soil properties: The high compost strategy will result in higher POx-C than the low compost strategy and POx-C outcomes will not differ by region because in all regions the high compost strategy will provide greater C input. The high compost strategy will result in higher post-season available N and P than the low compost strategy. The high compost strategy applies P in excess of crop removal, so soil P build-up is expected. Soil P status is not expected to change from the low compost strategy as it theoretically applies P at the same rate as P is removed with crop harvest. The high compost strategy is expected to have higher residual PAN after the crop is harvested, due to the higher likelihood of delayed PAN release (from a greater total N input than the low compost strategy), and this will be true across all regions. H Obj.2 D. Farm Economics: Given that specialty fertilizers are processed more than composts and manures, I expect them to be more expensive nutrient sources. Therefore, across all regions and treatments, I expect the low compost + N strategy to be more expensive than the high compost strategy.  Results from this study will help farms in humid temperate climate regions to use compost and organic fertilizers more effectively for improved crop production, economic, and environmental outcomes. Contextualized broadly, this work will enhance our ability as researchers to understand the consequences of imbalanced nutrient flows at the landscape level.   18 Chapter 2: Methods Field trials were established in the spring of 2018 on mixed vegetable farms that rely on organic amendments in three regions of south-west BC: the lower Fraser Valley, Pemberton Valley, and Vancouver Island  (Figure 2.1). A total of 20 different farms participated over the two-year study period, with 19 farms in the first year and 18 in the second year. Sixteen out of the 20 farms were certified organic, while 4 were using organic nutrient management practices, but were not certified organic. Not every farm site has a complete data set for each year (i.e. all four indicators); Table A 1 in the appendix lists which farms/plots are included in which analyses Figure 2.1 Map of farm locations in three regions of southwest British Columbia used in field trials in 2018-19.   19 and the reasons for missing data (i.e. farms harvested before our sampling or crop failure in the research plots). Farms in the lower Fraser Valley region have soils that are characterized as poorly drained and fine textured, and developed on fluvial deposits that are classified as Rego Humic Gleysols, Humic Luvic Gleysols, and Orthic Humic Gleysols (Government of British Columbia, 2018), with the exception of one urban farm site that has soils constructed of imported sand and organic matter (compost). Soils in the Pemberton Valley are similarly characterized as typically poorly to imperfectly drained, fine textured, and developed on fluvial deposits (Rego Gleysols and Gleyed Regosols) (Government of British Columbia, 2018). Of the three regions, the Vancouver Island region spans the largest geographic area and therefore has the most diverse soil types among the three regions included in this study. Participating farms in this region generally have soils that are imperfectly to poorly drained and developed on marine deposits (subgroups of Brunisols and Gleysols), with soil textures ranging from clay loam to sandy loam (Government of British Columbia, 2018) but are generally more coarse-textured than the other two regions. Soil particle analysis was carried out on pre-season samples at each farm site (shown in Table C 1 in the appendix, and summarized by region in Table 3.2); further details for each farm site, including the names of each primary soil series and associated texture, drainage class, coarse fragment content, mode of deposition of parent material, and soil classification according to the online BC Soil Information Finder Tool (Government of British Columbia, 2018) are shown in Table C 2 in the appendix. Both the lower Fraser Valley and Vancouver Island have moderate maritime climate (wet and cool winters, dry and warm summers), whereas the Pemberton Valley has a continental    20 climate, with a broader range of annual temperatures (colder winters and hotter summers) (Figure 2.2).                        21                        Figure 2.2 Historical (1981-2010) and yearly data for total monthly precipitation and average temperature from the (A) Vancouver International airport, (B) Pemberton airport, (C) Nanaimo, and (D) Sidney weather stations (Government of Canada, 2019). mid-season July 10, 2018 July 26, 2019 post-season Oct. 16, 2018 Oct. 3, 2019 pre-season April 24 to June 1 harvest July 26 to Sept. 13 mid-season July 13, 2018 July 22, 2019 post-season Oct. 9, 2018 Oct. 14, 2019 pre-season April 24 – May 1 harvest Aug. 7 – Sept.  18 post-season Sept. 25 + Oct. 2, 2018 Oct. 11/12, 2019  harvest July 24 – Oct. 29 pre-season April 23 – May 24 mid-season July 11/12, 2018 July 23/24, 2019   22 At each of the farm sites, the following three nutrient management strategies were evaluated: • High Compost (HC): compost applied to meet crop N removal; • Low Compost + N (LC+N): compost applied to meet crop P removal plus a feather meal fertilizer to meet crop N removal; • Typical (TYP): These treatments received varying combinations of organic fertilizers, composts, and manures, or no amendments applied, which reflected the typical nutrient application the farmer uses for their farm.  The amendments used for these treatments were determined by each farmer for the plots on their farm, and I simply quantified the inputs for my study. Each treatment (summarized in Table 2.1) was established in one plot per farm site, so each farm site in each year had a total of three plots. Plot size depended on the size of the farm but averaged 29.3 m2 and ranged from 6.3 to 100.0 m2. Three of the participating farms required me to move research plots to entirely different fields for the second year. In total 11 plots received treatments for two years and 23 plots for only one. Crops grown in 2018 include beet (Beta vulgaris L. subsp. Vulgaris), broccoli (Brassica oleracea L. var. botrytis L.), carrot (Daucus carota L. subsp. sativus), cauliflower (Brassica oleracea L. var. botrytis L.), potato (Solanum tuberosum L.), and pickling cucumber (Cucumus sativus L.), and in 2019 include cabbage (Brassica oleracea L. var. capitata), carrot, beet, onion (Allium cepa L. var. cepa), and potato.     23 Table 2.1 Characteristics of the three nutrient management strategies evaluated in my study, including the crop nutrient targeted for compost or feather meal (N or P) applications, and hypothetical outcomes in terms of N or P. strategy Compost Fertilizer N P target Rate Target high compost (HC N High - meet exceed low compost + N (LC+N) P Low N meet meet Typical (TYP) chosen by the farmer; specific to each farm ? ?   2.1 Compost application rate calculations Amendments were applied at rates to target crop-specific N and P removal, where estimates of crop N and P removal in harvests were determined from target or expected yields chosen by each farmer for their crop and nutrient concentrations from local data when available. When nutrient concentrations were not available, crop-specific recommended nutrient application rates from best available sources were used as target nutrient application rates instead. Data sources and nutrient application rates are listed in Table B 1 in the appendix and are summarized by general crop groups in Table 2.2.  Table 2.2 Estimated nitrogen (N) and phosphorus (P) crop removal rates (kg ha-1) based on estimated yields, averaged across major crop categories. Values shown are number in each category (n), mean   standard deviation (SD), minimum (min.) and maximum (max) values.   N P Crop n mean  SD min. max. mean  SD min. max. potato 10 73  27 21 116 12  5 4 20 carrot 7 66  22 40 97 12  4 7 18 beet 7 128  50 51 215 18  7 8 30 brassicas* 4 138  55 59 181 35  27 11 73 *brassicas include broccoli, cabbage, cauliflower, and kohlrabi    24 Composts were unique to each farm and either currently being used by the farmer or were regionally available, and therefore varied widely in their composition due to varied feedstocks and sources. All composts properties are listed in Table D 1 in the appendix and summarized by type in Table 3.4. Composts were applied on various spring and summer dates to match when the farm would be planting; see Table 3.5 for the mean and median amendment application rates and dates, and associated C, N, and P rates for amendment categories for both years of the study. All composts and fertilizers were weighed and broadcast by hand. Amendments were mixed into the soil either by hand or by the farmer with tractor-mounted equipment. 2.2 Compost analyses Compost samples were taken directly from compost piles at each farm during initial farm visits on various dates in the spring of both years of the study; sample dates are listed in Table D 1 in the appendix. Composite samples were collected by taking five subsamples from different locations on the pile, of roughly 0.5 L volume, from 0.5 m into the pile and mixed thoroughly. From this composite, a subsample was taken and immediately placed on ice in a cooler, then shipped and kept on ice until analysis at an external laboratory (Ministry of Environment Analytical Laboratory [MOE], Victoria, BC, Canada). Compost samples were analyzed for NH4+ and NO3-, total C, N, P, and K, pH, EC, and moisture content. Within 72 hours of sample collection, NH4+ and NO3- were measured using a 2 M potassium chloride (KCl) extraction (Maynard et al., 2008). Briefly, 5 g of field-moist soils were weighed into a 50 mL centrifuge tube with 25 mL of KCl. After shaking for 30 minutes, samples were filtered through Whatman #2 filters, then analyzed colorimetrically using an A2 Analyzer (Astoria-Pacific, Clackamas, USA) (Doane & Horwath, 2003; Weatherburn, 1967). Total P and K of composts were determined by microwave-assisted acid digestion using an ultraWave   25 microwave (Milestone, Sorisole, Italy) (Karam, 2008), then element concentrations were determined by ICP-OES on a Prodigy Spectrometer (Teledyne Leeman Labs, Mason, OH, USA). Total C and N were measured by combustion on a Flash 2000 Elemental Analyzer (Thermo Fischer Scientific Inc., Waltham, MA, USA) (Thermo Fisher Scientific, 2010). Electrical conductivity was measured using a 1:4 soil water ratio with 5 g of compost shaken with deionized water in a 50 mL centrifuge tube for 1 minute then centrifuged for 10 minutes(Hendershot, 2008a). After transfer to a 16 mL centrifuge tube, conductivity of the clean supernatant was read on a conductivity meter and small volume flow-through cell. Varying compost to water ratios were used to measure pH, and the ratios used are listed in Table D 1 in the appendix. First, deionized water was added to 5 g of compost and stirred. After resting for 30 minutes, the suspension was stirred again, and pH was measured with a pH meter (Hendershot, 2008b). Compost bulk density was measured following guidelines by the Washington State University - Puyallup Organic Farming Systems and Nutrient Management program (Washington State University, n.d.). First, a scale was tared to the weight of an empty 5-gallon bucket, then the bucket was filled 1/3 full of compost taken from a hole dug in the compost pile (not from the dried-out outer layer). Next, the bucket was dropped ten times from roughly 0.3 m onto a hard surface. The bucket was then filled to 2/3 full of compost, dropped ten times again, filled to full, and dropped ten times again. Finally, the bucket was filled to full again and the weight was taken. The compost bulk density was calculated using the following equation:  Equation 1: 𝑪𝒐𝒎𝒑𝒐𝒔𝒕 𝑩𝑫 =  𝑾𝒕𝟏𝟖.𝟗𝟑 𝒙 𝟏𝟎𝟎𝟎 Where Compost BD is in kg compost m-3, wt is the measured weight of the compost in kg, 18.93 L is the volume of a 5-gallon bucket, and 1000 is used to convert units of volume from L to m3.   26 Compost bulk density was only measured in 2019, and values are provided in Table D 1 in the appendix. In the HC treatment, compost was applied at a rate where the estimated rate of crop removal N was matched with the estimated in-season PAN from the compost. In the LC+N treatment, both compost and feather meal were used: compost was applied at a rate where the estimated rate of crop removal P was matched with total P from the compost, and feather meal was applied at a rate where its PAN matched the difference between N applied with the compost and the estimated crop removal N. If cover crops were present in the research plots, samples were taken and dried for biomass. Given the challenge of coordinating sampling with farmers, and that only two farms in 2018 and three in 2019 had cover crops in the plots, cover crop N input was not included in the estimate of N within each treatment. When present, a minimum of three cover crop samples per farm were collected randomly from within the research plots using a 0.25 cm by 0.25 cm quadrat. The quadrat was placed on the ground and all above-ground biomass within the quadrat was clipped and placed in a paper bag. Samples were oven-dried at 65C and the biomass dry weight was recorded. Cover crops were observed to be uniform across the three plots and are reported as farm averages (Table 2.3).  Estimated compost PAN was calculated as 15% of the compost organic N plus the compost inorganic N (NH4+ and NO3-). A 15% mineralization rate was used based on the literature and a conservative approach to ensure adequate N availability from a variety of composts and manures (Gale et al., 2006). These calculations were made using the following equation:  Equation 2: 𝑷𝑨𝑵 = (𝑻𝒐𝒕𝒂𝒍 𝑵 − 𝑰𝒏𝒐𝒓𝒈𝒂𝒏𝒊𝒄 𝑵 ) ∗ 𝟎. 𝟏𝟓 + 𝑰𝒏𝒐𝒓𝒈𝒂𝒏𝒊𝒄 𝑵   27 Table 2.3 Cover crop biomass dry weight (dry wt.) from five sites. REG ID date collected cover crop type dry wt. (Mg ha-1) ------------------------------------------------ 2018 ------------------------------------------------ Vancouver Island 54 Apr 23 fall rye, some forbs 2.5 Vancouver Island 49 May 15 rye, pea, vetch mix 5.4 ------------------------------------------------ 2019 ------------------------------------------------ Fraser Valley 39 Mar 27 UBC farm mix* 1.0 Fraser Valley 38 Apr 24 UBC farm mix 6.6* Vancouver Island 51 Apr 17 UBC farm mix 3.0 * The UBC farm cover crop mix was shared with all farmers in Fall 2018 to plant. The mix consists of annual ryegrass (Lolium multiflorum), crimson clover (Trifolium incarnatum), and Austrian winter pea (Pisum sativum L.) seeded at 4, 0.44 and 2 lbs 1000 ft-2(196, 22, and 98 kg ha-1), respectively.  Two different feather meal products were used in the LC+N strategy based on the region. A feather meal with reported 11% N (11-0-0, Natures Intent, Pacific Calcium Inc., Tonasket, WA, USA) was used on all farms in the lower Fraser Valley and Pemberton Valley and a 13% N feather meal (13-0-0, Gaia Green, Grand Forks, BC) was used for all farms on Vancouver Island. For both products, calculations were based on ‘guaranteed’ total N concentration reported by the manufacturer (on the bag label), and 100% of this N was assumed to be PAN during the growing season, (i.e. 100% mineralization). No adjustment was made for moisture content for these products.  Amendments were weighed and applied by hand to the HC and LC+N plots using shovels, a 5-gallon pale, and a field scale. Amendment application rates in the TYP treatments were quantified in two ways. If amendments were spread by hand, I weighed them and applied them by hand for the farmer, using the methods as for the HC and LC+N treatments. If a tractor-  28 mounted compost/manure spreader was used, then I used a tarp and 1 m x 1 m quadrat to measure the application rate. Briefly, I first covered the two research plots (HC and LC+N) with a heavy-duty poly tarp held in place with ground staples. The farmer then drove over the tarp while spreading manure at their desired rate (applying amendment directly onto the TYP plot as well). I then collected five, 1 m2, subsamples from the tarp. At each of these subsamples I placed the quadrat on the tarp and collected and weighed all of the amendment from within the quadrat, then I averaged the weight of the subsamples to find one rate (kg m-2) to represent what was spread on the TYP plot. 2.3 Soil analyses Soil samples were taken three times at all farms in 2018 (pre-season, mid-season, and post-season) and two times for all farms in 2019 (mid-season and post-season), except for at the three new farm sites in 2019 that were not included in the first year of the study, where pre-season samples were also collected in 2019. Sampling dates and baseline soil properties from pre-season samples by farm site and depth are included in Table C 1 in the appendix; these soil properties are summarized by region in Table 3.1 and Table 3.2. Pre-season samples were collected at two depths (0-15 cm and 15-30 cm) prior to applying amendments. Mid-season soil samples were collected within crop rows at one depth (0-15 cm) across four sampling days in July each year and were analyzed for POx-C. Post-season soil samples were collected at both depths (0-15 cm and 15-30 cm) after the crops had been harvested and were analyzed for PAN and Kelowna-extractable available P. All post-season sampling dates were prior to the latest sampling date appropriate for a post-harvest nitrate test (PHNT) according to provincial guidelines (Government of British Columbia, 2019), which   29 account for soil texture and local precipitation to determine latest sampling dates for accurate PHNT measurements; sampling dates are summarized with weather data in Figure 2.2.  Depending on conditions, soil samples were collected using either a soil auger (5.5 cm inner diameter) or probe (1.9 cm inner diameter). Ten to fifteen subsamples were taken from each plot when using a probe, or five subsamples when using an auger to account for differences in sampling volume. Subsamples were collected randomly and composited by depth and plot and mixed well before being shipped to the Ministry of Environment (MOE) Laboratory or transported to UBC for analysis. All pre-season soil samples were analyzed at the MOE laboratory. Within 72 hours of sample collection, NH4+ and NO3- were measured on a subsample using a 2 M potassium chloride (KCl) extraction using the same methods as described previously for composts. The remaining sample was dried (35C), ground, and sieved to <2 mm particle size prior to all other analyses. Percent sand, silt, and clay were determined using the hydrometer sedimentation method with water maintained at 25C and particles were dispersed using Calgon detergent prior to analysis (Kroetsch & Wang, 2008). Total C and N were measured by combustion on a Flash 2000 Elemental Analyzer (Thermo Fischer Scientific Inc., Waltham, MA, USA) (Thermo Fisher Scientific, 2010) and inorganic carbon was measured on a Primacs SNC-100 TN/TC Analyzer (Skalar, Breda, the Netherlands) (Skalar Analytical, 2019). Available P and potassium were measured from 2.5 g of soil with 25mL of Mehlich-3 extractant (Ziadi & Sen Tran, 2008). After filtration the element concentrations were determined by ICP-OES on a Prodigy Spectrometer (Teledyne Leeman Labs, Mason, OH, USA). Electrical conductivity was measured using a 1:2 soil water ratio with 10g of soil shaken with deionized water in a 50 mL centrifuge tube for 1 minute then centrifuged for 10 minutes (Hendershot, 2008a). After transfer to a 16 mL centrifuge   30 tube, conductivity of the clean supernatant is read on a conductivity meter and small volume flow-through cell. For pH a 1:1 soil water ratio with 10 g of soil was used. The soil and deionized water were combined and stirred, then after sitting for 30 minutes, the suspension was stirred again, and pH was measured with a pH meter (Hendershot, 2008b). All mid- and post-season samples were analyzed at UBC. The mid-season soil samples were air-dried and sieved to 2 mm, then 2.5 g of soil was combined with 18.0 mL of distilled water and 2.0 mL of 0.2 M potassium permanganate (KMnO4) solution adjusted to pH 7.2 (Weil et al., 2003). Samples were mixed on a reciprocal shaker for two minutes then placed in a dark cupboard for 10 minutes. After incubating, 0.5 mL of the supernatant was transferred to a new centrifuge tube with 49.5 mL distilled water. This solution was then transferred and analyzed on a 96-well plate with blanks of deionized water and solution standards on a TECAN Spark® spectrophotometer at 550 nm (TECAN Group Ltd., Mannedorf, Switzerland). For one farm site with high SOM, 1 g of soil was used instead of 2.5 g (i.e. Miller et al., 2018), and calculations were adjusted accordingly. From the absorbance reading, POX-C was calculated using the following equation: Equation 3:𝒎𝒈 𝑷𝑶𝑿𝑪𝒌𝒈 𝒔𝒐𝒊𝒍= (𝟎. 