PRODUCTION SYSTEM IMPACTS OF STACKED WEED MANAGEMENT TACTICS DURING THE ESTABLISHMENT PERIOD OF A HIGH-DENSITY, CERTIFIED ORGANIC APPLE ORCHARD A Thesis Presented to the Faculty of the Graduate School of Cornell University in Partial Fulfillment of the Requirements of the Degree of Master of Science by Kate Louise Brown August 2022 © 2022 Kate Louise Brown ABSTRACT In the United States, certified organic apples comprise over 11,000 ha and represent $9 billion in annual sales. Organic apple production in New York State and the Northeastern U.S. lags behind production in Washington State where more arid conditions limit pressure from pests, disease, and weeds. Weeds, which compete for resources and provide habitat for tree- girdling rodents, are regularly identified as a major barrier to more widespread adoption of certified organic apple production in the Northeastern U.S. In the present study, we sought to build on the existing body of work focused on efficacious, in-row weed management strategies for organic apple production by combining tactics. The first component of this project was evaluating the impacts on weed biomass, soil chemical, biological, and physical characteristics, apple trees nutrient status, and growth and productivity of the trees during the first four years after planting. The second component was an in-depth analysis of the weed species and changes to weed community assembly in response to the various weed management approaches during this time period. Eight rows of ‘Honeycrisp (Firestorm)’/‘Budagovsky.9’ apple trees were planted (0.9 m × 3.7 m) in a certified organic block at the Cornell University Agricultural Experiment Station Orchards in Ithaca, New York in 2015. Twelve combinations of weed management tactics were implemented in 2016 comprising a split plot experimental design with three main (Mulch, Cultivation, Control) and four split (Mowing, Ammoniated soap, Capric acid, Control) treatments. Between 2016 and 2019, the Mulch treatment improved soil health regardless of the split treatment with improvements to soil organic matter and soil respiration being notable compared to the other main treatments. Although the Mulch treatment also maintained less weed biomass than the other main treatments through 2018, regardless of split treatment, these factors did not optimize tree growth; instead, only the Cultivation treatment resulted in trees with a trunk cross- sectional area (TCSA) significantly greater than the weedy check. Over time, the split treatments became the significant factor influencing weed biomass and the weed community. Within the split treatment Control, an abundance of creeping perennials Solidago spp. and Symphytotrichum lanceolatum was observed regardless of main treatment. As a result, this treatment had significantly greater biomass than the other split treatments by sample period 2 in 2018 and 2019. The organic herbicide split treatments of Ammoniated soap and Capric acid increased biomass from monocot weeds compared to the split Control and Mowing treatments, but always maintained weed biomass significantly lower than the weedy check. Nutrient management was a challenge, as leaf tissue analysis revealed macronutrients were commonly deficient throughout the experiment. Although the Cultivation treatment increased apple leaf nitrogen content compared to the other main treatments, there was no statistically significant correlation between leaf nitrogen and tree growth. Site selection, site preparation, and choice of apple rootstock and scion combination for this study are the main factors that may have influenced apple tree response to these management practices. However, the results of this experiment affirm the relationship between weed management and several horticultural aspects of the orchard production system and underscores the challenge of balancing weed control and nutrient management during the establishment period of the orchard. This work highlights the need for continued research in this subject area to facilitate greater adoption of certified organic apple production. BIOGRAPHICAL SKETCH Kate Louise Brown was born to Amy and Eric Brown in July 1995 and grew up in the small town of Northford, Connecticut alongside her brother, Colton. Among numerous other childhood activities, Kate spent most of her time at a local horse farm where her love of horses grew, and her interest in agriculture began. She attended Lyman Hall High School where she actively participated in her agriculture education classes and served as President of the FFA chapter. After graduating high school in 2013, Kate attended Rutgers University where she delved into agricultural research and had her first experiences with outreach and education for commercial farmers. In 2014, she also achieved a long-time goal of earning her American FFA Degree from the National FFA Organization. In pursuit of a career in Cooperative Extension, Kate graduated from Rutgers in 2018 with her bachelor’s degree in Plant Science and soon moved to Ithaca, NY to begin her master’s research at Cornell University. Just before the onset of the COVID-19 pandemic in March 2020, Kate accepted a position with Rutgers Cooperative Extension as a county-based Program Associate in Commercial Agriculture. Since September 2020, Kate has worked for Rutgers while completing her master’s degree from Cornell. Following the successful defense of her thesis, Kate will continue in her role with Rutgers and explore faculty opportunities within Rutgers Cooperative Extension. iii To my parents, who believe in me more than I believe in myself. Their unwavering love and support are simply unmatched. All that I am, is because of them. iv ACKNOWLEDGEMENTS I would first like to acknowledge my advisor, Dr. Gregory Peck, and thank him for welcoming me to his lab group and to Cornell University in 2018. Through these four years, Greg’s patience and support was continuous, and I am grateful for the experience I have had, and lessons learned as a member of the Peck Lab Group. I would also like to acknowledge the Peck Lab technicians, David Zakalik and Michael Brown, as well as countless Cornell Orchards Interns and fellow Peck Lab members who spent hours in the hot summer sun helping to collect weed, soil, and leaf samples for my thesis research. Thank you, Mike, for making field work both a comedy show, and a math lesson rolled into one. Thank you, David, for the many Zoom writing sessions to help me see this thing through. I cannot overstate my gratitude for the “many hands” who have made this project and this thesis possible. Lastly, I would like to acknowledge the School of Integrative Plant Science for awarding me an Extension Outreach Assistantship which was critical in securing my spot as a Cornell graduate student. The project itself was supported in part by the Arthur Boller Research Fund and grants from the Toward Sustainability Fund. v TABLE OF CONTENTS Biographical Sketch ................................................................................................................ iii Acknowledgements ................................................................................................................. iv Table of Contents ......................................................................................................................v Chapter One. Introduction .............................................................................................................1 References ...............................................................................................................................10 Chapter Two. Reduced Weed Biomass and Increased Soil Health Did Not Promote Growth or Productivity During Establishment Years of an Organic, High-Density Apple Orchard in New York ...............................................................................................................................................13 Abstract .................................................................................................................................13 2.1 Introduction.....................................................................................................................15 2.2 Materials and Methods...................................................................................................17 Study Site ...........................................................................................................................17 Experimental Design ..........................................................................................................18 Pest and Disease Management ...........................................................................................19 Foliar-applied Fertilizer .....................................................................................................20 Percent Cover and Weed Biomass .....................................................................................20 Foliar Mineral Content .......................................................................................................21 Soil Physical, Chemical, and Biological Properties...........................................................21 Soil Volumetric Water Content .........................................................................................22 Tree Growth .......................................................................................................................22 Bloom and Crop Density ...................................................................................................23 Statistical Analysis .............................................................................................................23 2.3 Results ..............................................................................................................................24 Weather ..............................................................................................................................24 Percent Weed Cover ..........................................................................................................25 Aboveground Weed Biomass ............................................................................................26 Apple Tree Leaf Nutrient Content .....................................................................................29 Soil Mineral Content ..........................................................................................................30 Soil Health .........................................................................................................................33 Soil Volumetric Water Content .........................................................................................35 Trunk Cross-Sectional Area ...............................................................................................36 Bloom and Crop Density ...................................................................................................37 Multivariate Analyses ....................................................................................................... 38 Correlation Matrix .............................................................................................................38 Principal Component Analysis ..........................................................................................41 2.4 Discussion ........................................................................................................................44 Weed Control Efficacy ......................................................................................................44 Effects on Soil Health and Soil Mineral Content ...............................................................45 Availability and Uptake of Nitrogen..................................................................................47 Apple Leaf Nutrient Content .............................................................................................49 Tree Growth and Yield Response ......................................................................................50 2.5 Conclusions ......................................................................................................................52 vi References ..........................................................................................................................54 Chapter Three. Weed community changes in response to in-row management tactics across three growing seasons in an organic, high density apple orchard in New York State ...................57 Abstract .................................................................................................................................57 3.1 Introduction.....................................................................................................................59 3.2 Materials and Methods...................................................................................................61 Study Site ...........................................................................................................................61 Experimental Design ..........................................................................................................62 Weed Biomass ...................................................................................................................63 Statistical Analysis .............................................................................................................64 3.3 Results ..............................................................................................................................65 Total Weed Biomass ..........................................................................................................65 Monocot and Dicot Weed Biomass ...................................................................................66 Dicot Weeds by Life Cycle ................................................................................................69 Species Richness, Evenness, and Shannon Diversity Index ..............................................71 Multivariate Analysis .........................................................................................................72 Sample Period 1 and 2...................................................................................................76 Sample Period 3 ............................................................................................................76 Sample Period 4 ............................................................................................................77 3.4 Discussion ........................................................................................................................77 The effects of stacked weed management strategies on total weed biomass .....................77 Treatment effects on weed community diversity and assembly ........................................79 Implications for weed management in organic orchards in the Northeastern U.S. ...........81 3.5 Conclusions ......................................................................................................................82 References ..........................................................................................................................84 Chapter Four. Conclusions .........................................................................................................86 Appendix .......................................................................................................................................89 vii Chapter 1: Review of Literature Public concern over the increasing use of synthetic pesticides and fertilizers in the first half of the 20th century spurred growth and development of the organic farming industry in the 1960s and 1970s. After the U.S. Department of Agriculture finalized regulations for organic certification and established the National Organic Program (NOP) in 2002, the number of certified organic farms and planted area has continued to increase annually. Apples, being the top fruit of choice among American consumers (AGMRC, 2021; USDA-ERS, 2019), have experienced a substantial increase in certified organic production as well. In 2017, there were 728 farms comprising 11,052 ha of certified organic apple production in the U.S. (USDA-NASS, 2020) and this was double the number of farms and acreage reported in 2011 (USDA, 2012). Currently, 7% of the area planted with apple orchards in the U.S. is certified organic (AGMRC, 2021; USDA-NASS, 2017; USDA-NASS, 2020). Early demand and limited supply of fruit after the establishment of the NOP resulted in a price premium of 29% for certified organic apples compared to nonorganic counterparts between 2004 and 2010 (UDSA-ERS, 2017). More recently, demand for organic produce has continued to climb with organic produce sales topping $9.1 billion for the first time in 2021 (Linden, 2021; The Packer, 2022). Compared to 2019, organic produce dollar sales and volume increased more than 20% in the U.S. (Linden, 2021). In fact, apples, along with packaged salads and berries, accounted for two-thirds of all fresh organic produce dollar growth observed between 2020 and 2021 (The Packer, 2022). Increased accessibility of organic produce is one contributor to the growing demand (USDA-ERS); in 2021, 81.3% of retailers sold organic products in the Northeastern U.S (Linden, 2021). 1 Organic agriculture, by definition, is “a production system that…respond[s] to site- specific conditions by integrating cultural, biological, and mechanical practices that foster cycling of resources, promote ecological balance, and conserve biological diversity” (National Organic Program, 2000). Therefore, a central goal of certified organic is to limit negative environmental impacts from agriculture. Nonetheless, increased pressure from pests, disease, and weeds, and difficulty meeting crop fertility requirements are two challenges organic farmers face without synthetic pesticides and fertilizers in their arsenal. As input costs continue to increase and climate change prompts increasing pest and disease pressure, strategies are needed to improve the economic sustainability of organic cropping systems. At present, New York State produces the second largest apple crop in the U.S., with more than 20,000 hectares devoted to apple production (USDA-NASS, 2019); yet less than 1% of this acreage is utilized for certified organic apple production (USDA-NASS, 2020). Instead, most organic production is situated in Washington State in the rain shadow the Cascade Mountain Range where low annual precipitation limits pressure from pests and disease. In the humid and precipitous region of the Northeastern U.S., pests, disease, and weeds proliferate (Agnello et al., 2017). In the Northeastern U.S., weeds have been identified as a top priority for research by several studies (Agnello et al., 2017; Granatstein and Mullinix, 2008; Granatstein and Sánchez, 2009; Hoagland et al., 2008; Peck et al., 2010; Williams et al., 2015). In addition to the difficulty of managing weeds in organic cropping systems, there is a disparity in the cost and efficacy of weed management options between organic and conventional production systems. In conventional apple production, a wide array of herbicides is available to effectively control weeds in the tree row and weed control is among the lowest costs to production (Rowley et al., 2011; Merwin et al., 1995). By contrast, cultivation, mulches, and 2 mowing are commonly recommended practices for organic producers (Bradshaw, 2017; Merwin and Peck, 2009), but these practices can be less effective (Granatstein et al., 2010; Rowley et al., 2011) and more costly (Granatstein and Sanchez, 2009; Merwin et al., 1995; Peck, et al., 2010; Rowley et al., 2011) than synthetic herbicides used in conventional apple production. Organic apple growers lack an acceptable herbicide option that can be used as a single strategy to prevent crop-weed competition (Granatstein et al., 2014) and a recent survey of 41 organic apple growers from the Eastern U.S. indicated that mowing and mulching were the most widely used weed management practices (unpublished data). Tradeoffs among individual weed management strategies for organic apple production have been documented in several studies. For example, in-row application of wood chips or straw has consistently improved tree growth in newly established orchards by increasing and regulating soil moisture (Granatstein et al., 2010; Granatstein and Mullinix, 2008; Hoagland et al., 2008; Merwin and Stiles, 1994; TerAvest et al., 2010; Walsh et al., 1996; Yao et al., 2004). However, excessive soil moisture can result in increased tree mortality (Agnello et al., 2017; Merwin et al., 1992). Generally, increased or more consistent soil moisture under mulch is due to decreased evaporation from soil, but reduced weed pressure in mulched areas may also limit transpiration losses of soil water from the non-crop plants (Hoagland et al., 2008, Merwin et al., 1992, TerAvest et al., 2010, Walsh et al., 1996b). Mulches tend to preserve or improve soil health over time by increasing soil organic matter, soil biological activity, and soil structure (Glover et al., 2000; Haynes, 1980; Peck et al., 2011; St. Laurent et al., 2008; Yao et al., 2005). This is a particularly important for certified organic farms which are required to “maintain or improve the physical, chemical, and biological condition of the soil and minimize erosion” (National Organic Program Standard, 2000). Although wood mulch tends to increase soil organic 3 matter in heavier textured soils of New York (Peck et al., 2011; Yao et al., 2005), soil organic matter did not increase three years after application in Washington (Granatstein and Mullinix, 2008; Granatstein et al., 2014; Hoagland et al. 2008) or Arkansas (Choi et al., 2011). Wood chip mulch can also be a source of soil nutrients as it decomposes, as several studies found increased soil content of phosphorus, potassium, and calcium following mulch application (Atucha et al., 2011; Merwin and Stiles, 1994; Peck et al., 2011; Yao et al., 2005). Lastly, the high carbon to nitrogen ratio of organic mulches poses a risk for nitrogen immobilization (Magdoff and Van Es, 2009) and surface applied mulches can increase vole damage to trees (Merwin et al., 1994). There are a variety of cultivation implements designed for use in orchard systems, all with varying degrees of efficacy and tillage intensity. One of these tools that has been readily evaluated is the Wonder Weeder (Harris Manufacturing, Burbank, WA) – a ground-driven implement with four gangs of Lilliston spiders and a spring steel bar to control weeds directly in the tree row. When compared to wood chip mulch, weed pressure is often greater where the Wonder Weeder is used on a monthly basis during the growing season (Agnello et al., 2017; Peck et al., 2011; Granatstein et al., 2010). Although cultivation with this specific implement did not improve weed control relative to wood chip mulch, there are other benefits to this type of equipment. Merwin et al. (1994) used a rototiller – a PTO driven implement – to maintain a weed free area under the tree row and reported an 18% decline in soil organic matter after five years. Alternatively, the Wonder Weeder consistently maintains (Granatstein et al., 2014; Hoagland et al., 2008) or improves (Peck et al., 2011) soil organic matter. This is likely the result of lower tillage intensity compared to PTO-driven implements, shallow working depth of the implement, and retention of weed biomass on the soil surface which squanders erosion (Magdoff and Van Es, 2009; Harris Manufacturing, Burbank, WA). Cultivation also results in 4 mineralization of soil nitrogen (N), so N may become more available to trees for growth (Hoagland et al., 2008). Several OMRI-approved herbicides with a range of active ingredients have been evaluated for use in certified organic orchards with limited success. Agnello et al. (2017) evaluated acetic acid, limonene, soap, and caprylic acid products in a New York orchard by