𝟎𝟐𝒎𝒐𝒍𝑳− (𝒂 + 𝒃 𝒙 𝑨𝒃𝒔)) 𝒙 (𝟗𝟎𝟎𝟎𝒎𝒈 𝑪𝒎𝒐𝒍) 𝒙 (𝟎. 𝟎𝟐𝑳 𝒔𝒐𝒍𝒖𝒕𝒊𝒐𝒏𝒔𝒐𝒊𝒍 𝒘𝒕) Where 0.02 mol L-1 is the initial KMnO4 solution concentration, a and b are the intercept and slope, respectively, of the standard curve, Abs is the unknown sample absorbance measured on the spectrophotometer, 9000 mg C mol-1 is the milligrams of C oxidized by one mole of MnO4, 0.02 L is the volume of stock solution reacted, and wt is the weight of air-dried soil sample in kg. For all samples, wt = 0.0025 kg, except for the one farm site with high SOM, where wt = 0.0001 kg.    31 Post-season samples were transported to UBC and NH4+ and NO3- concentrations were measured on both sample depths (0-15 cm and 15-30 cm) by extracting 2.5 g of field-moist soils with 25 mL of 2 M potassium chloride (KCl) for 20 min on a reciprocal shaker (Maynard et al., 2008). Extracts were centrifuged for 5 min (5000 rpm) then filtered (Fisherbrand Q2 filters). Finally, NH4+ and NO3- concentrations were measured colorimetrically (following Weatherburn, 1967 and Doane & Horwath, 2003, respectively) using a spectrophotometer (Bio-Rad iMark, Hercules, CA, USA). After air-drying for two weeks, samples from only the surface depth (0-15 cm) were analyzed for Kelowna-extractable P (van Lierop, 1988) by extracting 2.5 g of air-dried soils with 25 mL of a solution of 0.015 M ammonium fluoride (NH4F) and 0.25 M acetic acid (CH3COOH). Extracts were centrifuged for 5 min (5000 rpm) then filtered (Fisherbrand Q2 filters). Finally, P concentrations were determined on a Varian 725-ES ICP-OES (Agilent Technologies, Mulgrave, Victoria, Australia). To determine soil water content, the weight of a field-moist soil sample was recorded before and after oven-drying for 48 hours at 105C.  2.3.1 Soil bulk density Soil bulk density was measured at the time of crop harvest for each farm site in 2019 in the 0-15 cm and 15-30 cm depths. Two undisturbed soil cores were taken from the center of each depth for a total of four samples per farm site. Samples were collected using cores with a 7.5 cm diameter and 7.5 cm depth using a double-cylinder, drop-hammer sampler and the samples were dried for 24 hours at 105C. Soils were sieved to 2 mm and the weight of both the fine earth and coarse fragment fractions were recorded. Because several farm sites had substantial coarse fragment content (i.e. > 40% coarse fragment content by weight), bulk density values have been calculated using the following three methods by Throop et al. (2012): (1) with the mass of all material in the total core volume, (2) with the mass of the fine earth fraction only in total core   32 volume, and (3) the mass of the fine earth fraction only in the volume of the fine earth fraction only (estimated by assuming a particle density of 2.65 g cm-3 to subtract the volume of coarse fragments from the total core volume). The first method is the most simple and functions well for soils with minimal coarse fragments; the second “hybrid” method is preferable for transforming soil nutrient concentrations to stocks; the third method most accurately describes the degree of soil structure and porosity within the fine earth fraction of each of the sites and is also most appropriate for soils with large amounts of coarse fragments. Bulk density values and percent coarse fragment content, averaged for each depth by farm site, are provided in Table C 3 in the appendix. 2.4 Crop yield sampling Depending on plot size, between two to ten crop biomass subsamples were taken from each plot. Subsamples were averaged and recorded as the weight of crop biomass per one bed metre (kg m-1). The total number of bed metre samples per plot was equal to or greater than 30% of the total in a given plot, minus the bed metres designated as buffers on the plot perimeter; harvest dates, plot sizes, and number of subsamples taken per plot are shown in Table B 1 in the appendix. Plot buffer widths varied between 0.5 and 2.5 m, depending on how plots were managed (i.e. small buffers were used for smaller farms with permanent raised bed systems, whereas large buffers were used for plots in fields that were managed with large tillage equipment, which result in substantial soil mixing between plots). Example sampling schemes for large and small farms are in Figure E 1 and Figure E 2 in the appendix. After determining the required buffer area and the total number of bed metres to sample    33 for a given plot (based on ~30% of the remaining area), a stratified sampling method was used to choose subsample locations in each plot. Subsamples were taken by placing a 1 m x 1 m quadrat on top of a crop bed or row at a pre-designated location determined by the stratified sampling plan, then all saleable crop biomass (i.e. potato tubers but not tops) between the two ends of the quadrat were harvested, weighed, and recorded. 2.5 Economics: estimation of input costs An economic indicator for each nutrient management strategy was calculated as the total cost of inputs for a given strategy for each farm. Because plot sizes vary between farms, final input costs used for statistical analysis are expressed on a dollars per hectare basis ($ ha-1), as a function of the input costs and their rate of application, extrapolated to a one hectare area. All costs are in Canadian dollars (CAD), unless noted otherwise. Specific input costs were collected from each farm and include both the amount paid for the amendment as well as any associated shipping or transportation costs. Most farms have amendments delivered to their farm and provided shipping costs; however, for farms that pick-up amendments locally, costs associated with the farmer’s time and vehicle mileage were valued at $20 hr-1 and $0.59 km-1, respectively, and were applied to an estimate of round-trip time and mileage provided by the farmer. Any inputs that did not include the two nutrients being studied, N or P (i.e. lime, micronutrients, etc.), were not included in the total cost because they were used at the same rate in all three treatments on any farm that applied them.  The input cost of each strategy was calculated using the compost costs (purchase and transport) specific to each farm as reported by the farmer and/or the fertilizer cost (either the project costs of purchasing the feather meal used in the LC+N treatment or the fertilizers specific to each farm as reported by the farmer), plus a farm-specific shipping cost for each farm. In the   34 case where the farm did not have shipping costs, the cost was estimated based on nearby farms or estimated mileage costs to the nearest available retailer.  2.6 Statistical analyses To complete objective 2 (assess outcomes of nutrient management strategies), the effect of nutrient management strategy on crop yield, POx-C, post-season available N (by ion and depth separately), available P, and input costs was analyzed using a linear mixed effects (LME) model. All analyses were performed in R (R Core Team, 2019). Analysis was run with the lme function in the nlme package version 3.1-143 (Pinheiro, 2019) using the maximum likelihood (ML) method for model comparisons and the restricted maximum likelihood (REML) method for reporting final model output. As the primary explanatory variable of interest, nutrient management strategy was included as a categorical fixed effect with three levels (HC, LC+N, and TYP), and year (2018 and 2019), region (lower Fraser Valley, Pemberton Valley, and Vancouver Island), and all interactions, were included as fixed effects to investigate if the impact of nutrient strategy on the dependent variables was different between years or regions (i.e. to consider interactions). Each year within one farm site was considered a block of treatments and was included as a nested random effect in the model to account for differences in the mean response of the measured outcomes (i.e. yield, POx-C, etc.) for each block (Crawley, 2013; Krzywinski & Altman, 2014a; Webster & Clark, 2018), and to account for autocorrelation of repeated measures where the same plots were sampled from in both years (Krzywinski et al. 2014). The LME model with the three main effects and all (two-way and three-way) interactions was considered the full / maximal model and was always used first to check for interactions between main effects. Then, model simplification was performed using Aikaike’s information   35 criterion (AIC) as described by Crawley (2013) and using conditional and marginal R2 values as described by Nakagawa and Schielzeth (2013) such that all output is reported from the minimally adequate model, based on principle of parsimony. In mixed models, the conditional R2 is the variance explained by the fixed effects, while the marginal R2 is the variance explained by whole model (both the fixed and random effects) (Nakagawa & Schielzeth, 2013). When there were significant interactions between fixed effects, the model was run separately for each fixed effect. When main effects were found significant in the LME model ANOVA, a post-hoc (Tukey method) test (Lenth, 2019) was used to determine significant differences between factor levels using the emmeans function (Lenth, 2019). Differences were determined to be significant for P-values <0.05, and marginally significant for P-values <0.10. ANOVA F and p-values are shown in Appendix F  . Assumptions of normality and homogeneity of variance were tested using the Shapiro-Wilk test and Bartlett test, respectively. If assumptions were not met, data was transformed using a square, square root, cube root, or log10 transformation. A principal component analysis (PCA) was performed using the FactoMineR package (Husson et al., 2020) to help explain variation in the measured outcomes and illustrate trade-offs by plotting them with pre-season, baseline compost and soil (0-15 cm) properties. All outcomes (i.e. yield, POx-C, etc.) were plotted as data relativized within the grouping of three treatments at each farm site. This was accomplished by dividing the observation from a given nutrient strategy within one farm site by the average of that farm site for that outcome. This allows the plotted outcomes in the PCA to vary by nutrient strategy, rather than the farm average. Compost or soil variables that are correlated (i.e. soil total N and total organic C) were not plotted together.   36 Data were plotted in boxplots which indicate the lower quartile (Q1) and upper quartile (Q3), such that the box covers the interquartile range (IQR = Q3 – Q1), representing the central 50% of the data (Krzywinski & Altman, 2014b). The Tukey-style whiskers extend to a maximum of 1.5 x IQR outside of the box, and open circles are data points outside of these bounds. Radar charts were generated using the fmsb package (Nakazawa, 2019) to illustrate trade-offs in outcomes among the nutrient management strategies. The five outcomes (yield, input costs, POx-C, available N, available P) were plotted on separate axes, and axis limits were set to the highest value among the three nutrient strategies.      37 Chapter 3: Results & Discussion 3.1 Inventory of regional characteristics 3.1.1 Soil properties Soil properties varied widely not only across the regions but within the regions. As expected, participating farms in the lower Fraser Valley were exceptionally high in soil P. Farms on Vancouver Island had the lowest average soil P among the three regions, while soil P in the Pemberton Valley was variable, but generally lower than the lower Fraser Valley. The only exception was one farm in the Pemberton Valley that has exceptionally high soil P (Table 3.1). As discussed previously, high soil P in the lower Fraser Valley is tightly linked to the intensive livestock industry in this region, which generates cheap P inputs (as manures) used by local farms. In contrast, the other two regions do not have these large animal populations and have noticeably lower soil P. Study by Kowalenko (2010) analyzed soils from 54 and 56 farms in the lower Fraser Valley and the Okanagan-Similkameen regions of BC, respectively, for Mehlich-3-extractable available P.  Farms in the lower Fraser Valley included in my study had higher available P (mean = 180 mg kg-1; Table 3.1) than reported in the study mentioned above (mean = 120 mg P kg-1). In contrast, the average available soil P found by Kowalenko (2010) on farms in the Okanagan-Similkameen region (mean = 70 mg P kg-1) was similar to soil P on farms in the Pemberton Valley region in my study (mean = 90 mg P kg-1). Average soil P on farms in the Vancouver Island region (mean = 49 mg P kg-1) in my study were lower than what I found in the other two regions, as well as those observed in the two regions studied by Kowalenko (2010).  I found low SOC on participating farms in the Delta area of the lower Fraser Valley region. Farms on Vancouver Island were consistently high in SOC, with the highest average   38 SOC among the three regions. Farms in the Pemberton region had variable, but generally low SOC; the farm with the lowest SOC was found in this region. The SOC on farms in my study (mean = 4.2%, min = 0.9%, max = 10.0%) (Table 3.1) is similar to those found in organic mixed vegetable fields in southwestern Ontario by Hargreaves et al. (2019), where SOM varied from 3.32 to 7.72%. Overall, farms in the lower Fraser Valley tended to have greater clay content while farms on Vancouver Island tended to have greater sand content (Table 3.2). Differences in soil texture can be a major determinant for the outcomes of agricultural management practices.  Soil texture for example can strongly influence the extent of field N losses through leaching. The risk of post-harvest nitrate leaching is generally greater on coarse-textured soils than from fine-textured soils. Coarse-textured soils, with a greater proportion of macropores compared to fine-textured soils, tend to have quicker drainage and decreased potential for denitrification (Di & Cameron, 2002).     39 Table 3.1 Chemical soil properties determined in the pre-season, averaged within regions, lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI), including total nitrogen (N), carbon (C), and inorganic carbon (IC), Mehlich-3 phosphorus (P), Mehlich-3 potassium (K), electrical conductivity (EC) and pH. Values are means  standard deviation and ranges with minimum – maximum.      N C IC P K EC pH REG Depth (cm) ------------------------------------ % ------------------------------------ -------------------- mg kg-1 --------------------   FV 0-15 0.30.3 0.0-0.7 3.93.1 1.3-10.0 0.10.1 0.0-0.2 18091 78-320 28567 217-421 0.30.1 0.1-0.5 6.40.5 5.4-7.1 15-30 0.20.2 0.0-0.5 3.22.5 1.1-7.7 0.10.1 0.0-0.2 14866 87-257 22560 159-361 0.20.1 0.1-0.5 6.30.6 5.2-6.9 PV 0-15 0.20.1 0.1-0.4 2.61.5 0.9-5.1 0.10.1 0.0-0.1 9098 11-277 16386 85-327 0.20.1 0.1-0.4 5.90.7 5.3-7.2 15-30 0.20.1 0.1-0.3 1.71.0 0.9-3.5 0.00.1 0.0-0.1 2619 7-61 9650 60-193 0.10.0 0.1-0.2 5.60.8 4.9-6.9 VI 0-15 0.30.1 0.2-0.4 4.50.6 3.4-5.7 0.00.0 0.0-0.1 4939 5-130 121104 26-338 0.20.1 0.1-0.3 5.90.8 4.7-6.9 15-30 0.20.1 0.1-0.3 2.91.2 1.5-4.5 0.00.0 0.0-0.1 2215 4-43 5844 20-123 0.10.0 0.1-0.1 5.60.6 4.6-6.4   Table 3.2 Soil particle size distribution determined on the pre-season samples, and coarse fragment content (% volume) determined on the bulk density samples, at the 0-15 cm and 15-30 cm depths, averaged within regions, lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI). Values are means  standard deviation and ranges with minimum – maximum.    Sand Silt Clay Coarse Fragments REG depth % FV 0-15 20  22 4-72 5717 19-69 24  7 9-31 1.91.9 0.3-4.6 15-30 20  22 4-72 5717 19-71 24  6 9-31 2.44.2 0.1-8.7 PV 0-15 18  9 6-29 694 64-76 13  8 7-27 0.20.1 0.1-0.3 15-30 18  9 5-31 694 63-76 13  9 6-30 0.10.1 0-0.2 VI 0-15 45  16 23-68 3913 21-56 16  6 8-26 6.56.9 0.42-20.6 15-30 45  16 24-69 3812 21-56 16  5 9-24 7.69.1 0.1-26.5      40 3.1.2 Compost characteristics Compost properties were likely influenced by feedstock availability within each region. They have been generally categorized within 8 types and are described in Table 3.3. I sampled 40 different composts during the study and the chemical properties of the 29 composts that were used are summarized in Table 3.4.  All composts/manures that I sampled in the lower Fraser Valley region were based on livestock manure (poultry and pig) and/or urban green bin products. This was not surprising given that a recent nutrient survey of crops and manures in southwest BC estimated that the lower Fraser Valley accounts for 88% and 65% of provincial poultry and swine manure production, respectively (Smukler et al., 2015).  Most farms in the Pemberton Valley use a commercial food scrap / yard trimming compost product from a local facility, although one farm makes their own compost (primarily for their hop production) and one farm was importing poultry manure from another region. In contrast, farms on Vancouver Island had the greatest diversity of compost types: this region had the greatest prevalence of ‘on farm’ composts and all ‘steer / horse’ composts that I sampled were in this region; no poultry manure was used on Vancouver Island. Given these regional differences in feedstocks, it was not surprising that compost total N % trended from high to low in the order of the lower Fraser Valley > Pemberton Valley > Vancouver Island.  I observed a similar trend in compost total P %, although surprisingly the Pemberton Valley and Vancouver were not much different (Table 3.4). The high P content in the lower Fraser Valley is directly linked to the pig and poultry manure composts used there. Smukler et al. (2015) also found high P contents of poultry manures in south-west BC (mean total P = 1.12%),   41 but reported lower total P for swine manures (mean total P = 0.18%) than I found for the two pig manure samples in my project (mean total P = 1.0%). In contrast, composts based on food scraps, yard wastes, and horse or steer manure, found in the Pemberton Valley and Vancouver Island had lower total P. Because of this, I observed a higher compost N:P in the Pemberton Valley, whereas the other two regions were similar, following the trend in N:P ratios in the order of food scrap > horse/steer > poultry similarly documented in a study in Germany by Moller (2018). Finally, I expected to find composts with the lowest C:N in the lower Fraser Valley. This is because composts and manures generally have a stable C content (around 40%), so their C:N inversely reflects total N (Sullivan et al., 2019); given the availability of manures with high N content for composting I expected to find the highest N (and therefore lowest C:N) composts and manures in the lower Fraser Valley. Instead, I found that compost C:N ranged from high to low in the following order: Vancouver Island > Fraser Valley > Pemberton Valley. Notably, mean compost C:N ratio in the lower Fraser Valley was increased by a pig manure-based product made with wood shavings, reducing the total N but increasing the total C content of this product. Yet, the total N (mean=1.5%) of the two pig manures that I sampled was higher than the 8 samples (mean total N = 0.86%) sampled in a manure and crop nutrient review by Smukler et al. (2015) in south-west BC. Although this previous review also reported high variability in the nutrient content of the 8 hog manure samples they analyzed. In contrast, the poultry manure and urban composts found in the lower Fraser Valley have lower C:N, similar to those in the Pemberton Valley (C:N~10). This C:N is in the lower range of what other researchers have found for food scrap compost C:N in Oregon, USA (mean C:N = 16.5; Gale et al., 2006) and in California, USA (mean C:N = 17.6; Lazicki et al., 2020).   42 Despite the lower overall C:N in the Pemberton Valley versus the lower Fraser Valley, composts in the Pemberton Valley still had lower estimated PAN. This indicates that these food scrap compost product found in the Pemberton Valley region were more stabilized, with lower initial mineral N, in contrast to poultry manure composts which had the highest fraction of total N in inorganic forms (NH4+ and NO3-). Lazicki et al. (2020) found similar differences, reporting 0.83% of total N as mineral N for plant-based composts, compared to 16.6% for manure-based composts.   Table 3.3 Descriptions of compost types used in the field trials. Composts that were produced by facilities whose main purpose is to generate and sell compost are indicated by ‘yes’ in the Commercial column; all other composts were sold by the farm or business generating the materials or were made on the farms that we worked with.  compost type Commercial description of primary feedstocks ‘fish / forest’ yes  fishing and forest industry wastes ‘offal’ no  slaughterhouse byproducts ‘on farm’ no  highly variably  mainly carbon sources (i.e. wood shavings, hay, straw, etc.), crop or weed residues, and some animal manure (i.e. sheep, beef, horse, etc.) ‘pig’ no  pig manure and wood shavings ‘poultry’ no  poultry manure and a carbon source (i.e. wood shavings, hay, straw, etc.).   all types of poultry. ‘steer / horse’ no  beef or horse manure with hay or straw. ‘urban’ yes  urban food scraps and yard wastes. ‘urban / manure’ yes  urban food scraps, yard wastes, and livestock manure.     43 Table 3.4 Compost properties averaged by type within region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)), including total carbon (C), total nitrogen (N), the proportion of total N estimated to be plant available N (PAN), total phosphorus (P), carbon to nitrogen ratio (C:N), and nitrogen to phosphorus ratio (N:P). Values shown are number of samples per group (n), means  standard deviations, minimum values (min), and maximum values (max).     C N PAN P C:N N:P    ------------------------------------------------------------- % -------------------------------------------------------------   REG type n mean SE min max Mean SE min max mean SE min max mean SE min max mean SE min max mean SE min max FV poultry 4 37.5  4.9 24.0 45.9 3.4  0.5 2.2 4.5 41.7  7.0 29.8 60.9 1.7 0.2 1.2 2.2 11.3  0.7 9.6 12.8 2.0  0.3 1.3 2.5 urban / manure 2 29.4  6.4 23.0 35.7 2.7  0.5 2.2 3.3 20.2  4.4 15.9 24.6 0.6 0.1 0.5 0.7 10.7  0.2 10.5 10.9 4.4  0.1 4.3 4.5 pig 2 36.8  7.3 29.5 44.0 1.5  0.2 1.2 1.7 25.7  5.4 20.3 31.1 1.0 0.3 0.7 1.3 25.1  0.8 24.2 25.9 1.5  0.2 1.3 1.7 region mean 8 35.3  3.2 23.0 45.9 2.7  0.4 1.2 4.5 32.3  5.0 15.9 60.9 1.3 0.2 0.5 2.2 14.6  2.3 9.6 25.9 2.5  0.4 1.3 4.5 PV urban 6 24.7  1.1 20.5 28.0 2.1  0.1 1.7 2.6 20.7  1.4 17.1 26.0 0.6 0.1 0.4 1.2 11.8  0.7 8.8 14.1 4.0  0.5 2.2 5.2 on farm 1 8.3 - - - 0.6 - - - 16.4 - - - 0.2 - - - 13.3 - - - 3.7 - - - poultry 1 41.2 - - - 4.4 - - - 44.5 - - - 2.3 - - - 9.3 - - - 1.9 - - - region mean 8 24.7  3.2 8.3 41.2 2.2  0.4 0.6 4.4 23.1  3.3 16.4 44.5 0.7 0.2 0.2 2.3 11.7  0.6 8.8 14.1 3.7  0.4 1.9 5.2 VI steer / horse 6 29.9  4.4 14.6 43.0 1.2  0.2 0.9 2.2 18.7  1.8 15.5 27.3 0.4 0.1 0.3 0.6 26.4  5.6 15.5 47.8 3.0  0.3 2.2 3.8 fish / forest 3 37.6  1.0 35.7 39.0 1.9  0.9 0.9 3.7 19.8  3.6 15.6 27.0 0.6 0.4 0.2 1.4 28.2  9.4 10.5 42.7 3.5  0.5 2.6 4.3 on farm 3 16.0  6.1 6.1 27.0 1.4  0.6 0.5 2.6 21.4  3.8 15.0 28.1 0.7 0.4 0.1 1.5 11.8  0.7 10.4 12.8 2.8  0.7 1.7 4.1 offal 1 37.0 - - - 1.6 - - - 16.4 - - - 1.2 - - - 23.1 - - - 1.3 - - - region mean 13 29.0  3.2 6.1 43.0 1.5  0.2 0.5 3.7 19.4  1.4 15.0 28.1 0.6 0.1 0.1 1.5 23.2  3.6 10.4 47.8 2.9  0.3 1.3 4.3   44 3.1.3 Carbon and nutrient application  As expected, differences in the amount of N and P applied to the TYP treatments among farms followed trends in the ease of access and cost of these nutrients among regions. Rates of C, N, PAN, and P applied to the research treatments are summarized by region in Table 3.5.   On average, I observed greater rates of total N applied to TYP treatments in the lower Fraser Valley and the Pemberton Valley than on Vancouver Island.  The lower total N observed in applications on Vancouver Island is likely due to the limited access of composting facilities to manure. However, PAN applications to TYP treatments were lower in the Pemberton Valley than in the lower Fraser Valley and on Vancouver Island, which were similar. The higher PAN values on Vancouver Island illustrate the farmers’ reliance on specialty fertilizers as inputs, compared to the other two regions. Lacking high N compost options their fertilization strategy uses specialty fertilizers with readily available N.    Surprisingly, there was only a slight trend  in the average amount of C applied to the TYP treatments among farms in my study; I observed decreasing C applications to the TYP treatments among regions in the order of lower Fraser Valley > Pemberton Valley > Vancouver Island, which mirrors compost availability among regions. This is likely because farms are more likely to use C inputs when they can also be credited as an N source. Specifically, this trend in C inputs among TYP treatments reflects the use of high PAN composts in the lower Fraser Valley, in contrast to the use of high N (but low C) fertilizers on Vancouver Island (used to top-up crop PAN requirements, given their low-PAN composts). In contrast, farmers in my study in the Pemberton Valley were less likely to use purchased fertilizers in addition to composts, while farmers in the lower Fraser Valley rarely used fertilizers (discussed in the next section).    45  Finally, TYP treatments in the lower Fraser Valley received the greatest rate of P application, followed by the Pemberton Valley and Vancouver Island. This is similar to regional trends in P applications in connection to livestock industries found in other studies. As previously discussed, farms tend to apply more P in regions where livestock industries provide this nutrient in relatively inexpensive manure and compost products derived from manure (Carr et al., 2019; Martin et al., 2007), compared to more expensive inputs like rock phosphate or specialty organic fertilizers.             46  Table 3.5 Total carbon (C), total nitrogen (N), plant available N (PAN) and total phosphorus (P) input rates averaged by region for each nutrient management strategy for all farms and both years of the study.    C N PAN P   ---------- Mg ha-1 ---------- --------------------------------------------------- kg ha-1 -------------------------------------------------- REG strategy mean SD min. max. mean SD min. max. mean SD min. max. mean SD min. max. Fraser Valley (n=10)                  HC 5.8 3.3 2.9 13.8 445 319 149 1268 126 66 46 263 165 53 92 283 LC+N 1.0 1.1 0.3 3.8 174 107 53 403 113 45 46 195 25 20 8 74 TYP 4.6 5.6 0 15.9 455 602 0 1608 152 179 0 530 140 140 0 369 Pemberton Valley (n=9)                  HC 6.4 3.8 2.7 14.6 542 287 301 1102 117 58 53 215 169 93 61 300 LC+N 0.7 0.4 0.3 1.3 165 80 77 291 118 64 52 215 18 8 8 29 TYP 4.2 6.1 0 14.6 381 512 0 1358 101 104 0 317 107 139 0 357 Vancouver Island (n=15)                  HC 10.1 9.3 1.3 36.8 435 246 87 1005 80 35 25 151 152 85 30 349 LC+N 1.0 0.9 0.1 2.8 121 49 26 220 88 35 22 151 13 5 3 20 TYP 3.6 6.2 0 24.6 262 264 0 1116 153 167 0 689 87 68 0 220 Average for all regions                 HC 7.4 5.5 2.3 21.7 474 284 179 1125 108 53 41 210 162 77 61 311 LC+N 0.9 0.8 0.2 2.6 153 79 52 305 107 48 40 187 19 11 6 41 TYP 4.2 6.0 0.0 18.4 372 459 8 1361 139 150 8 512 113 115 6 315   47 3.1.4 Amendment costs by region Use and cost of fertilizers and composts varied substantially among regions. Only one fertilizer type was used by farms in the lower Fraser Valley, while in the Pemberton Valley four were used, as part of the TYP nutrient management strategy in this study. In contrast, 24 fertilizers were used on Vancouver Island as part of the TYP nutrient management strategy (Figure 3.1). While the mean cost of fertilizers used in the TYP treatments in the Pemberton Valley and Vancouver Island were similar, the prices were more varied on Vancouver Island than in the Pemberton Valley (Figure 3.1). Figure 3.1 also shows the variability in fertilizer transportation costs among farms on Vancouver Island and in the Pemberton Valley, where differences in the LC+N treatment fertilizer were only caused by variable shipping costs. The variability in the fertilizer costs in the LC+N treatments in the Pemberton Valley and on Vancouver Island demonstrates the greater influence of shipping on fertilizer costs in these two regions. In contrast, my calculated fertilizer transportation costs in the lower Fraser Valley were very small compared to the cost of the fertilizer, so fertilizer costs in the LC+N treatments in this region did not vary among farms (Figure 3.1).  Similarly, I found relatively inexpensive compost products in the lower Fraser Valley (mean = $ 12 m-3) compared to the other two regions, and again comparable mean composts costs in the Pemberton Valley and on Vancouver Island (mean = $ 41 and $ 39 m-3, respectively), but with more variability in costs on Vancouver Island (Figure 3.2). I found that farmers on Vancouver Island reported lower costs for horse and beef manure purchased from neighboring farms, but much higher costs for commercially processed composts based on fish and forest industry wastes [data not shown].    48                         n=8 n=1 n=7 n=4 n=14 n=24 n=9 n=7 n=11 Figure 3.1 Boxplot of fertilizer costs (cost of product + cost of transport) ($ kg-1) used in the LC+N and TYP plots by region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)). The center line indicates the median, means are shown as black dots and sample size indicated by n. Figure 3.2 Boxplot of compost costs ($ cubic yard-1) by region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)). The center line indicates the median, means are shown as black dots and sample size indicated by n.    49 In comparison, Lynch et al. (2008) reported much lower prices for a commercial hog manure compost and pelletized poultry manure product from their study on organic potato production, N uptake and soil mineral N in Nova Scotia, Canada. While this can be partially explained by price inflation over time, this is unlikely the sole reason for the much higher costs I found. Specifically, while their estimate of compost cost (~$15 yard-1, assuming bulk density of 500 kg yard-1) is in the range I found in the lower Fraser Valley (Figure 3.2), where urban and agricultural compost feedstocks are common, I found more than double this cost for composts in the Pemberton Valley and Vancouver Island, as well as in my study overall (mean = $40 yard-1). Similarly, the fertilizer cost ($0.10 kg-1 fertilizer) reported by Lynch et al. (2008) was much lower than what I found for fertilizers used in the TYP treatments (overall mean = $2.30 kg-1 fertilizer). This is likely partly due to a lower average nutrient content of the fertilizer used by Lynch et al. (2008) (mean % N-P-K = 4.2 - 1.4 - 1.8), compared to the average across of all N and P fertilizers used in the TYP treatment in my study (mean % N-P-K = 5.5 - 2.3 - 1.1), although this would not solely account for 23-fold difference in average price. These high input prices (observed in my study) likely contribute to farmers across North America listing nutrient management as one of their top research priorities (Jerkin & Ory, 2016).          50 3.2 Crop Yield Yield responses to nutrient management strategy varied substantially across farms and across regions. When relativized to the farm average, yields ranged from 46 to 155% across treatments with some farms showing ranges among nutrient management strategies as high as 96% (46 vs. 144%). Averaged over regions and years, I did not find yield differences among nutrient management strategies. This supports my overarching hypothesis that yields would not be different given all the treatments were designed in theory to meet crop demand and given the heterogeneity of soils and compost characteristics across farms in my study. Similarly, in a three-year study in Virginia, there were no yield differences in vegetables grown using a high, N-based compost applications or low compost applications plus a (conventional) N fertilizer (Evanylo et al., 2008), presumably due to soil nutrient reserves as well as adequate supply of nutrients from the applications. However, yield response to nutrient management strategy varied by region and year (nutrient strategy by region by year interaction, p = 0.044) and when analyzed by region, the overall trend of no yield differences was consistent for the Pemberton Valley and Vancouver Island, but I did find differences in the lower Fraser Valley. When averaged over both years, I found higher yields in the HC  than the TYP treatment in the lower Fraser Valley (Figure 3.3), while yields in the LC+N treatment were not different than the other two nutrient management strategies. In addition, in this region the yield response to nutrient management strategy also varied by year (nutrient strategy by year interaction, p < 0.001) and when analyzed by year, differences among treatments were only found in 2019 and not 2018 (Figure 3.4). It is likely that I found yield differences in the lower Fraser Valley and not the other regions because the soils and composts of the farms in my study in the lower Fraser Valley were   51 more homogenous than those found in the other two regions (Table 3.1). All farms in this region were considered in the “high” or “excess” soil available P categories, as defined by the BC Ministry of Agriculture (Poon & Schmidt, 2010b) and composts in the lower Fraser Valley were consistently higher in N and P, and therefore unlikely to cause N immobilization (Gale et al., 2006).    The yield differences (TYP > HC) that I found in the lower Fraser Valley are likely related to PAN, rather than available P, given that these soils have high available P, and both the TYP and HC treatments supplied additional P. Of the four farms in my study in the lower Fraser Valley in 2019, three applied less PAN to the TYP treatment than the HC treatment (103 vs. 115, 42 vs. 87, and 0 vs. 46 kg PAN ha-1 applied to TYP and HC treatments, respectively), while one applied substantially more PAN (530 vs. 97 kg PAN ha-1 applied to the TYP and HC treatments, respectively). It is possible that the three former farms under-applied N, while the latter farm over-applied. The link between PAN application rates and yields is well established in the literature, for example, Evanylo et al. (2008) found reduced yields from a low (20% of crop PAN requirements) application, relative to a higher,  PAN application rate of poultry manure compost; the same study found a correlation of soil NO3- and corn earleaf N, providing further evidence for the link between the yield response and PAN supply. However, reduced yields from over-fertilizer was found by Reiter et al. (2012) when comparing four N fertilizer rates (0, 67, 134, 201, and 268 kg N ha-1) in potato production, where the middle rate (134 kg N ha-1) produced the highest yields in two of three study years (while the third year had all yields reduced from a wet spring and summer drought). Overall, these findings point to the importance of developing accurate nutrient budgets to help ensure application of the right rate of nutrients in order to achieve yield goals by avoiding under- or over-fertilization.    52 In contrast to the lower Fraser Valley, farms in my study in the Pemberton Valley and on Vancouver Island had much more variability in compost properties (i.e. total N and P and C:N) resulting in a diversity of responses of available soil P to the nutrient management strategies. Therefore, a PCA analysis was performed to identify the soil and compost characteristics related to enhanced yields within each nutrient strategy across all farms and regions.   53                      F=4.68, p=0.031 n=8 a ab b F=0.16, p=0.853 n=12 F=0.70, p=0.5163 n=7 Figure 3.3 Boxplot of yield (kg m-1) by nutrient management strategy within each region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)). ANOVA F and p-values refer to main effect of nutrient strategy within each region over both years. Boxplots with different letters represent significant differences among treatments using Tukey’s post-hoc test at p<0.05. The center line indicates the median, means are shown as black dots and sample size indicated by n. F=4.29, p=0.068 n=4 F=6.39, p=0.033 n=4 ab a b Figure 3.4 Boxplot of yield (kg m-1) in the lower Fraser Valley by nutrient management strategies within years. ANOVA F and p-values refer to main effect of nutrient strategy within year. Boxplots with different letters represent significant differences among treatments using Tukey’s post-hoc test at p<0.05. The center line indicates the median, means are shown as black dots and sample size indicated by n.   54 3.2.1 Crop yield and baseline soil and compost properties I observed a large variation in performance of nutrient management strategies in terms of relative yield across farms as well as across the three regions. Some differentiation among regions and nutrient management strategies was apparent when a PCA analysis was performed on relative yields plotted with pre-season soil and compost properties and grouped by region (Figure 3.5A). Dimensions 1 and 2 explained 47.0% (29.6 and 15.1%, respectively) of the variation. Compost N and P content were inversely related to compost C:N and contributed to the differentiation of relative yield by region along dimension 1. High positive loadings of compost total P and N, distinguished the relative yields in the lower Fraser Valley and Pemberton Valley regions from Vancouver Island, while high negative loadings of compost C:N were associated with Vancouver Island. Noticeably, relative yields in the HC and LC+N treatments were associated with some baseline soil and compost properties, whereas yields in the TYP were not. This can be explained by the standardization of the research treatments, and the high variability of TYP treatments, from no nutrient application, to rates much higher than the research treatments (Table 3.5). In contrast, relative yields in the HC and LC+N treatments showed some differentiation by soil and compost nutrient properties. Relative yields in the HC treatments were associated with high soil P and higher nutrient composts found in the lower Fraser Valley and Pemberton Valley regions. In contrast, relative yields in the LC+N treatments were more associated with high compost C:N and SOC, largely associated with the Vancouver Island region.  Relative yields in the LC+N treatments were likely associated with compost C:N because carbon based amendments with high C:N often immobilize N rather than mineralize  N (Gale et al., 2006). This process happens regardless of how much total N is applied, and some composts   55 in my study would be expected to immobilize N, based on their high C:N. In contrast, high N fertilizers, such as feather meal, are not expected to cause N immobilization given their C:N <10 (Gale et al., 2006; Lazicki et al., 2020). Mikhabelaa et al. (2005) found similar results with potatoes, where yields were N limited when a high C:N compost (23:1) was used, regardless of rate, and additional N from a fertilizer was needed to enhance N availability to the crop and increase yields. In my study, the feather meal would help overcome amendment N immobilization and increase yields relative to the HC treatments on farms using composts with higher C:N.  