making 5 applications per year – once every three weeks – over course of three years. In the first two years, the weed free area for these herbicides was 50-80% in August but the third year was less successful, with only 30-50% weed free area by September. While the level of weed control in the first two years was acceptable, the authors note that none of the herbicides effectively controlled Cirsium arvense (L.) Scop. – a difficult to control, rhizomatous perennial weed. Similarly, Rowley et al. (2011) evaluated two concentrations of acetic acid as well as lemon grass oil and clove oil in three orchards in Utah and found that these products often failed to provide adequate weed control when used independently. When these herbicides were combined with mulches like paper, straw, or wood chips, the total percentage of weed control after two years tended to be significantly greater than where no herbicide was applied. Unlike herbicides for conventional agriculture, all herbicides for organic agriculture are non-selective contact herbicides with limited to zero residual activity (Peck and Merwin, 2009). Because of this, organic herbicides must be used often to target small weeds less than 10-15 cm tall to be effective. Oftentimes the frequency at which you have to apply the material, or the concentration you have to use, makes it cost-prohibitive to utilize organic herbicides as the sole strategy. Mowing is a form of suppression for a vegetative cover or “living mulch” in the tree row. Maintaining vegetative cover in the tree row has the potential to enhance ecosystem services in the orchard through greater biodiversity, improved nutrient cycling, protection of soil and water 5 quality, direct weed competition, and more (Hammermeister, 2016; Granatstein and Mullinix, 2008). As a result, considerable effort has been made to identify and evaluate suitable groundcover species to co-exist with perennial tree fruit without appreciable reductions in yield. Leguminous living mulches are frequently studied because nitrogen fixation is expected to enhance nitrogen availability to trees. While soil nitrogen availability has been shown to increase where leguminous living mulches are grown (Granatstein and Mullinix, 2008), the level of competition has still reduced tree growth and yield in apple (Hoagland et al., 2008; TerAvest et al., 2010), black currant (Larsson et al., 1997), and cherry orchards (Sanchez et al., 2003). One alternative is the Swiss “Sandwich” system where perennial species are maintained in the immediate tree row and shallow tillage is utilized on either side of the tree rows (Hammermeister, 2016; Hoagland et al., 2008). In general, though, living mulch systems have been found to be better suited to established orchards where trees are less susceptible to resource competition. While weed pressure is not the only factor influencing tree growth, it’s clear that weed management strategies have implications for other factors influencing tree growth and productivity such as soil health, and resource competition. In addition to optimizing a strategy to control weeds, consideration must also be given to the appropriate timing for weed control actions. The importance of early season weed management in orchards has been demonstrated repeatedly (Breth, 2015; Merwin and Ray, 1997; Neilsen and Hogue, 1985). Specifically, Merwin and Ray (1997) showed a weed free area (WFA) of at least 2 m2 around ‘Imperial Gala’/‘Malling.26’ apple trees (3 m × 6 m) significantly increased tree size, cumulative yield, and yield efficiency when the WFA was maintained for 60-90 days starting in May. Similar results reported by Breth (2015) in high-density apple orchards support the work by Merwin and 6 Ray (1997). The importance of this early season weed control is related to the timing of nutrient uptake by apple trees; apple trees rely on N reserves rather than soil N for initial spring growth (Cheng and Fuchigami, 2002; TerAvest et al., 2010), but there is a period of rapid N uptake from bloom through the end of shoot growth each year (Cheng and Raba, 2009; TerAvest et al., 2010) that coincides with the recommended critical weed free period. Neilsen and Hogue (1985) also validated this with their work, where eliminating tree row vegetation through July 15 improved leaf N and prevented early yield reduction. Cumulative yield potential and economic return are maximized when early yields are achieved in modern, high-density orchards (Robinson et al., 2006; Robinson et al., 2007). For this reason, adequate weed control during the establishment years is pivotal as the nitrogen needs of young trees is higher than mature trees (Peck and Merwin, 2009; Stiles and Reid, 1991) and the nutrient release of organic fertilizers is less predictable than synthetic fertilizers (Granatstein and Sánchez, 2009; Hoagland et al., 2008). The various weed control tactics available for organic production have been evaluated individually but infrequently as stacked treatments (Rowley et al., 2011; Granatstein et al., 2014; Europe). The use of multiple tactics simultaneously could be a strategy to improve weed control. This concept, coined the ‘many little hammers’ approach to weed management by Liebman and Gallandt (1997), capitalizes on the synergistic effects of complementary weed management strategies and is part of an integrated approach to weed management (Swanton and Murphy, 1996). Integrated weed management also seeks to affect the weed community such that weeds are maintained below an economic threshold without the need to eradicate all weeds (Clements et al., 1994; Swanton and Weise, 1991). This could be advantageous for organic systems where the intention is to “promote ecological balance, and conserve biological diversity” (National Organic Program, 2000). In an agroecosystem, individual weed management tactics act as filter 7 that affect the weed community assembly, including the abundance and diversity (Ciaccia et al., 2022; Clements et al., 1994). In a practical sense, it is important to learn about the strengths and limitations of weed management strategies – both individually and in combination – as it relates to changes in community assembly. Some weed types or species are more competitive with crops or have characteristics that make them less vulnerable to different management strategies (Mohler et al., 2021; Swanton and Weise, 1991). With this knowledge, orchard managers can select the most efficacious weed management strategy for their given weed community and adapt management as the community dynamics change in response to management (Ciaccia et al., 2022). In the present study, we implemented twelve combinations of weed management strategies (3 main × 4 split) during the establishment period of a certified organic block of ‘Honeycrisp’/‘Budagovsky.9’ apple trees in New York State. The trees were planted in 2015 and the treatments were implemented between 2016 and 2019. During this time, we measured weed biomass and species diversity, soil fertility and soil health, leaf nutrient content, and apple tree growth and productivity. The main objective was to evaluate the efficacy of stacked weed management strategies and measure the impact of these combinations on various horticultural response variables in the first four seasons following planting. Our secondary objective was to document changes in the weed community in response to the various weed management strategies. Evaluations of stacked, or integrated, approaches to weed management are not widely available in the current literature for perennial tree fruit systems yet organic growers still struggle to control weeds. Our intention was to learn whether using more than one weed management strategy would have benefits for weed control without consequences for tree growth or soil health. It was also important for us to learn how weed communities change in response to 8 management to inform the utility of different weed tactics for orchard sites based on the existing weed community. An analysis of the horticultural responses such as treatment effects on total weed biomass, soil properties, fertility, and tree growth are provided in Chapter 2. An in-depth analysis of the weed biomass, weed community, and changes to community assembly over time is compiled in Chapter 3. 9 REFERENCES Agnello, A., K. Cox, J. Lordan, P. Francescatto, and T. Robinson. 2017. Comparative programs for arthropod, disease and weed management in New York organic apples. Insects 8(3):96. https://doi.org/10.3390/insects8030096. Apples. 2022 . .\ Bradshaw, T. 2017. Non-chemical weed management in apple orchards. University of Vermont Extension, Burlington, Vermont. Breth, D. 2015. Critical weed control requirements in young high density apple orchards. New York Fruit Quarterly. 23(4):11-15. Cheng, L. and L.H. Fuchigami. 2002. Growth of young apple trees in relation to reserve nitrogen and carbohydrates. Tree Physiology. 22(18):1297-1303. https://doi- org.proxy.library.cornell.edu/10.1093/treephys/22.18.1297. Cheng, L. and R. Raba. 2009. Accumulation of macro- and micronutrients and nitrogen demand- supply relationship of ‘Gala’/‘Malling 26’ apple trees grown in sand culture. J. Amer. Soc. Hort. Sci. 134(1):3-13. https://doi.org/10.21273/JASHS.134.1.3. Choi, H., C.R. Rom, and M. Gu. 2011. Effects of different organic apple production systems on seasonal nutrient variations of soil and leaf. Scientia Horticulturae. 129(1):9-17. https://doi-org.proxy.library.cornell.edu/10.1016/j.scienta.2011.02.009. Ciaccia, C., E. Testani, T. Amoriello, and D. Ceccarelli. 2022. Weed community evolution under diversification managements in a new planted organic apricot orchard. Agriculture, Ecosystems & Environment. 335(2022):108014. https://doi- org.proxy.library.cornell.edu/10.1016/j.agee.2022.108014. Clements, D.R., S.F. Weise, and C.J. Swanton. 1994. Integrated weed management and weed species diversity. Phytoprotection. 75(1):1-18. https://doi.org/10.7202/706048ar. Glover, J.D., J.P. Reganold, and P.K. Andrews. 2000. Systematic method for rating soil quality of conventional, organic, and integrated apple orchards in Washington State. Agriculture, Ecosystems & Environment. 80(1-2):29-45. https://doi- org.proxy.library.cornell.edu/10.1016/S0167-8809(00)00131-6. Granatstein, D. and E. Sánchez. 2009. Research knowledge and needs for orchard floor management in organic tree fruit systems. International Journal of Fruit Science. 9(3):257-281. https://doi.org/10.1080/15538360903245212. Granatstein, D. and K. Mullinix. 2008. Mulching options for Northwest organic and conventional orchards. HortScience. 43(1):45-50. https://doi.org/10.21273/HORTSCI.43.1.45. Granatstein, D., M. Wiman, E. Kirby, and K. Mullinix. 2010. Sustainability trade-offs in organic orchard floor management. Acta Hort. 873:115-122. Granatstein, D., P. Andrews, and A. Groff. 2014. Productivity, economics, and fruit and soil quality of weed management systems in commercial organic orchards in Washington State, USA. Org. Agr. 4(3):197-207. https://doi.org/10.1007/s13165-014-0068-0. Hammermeister, A.M. 2016. Organic weed management in perennial fruits. Scientia Horticulturae 208:28-42. https://doi.org/10.1016/j.scienta.2016.02.004. Haynes, R.J. 1980. Influence of soil management practice on the orchard agro-ecosystem. Agro- Ecosystems 6(1):3-32. https://doi-org.proxy.library.cornell.edu/10.1016/0304- 3746(80)90003-7. 10 Hoagland, L., L. Carpenter-Briggs, D. Granatstein, M. Mazzola, J. Smith, F. Peryea, and J.P. Reganold. 2008. Orchard floor management effects on nitrogen fertility and soil biological activity in a newly established organic apple orchard. Biol Fertil Soils. 45(1):11-18. https://doi.org/10.1007/s00374-008-0304-4. Larsson, L., B. Stenberg, and L. Torstensson. 1997. Effects of mulching and cover cropping on soil microbial parameters in the organic growing of black currant. Communications in Soil Science and Plant Analysis. 28(11-12):913-925. https://doi.org/10.1080/00103629709369842. Liebman, M. and E. Gallandt. 1997. Many little hammers: Ecological management of crop-weed interactions. In L.E. Jackson (ed.). Ecology in Agriculture. Academic Press. San Diego, California. Magdoff, F. and H. Van Es. 2009. Building soils for better crops: sustainable soil management. 3rd ed. Sustainable Agriculture and Research Education (SARE) program, Handbook series book 10. College Park, Maryland. Merwin, I.A. and J.A. Ray. 1997. Spatial and temporal factors in weed interference with newly planted apple trees. HortScience. 32(4):633-637. https://doi.org/10.21273/HORTSCI.32.4.633. Merwin, I.A. and W.C. Stiles. 1994. Orchard groundcover management impacts on apple tree growth and yield, and nutrient availability and uptake. J. Amer. Soc. Hort. Sci. 119(2):209-215. https://doi.org/10.21273/JASHS.119.2.209. Merwin, I.A., D.A. Rosenberger, C.A. Engle, D.L. Rist, and M. Fargione. 1995. Comparing mulches, herbicides, and cultivation as orchard groundcover management systems. HortTech. 5(2):151-158. https://doi.org/10.21273/HORTTECH.5.2.151. Merwin, I.A., W.F. Wilcox, and W.C. Stiles. 1992. Influence of orchard ground cover management on the development of Phytophthora crown and root rots of apple. Plant Dis. 76(2):199-205. Mohler, C.L., J.R. Teasdale, and A. DiTommaso. 2021. Manage weeds on your farm: a guide to ecological strategies, SARE handbook series. Sustainable Agriculture Research & Education (SARE), College Park, Maryland. Neilsen, G.H. and E.J. Hogue. 1985. Effect of orchard soil management on the growth and leaf nutrient concentration of young dwarf red delicious apple trees. Can. J. Soil. Sci. 65(2):309-315. https://doi-org.proxy.library.cornell.edu/10.4141/cjss85-034. Linden, T. 2022. Organic sales continue to impress through pandemic. Produce News. . Peck, G. and I.A. Merwin. 2009. A grower’s guide to organic apples. New York State IPM. Publication Number 223. Peck, G.M., I.A. Merwin, J.E. Thies, R.R. Schindelbeck, and M.G. Brown. 2011. Soil properties change during the transition to integrated and organic apple production in a New York orchard. Applied Soil Ecology. 48(1):18-30. https://doi- org.proxy.library.cornell.edu/10.1016/j.apsoil.2011.02.008. Peck, G.M., I.A. Merwin, M.G. Brown, and A.M. Agnello. 2010. Integrated and organic fruit production systems for ‘Liberty’ apple in the Northeast United States: a systems-based evaluation. Hort. Sci. 45(7):1038-1048. https://doi- org.proxy.library.cornell.edu/10.1016/j.apsoil.2011.02.008. Robinson, T.L., A.M. DeMarree, and S.A. Hoying. 2007. An economic comparison of five high density apple planting systems. Acta Hort. 732: 481-489. 11 Robinson, T.L., S.A. Hoying, and G.H. Reginato. 2006. The tall spindle apple production system. New York Fruit Quarterly. 14(2):21-28. Rowley, M.A., C.V. Ransom, J.R. Reeve, and B.L. Black. 2011. Mulch and organic herbicide combinations for in-row orchard weed suppression. International Journal of Fruit Science. 11(4):316–331. https://doi- org.proxy.library.cornell.edu/10.1080/15538362.2011.630295. Sanchez, J.E., C.E. Edson, G.W. Bird, M.E. Whalon, T.C. Willson, R.R. Harwood, K. Kizilkaya, J.E. Nugent, W. Klein, A. Middleton, T.L. Loudon, D.R. Mutch, and J. Scrimger. 2003. Orchard floor and nitrogen management influences soil and water quality and tart cherry yields. J. Amer. Soc. Hort. Sci. 128(2):277-284. https://doi.org/10.21273/JASHS.128.2.0277. St. Laurent, A., I.A. Merwin, and J.E. Thies. 2008. Long-term orchard groundcover management systems affect soil microbial communities and apple replant disease severity. Plant and Soil. 304(1):209-225. https://doi.org/10.1007/s11104-008-9541-4. Stiles, W.C. and W.S. Reid. 1991. Orchard nutrition management. Cornell Cooperative Extension. Information Bulletin 2019. Swanton, C.J. and S.D. Murphy. 1996. Weed science beyond the weeds: The role of integrated weed management (IWM) in agroecosystem health. Weed Sci. 44(2):437-445. https://doi.org/10.1017/S0043174500094145. Swanton, C.J. and S.F. Weise. 1991. Integrated weed management: The rationale and approach. Weed Technology 5(3):657-663. https://doi.org/10.1017/S0890037X00027512. TerAvest, D., J.L. Smith, L. Carpenter-Boggs, L. Hoagland, D. Granatstein, and J.P. Reganold. 2010. Influence of orchard floor management and compost application timing on nitrogen partitioning in apple trees. HortScience. 45(4):637-642. https://doi.org/10.21273/HORTSCI.45.4.637. USDA-NASS. 2012. Certified Organic Production Survey, 2011. United States Department of Agriculture, National Agricultural Statistics Service, Washington, DC. USDA-NASS. 2019. Fruits and Nuts: 2017 and 2012. United States Department of Agriculture, National Agricultural Statistics Service, Washington, DC. USDA-NASS. 2020. 2019 Organic Survey, 2017 Census of Agriculture Special Study. United States Department of Agriculture, National Agricultural Statistics Service, Washington, DC. U.S. organic produce sales top $9 billion in 2021, up 5.5% from 2020. 2022. The Packer. January 28, 2022. . Walsh, B.D., A.F. MacKenzie, S. Salmins and D.J. Buszard. 1996. Impact of soil management systems on organic dwarf apple orchards and soil aggregate stability, bulk density, temperature and water content. Canadian Journal of Soil Science 76(2):203-209. https://doi-org.proxy.library.cornell.edu/10.4141/cjss96-028. Williams, M.A., J.G. Strang, R.T. Bessin, D. Law, D. Scott, N. Wilson, S. Witt, and D.D. Archbold. 2015. An assessment of organic apple production in Kentucky. HortTech. 25(2):154-161. https://doi.org/10.21273/HORTTECH.25.2.154. Yao, S., I.A. Merwin, G.W. Bird, G.S. Abawi, and J.E. Thies. 2005. Orchard floor management practices that maintain vegetative or biomass groundcover stimulate soil microbial activity and alter soil microbial community composition. Plant and Soil. 271(1):377-389. https://doi.org/10.1007/s11104-004-3610-0. 12 Chapter 2: Reduced Weed Biomass and Increased Soil Health Did Not Promote Growth or Productivity During Establishment Years of an Organic, High-Density Apple Orchard in New York Abstract Prolific weed growth in the rainy, humid summers of New York coupled with poor efficacy of available weed management strategies remains a major barrier to adoption of organic apple production. In 2015, eight rows of ‘Honeycrisp (Firestorm)’/‘Budagovsky.9’ apple trees were planted (0.9 m × 3.7 m) in a NOFA-NY certified organic orchard block at the Cornell University Agricultural Experiment Station Orchards in Ithaca, NY. To test the effectiveness of stacked, in- row weed management strategies on percent weed cover and biomass, soil health, and tree growth and productivity, a split-plot experimental design was implemented in 2016. Main treatments included an untreated Control (main treatment Control), cultivation with a Wonder Weeder (Cultivation), and surface-applied wood chip mulch (Mulch). Split treatments included an untreated Control (split treatment Control), mowing with a string trimmer (Mowing), and capric/caprylic acid (Capric acid), and ammoniated soap of fatty acids (Ammoniated soap). The Capric acid and Ammoniated soap products are both approved for organic production in the U.S. All treatment combinations were implemented four times per season between 2016 and 2019, approximately monthly, from May to August, with weed sampling occurring the day before the treatments were implemented. Mulch was applied once, in spring 2016. Between 2017 and 2019, nearly all measured soil health parameters increased in the Mulch treatment relative to the main treatment Control. Notably, soil organic matter in the Mulch treatment increased from 4.81 in 2016 to 5.21% in 2019 while it declined from 4.46 to 4.06% in the Cultivation treatment during those years. Regardless of split treatment, the Mulch treatment maintained 16-45% less weed biomass than the other main treatments throughout the study; however, the split treatments 13 became a more significant factor in the latter years. Regardless of main treatment, the split treatments of Mowing, Capric acid, and Ammoniated soap all maintained less than 200 g‧m-2 weed biomass per sampling period which was significantly less than the split treatment Control at time points 3-4 in 2017 and time points 2-4 in 2018 and 2019. This was mainly due to an abundance of Symphytotrichum lanceolatum (Willd.) and Solidago spp. in treatment combinations with the split treatment Control. Tree growth was greatest in the Cultivation treatment where trunk cross-sectional area (TCSA) was 4.79 cm2 in fall 2019 – a 4-fold increase from the first measurement taken in spring 2016. Tree growth in the Cultivation treatment was greater than the Mulch treatment, regardless of split treatment, but the TCSA of trees in the Control × Capric acid or Control × Ammoniated soap treatments was similar to TCSA in the Cultivation treatment. Although TCSA was weakly positively correlated with leaf nitrogen (N) in every year, there was no significant positive or negative linear correlation between TCSA and leaf N, weed biomass, or other measured soil health assessments. These findings demonstrate there are multiple factors that all contribute to early tree growth and productivity. These data also highlight the challenge of maintaining adequate weed control and fertility during the establishment period of an organic apple orchard and underscore the need for further research to address this challenge. 14 2.1 Introduction There is growing demand for certified organic apples (Ag Marketing Research Center, 2021; Linden, 2021; Slattery et al., 2011; The Packer, 2022); a market niche to which less than 1% of New York apple orchard area is currently devoted (USDA, 2019; USDA, 2020). Although New York was the second largest U.S. producer of apples in 2021 (Ag Marketing Research Center), 85% of certified organic apple area is currently located in Washington state (USDA, 2020) where less than 50 cm of annual precipitation minimizes pest and disease pressure. Growers in the Northeastern U.S., however, must contend with high humidity and over 100 cm of annual precipitation that results in high disease pressure and shorter generation time for pests. Correspondingly, growers in the northeast contend with over 50 direct and indirect arthropod pests and more than 20 plant diseases (Agnello et al., 2017). Constrained by a large pest complex and frequent summer precipitation that also fosters weed germination and growth, certified organic apple growers often cite weeds as the most challenging pest to manage organically, which has been verified by many research studies from a wide range of locations and climates (Agnello et al., 2017; Granatstein and Mullinix, 2008; Granatstein and Sánchez, 2009; Hoagland et al., 2008; Peck et al., 2010; Williams et al., 2015). Failure to achieve adequate weed control can reduce nutrient and water availability, negatively impacting tree growth and fruit yield. Merwin and Ray (1997) found May through July was the critical period for weed control in orchards, and that weeds must be controlled within a 2 m2 area around each tree to prevent yield loss. Neilsen and Hogue (1985) also highlighted the importance of eliminating tree row vegetation through July 15 to improve leaf N and prevent early yield reduction. These studies were more recently validated for high-density apple orchards by Breth (2015). Weed management is most critical during the first five years 15 after planting, as young trees are more susceptible to competition from weeds (Peck and Merwin, 2009) and the nitrogen needs of young trees is higher (Hoagland et al., 2008; Stiles and Reid, 1991). Studies suggest that orchards may be less susceptible to resource competition once trees reach maturity (Atucha et al., 2011; Hoagland et al., 2008; Stefanelli et al., 2009). Cultivation, mowing, mulching, and hand weeding are often used in organic apple orchards to achieve an acceptable level of weed control and have been broadly studied (Bradshaw, 2017), but these strategies can come at a high cost compared to conventional herbicides (Merwin et al., 1995; Mia et al., 2020). For example, use of a cultivator or mower has associated time, labor, and equipment costs (Peck et al., 2010). Another consideration