The correlation between SOC and relative yields in the LC+N treatments may be related to soil P availability. For example, many farms in my study (on Vancouver Island and in the Pemberton Valley) were in categories of soil available P where yield responses would be expected from P applications (i.e. farms where the HC treatments would be expected to have performed better than the LC+N treatments due to higher P applications). However, Schneider et al. (2016) demonstrated that available P test values are not well correlated with yields on organic farms. Specifically, they found that when organic P (dairy manure) is the only P input to the soil, forage yields were maintained despite very low soil test P (Main et al. 2013). They suggested that these partial-extraction soil tests that are used to determine available P do not properly account for soil organic P, which can mineralize and supply P to crops during the growing season. Therefore, yields in the LC+N treatments may be less likely to be P limited on farms with low extractable-P but high SOM, and could even have enhanced crop productivity compared to an HC strategy when high C:N composts are used (by overcoming N limitations with feather meal, as discussed previously).   56                        A B Figure 3.5 Principal Component Analysis (PCA) biplots for relative yields by nutrient management strategy (Typical:TYP, high compost: HC, and low compost + N: LC), plotted with pre-season compost and soil properties and grouped by region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)), where: (A) includes all nutrient management strategies, and (B) shows only the TYP strategy plotted with the quantity of applied nutrients (applied total phosphorus, carbon, nitrogen, and plant-available nitrogen (NO3- + NH4+) = appTP, apptTC, appN, and appPAN, respectively)). Pre-season compost and soil properties include: compost plant-available N (NO3- + NH4+) = compost_PAN, compost C to N ratio = compost_CN, soil C to N ratio = soil_CN, soil total organic C content = soil_C, soil mehlich-3 phosphorus = soil_P, and soil sand, clay, and pH.   57 Finally, in this analysis (Figure 3.5A), high compost C:N is associated with high SOC, primarily and to a lesser extent, with sand on Vancouver Island. The importance of SOC for holding water and nutrients increases on sandier soils and thus could drive relative yields particularly when smaller amounts of compost are being added. The association between SOC and compost C:N on farms on Vancouver Island is unlikely due to greater C inputs in this region given that TYP application rates measured were similar to the lower Fraser Valley. (Table 3.5). This association is also surprising given that it is now believed that input C is sequestered more efficiently when the C to nutrient stoichiometry of an input is lower and more closely matches those required by microbes (Cotrufo et al., 2013; Kirkby et al., 2013). It is possible that other mechanisms are responsible for the association between SOC and compost C:N. One possibility is through tillage: the farmers in my study in the lower Fraser Valley reported that the high soil clay content requires extra tillage (lower Fraser Valley farmer, personal communication, April 18, 2018). This can destroy aggregates and result in SOC turnover (Denef et al., 2004) in this region. Another explanation for higher SOC on Vancouver Island is through cover crop use, given their known contribution to building SOC (Poepleau & Don, 2015). Of the three regions, the Vancouver Island region had the least difficulties in growing winter cover crops. I sampled cover crops at three of eight farms on Vancouver Island, and the remaining five farms that samples were not collected from, either had already plowed in the cover crop prior to sampling or had tarped their fields instead. The farm that used tarps for soil protection reported doing so due to concerns about winter cover crops contributing to wireworm populations in the following season.  In contrast, four out of six farms in the lower Fraser Valley did not have cover crops in the spring in the research treatments, which was reportedly due to their destruction by winter waterfowl (lower Fraser Valley farmer, personal communication, April 18, 2018). In addition,   58 the two farms in the lower Fraser Valley where I did collect cover crops were also the two farms in my study in this region that have higher SOC (7.2 and 10% SOC). While my study did not test a connection between cover crops and SOC accumulation on these farms, I did observe this trend between this management practice and SOC values. Another PCA analysis was performed on relative yields, with yields from the TYP treatments plotted with pre-season soil and compost properties as well as the quantity of nutrients applied for each treatment, including P, C, N, and PAN and grouped by region (Figure 3.5B). In this analysis, dimensions 1 and 2 explained 54.5% (29.9 and 24.6%, respectively) of the variation of the relative yields in the TYP treatments. Notably, dimension 1 has high positive loadings of relative yields in the TYP treatments and the quantity of nutrients applied (C, P, N, and PAN).  Together, these two PCA illustrate a lack of relationship between crop productivity in the TYP treatments and baseline compost and soil properties. Instead, yields in these treatments were more closely correlated with the quantity of nutrients applied, indicating that the TYP treatments were the most productive when farmers were adjusting their applications to account for characteristics of their system (i.e. compost and soil properties), rather than having outcomes determined by limitations such as low soil P or high C:N composts. Conversely, relative yields in the HC and LC+N treatments were not associated with the quantity of nutrients applied in these treatments (data not shown). Overall, these trends do not support my original hypothesis that there would not be yield differences between the LC+N and HC treatments, because both nutrient management strategies were designed to apply nutrients to meet crop demand. Instead these analyses illustrate how nutrient management decisions are highly farm-specific, and outcomes depend on a manager’s   59 ability to be flexible to take into consideration the characteristics of the soil, compost, and crop. Similarly, yields and N cycling efficiency across 13 organic tomato fields in California, USA, were maximized when management decisions reflected a field’s unique soil characteristics (termed ‘adaptive management’), rather than using a one size fits all (Bowles et al., 2015). 3.3 Soil properties 3.3.1 Permanganate oxidizable carbon Permanganate oxidizable carbon values were highly variable and reflected the heterogeneity of management practices (cover crops, C inputs, tillage, etc.) amongst farms. Despite being regarded as a management-sensitive indicator, POx-C did not differentiate among nutrient management strategies in my study when averaged over regions and years. However, POx-C followed the trend in SOC that I observed in my pre-season samples (Table 3.1), with values on Vancouver Island being numerically, but not statistically, greater than in the other two regions. Although POx-C values in Vancouver Island and the lower Fraser Valley regions were statistically different when the urban farm, with constructed soils (and high POx-C), in the lower Fraser Valley was removed from the dataset. Despite the variability I observed in POx-C, ranging from 247 to 1393 mg kg soil-1, my data is comparable to those found in similar systems elsewhere. In organic vegetable fields in southwestern Ontario, Canada (Hargreaves et al., 2019) and in New York, USA (Drinkwater, 2011; unpublished data cited in Culman et al., (2012), POx-C ranged from 661 to 1070 mg kg-1 and from 154 to 983 mg kg-1, respectively.  The high variation in relative POx-C across farms was driven by different soil types and input properties depending on the region. There is some differentiation of relative POx-C among the nutrient management strategies when analyzed in a PCA with pre-season soil and compost   60 properties (Figure 3.6). While POx-C outcomes in the HC treatments were not associated with compost properties, POx-C in the LC+N treatments was somewhat  associated with SOC and compost C:N. This is possibly due to the nutrient stoichiometry of the inputs in these treatments, where the LC+N treatments on farms with high C:N composts would have lower input C:N (than the HC treatments) because of the additional N from the feather meal. For example, Kirkby et al. (2013) demonstrated that input C is used more efficiently by microbes (less C is respired per unit input) when the input nutrient stoichiometry more closely matches microbial requirements. This is especially evident on Vancouver Island, where the average input C:N of the HC treatments is 23.2:1, compared to 8.3:1 in the LC+N treatments (closer to the microbial requirements of 8:1).                            Figure 3.6 Principal Component Analysis (PCA) biplot of the relative POx-C measured in each nutrient management strategy (typical = TYP, low compost + N = LC, and high compost = HC), plotted with pre-season soil and compost properties, including compost plant-available N (NO3- + NH4+) = compost_PAN, compost C to N ratio = compost_CN, soil C to N ratio = soil_CN, soil total organic C content = soil_C, soil mehlich-3 phosphorus = soil_P, and soil sand, clay, and pH, and grouped by region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)).     61 3.3.2 Post-season available nitrogen I found differences in post-season available N among nutrient management strategies in the upper depth (0-15 cm) and between years and regions in the lower depth (15-30 cm). When averaged over regions and years, there was higher post-season NH4+ (0-15 cm) in the TYP and HC treatments than in the LC+N treatments (Figure 3.7A). The impact of nutrient management strategy on post-season NH4+ was moderated by region (nutrient strategy by region interaction, p = 0.003), and when analyzed by region, the main effect of nutrient strategy was significant in the lower Fraser Valley and on Vancouver Island, but not in the Pemberton Valley (Figure 3.7B). In the lower Fraser Valley, I found the same trend of (numerically) less NH4+ (0-15 cm) in the LC+N treatments than the HC treatments, but this was not significant in the post-hoc test (HC – LC+N Tukey contrast, p=0.052. On Vancouver Island I found less NH4+ (0-15 cm) in the LC+N treatments than the TYP treatments. These findings align with my hypothesis that I would find greater residual available N in the HC treatments than the LC+N treatments. This can be explained by the larger application of total N (largely as organic N) in the HC treatment, which can continue to mineralize at the end of the season. As Maltais-Landry et al. (2019b) have pointed out, organic residues \are very likely to continue to be mineralized in the regions with mild and humid climate of the lower Fraser Valley and Vancouver Island. Averaged over region and years, post-season NO3- (0-15 cm) did not differ among nutrient management strategies (Figure 3.8A). However, similar to NH4+, the impact of nutrient strategy on post-season NO3- was moderated by region (nutrient strategy by region interaction, p = 0.010). When analyzed by region, the main effect of nutrient strategy was marginally significant in the lower Fraser Valley and on Vancouver Island (Figure 3.8B), with trends of greater residual NO3- (0-15 cm) in the HC treatments than in the LC+N treatments in the lower   62 Fraser Valley and in the TYP treatments than in the LC+N treatments on Vancouver Island, although neither of these differences were significant in the post-hoc test (HC – LC+N in the FV and TYP – LC+N Tukey contrasts, p=0.076 and 0.058, respectively) (Figure 3.8B). Similarly, when averaged over regions and years, I did not find differences in NO3- when averaged over both depths (0-30 cm) (Figure 3.9) and the impact of nutrient management strategy on post-season NO3- (0-30 cm) was moderated by region (nutrient strategy by region interaction, p = 0.044). When analyzed by region, the main effect of nutrient strategy was marginally significant (p<0.1) in the Pemberton Valley only, but treatments were not different in the post-hoc analysis (Figure 3.9). Despite no differences among nutrient strategies, and relatively low average post-season NO3- in the 87 plots (9.1 mg NO3--N kg-1 soil at 0-30 cm depth), I still found high residual NO3- in several plots. Specifically, there were 4 plots (2 from HC and 1 each from LC+N and TYP) with post-season NO3- greater than the threshold of 25 mg NO3--N kg-1 soil (corresponding to about 100 kg NO3--N ha-1). This is the threshold indicated by the Agricultural Environmental management Code of Practice from the Ministry of Environment (Government of British Columbia, 2019) that could require follow-up mandatory soil testing (depending on location and type of farm). This demonstrates that organic agriculture is not inherently without the possibility of environmental impact (Tuomisto et al., 2012). However, this is only 5% of all plots that I sampled. In contrast, a 2012 survey of 177 fields in the lower Fraser Valley (159 sampled before Oct 11 and the rest before Oct 24), found that 55% of fields had greater than 100 kg NO3--N ha-1 in the 0-60 cm depth (mean = 138 kg NO3--N ha-1) (Sullivan & Poon, 2012). More specifically, of the 30 vegetable fields they sampled, 73% were above this benchmark (mean = 173 kg NO3--N ha-1) (Sullivan & Poon, 2012). Given that the majority of farms in their study would not be managed organically (Statistics Canada, 2017), even though   63 organic farms can leave high residual NO3-, it is less frequent on the (primarily) organic farms in my study.  The post-season NO3- data was also relativized to farm averages and analyzed using a PCA with farm site baseline soil and compost properties and grouped by region (Figure 3.10). In this analysis, dimensions 1 and 2 explained 42.7% (33.6 and 19.1%, respectively) of the variation. Along dimension 1, high post-season NO3- (0-15 cm) in the HC treatments is associated with the high N (and high P) composts, primarily in the lower Fraser Valley region, and to a lesser extent in the Pemberton Valley. In contrast, high post-season NO3- in the LC+N and TYP treatments is not strongly associated with any pre-season soil or compost properties used in this analysis. These findings are congruent with other research in the lower Fraser Valley where post-harvest NO3- was more than double in vegetables fields that were manured compared to those that were not (Sullivan & Poon, 2012). Similarly, the meta-analysis by Norris and Congreves (2018) showed that despite the highly variable impact of C-based amendments on soil N, products that were higher in N such as poultry manures, tended to have increased risk for NO3- leaching.    64                     Figure 3.7 Boxplot of post-season NH4+-N (mg kg-1) (0-15 cm). (A) By nutrient management strategy, averaged over regions and years, and (B)  by nutrient management strategy within regions (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)) over years. ANOVA F and p-value refer to main effect of nutrient strategy. Boxplots with different letters represent significant differences among nutrient strategies using Tukey’s post-hoc test at p<0.05. The center line indicates the median, means are shown as black dots and sample size is indicated by n.  F=5.5, p=0.011 n=12  F=4.0, p=0.043 n=9 F=0.93, p=0.416 n=8  ab a b a F=5.72, p=0.006 n=29 b aA. B.   65                                Figure 3.8 Boxplots of post-season NO3--N (mg kg-1) (0-15 cm). (A) by nutrient management strategy, averaged over regions and years, and (B) by nutrient management strategy within regions (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)) over years. ANOVA F and p-value refer to main effect of nutrient strategy. Boxplots with different letters represent significant differences among nutrient strategies using Tukey’s post-hoc test at p<0.05. The center line indicates the median, means are shown as black dots and sample size is indicated by n. F=3.04, p=0.076 n=9 F=3.38, p=0.068 n=8   F=0.02, p=0.982 n=29 F=0.83, p=0.448 n=12 A. B.   66                                  F=0.04, p=0.959 n=12 F=2.99, p=0.074 n=8  F=2.62, p=0.104 n=9 Figure 3.9 Boxplot of post-season NO3--N (mg kg-1) (0-30 cm) by nutrient management strategy within regions averaged over years. ANOVA F and p-value refer to main effect of nutrient strategy within region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)). The center line indicates the median, means are shown as black dots and sample size is indicated by n. Red line indicates threshold  of 25 mg NO3- kg-1 for the 0-30 cm depth determined by the BC Ministry of Agriculture to trigger follow-up action.   67                   Figure 3.10 Principal Component Analysis (PCA) biplot of the relative NO3- (0-30 cm) measured in each nutrient management strategy (typical = TYP, low compost + N = LC, and high compost = HC), plotted with pre-season soil and compost properties, including compost plant-available N (NO3- + NH4+) = compost_PAN, compost C to N ratio = compost_CN, soil C to N ratio = soil_CN, soil total organic C content = soil_C, soil mehlich-3 phosphorus = soil_P, and soil sand, clay, and pH, and grouped by region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)).   68 3.3.3 Post-season available phosphorus When averaged over regions and years, the trend in post-season available P from each nutrient strategy closely matched the mean rates of applied total P to each treatment (as HC > TYP > LC+N; Figure 3.11; Table 3.5). However, only HC and LC+N treatments were significantly different (Figure 3.11). This supports my hypothesis that post-season available P would be greater in the HC than the LC+N treatments, given that the former is designed to apply P beyond what is removed by the crop, whereas the latter would theoretically have small effects on soil P status.                F=7.42, p=0.002 n=26 a b ab Figure 3.11 Boxplot of post-season Kelowna-extractable available P (mg kg-1) by nutrient management strategy, averaged over regions and years. ANOVA F and p-values refer to main effect of nutrient strategy. Boxplots with different letters represent significant differences among treatments using Tukey’s post-hoc test at p<0.05. The center line indicates the median, means are shown as black dots and sample size is indicated by n.     69 In contrast to my findings, Mkhabela and Warman (2005) did not find differences in available P (Mehlich-3 extractable) among plots receiving one or two years of compost at 1x, 2x, and 3x the rate of crop P removal by corn and potatoes. This difference in findings is likely because the HC treatments in my study received, on average, more than 8x the crop P removal that was applied in the LC+N treatments (values shown in Table 3.5), which is much greater than the rates trialed by Mkhabela and Warman (2005). Whereas, my results were similar to studies that showed high compost applications at 5x crop P removal and high manure applications at 4x crop P removal increased soil P in comparison to a low compost application (Evanylo et al., 2008; Maltais-Landry et al., 2019a). Overall, my study demonstrates that short term soil P build-up can occur rapidly with the large over application of P (8x crop removal) used in my study, while smaller over-application of P (i.e. 2-3x crop P removal) may not create measurable differences in soil P as quickly. Soil P outcomes from P application rates between these two extremes (i.e. 5x crop P removal) may depend more on the P dynamics specific to the amendment (Maltais-Landry et al., 2019a). Post-season available P was also relativized to the farm average and analyzed using a PCA with pre-season soil and compost properties and grouped by region (Figure 3.12). In this analysis, dimensions 1 and 2 explained 47.6% (31.6 and 16%, respectively) of the variation. Along dimension 2, high post-season available P (0-15 cm) in the HC treatments was correlated with high soil clay content, primarily in the lower Fraser Valley and the Pemberton Valley regions. In contrast, available P in the LC+N treatments was separated by a negative loading on dimension 1, and had a minor association with increasing compost C:N on Vancouver Island. Overall, post-season available P in the TYP treatments was not associated with any pre-season   70 soil or compost properties in this analysis, which is not surprising given the variability in P applications in these treatments (Table 3.5).                               