is the degradation of soil health associated with repeated cultivation events over the life of an orchard (Peck et al., 2011). According to the National Organic Program Standard §205.203, organic producers must “maintain or improve the physical, chemical, and biological condition of the soil and minimize erosion” – all aspects of soil health which can be degraded through cultivation. Lastly, management of persistent vine and perennial weeds may also necessitate hand weeding, which is a time consuming and, therefore, financially prohibitive practice on most farms. High density plantings are standard practice in newly established, conventionally managed orchards and provide multiple benefits such as higher productivity and profitability, as well as increased spray coverage and reduced drift (Robinson et al., 1991; Robinson et al., 2007). Dwarfing rootstocks which allow for high density orchards have smaller root systems and are more sensitive to resource competition from weeds. Weak scions (e.g., ‘Honeycrisp’) grafted onto dwarfing rootstocks (e.g., ‘Budagovsky.9’) are especially susceptible to weed competition. Limited availability of allowable herbicides under the USDA National Organic Program Standards and the risks of tree damage and spreading of fire blight (Erwinia amylovora) with 16 mechanical weed management in young, high-density orchards has ultimately limited adoption of these practices among certified organic orchards. In 2016, a split plot experimental design with three main treatments and four split treatments was implemented in a high-density, certified organic apple orchard in Ithaca, New York. The goal was to test the utility of in-row weed management strategies – used individually or in combination – during the establishment period of an apple orchard. Treatment effects on percent weed cover and biomass, soil health, foliar nutrient content, and tree growth and productivity were measured. We hypothesized the Mulch treatment would improve soil health parameters such as soil organic matter, respiration, and active carbon. We also expected “stacked” weed management treatments to improve weed control relative to individual treatments. Finally, we anticipated that treatments which improved soil health and reduced weed biomass would result in greater tree growth and productivity over the course of the study. 2.2 Materials and Methods Study Site The 0.16 ha site was located on NOFA-NY certified organic land at the Cornell University Agricultural Experiment Station orchards in Ithaca, NY (lat. 42.44519º, long. - 76.45912º). The soils at the site were 78% Hudson and Collamar silt loams and 22% Hudson silty clay loam, moderately well drained, with 2-6% slopes (Soil Science Division Staff, 2017). Semi-dwarf apple trees were previously grown at this site from 1981-2006 and the site was fallowed after these trees were removed. In fall 2014, the field was plowed and disked after one line of drainage tile was installed through a wet area at the north end of this field. No cover crop 17 was seeded, and no lime was applied as the soil pH was optimal at 6.80 (E. Shatt, personal communication). The eight tree rows associated with this study were located on the easternmost end of a block containing a total of 24 rows. All trees were planted in the spring of 2015 and the rows were planted to run from North to South. From East to West, the entire block included 15 NOFA-NY certified organic rows, then two organically managed buffer/transition rows, then seven conventionally managed rows. European black alder (Alnus glutinosa L.) flanked the north and east of the block, and a 10 m wide grass buffer strip separated the block from conventionally managed apples on the south side. A fallow field was to the west of the block for the duration of the study. Experimental Design Eight rows of ‘Honeycrisp (Firestorm)’/‘Budagovsky.9’ trees were planted in 2015, trained following the tall spindle system, and spaced 0.9 m between trees and 3.7 m between rows (Robinson et al., 2013; Robinson et al., 2006). A randomized, split-plot design with four complete blocks was implemented in 2016. Main treatments included Cultivation [Wonder Weeder, Harris Manufacturing, Burbank, WA], Mulch [surface-applied mix of hard and soft woods, 15 cm depth, sourced from the Cornell University Grounds Department], and the main treatment Control [untreated control/weedy check]. Split treatments included Mowing [hand-held string trimmer], Capric acid [Suppress® Herbicide EC, 6% solution (57 L a.i. ha-1), Westbridge Agricultural Products, Vista, CA], Ammoniated soap [Final-San-O®, (163 L a.i. ha-1), Certis USA, LLC, Columbia, MD], and the 18 split treatment Control [untreated control]. There were twelve total treatment combinations (3 main  4 split). Wood chip mulch was applied only once, in May 2016, while all other treatments were repeated four times per season within 1-3 days following each aboveground weed biomass sampling date (Table A1). Within the Cultivation treatment, cultivation preceded herbicide application and cultivation came after mowing. The split-plot experimental design was comprised of four complete blocks. Each block consisted of two tree rows which were divided into three main plots, one for each main treatment. Main plots were subsequently divided into four split-plots, one for each split treatment (Figure A1). Each of these split plots (n = 48) represented one experimental unit. Main plots were separated by six buffer trees while split plots were separated by three buffer trees. Percent cover, aboveground weed biomass, leaf tissue samples, soil samples, and soil volumetric water content were collected at the plot level. Bloom density, trunk cross-sectional area, and fruit yield were collected at the tree level, but the mean of these values per plot was used for analysis and comparison among treatments. In-row weeds were controlled within 0.5 m on each side of the tree row (Figure A2). A diverse stand of grass and broadleaf weeds in the inter-row space was mowed approximately four times per season. Every fall, a string trimmer was used to cut all weeds in the tree rows down to the soil surface following the onset of tree dormancy. Pest and Disease Management Pests and disease were managed according to USDA-NOP guidelines using local guidance (Peck and Merwin, 2009; Table A2). Common orchard pests we sought to manage 19 included oriental fruit moth (Grapholita molesta (Busck.)), codling moth (Cydia pomonella (L.)), plum curculio (Conotrachelus nenuphar (Herbst.)), apple maggot (Rhagoletis pomonella (Walsh)), European apple saw fly (Hoplocampa testudinea (Klug)), San Jose scale (Quadraspidiotus perniciosus (Comstock)), potato leafhopper (Empoasca fabae (Harris)), and Japanese beetle (Popillia japonica (Newman)). Diseases included fire blight (Erwinia amylovora (Burrill)), apple scab (Venturia inaequalis (Cooke)), powdery mildew (Podosphaera leucotricha (Ellis & Everhart)), and the complex of fungal species that cause sooty blotch and flyspeck. Foliar-applied fertilizer Nitrogen, magnesium, boron, and zinc were foliar-applied as needed based on annual leaf tissue analyses and according to USDA-NOP guidelines (Table A3). Percent cover and weed biomass Percent cover and aboveground weed biomass were measured one to three days prior to treatment application in 2017, 2018, and 2019. Sampling dates were approximately once monthly, between May and August of each year (Table A1). At each sampling date, one sampling location was selected per plot (n = 48). Each sample location was sampled only once per growing season and included the open, in-row area between two trees. Percent cover and aboveground weed biomass samples were collected from within a 0.25 m2 quadrat placed directly in the tree row at the randomly selected sample location (Figure A2). Percent cover was measured using the Canopeo smartphone application (Oklahoma State University, 2015) from the height at which the quadrat was no longer visible on the smartphone screen. All aboveground weed biomass rooted within the quadrat was then harvested at the ground level using scissors 20 and/or hand sheers. The bags were oven-dried at 70 ºC for seven days before the dry weight was measured. Weed species data is presented in Chapter 3. Foliar mineral content Approximately 90 days after full bloom in 2016-2019, a representative sample of fifty leaves was collected from each plot. Leaves were collected from along the mid-point of the current season’s growth and kept in paper bags. Prior to delivery to the Cornell Nutrient Analysis Laboratory (CNAL, Ithaca, NY), leaves were rinsed to remove residual foliar-applied minerals. If same day delivery to CNAL was not possible, leaves were stored in a walk-in cooler at 2 ºC overnight. Within 24 hours of sampling, leaves were delivered to CNAL where a combustion analysis was done to determine total carbon and nitrogen and inductively coupled plasma atomic emission spectroscopy (ICP-AES) was conducted to determine micro- and macronutrient content following standard protocols (VarioMax CNS; Elementar Analysensysteme GmbH, Langenselbold, Germany). Soil physical, chemical, and biological properties Soil samples were collected approximately 90 days after full bloom in 2017-2019. A 2- cm diameter soil core sampler was used to collect soil samples to a 15 cm depth. Soil samples for each plot were comprised of twelve soil cores (4 locations per plot × 3 soil cores per location). Within each plot, four locations were randomly selected wherein one soil core was collected from directly in the tree row, one from the east side of the tree row, and one from the west side of the tree row. Composite samples were kept in plastic, zip-top bags which contained all twelve soil cores from each plot. If same day delivery to CNAL was not possible, samples were stored in a walk-in cooler at 2 ºC overnight. Samples were delivered to the CNAL within 24 hours after 21 sampling and tested using the Standard Soil Health Analysis Package, as described by Moebius- Clune et al. (2017). This package includes soil pH, organic matter, modified Morgan extractable P, K, micronutrients, soil texture, active carbon, wet aggregate stability, soil respiration, total carbon, total nitrogen, predicted autoclave-citrate extractable (ACE) protein test, and predicted available water capacity. This array of tests provides insight to the chemical, physical, and biological characteristics of soil that are most sensitive to short-term variations in management (Moebius-Clune et al., 2017). We opted for this set of parameters to best understand how weed management strategies impact soil health during the establishment years of an orchard. Soil volumetric water content. In 2019, soil volumetric water content (VWC) was measured using a Field Scout TDR350 soil moisture meter (Spectrum Technologies, Inc., Aurora, Illinois) with 7.62 cm soil probes. Soil VWC was measured once per week from 6 June to 6 September 2019 for a total of 10 measurements. At each sampling date, soil VWC was collected from one location per plot, a location where weeds had not already been harvested. We anticipated soil VWC directly in the tree row would be higher than the edges of the tree row in the Cultivation treatment. Therefore, we collected one measurement from the tree row, one from the East side of the tree row, and one from the West side of the tree row for each plot, on each date, in every treatment. The average of these three measurements was used for statistical analysis. Tree growth Trunk cross-sectional area (TCSA) was measured after leaf fall in 2016-2019. Calipers were used to take two measurements of trunk diameter at 40-cm above the graft union of each 22 tree, with the second measurement being taken at a 180º turn from the first measurement. The average trunk cross-sectional area treatment was used for statistical analysis. Bloom and crop density The number of flower clusters per tree was recorded on approximately 5 May in 2018, 2019, and 2020. The number of flower clusters per tree was divided by the previous fall’s TCSA of each respective tree to determine bloom density. The number of fruits harvested, and the weight of those fruit, was recorded for each individual tree on 10 September 2018 and 10 September 2019. Pre-harvest fruit drops were also counted and included in the total number of fruits per tree, but not the weight of fruit per tree. The number of fruits per TCSA in each treatment was used for statistical analysis. Statistical analysis Data were analyzed in R (version 1.3.1056) with a linear mixed model which included main treatment (Cultivation, Mulch, Control), split treatment (Mowing, Capric acid, Ammoniated soap, Control) and their interactions as fixed effects. Block was a random effect. In the case of percent weed cover and weed biomass, time point was included as a fixed effect in the model. Therefore, an additional random effect term was Block × Time Point. Years were analyzed separately. Each dataset was subjected to an ANOVA to determine the significance of treatment and interaction effects, which were considered significant at the p = 0.05 level. Tukey’s HSD test was used to identify treatment differences. A correlation matrix was created in R using the corrplot package to visualize the linear relationship among variables. A principal component analysis (PCA) was conducted and included the following data: weed biomass, active soil carbon, soil respiration, ACE soil protein 23 index, soil organic matter, wet aggregate stability, available water capacity, total soil carbon and nitrogen, foliar nitrogen, and TCSA. The PCA was performed separately for each year from 2017-2019, with the weed biomass data log-transformed to normalize these data in 2017 and 2019. 2.3 Results Weather Daily temperature and monthly precipitation data for May through September of 2016 to 2019 was compared to the historic average between 1981 and 2010. Daily minimum, maximum, and average temperature was similar to historic values for most of the growing season in each year (Table A4). The months of May, August, and September of 2018 were the exception, with daily minimum temperatures being 3.9-4.8 º C higher than the historic average (NEWA; Northeast Regional Climate Center). Historic data suggests uniform monthly precipitation between 8.1 and 10.1 cm per month between May and September, resulting in an average of 46.6 cm of total precipitation during the growing season (Table A5). In 2017 and 2019, total precipitation between May and September was similar to the historic amount, but rainfall events in these years were less evenly distributed. Similar observations were made in 2018; however, there was nearly a 5 cm precipitation deficit by the end of this growing season relative to the historic values. Regardless of year, average daily temperature increased steadily from budbreak through sample period 4 before declining slightly in the interim between the last sample period and harvest (Table A6). Precipitation was usually greatest between sampling periods 3 and 4 with the exception of 2019, when rainfall was greater between sample period 4 and harvest relative to the other intervals. 24 Percent weed cover Table 1. Percent weed cover in a certified organic orchard of In every year, main ‘Honeycrisp’/‘Budagovsky.9’ apple trees grown in Ithaca, NY. These data represent the average value for each main treatment, split treatment, and sample period in a given year. treatment, split treatment, and Percent weed cover 2017 2018 2019 sample period each had significant Main treatment Cultivation 52.6 A 70.3 A 77.5 A effects on percent weed cover Mulch 36.6 B 55.9 B 70.2 B Main Control 60.3 A 70.4 A 74.9 AB Split treatment (Table 1). The Mulch treatment had Ammoniated soap 43.6 BC 56.0 B 64.9 B Mowing 62.5 A 78.0 A 85.0 A less weed cover than the Cultivation Capric acid 39.6 C 45.2 C 56.5 B Split Control 53.7 AB 82.9 A 90.4 A Sample Period treatment for the duration of the 1 39.9 B 54.3 B 67.6 B 2 49.8 AB 65.1 A 79.7 A experiment, but neither was 3 53.8 AB 73.5 A 72.2 AB 4 55.9 A 69.1 A 77.3 AB Statistical significance significantly different from the Main treatment *** *** * Split treatment *** *** *** Sample Period * *** ** main treatment Control in 2019 Main × Split ns ns ns Main × Sample Period ns ns ns when the average percent cover in Split × Sample Period ns * ns Main × Split × Sample Period ns ns ns Different letters, ‘A,’ ‘B,’ ‘C,’ indicate mean separation at p ≤ 0.05 using all the main treatments was 70- Tukey’s honestly significant difference test. Letters indicating mean separation are specific to either the main treatments, the split treatments, or the time points and letters. 78%, regardless of split treatment ns: nonsignificant, *: significant at p ≤ 0.05, **: significant at p ≤ 0.01, ***: significant at p ≤ 0.001. or time point. Among the split treatments, the Control and Mowing treatments resulted in percent cover that was 90% and 85%, respectively, in 2019. Capric acid and Ammoniated soap split treatments reduced percent cover relative to the Mowing or split treatment Control and always maintained percent cover less than 65%. These differences in percent cover among the split treatments were observed regardless of main treatment or sample period. Across all treatments, percent cover was lowest during sample period 1 then increased significantly at sample period 2. Percent cover never increased more than 6% between sample periods 2 and 4. 25 Aboveground weed biomass Aboveground weed biomass was significantly affected by main treatment and there was also an interaction between split treatment and sample period in each year (Table A7). Despite annual increases in aboveground * α = 0.05 weed biomass in this study, the Mulch treatment maintained 50- * * 70 g‧m-2 less weed biomass * * compared to the main treatment * * * Control while weed biomass in the Cultivation treatment was Figure 1. Aboveground weed biomass collected from within a 0.5 m2 quadrat during four sample periods between May and August in 2017, 2018, and 2019 always similar to the Control, in a certified organic orchard of ‘Honeycrisp’/‘Budagovsky.9’ apple trees grown in Ithaca, NY. regardless of split treatment. Regardless of main treatment, the split treatments of Ammoniated soap, Capric acid, and Mowing all significantly reduced weed biomass compared to treatment combinations with the split treatment Control and this was observed during most sample periods. Weed biomass in the weedy check (Control treatment) as well as Mulch and Cultivation × split treatment Control combinations was greater than the other split treatments by sample period 3 in 2017, and by sample period 2 in 2018 and 2019 (Figure 1). By sample period 4 in 2019, the average weed biomass in the weedy check and Mulch and Cultivation × split treatment Control was nearly 950 g‧m-2 while combinations with the other split treatments yielded an average of 120 g‧m-2. Rapid growth of Solidago spp. and S. lanceolatum were the main contributors to differences in weed biomass among the split treatment. 26 Table 2. Foliar mineral content of leaf tissue samples from a certified organic ‘Honeycrisp’/‘B.9’ apple orchard in Ithaca, NY. Macronutrients Micronutrients Ca C Cu K Mg N P B Fe Mn Zn (mg‧g-1) (mg‧g-1) (mg‧kg-1) (mg‧g-1) (mg‧g-1) (mg‧g-1) (mg‧g-1) (mg‧kg-1) (mg‧kg-1) (mg‧kg-1) (mg‧kg-1) 2016 Main treatments Cultivation 13.14 472.74 X 8.97 X 10.18 Y 2.22 18.09 X 3.02 Y 27.18 Y 83.54 29.89 39.80 Mulch 12.80 470.30 XY 8.45 XY 11.78 X 2.08 17.34 X 3.66 X 28.76 X 80.76 29.31 42.66 Main 12.40 469.91 Y 8.14 Y 11.38 X 2.09 16.20 Y 3.63 X 28.68 X 83.18 31.49 Control 47.22 Split treatments Ammoniated 12.71 469.76 8.54 9.76 b 2.21 17.74 ab 3.04 b 27.63 82.15 25.02 b soap 36.01 b Mowing 12.77 471.25 8.51 12.17 a 2.08 16.36 b 3.97 a 28.90 84.42 29.27 ab 48.66 a Capric acid 12.86 471.68 8.53 10.35 b 2.13 18.37 a 2.77 b 27.39 78.97 37.36 a 39.57 ab Split control 12.77 471.25 8.51 12.17 a 2.08 16.36 b 3.97 a 28.90 84.42 29.27 ab 48.66 a Statistical significance of main and split treatments Main ns * ** ** ns *** * * ns ns ns Split ns ns ns *** ns *** *** ns ns * ** Main × Split ns ns * ns * ns ns ns ns ns ns 2017 Main treatments Cultivation 13.26 X 458.90 X 27.14 11.58 Y 2.84 18.13 X 3.48 35.57 96.07 33.79 143.31 Y Mulch 11.56 Y 457.07 XY 24.08 13.01 XY 2.72 16.61 Y 3.80 35.53 107.14 44.70 166.15 X Main Control 13.62 X 454.93 Y 26.03 13.48 X 2.95 16.97 Y 4.03 36.79 96.16 44.38 171.55 X Split treatments Ammoniated soap 12.63 455.96 25.30 12.11 2.81 ab 17.50 ab 3.58 37.15 104.05 33.83 156.85 Mowing 12.47 457.57 28.34 13.75 3.17 a 16.88 ab 4.16 37.11 101.15 45.76 174.14 Capric acid 13.28 457.36 24.89 11.92 2.64 b 18.20 a 3.46 35.18 99.61 44.73 151.89 Split control 12.88 456.99 24.47 12.97 2.73 b 16.36 b 3.88 34.41 94.34 39.51 158.46 Statistical significance of main and split treatments Main * ** ns * ns ** ns ns ns ns ** Split ns ns ns ns ** ** ns ns ns ns ns Main × Split ns ns ns * ns * ns ns ns ns ns 2018 Main treatments Cultivation 12.57 457.57 54.99 10.91 2.50 18.44 2.22 Y 34.41 70.33 33.17 271.61 Mulch 11.88 456.19 54.56 12.42 2.42 17.60 3.11 X 34.17 68.26 43.60 269.18 Main Control 11.95 455.61 58.51 12.03 2.51 17.09 3.09 X 35.68 69.13 45.10 294.12 Split treatments Ammoniated soap 12.46 455.77 56.54 11.61 2.43 17.53 2.48 b 33.79 69.84 36.43 274.54 Mowing 11.78 455.72 55.44 11.96 2.50 17.30 3.15 a 35.57 70.15 44.41 290.44 27 Capric acid 12.27 458.44 55.87 11.35 2.38 18.47 2.23 b 33.94 67.86 42.40 263.42 Split control 12.03 455.90 56.23 12.23 2.60 17.53 3.37 a 35.73 69.12 39.24 284.81 Statistical significance of main and split treatments Main ns ns ns ns ns ns *** ns ns ns ns Split ns ns ns ns ns * *** ns ns ns ns Main × Split ns ns ns ns ns ns ns ns ns ns ns 2019 Main treatments Cultivation 6.50 X 469.27 33.06 9.17 Y 2.09 X 15.88 X 1.82 Y 47.84 59.30 Y 40.64 144.27 Mulch 5.64 Y 462.58 34.50 10.01 X 1.92 Y 14.69 Y 2.18 X 48.52 78.74 X 43.72 147.63 Main Control 6.61 X 452.87 35.90 10.11 X 2.09 X 15.41 XY 2.20 X 46.61 67.62 XY 47.15 152.43 Split treatments Ammoniated soap 6.36 456.24 33.57 9.65 2.02 ab 15.42 2.09 47.35 72.71 42.06 141.21 Mowing 6.33 462.14 34.14 9.80 2.16 a 15.43 2.18 48.39 66.67 45.21 146.83 Capric acid 5.99 464.99 34.06 9.35 1.90 b 15.54 1.94 47.78 69.62 48.49 145.39 Split control 6.31 462.92 36.19 10.25 2.06 ab 14.93 2.05 47.11 65.21 39.59 159.00 Statistical significance of main and split treatments Main *** ns ns ** *** ns ** ns ** ns ns Split ns ns ns ns ** ns ns ns ns ns ns Main × Split * ns ns ns ns ns ns ns ns ns ns 1 Means in the same column followed by the same letter are not significantly different according to Tukey’s HSD test (P≤0.05). Uppercase letters X and Y designate significant differences among main treatments. Lowercase letters a and b designate significant differences among split treatment plots. 