Figure 3.12 Principal Component Analysis (PCA) biplot of the relative post-season available P (0-15 cm) measured in each nutrient management strategy (typical = TYP, low compost + N = LC, and high compost = HC), plotted with pre-season soil and compost properties, including compost plant-available N (NO3- + NH4+) = compost_PAN, compost C to N ratio = compost_CN, soil C to N ratio = soil_CN, soil total organic C content = soil_C, soil mehlich-3 phosphorus = soil_P, and soil sand, clay, and pH, and grouped by region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)).   71 3.4 Input costs Input costs were highly variable, from $0 in TYP treatments with no nutrient application, to $34,977 ha-1 in a TYP treatment on Vancouver Island. Overall, mean input costs across all treatments, regions, and years was $4,959 ha-1. Averaged over regions and years I did not find differences in input costs among the nutrient management strategies. However, outcomes of nutrient strategies varied by region (nutrient strategy by region interaction, p = 0.002) and when analyzed by region I found differences in the lower Fraser Valley but not in the Pemberton Valley or on Vancouver Island (Figure 3.13). In the lower Fraser Valley, the TYP treatments had lower input costs than the HC and LC+N treatments (Figure 3.13). However, despite the LC+N treatments being more expensive than the TYP treatments in the lower Fraser Valley, the LC+N strategy had the least variability among regions, given that it is less dependent on the highly fluctuating cost of compost. I also found a trend of input costs among regions in the order of Vancouver Island > Pemberton Valley > lower Fraser Valley (mean input costs = $7,312, $4,361, and $1,959 ha-1, respectively). However, while the main effect of region was significant (p=0.009), these region means were not significantly different in the Tukey post-hoc test. Overall the values I estimated in my study were much higher than the $700 ha-1 reported for the pelletized poultry manure and pig manure applications used in potato field trial in Truro, Nova Scotia by Lynch et al. (2008). Although the input costs for all HC treatments, and several TYP treatments, using the cheaper poultry manure-based amendments in the lower Fraser Valley in my study were less than $800 ha-1, these represent only 6 out of the 72 treatments with associated input costs in my study. The difference in fertility costs between organic and conventional production in California has been found to be greater for vegetables studied than fruit or field crops (Klonsky, 2012). Specifically, organic inputs for broccoli and lettuce were   72 more than 2x the costs of conventional (~$1,561 and $2,247 ha-1for broccoli and lettuce, respectively, and assuming $1 USD ~ $1 CAD in 2011) and are more similar to the costs found in my study. Overall, my observations highlight the range in fertility costs for organic farms in southwest BC, which largely depend on the regional availability of composts and manures, as well as organic fertilizers.                    a  a  b  F=8.71, p=0.005 n=7  F=0.01, p=0.988 n=7 F=2.4, p=0.116 n=11 Figure 3.13 Boxplot of input costs ($ ha-1) by nutrient management strategy averaged over years within regions. ANOVA F and p-values refer to main effect of nutrient strategy within each region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)). Boxplots with different letters represent significant differences between nutrient strategies using Tukey’s post-hoc test at p<0.05. The center line indicates the median, means are shown as black dots and sample size is indicated by n.   73 3.4.1 Cost per unit yield A PCA was performed on the cost per unit yield for each nutrient strategy and analyzed with select pre-season soil and compost properties (Figure 3.14) and dimensions 1 and 2 explained 58.7% (45 and 13.7%, respectively) of the variation. In this analysis, the Vancouver Island region is differentiated by high negative loadings of input costs for the HC and TYP treatments. This likely due to the combination of the high cost of compost (for the HC treatments) and overall inputs (for the TYP treatments) in this region, and given that the relative yield PCA showed that HC treatments appeared to perform less well on Vancouver Island due to high compost C:N in this region (Figure 3.5A). In contrast, the relative input costs per unit yield in the LC+N treatments and compost PAN were correlated with high positive loadings on dimension 1, presumably because the high PAN composts (i.e. poultry manures) in my study were also the cheapest.               74                        Figure 3.14 Principal Component Analysis (PCA) biplot of the relative cost per unit yield by nutrient management strategy (typical = TYP, low compost + N = LC, and high compost = HC), plotted with pre-season soil and compost properties, including compost plant-available N (NO3- + NH4+) = compost_PAN, compost C to N ratio = compost_CN, soil C to N ratio = soil_CN, soil total organic C content = soil_C, soil mehlich-3 phosphorus = soil_P, and soil sand, clay, and pH, and grouped by region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)).     75 3.5 Trade-offs of nutrient management strategies Tradeoffs among nutrient management strategies, in terms of crop yields), soil POx-C, post-season available N and P, and input costs were largely region specific. When averaged across regions and years, nutrient strategies did not differ in terms of yields, POx-C, post-season NO3-, or input costs, but the HC treatments did have greater post-season available P than the LC+N treatments, which is an environmental concern. This is illustrated in the radar chart in Figure 3.15A, showing the primary trade-offs for all regions; this figure also shows that numerically, but not statistically, the HC treatments had greater post-season NO3- than the other treatments, while costs per unit yield were lower in the LC+N treatments, while the LC+N treatments had lower input costs. In the lower Fraser Valley (Figure 3.15B), input costs in the TYP treatments were lower than the HC and LC+N treatments, but the HC treatments had greater yields than the TYP treatments in one year in this region, but also greater post-season available P than the LC+N treatments. This represents a trade-off between crop yield and potential environmental impacts with the HC strategy in this region. The HC treatments also had numerically, but not statistically, higher post-season NO3- than the other two treatments.  In contrast, in the Pemberton Valley, there was numerically, but not statistically, greater post-season NO3- in the LC treatments, and higher input costs in the TYP treatments, than the other two treatments (Figure 3.15C). Along with the overall trend across all regions, LC+N treatments had statistically lower post-season P than the other two treatments. This highlights the possibility that nutrient management strategies can have both good environmental outcomes (i.e. lower soil P) and poor environmental outcomes (i.e. high post-season NO3-) at the same time.   76 While nutrient strategies had similar yields and post-season NO3- on Vancouver Island, the LC+N strategy performed well in other metrics (Figure 3.15D). Specifically, LC+N treatments not only had lower post-season available P than HC treatments, they also had numerically, but not statistically, lower input costs than both the HC and TYP treatments. Overall, these results highlight the importance of assessing management practices with multiple, and often competing, end-results, and the need for region- and farm-specific management decisions that can be flexible to system-specific input and soil properties.                    77                        A. B. C. D. Figure 3.15 Measured outcomes scaled to the maximum value observed among treatments, expressed as the mean within each treatment. Measured outcomes include: yield, input costs, post-season NO3- (0-30 cm depth) (NO3), post-season available P (0-15 cm depth) (soil_P), and permanganate oxidizable carbon (POXC). Measured outcomes with significant differences at alpha <0.05 between treatments are indicated with an asterisk (*).   78 Chapter 4: Conclusion 4.1 General conclusions Nutrient management strongly influences the sustainability of crop production. Organic sources of N and P can be expensive and therefore play a large role in the economic viability of organic farms. In addition, N and P deficits can reduce yields, while their excess can lead to environmental impacts through degraded air and water quality and/or greenhouse gas emissions. Since nutrient management is highly region-specific due to varying input availability, climate, and soils, in this thesis I inventoried three different regions of southwest BC in terms of compost and soil properties, amendment use and costs, and nutrient management strategies. I also assessed three nutrient management strategies in terms of their impacts to crop yields, a select set of soil properties, and input costs across 20 organic vegetable farms in these three regions. I explored possible  explanations for the variation in performance of each nutrient management strategy across these regions using PCA.  Finally, I examined potential tradeoffs among the nutrient management strategies by compiling the analysis into radar graphs. I found high soil P and relatively inexpensive, high N and P composts and manures based on livestock manures (poultry and pig) or urban wastes on farms in the lower Fraser Valley. Farms in my study on Vancouver Island have high SOC and tend to use composts with high C:N ratio based on steer or horse manure, or from commercial composts made with forest and fishing industry byproducts. Farms in the Pemberton Valley tend to have low soil P and largely use commercial composts based on food scraps and yard wastes. My hypothesis that yields would not be different was generally supported by my findings. I did; however, find yield differences in the lower Fraser Valley in 2018, where HC treatments were more productive than the TYP treatments, which was likely due to N dynamics. Although I   79 expected to find greater POx-C in the HC treatments than the LC+N treatments, I found no differences among nutrient strategies.  As expected, the HC treatments had greater post-season NH4+ than the LC+N and TYP treatments, indicating that organic N from the compost was still mineralizing after harvest. I did not; however, find differences in post-season NO3- among nutrient strategies, indicating the mineralization was either happening very slowly, or NO3- was being taken up by cover crops, or already lost through leaching. The PCA indicated a plausible link between high N composts in the lower Fraser Valley and the Pemberton Valley and greater residual NO3- in the HC treatments. As hypothesized, post-season, Kelowna-extractable available P was greater in the HC treatments than in the LC+N treatments due to greater P application in these treatments. Finally, input costs were lower for the TYP treatments than the HC and LC+N treatments in the lower Fraser Valley, where farmers have ready access to inexpensive nutrients, and soils with high P reserves. Costs of the LC+N treatments were the least variable among regions.  While I expected that there were likely to be clear trade-offs among these nutrient strategies my data did not support this.  The TYP treatment did not show clear differences from the other two treatments, likely because of the high variability in the approach.  Alternatively, across the regions post-season available P was significantly lower in the LC+N than the HC treatment, indicating environmental impacts can be reduced without incurring yield or input cost trade-offs.   4.2 Strengths and limitations of the research My study took place on 20 working, organic vegetable farms, which was both a benefit and a challenge. This provided the opportunity to assess the soil, compost, and nutrient management characteristics of working farms. In addition, the combinations of composts, soils,   80 climates, and crops that I studied actually reflect what is being used by organic farmers. In this way, my results are more relevant and specific to farmers and policy makers in these regions of our province. A major limitation of this type of study is the lack of consistency in the use of plots on farms. Specifically, because new plots were used on farms in the second year of the study, about 2/3 of my plots were studied for one year, while 1/3 of the plots were used for two years. This means treatment effects were observed after one or two years of application, because the dataset is mixed. In addition, a limitation is incomplete datasets for each farm. For example, some farms have all but one outcome measured (i.e. only post-season samples are missing because flags were moved after harvest).  Finally, a major limitation was the omission of the N contribution of cover crops. This choice was made due to time and logistical constraints of the project, and because farms were so varied in their use of cover crops. It is possible that large N contributions could have masked the effects of nutrient strategies if it created an overprovision of this nutrient in all strategies.  4.3 Management implications for farmers Overall, the results of this thesis serve to support the observations of other researchers (i.e. Bowles et al., 2015) that management decisions are most effective (in terms of economics, yields, etc.) when tailored to specific soils and farm systems. For example, when high C:N composts were used, the PCA suggested greater benefits in terms of POx-C and yields when a reduced compost application was combined with a N fertilizer (i.e. feather meal), presumably by lowering the input C:N. For yields, this combination was especially effective when farms also had high SOC, which presumably provides additional P cycling, despite instances of low soil test P. Finally, this strategy (LC+N) did not appear to have environmental tradeoffs (i.e. high post-  81 season available N or P), according to the PCA. In these systems where the LC+N strategy performed well, farms may maintain SOC through use of winter cover crops.  Given that environmental costs are not paid by the farmer, but yield penalties and input costs are, it is difficult to reason that farmers in the lower Fraser Valley should change their practices (to decrease available P and post-season available N) from a purely economic standpoint. Farms in this region rely on cheap manure and composts as economical sources of C and N, and they struggle to build SOC with cover crops because they are heavily grazed by waterfowl. Given that mandatory soil testing is being phased-in here in BC, farmers should begin consistent monitoring for available P and post-season NO3- to understand how they compare to these provincial standards. These regulations may provide the economic incentive to change management practices.  Given the need to reduce soil P in this region, farms will need to plan to reduce manure and/or compost applications. To do so, they could begin enhanced cover crop trials in their systems now. Given that shoulder season (pre- and post-season) cover crops are challenging to implement in parts of the lower Fraser Valley, farms could theoretically build SOC through a biannual rotation of cover crops with vegetable production (i.e. every second year a field is planted in cover crop for the summer), though this is unlikely given the high cost of land prices and associated opportunity cost from planting non-cash cover crops for a season. Instead, farms could focus on early establishment of winter cover crops, possibly using inter-cropping methods (relay cropping) to seed cover crops into still-standing crops. This will help to capture post-season available N and prevent leaching, while still using high nutrient manures, which contribute to SOC over time. Eventually, mandatory nutrient management planning may make high N fertilizers or more elaborate cover crop use economically feasible, and farms in this   82 region will (theoretically) have years of soil P reserves to rely on, while they focus on N and SOC management primarily. Finally, while results were not definitive in relation to farms in the Pemberton Valley, it is likely that farms in this region are heading in the direction of the high soil P that is currently found in the lower Fraser Valley. Specifically, farms in Pemberton Valley tend to use a locally produced compost product that provides C, N, and P all in one application as part of their typical nutrient management strategy. Farms here may have more economic incentive to use reduced compost applications. Specifically, compost there is more expensive than in the lower Fraser Valley, but does not have the same waterfowl problem that impedes cover crop viability. Therefore, farms in this region would benefit from employing a mix of cover crops, fertilizers, and compost, depending on how their viability, costs, and availability, respectively fluctuate from season to season, in order to make the context-specific decisions for enhanced outcomes. Given that the yield differences I found in the lower Fraser Valley in 2018 were likely due to both the under- and over-application of N, all farms would benefit from basic nutrient budgets to avoid excessive deficits or surpluses of this nutrient in a given season. In contrast, it appears that soil P can be managed with a long-term approach, especially on farms with high soil test P and/or SOC. 4.4 Directions for future research Future research should include a more substantial economic analysis component as the farm economic outcome in my study is based on input costs alone and only included amendment and shipping costs. Given the economics of a nutrient management strategy is a key driver for farmer decision making future research should account for important variables beyond input costs, such as labor, crop quality, and cost of nutrient analyses.   83 While my study did not quantify the input of cover crops directly, it does point to their importance in several ways. First, while this research did not test for a relationship between cover crops and SOC, I did observe that farms which were able to establish winter cover crops tended to have higher SOC. In addition, fall-planted winter cover crops can capture post-season available N and prevent it from leaching as nitrate. This could help reduce NO3- leaching from the four plots in which I found high post-season NO3- (greater than ~100 kg N ha-1) and the several other plots that I found NO3- in. In this way, future research should focus on development of practical methods of incorporating cover crops into these intensive farming systems. In the lower Fraser Valley specifically (where I also tended to find the highest post-season NO3-), cover crop options will need to be developed to combat waterfowl destruction. These developments could have widespread environmental benefits in the lower Fraser Valley region due to its high concentration of intensive crop production. Future research should also include a focus on developing soil tests for nutrient management planning that are specific to systems relying largely on carbon-based amendments (i.e. organic farms) and specialty organic fertilizers. For example, soil test methods of quality and/or quantity of SOC and/or microbe populations that can predict N and P supply to crops, such that N and P input requirements can be developed from them are much needed in the organic sector. Thus far, research has largely focused on quantifying the amount and availability of N and P to apply (i.e. from inputs like composts, manures, cover crops, and specialty fertilizers). However, applying these to target ‘agronomic’ rates based on conventional soil tests does not take into account the complexity of organic farming systems. Input-intensive, conventional farming systems rely less on N and P mineralization from SOC, cover crops, composts, and manures, so these tests are less useful in systems that do. Yet, it is clear that the   84 most effective nutrient management strategies are those that are tailored to a specific farming system and associated soil and input characteristics, so tests with predictive capacity are necessary. Therefore, soil tests for organic farms need to be developed into economically feasible and accessible testing services that can be realistically used by farmers as a nutrient management tool.                85 Bibliography Ackroyd, V. J., Cavigelli, M. A., Spargo, J. T., Davis, B., Garst, G., & Mirsky, S. B. (2019). Legume cover crops reduce poultry litter application requirements in organic systems. Agronomy Journal, 111(5), 2361–2369. doi:10.2134/agronj2018.09.0622 Agriculture and Agri-Food Canada. (2020). Statistical overview of the Canadian vegetable industry 2019. https://www.agr.gc.ca/eng/horticulture/horticulture-sector-reports/statistical-overview-of-the-canadian-vegetable-industry-2019/?id=1599696046871#a1.1 Berry P.M. Sylvester-Bradley, R., Philipps, L., Hatch, D. J., Cuttle, S. P., Rayns, F. W., & Gosling, P. (2002). Is the productivity of organic farms restricted by the supply of available nitrogen? Soil Use and Management, 18(3), 248–255. doi:10.1079/sum2002129 Berry, P. M., Stockdale, E. A., Sylvester-Bradley, R., Philipps, L., Smith, K. A., Lord, E. I., Watson, C. A., & Fortune, S. (2003). N, P and K budgets for crop rotations on nine organic farms in the UK. Soil Use and Management, 19(2), 112–118. doi:10.1079/sum2003176 Bittman, S., & Kowalenko, C. G. (2004). Advanced silage corn management: A production guide for coastal British Columbia and the Pacific Northwest (S. Bittman & C. G. Kowalenko (eds.)). Pacific Field Corn Association. Bowles, T. M., Hollander, A. D., Steenwerth, K., & Jackson, L. E. (2015). Tightly-coupled plant-soil nitrogen cycling: Comparison of organic farms across an agricultural landscape. PLoS ONE, 10(6), 1–25. doi:10.1371/journal.pone.0131888 Brennan, E. B. (2017). Can we grow organic or conventional vegetables sustainably without cover crops? HortTechnology, 27(2), 151–161. doi:10.21273/HORTTECH03358-16 Bünemann, E. K., Bongiorno, G., Bai, Z., Creamer, R. E., De Deyn, G., de Goede, R., Fleskens, L., Geissen, V., Kuyper, T. W., Mäder, P., Pulleman, M., Sukkel, W., van Groenigen, J. W., & Brussaard, L. (2018). Soil quality – A critical review. Soil Biology and Biochemistry, 120(January), 105–125. doi:10.1016/j.soilbio.2018.01.030 Canada Organic Trade Association. (2017). ORGANIC AGRICULTURE IN CANADA: BY THE NUMBERS. https://www.ota.com/sites/default/files/Org_Ag_Canada_overview_17.03.03-FINAL.pdf Carr, P. M., Cavigelli, M. A., Darby, H., Delate, K., Eberly, J. O., Fryer, H. K., Gramig, G. G., Heckman, J. R., Mallory, E. B., Reeve, J. R., Silva, E. M., Suchoff, D. H., & Woodley, A. L. (2020). Green and animal manure use in organic field crop systems. Agronomy Journal, 112(2), 648–674. doi:10.1002/agj2.20082    86 Carr, P. M., Cavigelli, M. A., Darby, H., Delate, K., Eberly, J. O., Gramig, G. G., Heckman, J. R., Mallory, E. B., Reeve, J. R., Silva, E. M., Suchoff, D. H., & Woodley, A. L. (2019). Nutrient cycling in organic field crops in Canada and the United States. Agronomy Journal, 111(6), 2769–2785. doi:10.2134/agronj2019.04.0275 Culman, S. W., Snapp, S. S., Freeman, M. A., Schipanski, M. E., Beniston, J., Lal, R., Drinkwater, L. E., Franzluebbers, A. J., Glover, J. D., Grandy, A. S., Lee, J., Six, J., Maul, J. E., Mirksy, S. B., Spargo, J. T., & Wander, M. M. (2012). Permanganate Oxidizable Carbon Reflects a Processed Soil Fraction that is Sensitive to Management. Soil Science Society of America Journal, 76(2), 494–504. doi:10.2136/sssaj2011.0286 Crawley, M. J. (2013). The R Book (2nd ed.). John Wiley & Sons, Ltd. Daryanto, S., Jacinthe, P. A., Fu, B., Zhao, W., & Wang, L. (2019). Valuing the ecosystem services of cover crops: barriers and pathways forward. Agriculture, Ecosystems and Environment, 270–271(October 2018), 76–78. doi:10.1016/j.agee.2018.10.021 De Ponti, T., Rijk, B., & Van Ittersum, M. K. (2012). The crop yield gap between organic and conventional agriculture. Agricultural Systems, 108, 1–9. doi:10.1016/j.agsy.2011.12.004 Denef, K., Six, J., Merckx, R., & Paustian, K. (2004). Carbon Sequestration in Microaggregates of No-Tillage Soils with Different Clay Mineralogy. Soil Science Society of America Journal, 68(6), 1935–1944. doi:10.2136/sssaj2004.1935 Di, H. J., & Cameron, K. C. (2002). Nitrate leaching in temperate agroecosystems: sources , factors and mitigating strategies. Nutrient Cycling in Agroecosystems, 46(2), 237–256. Eghball, B. (2002). Soil Properties as Influenced by Phosphorus- and Nitrogen-Based Manure and Compost Applications. Agronomy Journal, 94(1), 128. doi:10.2134/agronj2002.0128 Evanylo, G., Sherony, C., Spargo, J., Starner, D., Brosius, M., & Haering, K. (2008). Soil and water environmental effects of fertilizer-, manure-, and compost-based fertility practices in an organic vegetable cropping system. Agriculture, Ecosystems and Environment, 127(1–2), 50–58. doi:10.1016/j.agee.2008.02.014 Gale, E. S., Sullivan, D. M., Cogger, C. G., Bary, A. I., Hemphill, D. D., & Myhre, E. A. (2006). Estimating Plant-Available Nitrogen Release from Manures, Composts, and Specialty Products. Journal of Environment Quality, 35(6), 2321. doi:10.2134/jeq2006.0062 Gomiero, T., Pimentel, D., & Paoletti, M. G. (2011). Environmental impact of different agricultural management practices: Conventional vs. Organic agriculture. Critical Reviews in Plant Sciences, 30(1–2), 95–124. doi:10.1080/07352689.2011.554355   87 Government of British Columbia. (2012). 2011 Census of Agriculture: British Columbia highlights. https://www2.gov.bc.ca/assets/gov/farming-natural-resources-and-industry/agriculture-and-seafood/statistics/census/agricensus_2011_bchighlights.pdf Government of British Columbia. (2017). Agriculture in brief. https://www2.gov.bc.ca/assets/gov/farming-natural-resources-and-industry/agriculture-and-seafood/statistics/census/census-2016/aginbrief_2016_all_province_region_regional_districts.pdf Government of British Columbia. (2018). BC Soil Information Finder Tool. https://www2.gov.bc.ca/gov/content/environment/air-land-water/land/soil/soil-information-finder Government of British Columbia. (2019). Post-harvest nitrate testing. https://www2.gov.bc.ca/gov/content/industry/agriculture-seafood/agricultural-land-and-environment/soil-nutrients/nutrient-management/what-to-apply/soil-nutrient-testing/post-harvest-nitrate-testing#SoilSampling Government of British Columbia. (n.d.-a). Agricultural Environmental Management. Retrieved April 2, 2020, from https://www2.gov.bc.ca/gov/content/environment/waste-management/industrial-waste/agriculture Government of British Columbia. (n.d.-b). Field vegetables. Retrieved October 1, 2020, from https://www2.gov.bc.ca/gov/content/industry/agriculture-seafood/animals-and-crops/crop-production/field-vegetables Government of Canada. (2019). Past weather and climate – Historical data. https://climate.weather.gc.ca/historical_data/search_historic_data_e.html Gregorich, E. G., Carter, M. R., Angers, D. A., Monreal, C. M., & Ellert, B. H. (1994). Towards a minimum data set to assess soil organic matter quality in agricultural soils. Canadian Journal of Soil Science. Haney, R. L., Brinton, W. H., & Evans, E. (2008). Estimating soil carbon, nitrogen, and phosphorus mineralization from short-term carbon dioxide respiration. Communications in Soil Science and Plant Analysis, 39(17–18), 2706–2720. doi:.1080/00103620802358862 Haney, R. L., Hons, F. M., Sanderson, M. A., & Franzluebbers, A. J. (2001). A rapid procedure for estimating nitrogen mineralization in manured soil. Biology and Fertility of Soils, 33(2), 100–104. doi:10.1007/s003740000294 Hendershot, W. H., Lalande, H., & Duquette, M. (2008a). Electrical Conductivity and Exchangeable Acidity. In M. R. Carter & E. G. Gregorich (Eds.), Soil Sampling and Methods of Analysis (2nd ed., pp. 161–172). Canadian Society of Soil Science.   88 Hendershot, W. H., Lalande, H., & Duquette, M. (2008b). Soil Reaction and Exchangeable Acidity. In M. Carter & E. G. Gregorich (Eds.), Soil Sampling and Methods of Analysis (2nd ed., pp. 173–178). Canadian Society of Soil Science. Hargreaves, S. K., Dejong, P., Laing, K., McQuail, T., & Van Eerd, L. L. (2019). Management sensitivity, repeatability, and consistency of interpretation of soil health indicators on organic farms in southwestern Ontario. Canadian Journal of Soil Science, 99(4), 508–519. doi:10.1139/cjss-2019-0062 Hurisso, T. T., Culman, S. W., Horwath, W. R., Wade, J., Cass, D., Beniston, J. W., Bowles, T. M., Grandy, A. S., Franzluebbers, A. J., Schipanski, M. E., Lucas, S. T., & Ugarte, C. M. (2016). Comparison of Permanganate-Oxidizable Carbon and Mineralizable Carbon for Assessment of Organic Matter Stabilization and Mineralization. Soil Science Society of America Journal, 80(5), 1352–1364. doi:10.2136/sssaj2016.04.0106 Husson, F., Josse, J., Le, S., & Mazet, J. (2020). Package ‘FactoMineR’ (2.3). http://factominer.free.fr Idowu, O. J., Van Es, H. M., Abawi, G. S., Wolfe, D. W., Ball, J. I., Gugino, B. K., Moebius, B. N., Schindelbeck, R. R., & Bilgili, A. V. (2008). Farmer-oriented assessment of soil quality using field, laboratory, and VNIR spectroscopy methods. Plant and Soil, 307(1–2), 243–253. doi:10.1007/s11104-007-9521-0 Jerkins, D., & Ory, J. (2016). National organic research agenda. Organic Farming Research Foundation. https://ofrf.org/wp-content/uploads/2019/09/NORA_2016_final9_28.pdf Karam, A. (2008). Total element analysis (elements other than nitrogen, carbon, oxygen, and hydrogen). In M. R. Carter & E. G. Gregorich (Eds.), Soil Sampling and Methods of Analysis (2nd ed., pp. 335–340). Canadian Society of Soil Science. Kirkby, C. A., Richardson, A. E., Wade, L. J., Batten, G. D., Blanchard, C., & Kirkegaard, J. A. (2013). Carbon-nutrient stoichiometry to increase soil carbon sequestration. Soil Biology and Biochemistry, 60, 77–86. doi:10.1016/j.soilbio.2013.01.011 Kleinman, P. J. A., Sharpley, A. N., McDowell, R. W., Flaten, D. N., Buda, A. R., Tao, L., Bergstrom, L., & Zhu, Q. (2011). Managing agricultural phosphorus for water quality protection: Principles for progress. Plant and Soil, 349(1–2), 169–182. doi:10.1007/s11104-011-0832-9 Klonsky, K. (2012). Comparison of production costs and resource use for organic and conventional production systems. American Journal of Agricultural Economics, 94(2), 314–321. doi:10.1093/ajae/aar102 Kowalenko, C. G. (2010). Relationships between extraction methods for soil nutrient testing in British Columbia. https://www2.gov.bc.ca/assets/gov/farming-natural-resources-and-  89 industry/agriculture-and-seafood/agricultural-land-and-environment/soil-nutrients/nutrient-management/relationships_between_extraction_methods.pdf?bcgovtm=2free Kroetsch, D., & Wang, C. (2008). Partizle size distribution. In M. R. Carter & E. G. Gregorich (Eds.), Soil Sampling and Methods of Analysis (2nd ed., pp. 713–725). Taylor & Francis Group. Krzywinski, M., & Altman, N. (2014a). Points of significance: Analysis of variance and blocking. Nature Methods, 11(7), 699–700. doi:10.1038/nmeth.3005 Krzywinski, M., & Altman, N. (2014b). Visualizing samples with box plots. Nature Methods, 11(2), 119–120. doi:10.1038/nmeth.2813 Krzywinski, M., Altman, N., & Blainey, P. (2014). Points of Significance: Nested designs. Nature Methods, 11(10), 977–978. doi:10.1038/nmeth.3137 Lazicki, P., Geisseler, D., & Lloyd, M. (2020). Nitrogen mineralization from organic amendments is variable but predictable. Journal of Environmental Quality, 49(2), 483–495. doi:10.1002/jeq2.20030 Lenth, R. (2019). emmeans: Estimated Marginal Means, aka Least-Squares Means (R package version 1.4.3.01). Lynch, D. H., Zheng, Z., Zebarth, B. J., & Martin, R. C. (2008). Organic amendment effects on tuber yield, plant N uptake and soil mineral N under organic potato production. Renewable Agriculture and Food Systems, 23(3), 250–259. doi:10.1017/S1742170508002330 Main, M., Lynch, D. H., Paul Voroney, R., & Juurlink, S. (2013). Soil phosphorus effects on forage harvested and nitrogen fixation on Canadian organic dairy farms. Agronomy Journal, 105(3), 827–835. doi:10.2134/agronj2012.0500 Maltais-Landry, G., & Frossard, E. (2015). Similar phosphorus transfer from cover crop residues and water-soluble mineral fertilizer to soils and a subsequent crop. Plant and Soil, 393(1–2), 193–205. doi:10.1007/s11104-015-2477-6 Maltais-Landry, G., Scow, K., Brennan, E., Torbert, E., & Vitousek, P. (2016). Higher flexibility in input N: P ratios results in more balanced phosphorus budgets in two long-term experimental agroecosystems. Agriculture, Ecosystems and Environment, 223, 197–210. doi:10.1016/j.agee.2016.03.007 Maltais-Landry, G., Scow, K., Brennan, E., & Vitousek, P. (2015). Long-term effects of compost and cover crops on soil phosphorus in two California agroecosystems. Soil Science Society of America Journal, 79(2), 688–697. doi:.2136/sssaj2014.09.0369   90 Maltais-Landry, G., & Crews, T. E. (2020). Hybrid systems combining manures and cover crops can substantially reduce nitrogen fertilizer use. Agroecology and Sustainable Food Systems, 44(6), 683–695. doi:10.1080/21683565.2019.1649785 Maltais-Landry, G., Nesic, Z., Grant, N., Godinez, M., Thompson, B., Hsu, L. Y., & Smukler, S. M. (2019a). Quantifying trade-offs among on-farm and off-farm fertility sources to make vegetable organic farming systems more sustainable. Agriculture, Ecosystems and Environment, 286(May), 106657. doi:10.1016/j.agee.2019.106657 Maltais-Landry, G., Nesic, Z., Grant, N., Thompson, B., & Smukler, S. M. (2019b). Greater nitrogen availability, nitrous oxide emissions, and vegetable yields with fall-applied chicken relative to horse manure. Agronomy, 9(8). doi:.3390/agronomy9080444 Martin, R. C., Lynch, D. H., Frick, B., & van Straaten, P. (2007). Phosphorus status on Canadian organic farms. Journal of the Science of Food and Agriculture, 87, 2737–2740. doi:10.1002/jsfa.3077 Maynard, D. G., Kalra, Y. P., & Crumbaugh, J. A. (2008). Nitrate and Exchangeable Ammonium Nitrogen. In M. R. Carter & E. G. Gregorich (Eds.), Soil Sampling and Methods of Analysis (2nd ed., pp. 71–80). Canadian Society of Soil Science. McGowen, E. B., Sharma, S., Deng, S., Zhang, H., & Warren, J. G. (2018). An Automated Laboratory Method for Measuring CO 2 Emissions from Soils . Agricultural & Environmental Letters, 3(1), 180008. doi:10.2134/ael2018.02.0008 Miller, K., Aegerter, B. J., Clark, N. E., Leinfelder-Miles, M., Miyao, E. M., Smith, R., Wilson, R., & Geisseler, D. (2018). Relationship between soil properties and nitrogen mineralization in undisturbed soil cores from California Agroecosystems. Communications in Soil Science and Plant Analysis, 50(1), 77–92. doi:10.1080/00103624.2018.1554668 Mkhabela, M. S., & Warman, P. R. (2005). The influence of municipal solid waste compost on yield, soil phosphorus availability and uptake by two vegetable crops grown in a Pugwash sandy loam soil in Nova Scotia. Agriculture, Ecosystems and Environment, 106(1), 57–67. doi:10.1016/j.agee.2004.07.014 Moore, V. M., Mitchell, P. D., Silva, E. M., & Barham, B. L. (2016). Cover crop adoption and intensity on Wisconsin’s organic vegetable farms. Agroecology and Sustainable Food Systems, 40(7), 693–713. doi:10.1080/21683565.2016.1181694 Möller, K. (2018). Soil fertility status and nutrient input–output flows of specialised organic cropping systems: a review. Nutrient Cycling in Agroecosystems, 112(2), 147–164. doi:10.1007/s10705-018-9946-2   91 Morrow, J. G., Huggins, D. R., Carpenter-Boggs, L. A., & Reganold, J. P. (2016). Evaluating Measures to Assess Soil Health in Long-Term Agroecosystem Trials. Soil Science Society of America Journal, 80(2), 450–462. doi:.2136/sssaj2015.08.0308 Nakagawa, S., & Schielzeth, H. (2013). A general and simple method for obtaining R2 from generalized linear mixed-effects models. Methods in Ecology and Evolution, 4(2), 133–142. doi:10.1111/j.2041-210x.2012.00261.x Nakazawa, M. (2019). Package ‘fmsb’ (0.7.0). http://minato.sip21c.org/msb/ Nelson, N. O., & Janke, R. R. (2007). Phosphorus sources and management in organic production systems. HortTechnology, 17(4), 442–454. doi:10.21273/horttech.17.4.442 Ninh, H. T., Grandy, A. S., Wickings, K., Snapp, S. S., Kirk, W., & Hao, J. (2014). Organic amendment effects on potato productivity and quality are related to soil microbial activity. Plant and Soil, 386(1–2), 223–236. doi:10.1007/s11104-014-2223-5 Norris, C. E., & Congreves, K. A. (2018). Alternative management practices improve soil health indices in intensive vegetable cropping systems: A review. Frontiers in Environmental Science, 6(JUN). doi:10.3389/fenvs.2018.00050 Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., & R Core Team. (2019). _nlme: Linear and Nonlinear Mixed Effects Models_ (R package version 3.1-143). url: https://CRAN.R-project.org/package=nlme Poepleau, C., & Don, A. (2015). Carbon sequestration in agricultural soils via cultivation of cover crops - A meta analysis. Agriculture Ecosystems & Environment, 200, 33–41. doi:10.1016/j.agee.2014.10.024 Poon, D., & Schmidt, O. (2010b). Phosphorus considerations for nutrient management: Nutrient management factsheet - No. 6 in series. Government of British Columbia. https://www2.gov.bc.ca/assets/gov/farming-natural-resources-and-industry/agriculture-and-seafood/agricultural-land-and-environment/soil-nutrients/nutrient-management/631500-4_phosphorus_considerations_factsheet_no6_sep2010.pdf R Core Team. (2019). R: A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.r-project.org/ Reiter, M. S., Rideout, S. L., & Freeman, J. H. (2012). Nitrogen Fertilizer and Growth Regulator Impacts on Tuber Deformity, Rot, and Yield for Russet Potatoes. International Journal of Agronomy, 2012, 1–7. doi:10.1155/2012/348754 Schneider, K. D., Cade-Menun, B. J., Lynch, D. H., & Voroney, R. P. (2016). Soil Phosphorus Forms from Organic and Conventional Forage Fields. Soil Science Society of America Journal, 80(2), 328–340. doi:10.2136/sssaj2015.09.0340   92 Schneider, K. D., Lynch, D. H., Bünemann, E. K., & Voroney, R. P. (2017). Vegetative composition, arbuscular mycorrhizal fungi root colonization, and biological nitrogen fixation distinguish organic and conventional perennial forage systems. Agronomy Journal, 109(4), 1697–1706. doi:10.2134/agronj2016.12.0700 Schneider, K. D., Thiessen Martens, J. R., Zvomuya, F., Reid, D. K., Fraser, T. D., Lynch, D. H., O’Halloran, I. P., & Wilson, H. F. (2019). Options for Improved Phosphorus Cycling and Use in Agriculture at the Field and Regional Scales. Journal of Environmental Quality, 48(5), 1247–1264. doi:10.2134/jeq2019.02.0070 Seufert, V., Ramankutty, N., & Foley, J. A. (2012). Comparing the yields of organic and conventional agriculture. Nature, 485(7397), 229–232. doi:10.1038/nature11069 Seufert, V., Ramankutty, N., & Foley, J. A. (2012). Comparing the yields of organic and conventional agriculture. Nature, 485, 229–234. doi:10.1038/nature11069 Sharifi, M., Lynch, D. H., Hammermeister, A., Burton, D. L., & Messiga, A. J. (2014). Effect of green manure and supplemental fertility amendments on selected soil quality parameters in an organic potato rotation in Eastern Canada. Nutrient Cycling in Agroecosystems, 100(2), 135–146. doi:10.1007/s10705-014-9633-x