28 Apple tree leaf nutrient Table 3. Ratio of foliar carbon to foliar nitrogen (C:N) of apple leaf tissue samples collected between 2016 and 2019 from a certified content organic ‘Honeycrisp’/‘B.9’ apple orchard in Ithaca, NY. Ratio of apple leaf carbon to nitrogen 2016 2017 2018 2019 Regardless of split Main treatments Cultivation 26.19 Y 25.40 Y 24.98 29.63 Y Mulch 27.45 XY 27.77 X 26.07 31.60 X treatment, foliar carbon (C) and Main Control 29.30 X 27.15 XY 26.82 29.48 Y Split treatments nitrogen (N) were generally Ammoniated soap 26.61 b 26.41 ab 26.13 29.78 Mowing 29.04 a 27.32 ab 26.55 30.08 greater in the Cultivation Capric acid 25.89 b 25.25 b 24.96 29.98 Split control 29.04 a 28.12 a 26.18 31.10 treatment than the main Statistical significance of main and split treatments Main treatment *** * ns ** Split treatment *** * ns ns treatment Control, with Main × Split treatment ns * ns * 1 Means in the same column followed by the same letter are not significantly statistically significant different according to Tukey’s HSD test (P≤0.05). Uppercase letters X and Y designate significant differences among main treatments. Lowercase letters a and b designate significant differences among split treatment plots. differences occurring in 2016 and 2017 for both elements (Table 2). Split treatment had no effect on foliar C, but foliar N in the Capric acid split treatment was significantly greater than the split treatment Control in the first two years, regardless of main treatment. Despite these statistical differences, foliar N was deficient in all treatments, in all years, as no treatment combination ever reached the minimum threshold of 22.0 mg‧g-1 for young bearing apple trees (Table 2). Treatment differences in the ratio of foliar C:N were the opposite of trends for foliar N (Table 3). The Mulch treatment had the highest C:N ratio and the difference was significant relative to the Cultivation treatment in 2017 and 2019, and relative to the main treatment Control in 2019, regardless of split treatment. Likewise, among the split treatments, the Capric acid split treatment had a lower C:N ratio than the split treatment Control with the difference being significant in 2016 and 2017, regardless of main treatment. Treatment effects on foliar phosphorus (P) and potassium (K) were similar in all years (Table 2). The Cultivation treatment had 0.4 to 0.9 mg‧g-1 less foliar P and 1-2 mg‧g-1 less K than 29 the other main treatments and this was significant in 3 of the 4 sample years. Among the split treatments, Capric acid and Ammoniated soap had less foliar P and K than the split treatment Control and Mowing treatment, but this was only significant for both elements in 2016 and for P in 2018. Although the trends for foliar P and K were similar, foliar K was deficient in nearly all treatments, regardless of year, while foliar P was always adequate. Foliar content of calcium was often inadequate and especially in the Mulch treatment which had 1 to 2 mg‧g-1 less foliar Ca than the other main treatments in 2017 and 2019. Foliar Ca content was not affected by split treatment. Similar to foliar K, foliar magnesium (Mg) was deficient in all treatments regardless of year, with all treatment combinations having less than 3.5 mg‧g-1 foliar Mg (Table 3). With the exception of manganese (Mn), which was deficient in 71-94% of experimental units each year, foliar content of micronutrients was satisfactory. Statistical differences among main or split treatments were rare, and never occurred for more than one year. Soil mineral content Table 4. Ratio of soil carbon to soil nitrogen (C:N) of soil samples (15 cm depth) collected from 2017-2019 in a certified organic apple orchard of ‘Honeycrisp’/’Budagovsky.9’ trees in Ithaca, NY. Total soil C increased Ratio of soil carbon to nitrogen 2017 2018 2019 Main treatments over time in the Mulch treatment, Cultivation 11.63 Y 15.74 Y 10.35 Z Mulch 12.66 X 17.36 X 14.50 X from 31.8 mg‧g-1in 2017 to 36.7 Main Control 11.38 Y 16.36 Y 11.99 Y Split treatments mg‧g-1in 2019 (Table 4). In the Ammoniated soap 11.68 17.03 12.56 Mowing 12.07 16.24 12.09 Capric acid 11.95 16.26 12.30 Cultivation treatment and the Split Control 11.86 16.40 12.17 main treatment Control, soil C Statistical significance of main and split treatments Main treatment *** ** *** Split treatment ns ns ns was relatively unchanged, Main × Split treatment ns ns ns 1 Means in the same column followed by the same letter are not significantly -1 different according to Tukey’s HSD test (P≤0.05). Uppercase letters X, Y, and averaging 28 mg‧g , over the Z designate significant differences among main treatments. three years. As a result, soil C in the Mulch treatment was significantly greater than the other main treatments in 2018 and 2019. Total soil N did not vary by main or split treatment at any 30 time during the course of the study. Therefore, the ratio of soil C:N was closely related to total soil C. In the Mulch treatment, the ratio of soil C:N was 6-21% greater than the main treatment Control. Alternatively, soil C:N in the Cultivation treatment was similar to the main treatment Control except in 2019 when the ratio was 14% lower in the Cultivation treatment compared to the Control. Soil P, K, and Mn followed a similar trend wherein the Mulch treatment resulted in significantly greater content of each mineral compared to Cultivation and the main treatment Control, regardless of year (Table 5). Between 2017 and 2019, Mulch increased soil P by 21 to 38%, soil K by 21 to 28%, and soil Mn by 44 to 57% compared to the other main treatments. Soil Mg content was not affected by main treatment and averaged 329 mg‧kg-1 among the main treatments over the course of the study. Soil Fe responded similarly except in 2017 when soil Fe in the Mulch treatment was 10.5 mg‧kg-1, which was double the soil Fe content in the other main treatments. Soil Zn was inconsistently affected by main treatment. In 2018, soil Zn in the Cultivation treatment was 33% lower than the main treatment Control at 1.13 mg‧kg-1. In 2017 and 2019, though, soil Zn in the Cultivation and Mulch treatment was not significantly different from the Control. The differences in soil mineral content among the main treatments were observed regardless of split treatment. 31 Table 5. Macro- and micronutrient content of soil samples (15 cm depth) collected from a certified organic apple orchard of ‘Honeycrisp’/‘Budagovsky.9’ trees in Ithaca, NY. Soil Macronutrients Soil Micronutrients C K Mg N P Fe Mn Zn Year (mg‧g-1 ) (mg‧kg-1) (mg‧kg-1) (mg‧g-1) (mg‧kg-1) (mg‧kg-1) (mg‧kg-1) (mg‧kg-1) 2017 Main treatment Cultivation 29.48 X 114.83 Y 336.32 2.53 8.30 Y 4.63 Y 15.71 Y 1.26 Mulch 31.76 X 133.53 X 326.20 2.51 9.67 X 10.54 X 23.31 X 1.36 Main Control 27.2 Y 106.34 Y 310.24 2.39 7.69 Y 5.69 Y 13.94 Y 1.40 Split treatment Ammoniated soap 28.66 123.89 349.66 a 2.45 8.78 6.82 17.32 1.10 b Mowing 29.07 110.85 340.31 ab 2.41 7.60 6.92 17.11 1.23 ab Capric acid 29.99 114.25 296.64 b 2.53 8.70 6.94 18.82 1.42 ab Split Control 30.21 123.95 310.41 ab 2.54 9.14 7.14 17.37 1.62 a Statistical significance of main and split treatments Main treatment *** *** ns ns *** *** *** ns Split treatment ns ns ** ns ns ns ns * Main × Split treatment ns ns ns * ns ns ns ns 2018 Main treatment Cultivation 27.25 Y 191.45 Y 369.04 1.73 5.66 Y 1.52 4.67 1.13 Y Mulch 32.10 X 239.68 X 374.43 1.86 7.00 X 3.48 8.11 1.43 XY Main Control 28.17 Y 182.12 Y 348.94 1.73 5.15 Y 2.48 5.17 1.50 X Split treatment Ammoniated soap 28.54 214.64 386.93 1.68 5.83 1.81 4.98 1.11 Mowing 28.66 194.94 383.92 1.77 5.48 2.35 5.56 1.43 Capric acid 29.45 213.31 340.12 1.81 6.48 3.15 7.43 1.42 Split Control 30.03 194.78 345.58 1.84 5.95 2.66 5.97 1.46 Statistical significance of main and split treatments Main treatment *** *** ns ns *** ns ns * Split treatment ns ns ns ns ns ns ns ns Main × Split treatment ns ns ns ns ns ns ns ns 2019 Main treatment Cultivation 27.22 Y 113.14 Y 309.89 2.78 6.36 Y 3.27 16.73 Y 2.27 Y Mulch 36.65 X 134.52 X 310.03 2.58 8.66 X 4.57 24.31 X 3.26 X Main Control 28.01 Y 103.97 Y 282.52 2.41 6.15 Y 4.36 17.12 Y 3.04 XY Split treatment Ammoniated soap 29.27 121.36 316.27 2.42 6.51 3.39 17.67 2.85 Mowing 30.69 117.86 319.51 2.61 6.92 4.03 19.70 2.71 Capric acid 31.20 119.04 280.46 2.66 7.02 4.05 19.61 2.93 Split Control 31.34 110.58 287.01 2.68 7.78 4.79 20.56 2.93 Statistical significance of main and split treatments Main treatment *** *** ns ns *** ns *** ** Split treatment ns ns * ns ns ns ns ns Main × Split treatment ns ns ns ns ns ns ns ns 1 Means in the same column followed by the same letter are not significantly different according to Tukey’s HSD test (P≤0.05). Uppercase letters X, Y, and Z designate significant differences among main treatments. Lowercase letters a and b designate significant differences among split treatment plots. 32 Soil health The Mulch treatment generally had the highest values for all aspects of soil health measured in this study (Table 6). Soil organic matter (OM) in the Mulch treatment increased from 4.81 to 5.21%, between 2017 and 2019 while soil OM in the Cultivation treatment declined from 4.46 to 4.06%, over this same period. Wet aggregate stability in the Mulch and Cultivation treatments remained similar to the main treatment Control through 2019, but the Mulch treatment did have significantly higher aggregate stability than Cultivation in the final year. The Mulch and Cultivation treatments were further differentiated by soil respiration which was 1.46 mg CO2‧g -1 in the Mulch treatment and 0.97 mg CO ‧g-12 in the Cultivation treatment in 2019. Compared to the main treatment Control, soil respiration in the Mulch treatment was 29% greater while, in the Cultivation treatment, soil respiration was 14% lower. Like soil mineral content, these differences in soil health among the main treatments were observed regardless of split treatment. 33 Table 6. Soil biophysiochemical properties determined from soil samples (15 cm depth) collected from a certified organic apple orchard of ‘Honeycrisp’/‘Budagovsky.9’ trees in Ithaca, NY. Available Wet Soil ACE soil Soil Active water aggregate organic protein respiration Year Carbon Soil pH capacity stability matter index (mg -1 -1 -1 (mg‧g -1) (g‧g ) (%) (%) (mg‧g ) CO2‧g ) 2017 Main treatment Cultivation 0.35 XY 26.65 4.46 XY 9.99 1.04 Y 0.72 Y 6.63 Mulch 0.37 X 22.79 4.81 X 11.67 1.22 X 0.81 X 6.76 Main Control 0.33 Y 23.49 4.18 Y 9.86 0.99 Y 0.68 Y 6.53 Split treatment Ammoniated soap 0.35 19.72 b 4.51 10.53 1.03 b 0.76 6.76 Mowing 0.35 23.66 b 4.31 9.79 1.06 b 0.70 6.69 Capric acid 0.35 22.78 b 4.47 10.08 1.00 b 0.73 6.52 Split Control 0.35 31.07 a 4.64 11.62 1.25 a 0.75 6.59 Statistical significance of main and split treatments Main treatment ** ns *** * ** *** ns Split treatment ns *** ns ns ** ns ns Main × Split ns ns ns ns ns ns ns treatment 2018 Main treatment Cultivation 0.33 Y 15.78 Y 4.78 Y 10.45 Y 0.91 Y 0.77 Y 6.85 Mulch 0.36 X 21.60 X 5.33 X 13.03 X 1.12 X 0.92 X 6.99 Main Control 0.33 Y 19.41 XY 4.66 Y 10.78 Y 1.06 XY 0.82 Y 6.79 Split treatment Ammoniated soap 0.34 17.01 4.87 11.24 0.88 b 0.84 7.02 Mowing 0.33 20.24 4.87 10.88 1.02 ab 0.81 6.89 Capric acid 0.35 18.15 4.98 11.96 1.04 ab 0.85 6.79 Split Control 0.33 20.33 4.97 11.59 1.18 a 0.84 6.81 Statistical significance of main and split treatments Main treatment *** ** *** ** ** *** ns Split treatment ns ns ns ns ** ns ns Main × Split ns ns ns ns ns ns ns treatment 2019 Main treatment Cultivation 0.33 Y 29.57 Y 4.06 Y 10.35 Y 0.97 Z 0.90 Y 6.74 Mulch 0.35 X 36.38 X 5.21 X 16.24 X 1.46 X 1.07 X 6.72 Main Control 0.34 Y 32.29 Y 4.09 Y 11.25 Y 1.13 Y 0.88 Y 6.59 Split treatment Ammoniated soap 0.33 29.25 b 4.26 11.79 1.07 b 0.94 6.83 Mowing 0.34 33.28 ab 4.49 12.53 1.20 b 0.94 6.70 Capric acid 0.34 32.67 ab 4.51 12.82 1.04 b 0.94 6.62 Split Control 0.34 35.80 a 4.56 13.34 1.42 b 0.96 6.58 Statistical significance of main and split treatments Main treatment ** *** *** *** *** *** ns Split treatment ns * ns ns *** ns ns Main × Split ns ns ns ns ns ns ns treatment 1 Means in the same column followed by the same letter are not significantly different according to Tukey’s HSD test (P≤0.05). Uppercase letters X, Y, and Z designate significant differences among main treatments. Lowercase letters a and b designate significant differences among split treatment plots. 34 Soil volumetric water content Main treatment significantly influenced soil volumetric water content (VWC) during 6 of the 10 sampling dates in 2019 (Table A8, Figure 2). On 6, 13, and 25 June soil VWC in the Mulch treatment was 6-10% greater than the main treatment Control. Soil VWC was similar among Figure 2. Soil volumetric water content (VWC) measured once weekly (7.6 cm depth) in the tree row of a certified organic orchard of ‘Honeycrisp’/ ‘Budagovsky.9’ apple trees in Ithaca, NY, between 6 June and 6 September 2019. the main treatments when sampled between 8 July and 8 August. On 19 and 26 August and 6 September, soil VWC in the Mulch treatment was similar to the main treatment Control while the Cultivation treatment reduced soil VWC 17-21% compared to the Control. From 8 July through 6 September, split treatment was a significant factor in soil VWC (Table A8, Figure 2). At sampling dates in this time period soil VWC was not significantly different among the Mowing, Capric acid, or Ammoniated soap treatments. However, soil VWC was 21-55% greater in these treatments compared to the split treatment Control, regardless of main treatment. 