Silva, E. M., & Moore, V. M. (2017). Cover crops as an agroecological practice on organic vegetable farms in Wisconsin, USA. Sustainability (Switzerland), 9(1). doi:10.3390/su9010055 Skalar Analytical. (2019). Primacs SNC-100 TN/TC Analyzer Manual (Skalar Analytical (ed.)). Smukler, S. M., Lewis, D., Robinson, C., & Mcconville, N. (2015). Southwest British Columbia Crop and Manure Nutrients A Review of Current Data Availability and Need. https://www2.gov.bc.ca/assets/gov/farming-natural-resources-and-industry/agriculture-and-seafood/agricultural-land-and-environment/soil-nutrients/nutrient-management/bc_manure_and_crop_nutrients_report.pdf?bcgovtm=buffer Statistics Canada. (2017). Growing opportunity through innovation in agriculture. https://www150.statcan.gc.ca/n1/pub/95-640-x/2016001/article/14816-eng.htm Statistics Canada. (2020). Experimental estimates of organic fruit and vegetable production, 2019. https://www150.statcan.gc.ca/n1/daily-quotidien/200715/dq200715c-eng.htm Sullivan, C. S., & Poon, D. (2012). Fraser Valley Soil Nutrient Survey 2012. https://www2.gov.bc.ca/assets/gov/farming-natural-resources-and-industry/agriculture-and-seafood/agricultural-land-and-environment/soil-nutrients/600-series/fvsns2012_report-aug2016.pdf?bcgovtm=buffer   93 Sullivan, D. M., Andrews, N., Sullivan, C., Brewer, L. J., Sullivan, D., & Sullivan, D. M. (2019). OSU Organic Fertilizer & Cover Crop Calculator: Predicting Plant-available Nitrogen Why use the OSU calculator? OREGON STATE UNIVERSIT Y EXTENSION SERVICE (Issue May). www.oregon.gov/oda/programs/pesticides/ Thermo Fisher Scientific. (2010). Flash 2000 Elemental Analyzer Operating Manual (Thermo Fisher Scientific (ed.)). Thermo Fisher Scientific. Throop, H. L., Archer, S. R., Monger, H. C., & Waltman, S. (2012). When bulk density methods matter: Implications for estimating soil organic carbon pools in rocky soils. Journal of Arid Environments, 77(1), 66–71. doi:10.1016/j.jaridenv.2011.08.020 Tuomisto, H. L., Hodge, I. D., Riordan, P., & Macdonald, D. W. (2012). Does organic farming reduce environmental impacts? - A meta-analysis of European research. Journal of Environmental Management, 112(834), 309–320. doi:10.1016/j.jenvman.2012.08.018 Wade, J., Culman, S. W., Hurisso, T. T., Miller, R. O., Baker, L., & Horwath, W. R. (2018). Sources of Variability that Compromise Mineralizable Carbon as a Soil Health Indicator. Soil Science Society of America Journal, 82(1), 243. doi:10.2136/sssaj2017.03.0105 Wade, J., Culman, S. W., Sharma, S., Mann, M., Demyan, M. S., Mercer, K. L., & Basta, N. T. (2019). How does phosphorus restriction impact soil health parameters in midwestern corn–soybean systems? Agronomy Journal, 111(4), 1682–1692. doi:10.2134/agronj2018.11.0739 Washington State University. (n.d.). Calculating compost bulk density. Retrieved April 5, 2020, from https://puyallup.wsu.edu/soils/bulkdensity/ Wassenaar, L. (1995). Evaluation of the origin and fate of nitrate in the Abbotsford Aquifer using the isotopes of 15N and 18O in NO3. Applied Geochemistry, 10, 391–405. doi:10.1016/0883-2927(95)00013-A Watson C.A.*, Bengtsson, H., Ebbesvik, M., Løes, A.-K., Myrbeck, A., Salomon, E., Schroder, J., & Stockdale, E. A. (2002). A review of farm-scale nutrient budgets for organic farms as a tool for management of soil fertility. Soil Use and Management, 18(3), 264–273. doi:10.1079/sum2002127 Weil, R. R., Stine, M. A., Islam, K. R., Gruver, J. B., & Samson-Liebig, S. E. (2003). Estimating active carbon for soil quality assessment : A simplified method for laboratory and field use Estimating active carbon for soil quality assessment : A simplified method for laboratory and field use. American Journal of Alternative Agriculture, 18(1), 3–17. doi:10.1079/AJAA200228 White, K. E., Brennan, E. B., Cavigelli, M. A., & Smith, R. F. (2020). Winter cover crops increase readily decomposable soil carbon, but compost drives total soil carbon during   94 eight years of intensive, organic vegetable production in California. PLoS ONE, 15(2), 1–26. doi:10.1371/journal.pone.0228677 Wilson, C., Zebarth, B. J., Burton, D. L., & Goyer, C. (2018). Short-Term Effects of Diverse Compost Products on Soil Quality in Potato Production. Soil Science Society of America Journal, 82(4), 889–900. doi:10.2136/sssaj2017.10.0345 Zhang, H., Hunt, D. E., & Bittman, S. (2019). Animal-Based Organic Amendments and Their Potential for Excessive Nitrogen Leaching and Phosphorus Loading. Agronomy Journal. doi:10.2134/agronj2018.12.0812 Ziadi, N., & Sen Tran, T. (2008). Mehlich 3-Extractable Elements. In M. R. Carter & E. G. Gregorich (Eds.), Soil Sampling and Methods of Analysis (2nd ed., pp. 81–88). Canadian Society of Soil Science.      95 Appendices Appendix A  - Datasets Table A 1 List of all farms by identifier (ID) that participated in the project and which datasets they contributed data to. Numbers (1 or 2) indicate the number of years a plot was used for, while a dash (-) indicates data from that plot was not included in a given data set. Datasets include both the linear mixed effect model (LME) and the PCA analysis for the outcome variables Yield, POx-C, post-season residual available N (NH4- and NO3- (0-15 and 15-30 cm), and NO3- (0-30 cm) depth for LME and NO3- (0-30 cm) for PCA) and P, and economics (input costs ($ kg-1)for LME and cost per yield for PCA).  ID Year Yield POx-C N P Costs Notes LME PCA LME PCA LME PCA LME PCA LME PCA 35 2018 1 Y 1 X 1 N 1 P 1 C  36 2018 - * - * 1 X 1 N 1 P 1 - * * Different plant spacing among plots so yields not measured. 36 2019 - - - - - - - - - - Crop failure. 37 2019 1 Y 1 X 1 N 1 P 1 C  38 2018 1 - ** 1 - ** 1 - ** 1 - ** - * - ** * Non-profit organization; input decisions based on donations, not management & economics. ** Excluded from PCA because of high SOM and high sand content. 38 2019 2 - ** 2 - ** 2 - ** 2 - ** - * - ** * Non-profit organization; and input decisions based on donations, not management & economics. ** Excluded from PCA because of high SOM and high sand content. 39 2018 1 Y 1 X 1 N 1 P 1 C  39 2019 2 Y 2 X 2 N 2 P 2 C  40 2018 1 Y 1 X 1 N 1 P 1 C  402 2019 1 Y 1 X 1 N 1 P 1 C  41 2018 - - - - - - - - - - Crop failure. 412 2019 1 Y 1 X 1 N 1 P 1 C  43 2018 1 Y 1 X 1 N 1 P 1 C  432 2019 1 Y 1 X 1 N 1 P 1 C  45 2018 - * - * 1 X 1 N 1 P 1 - * Yield data not collected. 45 2019 2 Y 2 X 2 N 2 P 2 C    96 ID Year Yield POx-C N P Costs Notes LME PCA LME PCA LME PCA LME PCA LME PCA 46 2018 1 Y 1 X 1 N 1 P 1 C  46 2019 2 Y 2 X 2 N 2 P 2 C  48 2018 1 Y 1 X 1 N 1 P 1 C  48 2019 - - - - - - - - - - Poor & uneven germination; data excluded. 49 2018 1 Y 1 X 1 N - * - 1 C * Available P data not included; excessive heterogeneity among plots in pre-season soil P. 49 2019 2 Y 2 X 2 N - * - 2 C * Available P data not included; excessive heterogeneity among plots in pre-season soil P. 50 2018 1 Y 1 X 1 N 1 P - * - * * On-farm compost does not reflect commercial costs. 50 2019 2 Y 2 X 2 N 2 P 2 C  51 2018 1 Y 1 X 1 N 1 P 1 C  51 2019 2 Y 2 X 2 N 2 P 2 C  53 2018 1 Y 1 X 1 N 1 P 1 C  53 2019 2 Y 2 X 2 N 2 P 2 C  54 2018 1 Y 1 X 1 N - * - * 1 C * No post-season available P data; sample was missed on first round of soil P analysis, and second round was not run due to COVID-19. 54 2019 2 Y 2 X 2 N 2 P 2 C  55 2018 - - - - - - - - - - Crop failure. 55 2019 1 ** Y 1 ** X - * - * - * - * 2 C * Plots were tilled and flags removed before post-season sampling. ** Plots moved to new location so only received only one year of application. However, pre-season sampling not re-done because of proximity to old plots. 56 2018 - * - * 1 X 1 N 1 P - ** - * Plots harvested by farm volunteers before I could harvest for crop yields. ** On-farm compost does not reflect commercial costs 57 2018 1 Y 1 - * 1 N 1 P - ** - ** * Excluded from because of low POx-C outlier. (Figure of PCA including site 57 to illustrate outlier effect is shown in Figure F 1 below) ** On-farm compost does not reflect commercial costs. Total 1 18 17 21 19 20 19 18 17 16 16  2 9 8 9 8 9 8 8 7 9 7  1&2 27 25 30 27 29 27 26 24 25 23    97 Appendix B  - Crop characteristics  Table B 1 Crop-specific target nutrient (N rate and P rate) application rates (kg ha-1) for each site and year, including plot sizes (m2), % N dry weight (N), % P dry weight (P), and data sources: (a) Smukler et al. 2015; (b) OSU Prod. Guide; (c) Maltais-Landry, 2019a; and (d) BC Nutrient Management Calculator.  ID year crop N N rate P P rate data source harvest date plot size n 35 2018 potato 1.42 85 0.42 14 a 13-Sept 44.8 6 36 2018 kohlrabi n/a 140 n/a 73 b - 8.2 - 2019 beet 1.48 105 0.223 16 c - 8.2 - 37 2019 potato 1.42 64 0.24 11 a 19-Aug 13.7 5 38 2018 beet 1.48 202 0.223 30 c 31-July 6.3 2 2019 carrot 1.51 97 0.28 18 a 13-Sept 6.3 3 39 2018 carrot 1.51 59 0.28 11 a 17-Aug 18.6 6 2019 potato 1.42 46 0.24 8 a 26-July 18.6 8 40 2018 potato 1.42 101 0.24 17 a 6-Aug 36.0 6 402 2019 potato 1.42 116 0.24 20 a 26-July 100.0 10 41 2018 beet n/a 164 n/a 22 d - 64.7 - 412 2019 potato 1.42 87 0.24 15 a 26-July 65.3 6 43 2018 potato 1.42 63 0.24 10 a 11-Sept 21.9 4 432 2019 beet 1.48 82 0.223 12 c 18-Sept 27.4 4 44 2018 carrot 1.51 40 0.28 7 a - 36.6 - 45 2018 broccoli n/a 170 n/a 29 d - 7.7 - 2019 beet 1.48 51 0.223 8 c 8-Aug 7.7 3   98 ID year crop N N rate P P rate data source harvest date plot size n 46 2018 beet n/a 215 n/a 28 d 11-Sept 43.9 6 2019 carrot 1.51 78 0.28 15 a 18-Sept 43.9 7 48 2018 carrot 1.51 87 0.28 16 a 9-Aug 13.7 5 2019 beet 1.48 103 0.22 16 c - 13.7 - 49 2018 beet n/a 100 n/a 13 d 23-Oct 53.4 4 2019 cabbage n/a 59 n/a 11 d 29-Oct 53.4 9 50 2018 beet n/a 151 n/a 20 d 9-Aug 7.0 2 2019 onion n/a 161 n/a 34 d 8-Sept 7.0 3 51 2018 potato 1.42 21 0.24 4 a 10-Aug 19.5 3 2019 beet 1.48 105 0.22 16 c 2-Aug 19.5 4 53 2018 carrot 1.51 43 0.28 8 a 2-Oct 38.1 4 2019 beet 1.48 131 0.22 20 c 2-Aug 38.1 5 54 2018 cucumber n/a 94 n/a 20 d n/a 25.6 n/a 2019 onion n/a 80 n/a 17 d 15-Aug 25.6 8 55 2018 potato 1.42 71 0.24 12 a - 13.9 - 2019 potato 1.42 76 0.24 13 a 24-July 13.5 8 56 2018 carrot 1.51 58 0.28 11 a - 6.8 - 57 2018 cauliflower n/a 181 n/a 27 d 7-Aug 15.7 4       99 Appendix C  - Soil properties  C.1 Baseline soil properties Table C 1 Spring baseline soil properties and sampling date by farm site identifier (ID) within region (REG) (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)) including percent sand, silt, clay, total nitrogen (N), total carbon (C), total inorganic carbon (IC), Mehlich-3 extractable phosphorus (M3P), Mehlich-3 extracted potassium (M3K), electrical conductivity (EC) and pH.  REG ID Sample Date Depth Texture Sand Silt Clay N C IC M3P M3K EC pH cm ----------------------- % ----------------------- --- mg kg-1 --- mS cm-1  ----------------------------------------------------------------------------------- 2018 ----------------------------------------------------------------------------------- FV 35 June 1 0-15 Silt Loam 3.7 69.1 27.1 0.1 1.3 0 204.3 232.7 0.3 6.8 15-30 Silt Loam 3.7 70.5 25.9 0.1 1.1 0 168.3 188.7 0.2 6.7 36 May 11 0-15 Silt Loam 12.9 65.4 21.7 0.1 1.3 0 77.6 223.7 0.5 6.3 15-30 Silt Loam 12.4 64.6 22.9 0.1 1.1 0 102.1 159.4 0.5 6.3 38 May 10 0-15 Sandy Loam 72 19 9.1 0.7 10 0.1 295.3 216.7 0.3 6.4 15-30 Sandy Loam 72 18.9 9 0.5 7.7 0.1 225.2 213.1 0.3 6.5 39 May 11 0-15 Silt Loam 15.3 62.4 22.3 0.6 7.4 0.2 127.5 420.8 0.2 6.4 15-30 Silt Loam 19.1 58.3 22.6 0.5 6.5 0.2 86.9 361.3 0.2 6.4 40 May 4 0-15 Silt Loam 8.9 64.7 26.4 0.2 2.4 0 200.8 318.3 0.2 5.4 15-30 Silt Loam 6.4 67.2 26.4 0.2 2.1 0 161.2 219.6 0.2 5.2 41 June 1 0-15 Loam 28.4 46.5 25.2 0.2 2.6 0 118.7 315.3 0.2 6.8 15-30 Loam 24.9 49 26 0.2 1.7 0 90.3 208 0.2 6.6 PV 43 April 30 0-15 Silty Clay Loam 5.9 67 27.1 0.4 5.1 0 51.4 160.7 0.2 5.6 15-30 Silty Clay Loam 4.6 65.9 29.5 0.3 3.5 0 22 75 0.2 5.2 45 April 30 0-15 Silt Loam 27.4 65.4 7.2 0.3 3.5 0.1 277.1 327.3 0.4 5.9 15-30 Silt Loam 25.9 66.5 7.6 0.2 2.2 0 60.7 193.3 0.1 4.9 46 May 1 0-15 Silt Loam 20.5 70.2 9.3 0.2 2.6 0 69.4 134 0.1 5.3 15-30 Silt Loam 18.9 71.8 9.3 0.1 1.6 0 14.2 69.9 0.1 5.2 57 May 1 0-15 Silt Loam 11 75.9 13.1 0.1 0.9 0 25.4 85.4 0.1 5.3   100 REG ID Sample Date Depth Texture Sand Silt Clay N C IC M3P M3K EC pH cm ----------------------- % ----------------------- --- mg kg-1 --- mS cm-1  15-30 Silt Loam 11.4 75.5 13.1 0.1 0.9 0 23.2 77 0.1 5.3  VI 48 April 24 0-15 Silt Loam 22.9 54.3 22.8 0.4 4.9 0 19.1 84.3 0.1 6 15-30 Silt Loam 23.6 52.4 24.1 0.2 3.1 0 7.8 22.1 0.1 5.6 49 May 15 0-15 Loam 45.1 47.1 7.8 0.3 4.4 0 66.3 148.7 0.1 6.5 15-30 Loam 48.7 42.8 8.6 0.1 1.8 0 19.7 83.3 0.1 6.4 50 April 24 0-15 Loam 46.8 35.6 17.6 0.4 4.9 0 4.5 50.7 0.1 5.7 15-30 Loam 45.6 35.5 18.8 0.3 3.2 0 4 30.3 0.1 5.9 51 April 25 0-15 Clay Loam 29.9 43.9 26.2 0.3 4 0 43.9 205.5 0.1 5.8 15-30 Clay Loam 35.5 44.4 20.1 0.2 2 0 22 90.8 0.1 5.8 53 April 25 0-15 Silt Loam 26.1 56 17.9 0.3 4.5 0 130.1 338.1 0.2 6 15-30 Silt Loam 25.2 56.1 18.7 0.1 1.5 0 42.7 123.1 0.1 5.7 54 April 23 0-15 Loam 47.2 35.4 17.5 0.2 3.4 0.1 5 46 0.3 6.9 15-30 Loam 46.1 35.2 18.7 0.1 1.6 0 4.3 19.7 0.1 5.8 55 May 24 0-15 Sandy Loam 68.4 20.9 10.7 0.3 4.4 0 51 25.7 0.2 4.7 15-30 Sandy Loam 69.3 20.5 10.2 0.3 4 0 38.1 19.6 0.1 4.6 56 April 24 0-15 Loam 46.5 36.5 17 0.4 5.7 0.1 65.7 160.6 0.2 6.5 15-30 Loam 45.2 38.4 16.4 0.3 4.5 0.1 18.4 112.2 0.1 6.2 ----------------------------------------------------------------------------------- 2019 ----------------------------------------------------------------------------------- FV 402 May 10 0-15 Silty Clay Loam 6.1 63 31 0.2 3.4 0.1 320 283.3 0.1 6 15-30 Silty Clay Loam 6.1 63 31 0.2 3 0.1 256.7 206.7 0.1 5.8 412 April 24 0-15 Silt Loam 11.3 63.3 25.3 0 2.5 0.1 97 270 0.2 7.1 15-30 Silt Loam 12.3 62.7 25.3 0 2.3 0.1 92 240 0.2 6.9 PV 37 April 29 0-15 Silt Loam 29.3 64 7.2 0.2 2.1 0.1 108.3 163.3 0.1 7.2 15-30 Silt Loam 30.7 63.3 5.9 0.1 1 0.1 28.7 103 0.1 6.9 432 April 30 0-15 Silt Loam 15 70.7 14.3 0.2 1.5 0.1 10.5 110 0.1 5.9 15-30 Silt Loam 17.7 69.3 13 0.1 1 0.1 7 60 0.1 6.1    101 C.2 General soil properties Table C 2 Physical soil properties by farm site identifier (ID) within region (REG) (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)) from the BC Soil Information Finder Tool (SIFT) (Government of British Columbia, 2018)  REG Site ID SIFT ID Primary Soil Name Texture Drainage CF PM Classification FV 35 74307 Westham Silt Loam Poorly Drained 0 Fluvial Rego Humic Gleysol 41/412 77852 Ladner Silt Loam Poorly Drained 0 Fluvial Humic Luvic Gleysol 36 77249 Ladner Silt Loam Poorly Drained 0 Fluvial Humic Luvic Gleysol 40 74722 Spetifore Silty Clay Loam Poorly Drained 1 Fluvial Rego Humic Gleysol 402 77669 Westham Silt Loam Poorly Drained 0 Fluvial Rego Humic Gleysol 39 72667 Hjorth Silt Loam Poorly Drained 0 Fluvial Orthic Humic Gleysol PV 44 80609 Sankey Silty Clay Loam Poorly Drained 0 Fluvial Rego Gleysol 45 80075 Sankey Silty Clay Loam Poorly Drained 0 Fluvial Rego Gleysol 43 80077 Sankey Silty Clay Loam Poorly Drained 0 Fluvial Rego Gleysol 432 80253 Sankey Silty Clay Loam Poorly Drained 0 Fluvial Rego Gleysol 37 80231 Wolverine Loam Imperfectly Drained 0 Fluvial Gleyed Regosol 46 80253 Sankey Silty Clay Loam Poorly Drained 0 Fluvial Rego Gleysol 57 79853 Shantz Loam Poorly Drained 0 Fluvial Rego Gleysol VI 51 41345 Saanichton Silt Loam Imperfectly Drained 10 Marine Orthic Sombric Brunisol   102 REG Site ID SIFT ID Primary Soil Name Texture Drainage CF PM Classification 53 110260 Tagner Silt Loam Poorly Drained 0 Marine Orthic Humic Gleysol 49 41432 Mill Bay (70) Silt Loam Moderately Well Drained 10 Marine Duric Dystric Brunsiol 54 126773 Fairbridge (100) Silt Loam Imperfectly Drained 0 Marine Gleyed Eluviated Dystric Brunisol 50 122348 Cowichan (70) Silt Loam Poorly Drained 0 Marine Humic Luvic Gleysol 55 42811 Baynes (80) Loamy Sand Imperfectly Drained 38 Fluvial GleyedDystric Brunisol 48 43061 Pluntledge Loam Imperfectly Drained 0 Fluvial Gleyed Sombric Brunisol 56 43018 Suffolk Silt Loam Imperfectly Drained 24 Marine Gleyed Dystric Brunisol                103 C.3 Bulk density Table C 3 Soil bulk density (BD, g cm-3), using three calculation methods, and coarse fragment content (CF, %) by volume (vol) and by weight (wt.) for each farm site, by farm site identifier (ID) with region (REG) (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)).   BD CF REG ID Depth (cm) standard adjusted hybrid vol. wt. FV 38 0-15 0.67 0.57 0.55 4.64 18.4 38 15-30 0.98 0.82 0.75 8.73 23.5 39 0-15 0.69 0.69 0.68 0.26 1.0 39 15-30 0.79 0.79 0.79 0.05 0.2 402 0-15 1.13 1.10 1.08 1.63 3.8 402 15-30 1.20 1.20 1.19 0.39 0.9 412 0-15 1.21 1.19 1.18 1.10 2.5 412 15-30 1.33 1.33 1.32 0.39 0.8 PV 37 0-15 1.09 1.09 1.08 0.29 0.7 37 15-30 1.32 1.32 1.32 0.00 0.0 45 0-15 0.86 0.86 0.85 0.34 1.1 45 15-30 1.19 1.19 1.18 0.22 0.5 46 0-15 1.04 1.04 1.04 0.10 0.3 46 15-30 1.33 1.33 1.32 0.05 0.1 432 0-15 1.10 1.10 1.10 0.12 0.3 432 15-30 1.23 1.23 1.23 0.05 0.1 VI 48 0-15 0.77 0.77 0.76 0.42 1.5 48 15-30 0.98 0.98 0.98 0.10 0.3 49 0-15 1.13 0.73 0.58 20.64 48.5 49 15-30 1.44 1.00 0.74 26.51 48.8   104  BD CF REG ID Depth (cm) standard adjusted hybrid vol. wt. 50 0-15 0.96 0.91 0.87 3.33 9.2 50 15-30 1.04 0.99 0.96 3.04 7.8 51 0-15 1.01 0.85 0.77 8.87 23.4 51 15-30 1.16 0.98 0.87 11.01 24.8 53 0-15 0.58 0.51 0.49 3.25 14.9 53 15-30 1.24 1.12 1.04 7.59 16.2 54 0-15 1.00 0.88 0.82 6.90 18.5 54 15-30 1.32 1.27 1.23 3.67 7.3 55 0-15 0.82 0.79 0.78 1.85 6.3 55 15-30 1.19 1.17 1.15 1.48 3.3                    105 Appendix D  - Amendments D.1 Composts Table D 1 Compost properties by farm site (ID) within region (REG) (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)) including total nitrogen (N), carbon (C), phosphorus (P) and potassium (K), ammonium (NH4+), nitrate (NO3-), estimated plant available nitrogen (PAN), carbon to nitrogen ratio (C:N), nitrogen to phosphorus ratio (N:P), moisture content (MC), electrical conductivity (EC) and pH. Sample date (SMPL) and application date (APPL) are also shown.        BD C P K N NH4+ NO3- PAN C:N N:P MC EC g:mL pH g:mL REG ID SMPL APP type kg m-3 ----------- % ----------- ---- mg kg-1 ---- % - - % - - - - ------------------------------------------------------------------------------------ 2018 ------------------------------------------------------------------------------------ FV 35 27-Apr 01-Jun poultry litter  43.2 1.8 2.0 4.5 7851 2 29.8 9.6 2.5 45.6 16.7 5:20 6.0 5:20 36 18-Apr 11-May food / yard / manure  31.6 0.6 1.2 2.8 1761 291 21.2 11.2 4.6 63.0 6.1 5:20 7.1 10:30 38 16-Apr 10-May food / yard / manure  35.7 0.7 0.4 3.3 323 13 15.9 10.9 4.5 65.8 1.7 5:20 7.5 5:20 39 16-Apr 11-May pig manure  29.5 0.7 1.5 1.2 9 756 20.3 24.2 1.7 68.8 9.0 5:20 7.0 5:15 40 18-Apr 27-Apr food / yard  21.1 0.3 0.4 1.4 6 73 15.5 14.7 5.3 53.0 0.7 5:20 7.4 10:25 40 18-Apr 27-Apr 50:50 food scrap broiler litter  29.0 0.9 1.0 2.3 2659 37 24.8 12.4 2.6 56.5 3.7 5:20 7.4 10:25 40 18-Apr 27-Apr poultry litter  45.9 1.5 2.3 3.9 20905 74 60.9 11.8 2.5 59.0 21.4 5:20 6.0 5:25 41 18-Apr 14-Jun poultry litter  39.5 1.1 1.4 2.1 6522 0 41.2 18.7 2.0 68.1 8.3 5:20 6.9 5:25 PV 43 30-Apr n/a poultry litter  37.5 3.4 0.6 2.9 306 1728 20.9 12.7 0.9 65.4 5.2 5:20 6.5 10:10 43 18-May 18-May food / yard  24.5 0.5 0.9 1.7 1853 0 24.0 14.1 3.7 38.8 4.1 5:20 6.8 5:10   106      BD C P K N NH4+ NO3- PAN C:N N:P MC EC g:mL pH g:mL REG ID SMPL APP type kg m-3 ----------- % ----------- ---- mg kg-1 ---- % - - % - - - - 44 01-May 12-Jun food / yard  15.9 0.4 0.7 1.3 556 158 19.6 12.1 3.2 48.4 5.8 5:20 7.6 5:15 45 30-Apr 12-Jun food / yard  20.5 0.5 0.8 1.8 433 335 18.6 11.4 4.0 50.1 3.0 5:20 7.6 10:10 46 01-May 12-Jun poultry litter  41.2 2.3 2.1 4.4 15299 7 44.5 9.3 1.9 52.3 17.2 5:20 5.9 5:20 47 01-May n/a food / yard  28.5 0.7 0.8 2.6 82 890 18.1 10.8 4.0 51.2 4.7 5:20 7.2 5:15 57 01-May 18-May on farm  8.3 0.2 0.5 0.6 1 104 16.4 13.3 3.7 44.5 1.5 5:20 6.8 10:10 VI 48 24-Apr 16-May steer / horse  22.6 0.4 0.4 1.1 1 189 16.4 20.2 2.7 57.2 1.8 5:20 7.1 5:15 49 15-May 07-Jun steer / horse  34.5 0.6 0.5 2.2 15 111 15.5 15.5 3.6 78.2 2.0 5:20 7.1 5:26 50 24-Apr 24-May on farm  6.1 0.1 0.2 0.5 0 0 15.0 12.3 4.1 31.8 0.2 10:20 6.1 10:10 51 25-Apr n/a on farm  30.0 1.1 1.5 2.0 1408 1723 28.4 15.1 1.8 61.4 8.8 5:20 6.7 5:15 51 15-May 15-May steer / horse  14.5 0.3 0.8 0.9 337 132 19.7 17.1 2.5 51.2 2.0 5:20 7.8 5:12 53 25-Apr 15-May steer / horse  38.9 0.3 1.1 1.0 1394 40 27.3 39.4 3.8 64.6 4.0 5:20 7.9 2.5:10 54 23-Apr 30-May fish / forest  35.7 0.3 0.1 1.1 1 77 15.6 31.4 4.3 54.2 0.9 5:20 5.0 5:10 55 16-May 24-May fish / forest  24.7 0.6 0.4 2.0 1062 878 23.2 12.3 3.6 53.2 3.9 5:20 4.9 5:13 56 16-May 24-May on farm  14.8 0.4 1.0 1.2 10 799 21.0 12.8 2.7 53.7 5.8 5:20 7.2 5:11 ------------------------------------------------------------------------------------ 2019 ------------------------------------------------------------------------------------ FV 35 01-May n/a poultry litter 255 38.0 0.3 0.2 1.4 8400 3 66.0 27.1 4.4 57.0     36 18-Jun 02-Jul fish / forest 375 28.0 0.3 0.2 1.1 10 740 20.8 25.5 4.4 56.0       107      BD C P K N NH4+ NO3- PAN C:N N:P MC EC g:mL pH g:mL REG ID SMPL APP type kg m-3 ----------- % ----------- ---- mg kg-1 ---- % - - % - - - - 37 30-Apr 03-Jun food / yard 488 28.0 0.5 0.8 2.4 670 690 19.8 11.7 5.2 46.0     38 22-May 19-Jun food / yard / manure 467 23.0 0.5 1.2 2.2 2200 280 24.6 10.5 4.3 53.0     39 27-Mar 19-Apr pig manure 129 44.0 1.3 3.5 1.7 220 3000 31.1 25.9 1.3 76.0     402 24-Apr 10-May poultry litter 316 37.0 2.2 2.5 2.9 9400 210 43.2 12.8 1.3 66.0     412 24-Apr 22-May poultry litter 433 24.0 1.2 1.4 2.2 4600 5 32.8 10.9 1.8 59.0     PV 432 07-May 11-Jun food / yard 438 27.0 0.4 0.9 2.2 500 370 18.4 12.3 5.1 43.0     45 30-Apr 03-Jun food / yard 550 25.0 0.5 1.1 2.0 35 470 17.1 12.5 3.8 47.0     46 30-Apr 26-Jun food / yard 415 23.0 1.2 0.8 2.6 53 3300 26.0 8.8 2.2 41.0     57 30-Apr 15-Jun on farm 818 4.3 0.2 0.5 0.4 1 190 19.4 11.6 2.3 34.0     VI 48 01-May 21-May fish / forest 512 39.0 1.4 2.1 3.7 5200 2 27.0 10.5 2.6 40.0     49 16-Apr 12-Jun steer / horse 224 26.0 0.6 0.8 1.4 3 200 16.2 18.6 2.2 73.0     50 15-Apr 21-May on farm 964 5.1 0.1 0.3 0.5 0 10 15.2 11.3 3.8 30.0     51 17-Apr 21-May on farm 376 27.0 1.5 2.5 2.6 3400 600 28.1 10.4 1.7 40.0     53 17-Apr 02-May steer / horse 198 43.0 0.3 1.0 0.9 46 170 17.0 47.8 3.0 70.0     54 16-Apr 08-May fish / forest 337 38.0 0.2 0.4 0.9 4 180 16.8 42.7 3.7 56.0     55 16-Apr 02-May offal n/a 37.0 1.2 0.4 1.6 190 75 16.4 23.1 1.3 64.0        108  D.2 Specialty fertilizers Table D 2 Nutrient content of the specialty fertilizers used in the low compost + N plots and typical plots by farm site (ID) within regions (REG) (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)). From the product label guaranteed analysis for total nitrogen (N), total phosphorus (P), and total potassium (K).   REG ID Year Description N P K  ---------------------------------------------------------- Low Compost + N ---------------------------------------------------------- FV/PV all both Natures Intent feathermeal (11-0-0) 11.0 0 0 VI all both Gaia Green feathermeal (13-0-0) 13.0 0 0 ---------------------------------------------------------------- Typical ---------------------------------------------------------------- ISL 48 2018 blend (4-7.5-5) 4.0 3.3 4.2 2018 enterra (3-2-5) 3.0 0.9 4.2 2019 Hi-P (0-12-0) 0.0 5.2 0.0 2019 Bloodmeal (10-0-0) 10.0 0.0 0.0 49 2019 feathermeal (12-0-0) 12.0 0.0 0.0 50 2018 natures intent fish bonemeal 4-13-0 4.0 5.7 0.0 2018 alfalfa pellets 3-0.6-2.9 3.0 0.3 2.4 2019 pro-mix (7-3-3) 7.0 1.3 2.5 51 2018 natures intent fish bonemeal 4-13-0 4.0 5.7 0.0 53 2018 Bloodmeal (12-0-0) 12.0 0.0 0.0 2018 Hi-N (10-3-0) 10.0 1.3 0.0 2019 Hi-N (10-3-0) 10.0 1.3 0.0   109 REG ID Year Description N P K 2019 bloodmeal (12-0-0) 12.0 0.0 0.0 54 2018 natures intent fish bonemeal 4-13-0 4.0 5.7 0.0 2018 bat guano (0-12-0) 0.0 5.2 0.0 2018 alfalfa pellets (3-2-2) 3.0 0.9 1.7 2019 fish bonemeal (4-13-0) 4.0 5.7 0.0 2019 alfalfa pellets (3-2-2) 3.0 0.9 1.7 2019 guano (0-12-0) 12.0 0.0 0.0 55 2018 natures intent fish bonemeal 4-13-0 4.0 5.7 0.0 2018 rock phosphate (0-12-0) 0.0 5.2 0.0 2018 alfalfa pellets (3-0-2) 3.0 0.0 1.7 2019 Hi-N (10-3-0) 10.0 1.3 0.0 2019 Hi-P (0-12-0) 0.0 5.2 0.0 LFV 36 2018 bloodmeal (12-0-0) 12.0 0.0 0.0 38 2018 Enterra (3-2-5) 3.0 0.9 4.2 2019 Enterra (3-2-5) 3.0 0.9 4.2 PV 37 2019 Hi-N (10-3-0) 10.0 1.3 0.0 43 2018 Hi-P (0-12-0) 0.0 5.2 0.0 45 2018 Gaia Green mix (4-4-4) 4.0 1.7 3.3 2019 rock phosphate (0-3-0) 0.0 1.3 0.0 57 2018 mixed fertilizer (10-3-0) 10.0 1.3 0.0 2019 fertilizer blend (6-2-7) 6.0 0.9 5.8     110 D.3 Application rates Table D 3 Application rates for each farm. Application rates by amendment include the rate the amendment was applied at, as well as associated nutrients, including carbon (C), nitrogen (N), plant available N (PAN), and phosphorus (P) by amendment within each strategy, and the total nutrient application by strategy for N, PAN and P. The C application rate by strategy is taken as the C application rate from the compost product, if compost was used.       by amendment by strategy REG ID year strategy amendment rate  C N PAN P N PAN P ----- Mg ha-1 ----- -------------------- kg ha-1 -------------------- FV 35 2018 HC compost 8.0 3.5 362 108 145 362 108 145 LC compost 2.4 1.0 106 32 43 183 108 43 feathermeal (11-0-0) 0.7 0.0 76 76 0    TYP compost 11.2 4.9 507 151 203 507 151 203 36 2018 HC compost 23.5 7.4 662 140 145 662 140 145 LC compost 12.0 3.8 336 71 74 403 138 74 feathermeal (11-0-0) 0.6 0.0 67 67 0    TYP compost 2.6 0.8 73 15 16 97 39 16 bloodmeal (12-0-0) 0.2 0.0 24 24 0    38 2018 HC compost 38.6 13.8 1268 201 283 1268 201 283 LC compost 4.0 1.4 132 21 29 306 195 29 feathermeal (11-0-0) 1.6 0.0 175 175 0    TYP compost 44.5 15.9 1460 232 326 1608 379 369 Enterra (3-2-5) 4.9 0.0 148 148 43    2019 HC compost 18.0 4.1 396 97 92 396 97 92 LC compost 3.5 0.8 77 19 18 155 97 18 feathermeal (11-0-0) 0.7 0.0 78 78 0    TYP compost 58.4 13.4 1285 316 298 1499 530 360 enterra (3-2-5) 7.1 0.0 214 214 62      111      by amendment by strategy REG ID year strategy amendment rate  C N PAN P N PAN P ----- Mg ha-1 ----- -------------------- kg ha-1 -------------------- 39 2018 HC compost 23.9 7.0 291 59 170 291 59 170 LC compost 1.5 0.5 19 4 11 72 57 11 feathermeal (11-0-0) 0.5 0.0 53 53 0    TYP - 0.0 0.0 0 0 0 0 0 0 2019 HC compost 8.7 3.8 149 46 114 149 46 114 LC compost 0.6 0.3 10 3 8 53 46 8 feathermeal (11-0-0) 0.4 0.0 43 43 0    TYP - 0.0 0.0 0 0 0 0 0 0 40 2018 HC compost 11.1 5.1 432 263 171 432 263 171 LC compost 1.1 0.5 42 25 17 145 129 17 feathermeal (11-0-0) 0.9 0.0 103 103 0    TYP compost 9.9 4.6 386 235 153 386 235 153 2019 HC compost 9.2 3.4 267 115 203 267 115 203 LC compost 0.9 0.3 26 11 20 130 115 20 feathermeal (11-0-0) 0.9 0.0 104 104 0    TYP compost 8.2 3.0 239 103 181 239 103 181 41 2018 HC compost 17.0 6.7 359 148 184 359 148 184 LC compost 1.8 0.7 38 16 19 184 162 19 feathermeal (11-0-0) 1.3 0.0 146 146 0    TYP compost 4.1 1.6 86 36 44 86 36 44 2019 HC compost 12.1 2.9 266 87 145 266 87 145 LC compost 1.2 0.3 27 9 15 105 87 15 feathermeal (11-0-0) 0.7 0.0 78 78 0    TYP compost 5.9 1.4 130 42 71 130 42 71 PV 37 2019 HC compost 13.4 3.7 321 64 61 321 64 61   112      by amendment by strategy REG ID year strategy amendment rate  C N PAN P N PAN P ----- Mg ha-1 ----- -------------------- kg ha-1 -------------------- LC compost 2.3 0.7 56 11 11 108 64 11 feathermeal (11-0-0) 0.5 0.0 52 52 0    TYP Hi-N (10-3-0) 1.4 0.0 139 139 18 139 139 18 43 2018 HC compost 21.8 5.3 379 91 103 379 91 103 LC compost 2.6 0.6 45 11 12 101 66 12 feathermeal (11-0-0) 0.5 0.0 55 55 0    TYP compost 8.2 2.0 143 34 39 143 34 65 Hi-P (0-12-0) 0.5 0.0 0 0 27    2019 HC compost 20.4 5.5 448 82 88 448 82 88 LC compost 2.9 0.8 64 12 12 134 82 12 feathermeal (11-0-0) 0.6 0.0 71 71 0    TYP compost 14.6 4.0 322 59 63 322 59 63 45 2018 HC compost 51.1 10.5 916 171 231 916 171 231 L compost 6.4 1.3 115 21 29 258 164 29 feathermeal (11-0-0) 1.3 0.0 143 143 0    TYP compost 71.3 14.6 1279 238 322 1358 317 357 Gaia Green mix (4-4-4) 2.0 0.0 79 79 34    2019 HC compost 15.3 3.8 307 53 80 307 53 80 LC compost 1.5 0.4 30 5 8 77 52 8 feathermeal (11-0-0) 0.4 0.0 47 47 0    TYP compost 58.2 14.5 1163 199 302 1163 199 324 rock phosphate (0-3-0) 1.6 0.0 0 0 22    46 2018 HC compost 10.9 4.5 483 215 250 483 215 250 LC compost 1.3 0.5 56 25 29 246 215 29   113      by amendment by strategy REG ID year strategy amendment rate  C N PAN P N PAN P ----- Mg ha-1 ----- -------------------- kg ha-1 -------------------- feathermeal (11-0-0) 1.7 0.0 190 190 0    TYP compost 5.7 2.4 252 112 130 252 112 130 2019 HC compost 11.6 2.7 301 78 139 301 78 139 LC compost 1.2 0.3 31 8 15 102 78 15 feathermeal (11-0-0) 0.6 0.0 70 70 0    TYP compost 0.0 0.0 0 0 0 0 0 0  57 2018 HC compost 175.6 14.6 1102 181 300 1102 181 300 LC compost 16.1 1.3 101 17 27 291 207 27 feathermeal (13-0-0) 1.5 0.0 190 190 0    TYP mixed fert (10-3-0) 0.2 0.0 23 23 3 23 23 3 2019 HC compost 167.2 7.2 619 120 268 619 120 268 LC compost 13.9 0.6 51 10 22 171 130 22 feathermeal (12-0-0) 1.0 0.0 120 120 0    TYP fert blend (6-2-7) 0.5 0.0 27 27 4 27 27 4 VI  48 2018 HC compost 47.2 10.7 528 87 194 - - - LC compost 3.9 0.9 44 7 16 120 83 16 feathermeal (13-0-0) 0.6 0.0 76 76 0    TYP blend (4-7.5-5) 1.4 0.0 54 54 44 103 103 59 enterra (3-2-5) 1.6 0.0 49 49 14    2019 HC compost 10.3 4.0 381 103 144 - - - LC compost 1.1 0.4 41 11 15 135 106 15 feathermeal (13-0-0) 0.7 0.0 95 95 0    TYP ‘Hi-P’ (0-12-0) 0.6 0.0 0 0 30 80 80 30 bloodmeal (10-0-0) 0.8 0.0 80 80 0      114      by amendment by strategy REG ID year strategy amendment rate  C N PAN P N PAN P ----- Mg ha-1 ----- -------------------- kg ha-1 -------------------- 49 2018 HC compost 24.4 8.4 543 84 152 543 84 152 LC compost 2.1 0.7 47 7 13 137 97 13 feathermeal (13-0-0) 0.7 0.0 90 90 0    TYP compost 10.4 3.6 231 36 65 231 36 65 2019  HC compost 26.0 6.7 363 59 166 363 59 166 LC compost 1.7 0.4 24 4 11 80 60 11 feathermeal (13-0-0) 0.4 0.0 56 56 0    TYP compost 7.5 2.0 106 17 48 226 138 48 feathermeal (12-0-0) 1.0 0.0 121 121 0    50 2018 HC compost 201.2 12.3 1005 151 248 1005 151 248 LC compost 16.3 1.0 81 12 20 220 151 20 feathermeal (13-0-0) 1.1 0.0 139 139 0    TYP fish bonemeal (4-13-0) 0.8 0.0 34 34 48 92 92 53 alfalfa pellets (3-0.6-2.9) 2.0 0.0 59 59 5    2019  HC compost (sea soil) 11.9 3.3 131 27 30 131 27 30 LC compost (sea soil) 1.8 0.5 19 4 4 157 141 4 feathermeal (13-0-0) 1.1 0.0 137 137 0    TYP compost (sea soil) 7.4 2.1 82 17 19 288 170 62 compost (on-farm) 14.0 0.7 63 10 17    pro-mix (7-3-3) 2.0 0.0 143 143 27    51 2018 HC compost 4.4 1.3 87 25 48 87 25 48 LC compost 0.3 0.1 5 2 3 26 22 3 feathermeal (13-0-0) 0.2 0.0 20 20 0      115      by amendment by strategy REG ID year strategy amendment rate  C N PAN P N PAN P ----- Mg ha-1 ----- -------------------- kg ha-1 -------------------- TYP fish bonemeal (4-13-0) 0.6 0.0 25 25 35 25 25 35 2019 HC compost 8.6 2.3 225 63 130 225 63 130 LC compost 0.6 0.2 17 5 10 114 102 10 feathermeal (13-0-0) 0.8 0.0 98 98 0    TYP compost 14.7 4.0 381 107 220 381 107 220 53 2018 HC compost 16.0 6.2 159 43 41 159 43 41 LC compost 3.1 1.2 31 8 8 65 42 8 feathermeal (13-0-0) 0.3 0.0 34 34 0    TYP Bloodmeal (12-0-0) 1.3 0.0 157 157 0 289 289 17 Hi-N (10-3-0) 1.3 0.0 131 131 17    2019 HC compost 85.6 36.8 771 131 257 771 131 257 LC compost 6.6 2.8 59 10 20 180 131 20 feathermeal (13-0-0) 0.9 0.0 121 121 0    TYP compost 57.2 24.6 515 88 172 1116 689 207 Hi-N (10-3-0) 2.7 0.0 273 273 36    bloodmeal (12-0-0) 2.7 0.0 328 328 0    54 2018 HC compost 53.0 18.9 603 94 141 603 94 141 LC compost 7.3 2.6 83 13 20 165 94 20 feathermeal (13-0-0) 0.6 0.0 81 81 0    TYP compost 13.4 4.8 153 24 36 259 130 175 fish bonemeal (4-13-0) 1.1 0.0 43 43 61    bat guano (0-12-0) 1.2 0.0 0 0 61    alfalfa pellets (3-2-2) 2.1 0.0 64 64 19      116      by amendment by strategy REG ID year strategy amendment rate  C N PAN P N PAN P ----- Mg ha-1 ----- -------------------- kg ha-1 -------------------- 2019 HC compost 53.7 20.4 478 80 129 478 80 129 LC compost 7.0 2.7 62 10 17 133 82 17 feathermeal (13-0-0) 0.5 0.0 71 71 0    TYP compost 18.2 6.9 162 27 44 381 246 82 fish bonemeal (4-13-0) 0.4 0.0 14 14 20    alfalfa pellets (3-2-2) 2.1 0.0 64 64 19    guano (0-12-0) 1.2 0.0 141 141 0    55 2018 HC compost 23.1 5.7 462 107 130 462 107 130 LC compost 2.2 0.5 43 10 12 109 76 12 feathermeal (13-0-0) 0.5 0.0 66 66 0    TYP compost 10.4 2.6 208 48 58 251 92 126 fish bonemeal (4-13-0) 0.6 0.0 25 25 35    rock phosphate (0-12-0) 0.6 0.0 0 0 33    alfalfa pellets (3-0-2) 0.6 0.0 19 19 0    2019 HC compost 29.1 10.8 465 76 349 465 76 349 LC compost 1.1 0.4 17 3 13 90 76 13 feathermeal (13-0-0) 0.6 0.0 73 73 0    TYP compost 7.7 2.9 123 20 92 201 98 118 Hi-N (10-3-0) 0.8 0.0 78 78 10    Hi-P (0-12-0) 0.3 0.0 0 0 16    56 2018 HC compost 27.9 4.1 321 67 117 321 67 117 LC compost 2.5 0.4 29 6 11 86 63 11 feathermeal (13-0-0) 0.4 0.0 57 57 0      117      by amendment by strategy REG ID year strategy amendment rate  C N PAN P N PAN P ----- Mg ha-1 ----- -------------------- kg ha-1 -------------------- TYP - 0.0 0.0 0 0 0 0 0 0   118  Appendix E  - Yield sampling  Figure E 1 Example crop biomass sampling scheme for a large farm, where amendments for each nutrient management strategy were applied in plots that span many crop rows in one field. For this example, 32 crop row meters were used as buffers, then, six bed meters were harvested as subsamples using a quadrat (representing ≥ 30% of the remaining six bed meters within the harvestable area).    Figure E 2 Example crop biomass sampling scheme for a small farm, where amendments for each nutrient management strategy were applied directly to one crop bed. For this example, two bed meters were used as end-of-row buffers, then, two bed meters were harvested as subsamples using a quadrat (representing ≥ 30% of the remaining six bed meters within the harvestable area).      119 Appendix F  - Results Table F 1 ANOVA results by outcome variable. Results reported for the minimally adequate mixed effect model after model simplification was performed from the full model, Y ~ Tmt*Year*Region, random = ~1|Site/Year, for each outcome variable, where Y = the measured outcome variables including POx-C, yield, post-season available P and N (as NH4+ and NO3-, by depth), and input costs. Tmt: the fixed effect of nutrient strategy. nT = number of observations for an outcome variable, nt = number of observations per nutrient management strategy.   numDF denDF F-value p-value Yield nT = 81, nt = 27 Tmt 2 42 0.98 0.383 Year 1 5 2.27 0.192 Region 2 16 0.60 0.559 Tmt x Year 2 42 0.66 0.521 Tmt x Region 4 42 1.26 0.230 Year x Region 2 5 0.14 0.871 Tmt x Year x Region 4 42 2.69 0.044 POx-C w/ urban farm nT = 90, nt = 30 Tmt 2 58 2.03 0.141 POx-C w/o urban farm nT = 81, nt = 27 Tmt 2 54 1.76 0.182 Region 2 17 5.33 0.020 NH4+, 0-15 cm nT = 87, nt = 29 Tmt 2 46 5.72 0.006 Year 1 6 4.54 0.077 Region 2 17 3.76 0.045 Tmt x Year 2 46 0.50 0.607 Tmt x Region 4 46 2.60 0.048 Year x Region 2 6 1.36 0.325 Tmt x Year x Region 4 46 1.94 0.120 NH4+, 15-30 cm nT = 87, nt = 29 Tmt 2 56 0.37 0.690 Year 1 6 14.06 0.010 Region 2 17 8.85 0.002 Year x Region 2 6 4.86 0.056 NO3-, 0-15 cm nT = 87, nt = 29 Tmt 2 52 0.02 0.982 Region 2 17 2.03 0.163   120 Tmt x Region 4 52 3.72 0.010 NO3-, 15-30 cm nT = 87, nt = 29 Tmt 2 52 0.78 0.463 Region 2 17 3.06 0.073 Tmt x Region 4 52 1.54 0.206 NO3-, 0-30 cm nT = 87, nt = 29 Tmt 2 50 0.30 0.741 Year 1 8 0.75 0.411 Region 2 17 2.39 0.121 Tmt x Year 2 50 1.00 0.377 Tmt x Region 4 50 2.65 0.044 Available P nT = 78, nt = 26 Tmt 2 Tmt 7.42 0.002 Year 1 Year 3.08 0.130 Region 2 Region 5.54 0.015 Input Costs nT = 75, nt =25 Tmt 2 44 0.47 0.627 Year 1 4 1.05 0.364 Region 2 15 6.51 0.009 Tmt x Region 4 44 5.03 0.002 Year x Region 2 4 6.07 0.061   Table F 2 ANOVA results for yield data by region. Results reported for the minimally adequate mixed effect model after model simplification was performed from the full model, Y ~ Tmt*Year, random = ~1|Site/Year, where Y = yield. Tmt: the fixed effect of nutrient strategy. nT = number of observations for an outcome variable, nt = number of observations per nutrient management strategy.  Yield numDF denDF F-value p-value Vancouver Island nT = 36, nt = 12 Tmt 2 22 0.16 0.853 Pemberton Valley nT = 21, nt = 7 Tmt 2 12 0.70 0.516 Lower Fraser Valley nT = 24, nt = 8 Tmt 2 12 4.68 0.031 Year 1 6 0.12 0.741   121 Tmt x Year 2 12 7.01 0.010    Table F 3 ANOVA output for yield data by year for the mixed effect model, Y ~ Tmt, random = ~1|Site/Year, where Y = yield. Tmt: the fixed effect of nutrient strategy. nT = number of observations for an outcome variable, nt = number of observations per nutrient management strategy.  Yield, lower Fraser Valley numDF denDF F-value p-value 2018 nT = 12, nt = 4 Tmt 2 6 4.29 0.068 2019 nT = 12, nt = 4 Tmt 2 6 6.39 0.033    Table F 4 ANOVA results for post-season NH4+ (0-15 cm) data by region. Results reported for the minimally adequate mixed effect model after model simplification was performed from the full model, Y ~ Tmt*Year, random = ~1|Site/Year, where Y = NH4+ (0-15 cm). Tmt: the fixed effect of nutrient strategy. nT = number of observations for an outcome variable, nt = number of observations per nutrient management strategy.  NH4+, 0-15 cm numDF denDF F-value p-value Vancouver Island nT = 36, n = 12 Tmt 2 22 5.50 0.011 Year 1 4 3.10 0.155 Pemberton Valley nT = 24, n = 8 Tmt 2 14 0.93 0.416 Lower Fraser Valley nT = 27, n = 9 Tmt 2 14 3.96 0.043 Year 1 1 13.61 0.169 Tmt x Year 2 14 1.89 0.188        122  Table F 5 ANOVA results for post-season N03- (0-15 cm) data by region. Results reported for the minimally adequate mixed effect model after model simplification was performed from the full model, Y ~ Tmt*Year, random = ~1|Site/Year, where Y = N03- (0-15 cm) Tmt: the fixed effect of nutrient strategy. nT = number of observations for an outcome variable, nt = number of observations per nutrient management strategy.  NO3-, 0-15 cm numDF denDF F-value p-value Vancouver Island nT = 36, n = 12 Tmt 2 22 0.83 0.448 Pemberton Valley nT = 24, n = 8 Tmt 2 12 3.38 0.068 Year 1 6 1.51 0.434 Tmt x Year 2 12 2.60 0.116 Lower Fraser Valley nT = 27, n = 9 Tmt 2 16 3.04 0.076     Table F 6 ANOVA results for post-season N03- (0-30 cm) data by region. Results reported for the minimally adequate mixed effect model after model simplification was performed from the full model, Y ~ Tmt*Year, random = ~1|Site/Year, where Y = N03- (0-30 cm) Tmt: the fixed effect of nutrient strategy. nT = number of observations for an outcome variable, nt = number of observations per nutrient management strategy.  NO3-, 0-30 cm numDF denDF F-value p-value Vancouver Island nT = 36, nt = 12 Tmt 2 22 0.04 0.959 Pemberton Valley nT = 24, nt = 8 Tmt 2 14 2.99 0.082 Year 1 1 2.14 0.381 Lower Fraser Valley nT = 24, nt = 8 Tmt 2 16 2.62 0.104       123  Table F 7 ANOVA results for input cost data by region. Results reported for the minimally adequate mixed effect model after model simplification was performed from the full model, Y ~ Tmt*Year, random = ~1|Site/Year, where Y = input cost. Tmt: the fixed effect of nutrient strategy. nT = number of observations for an outcome variable, nt = number of observations per nutrient management strategy.  Input Costs numDF denDF F-value p-value Vancouver Island nT = 33, nt = 11 Tmt 2 20 2.40 0.116 Pemberton Valley nT = 21, n = 7 Tmt 2 12 0.01 0.987 Year 1 1 8.90 0.205 Lower Fraser Valley nT = 21, n = 7 Tmt 2 12 8.71 0.005         Figure F 1 Principal Component Analysis (PCA) biplot of the relative POx-C, including Farm Site 57, measured in each nutrient management strategy (TYP: typical, LC: low compost + N, HC: high compost), with pre-season soil and compost properties (soil pH, clay and sand content, soil_C: soil organic carbon, soil_P: Mehlich-3 P, compost_TP: compost total P, compost_CN: compost C to N ratio, compost_PAN: compost plant available N (NO3- + NH4+), compost_TN: compost total N), and grouped by region (lower Fraser Valley (FV), Pemberton Valley (PV), and Vancouver Island (VI)).  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0394816/manifest

Comment

Related Items