35 Trunk cross-sectional area (TCSA) In spring 2016, prior to the initial α = 0.05 * application of weed * control treatments, TCSA * was uniform and the average TCSA was 1.20 cm2 (Table A9). Between spring 2016 and winter Figure 3. Trunk cross-sectional area at 40 cm above the graft union, between spring (S) 2016 and winter (W) 2019, for ‘Honeycrisp’/’Budagovsky.9’ apples 2019, trees in the trees grown in a certified organic orchard in Ithaca, NY. Cultivation treatment grew 42% more than those in the main treatment Control and 48% more than those in the Mulch treatment. These percentage differences are based on the average TCSA within the four split treatments of each main treatment. Within the Cultivation treatment, final TCSA in winter 2019 was similar across all split treatments with an average TCSA of 4.79 cm2 (Figure 3). Regardless of split treatment, trees in the Mulch treatment grew similarly between 2016 and 2019 and had a final average TCSA of 3.78 cm2. Conversely, within the main treatment Control, trees in the Capric acid and Ammoniated soap split treatments grew 279% while trees in the Mowing and split treatment Control grew 221% over the course of the study. Ultimately, trees in the main treatment Control × Capric acid and Ammoniated soap treatments had a TCSA of 4.49 cm2 which was statistically similar to the final TCSA measurement in the Cultivation main treatment. Additionally, trees in the Control × Capric acid or Ammoniated soap treatments were 0.80 cm2 larger than those in the Mowing treatment or split treatment Control (Figure 3). 36 Bloom and crop density Table 7. Flower density (clusters‧TCSA-1) and crop density (fruit‧TCSA-1) in a In 2018, bloom certified organic orchard of ‘Honeycrisp’/‘Budagovsky.9’ apple trees grown in Ithaca, NY. These data represent the mean values of both main and split density was similar among treatments across the four blocks in each year. Flower Density Crop Density -1 the main treatments, (clusters‧TCSA ) (fruit‧TCSA -1) 2018 20191 2018 20191 Main treatment regardless of split Cultivation 2.77 0.37 X 0.94 0.24 X Mulch 2.72 0.04 Y 0.96 0.03 Y Main Control 2.34 0.21 XY 1.02 0.13 XY treatment, and the average Split treatment Ammoniated soap 3.02 0.16 1.15 0.1 density was 2.61 Mowing 2.00 0.14 0.81 0.13 Capric acid 2.83 0.34 0.91 0.22 Split Control 2.44 0.17 0.97 0.06 clusters‧TCSA-1 (Table 7). Statistical significance of main and split treatments Main treatment ns *** ns *** Split treatment ns ns ns ns Split treatment did not Main treatment × split treatment ** ns *** ns 1 Different letters, ‘X’ and ‘Y’ indicate mean separation at p ≤ 0.05 using Tukey’s impact bloom density, but honestly significant difference test. Mean separation is within column and letters should not be compared across columns. ns: nonsignificant, *: significant at p ≤ 0.05, **: significant at p ≤ 0.01, ***: there was a two-way significant at p ≤ 0.001. interaction between main and split treatment in 2018. Within the Mulch treatment, the Ammoniated soap split treatment had 4.65 clusters‧TCSA-1 which was significantly more than the 0.53 clusters‧TCSA-1 in the split treatment Control. Bloom density did not differ among the split treatments in the Cultivation treatment or the main treatment Control in 2018. In 2019, bloom density was significantly impacted by main treatment only; the Cultivation treatment had 0.37 clusters‧TCSA-1 compared to 0.04 in the Mulch treatment. Neither treatment resulted in a bloom density which was significantly different from the main treatment Control (Table 7). Similar to bloom density, there was a two-way interaction between main and split treatment regarding crop density in 2018 (Table 7). Within the Mulch treatment, the Ammoniated soap split treatment resulted in a crop density of 1.62 fruits‧TCSA-1 which was 4 times greater than the split treatment Control. Crop density was reduced in the main treatment Control × Capric acid treatment compared to the weedy check – 0.56 versus 1.65 fruits‧TCSA-1, 37 respectively. Crop density was statistically similar among all split treatments in the Cultivation treatment with an average of 0.95 fruits‧TCSA-1. In 2019, main treatment was the only significant factor for crop density. Within the Cultivation treatment, crop density was 0.25 fruits‧TCSA-1 which was more than the 0.03 fruits‧TCSA-1 in the Mulch treatment; neither treatment resulted in crop density significantly different from the main treatment Control (Table 7). Multivariate Analyses Correlation matrix A correlation matrix was created to evaluate the relationships among the dependent variables. There were few instances in which two variables had a significantly positive or negative correlation. In fact, no variables were significantly negatively correlated, and all cases of significant positive correlations were in the case of the soil health variables to one another (Figure 4). Though not always significant, all variables related to soil health generally had a positive correlation to one another (Figure 4). For example, total soil carbon and active soil carbon were positively correlated in 2018 and 2019 (r2 = 0.82 in both years). Soil organic matter was positively correlated with soil respiration and total soil carbon in 2019 and, in addition to those, was positively correlated with total soil nitrogen and the ACE soil protein index in 2017. Total soil carbon was strongly positively correlated with total soil N in 2017 and 2018, but not in 2019. Lastly, the ACE soil protein index was positively correlated with active carbon in 2018 and total soil carbon in 2019. Interestingly, trunk cross-sectional area was often negatively correlated with soil health variables to some degree. There was no statistically significant relationship between 38 TCSA and leaf nitrogen in any year (r2 = 0.33 to 0.58) and no significant correlation was found between TCSA and weed biomass, nor between leaf nitrogen and weed biomass. 39 1 2017 2018 2019 Figure 4. Correlation matrices for visualization of linear relationships among all variables measured during three study years – 2017 (left), 2018 (middle), 2019 (right) – in a certified organic orchard of ‘Honeycrisp’/‘Budagovsky.9’ apple trees grown in Ithaca, NY. Weed biomass data log-transformed in 2017 and 2019 to normalize data in those years 40 Principal component analysis A principal component analysis was conducted to explore the variability among our datasets, and biplots were created to visualize the effects of our treatments on that variability. Species scores indicated which variables were most strongly related to each principal component, with a species score near ±1 being most significantly related to each principal component. There was a significant separation between the Mulch treatment and the other main treatments along principal component 1 (PC1) in every year (Table 8). Regardless of year, PC1 was most strongly correlated with soil health parameters including total soil C, total soil N (2017 and 2018), organic matter, ACE protein index, respiration (2017 and 2019), active carbon, and wet aggregate stability (2018 and 2019). There was no significant separation among the split treatments along PC1 except in 2019 when the Ammoniated soap split treatment was significantly different from the split treatment Control. PC1 explained at least 42% of the variation in our dataset in every year (Figure 5). The PCA did not reveal any main × split treatment interaction effects for PC1 or PC2; however, there was significant separation among the main treatments and, separately, among the split treatments for PC2 in every year (Table 8). In 2017, PC2 was most strongly correlated with TCSA and leaf N. There was significant separation between the Cultivation and Mulch treatments in this year, as well as between the split treatment Control and Mowing split treatment. In 2018, PC2 was most strongly correlated with leaf N. Among the main treatments, there was significant separation between the Cultivation treatment and the main treatment Control while, among the split treatments, there was a significant separation between the Capric acid and Mowing split treatments in this year. Lastly, in 2019, PC2 was most strongly correlated 41 with volumetric water content and the log-transformed weed biomass at time point 4. Among the main treatments, there was a significant separation between the Mulch treatment and the main treatment Control while, among the split treatments, the split treatment Control was significantly different from all the other split treatments. Table 8. Mean separation of position of main and split treatments along principal components 1 and 2. Principal component analysis included weed biomass, TCSA, leaf N, soil N, soil C, and soil health parameters measured across three years in a certified organic orchard of ‘Honeycrisp (Firestorm)’/ ‘Budagovsky.9’ apple trees in Ithaca, NY. 2017 2018 2019 Principal Principal Principal Principal Principal Principal component 1 component 2 component 1 component 2 component 1 component 2 Main treatment Main Control 0.358 Y 0.034 XY 0.179 Y 0.286 Y 0.268 Y -0.196 X Cultivation 0.081 Y -0.392 X 0.443 Y -0.317 X 0.534 Y -0.059 XY Mulch -0.439 X 0.358 Y -0.622 X 0.031 XY -0.802 X 0.254 Y Split treatment Split Control -0.358 -0.281 a -0.197 0.138 ab -0.290 a -0.987 a Mowing 0.165 0.331 b -0.065 0.264 b -0.032 ab 0.233 b Capric acid 0.093 -0.203 ab 0.052 -0.321 a 0.057 ab 0.426 b Ammoniated soap 0.100 0.153 ab 0.211 -0.081 ab 0.265 b 0.327 b Statistical significance of main and split treatments Main treatment ** ** *** ** *** ** Split treatment ns * ns ** * *** Main × Split ns ns ns ns ns ns treatment 1 Different letters, ‘X’ and ‘Y’ indicate mean separation of the main treatments at p ≤ 0.05 using Tukey’s honestly significant difference test. Different letters, ‘a’ and ‘b’ indicate mean separation of the split treatments at p ≤ 0.05 using Tukey’s honestly significant difference test. Mean separation is within column and letters should not be compared across columns. ns: nonsignificant, *: significant at p ≤ 0.05, **: significant at p ≤ 0.01, ***: significant at p ≤ 0.001. 42 Figure 5. Biplots of principal component analysis for variables measured in a certified organic orchard of ‘Honeycrisp’/‘Budagovsky.9’ apple trees grown in Ithaca, NY. Ellipses represent main treatments (top row) and split treatments (bottom row). 43 2.4 Discussion Weed Control Efficacy When used as individual treatments (i.e., the split treatment Control), both the Mulch and Cultivation treatments resulted in similar weed biomass as the main treatment Control in 2018 and 2019. Stacking the Mulch and Cultivation treatments with either Mowing, Capric acid, or Ammoniated soap significantly reduced weed biomass relative to these main treatments × split treatment Control. However, weed biomass in the Mulch and Cultivation stacked treatments was always similar to the main treatment Control × Mowing, Capric acid, or Ammoniated soap. Therefore, stacking significantly improved weed control in the Mulch and Cultivation treatments, but the split treatments were similarly effective regardless of the main treatment. The weed management treatments we evaluated in our study are commonly included in other weed control studies; however, these treatments tend to be compared as individual practices rather than as combinations. In New York, for example, Agnello et al. (2017) found bark mulch or monthly applications of organic soap, limonene, or caprylic acid yielded the most weed free area while Peck et al. (2011) found bark mulch plus a single glyphosate application provided adequate weed suppression for two years. In the drier climate of Washington, Granatstein et al. (2010) also reported on the efficacy of wood mulch for weed control in orchards. Many of these studies also confirm that weed pressure was greater where the Wonder Weeder was used for repeated cultivation relative to where wood mulch was applied (Agnello et al., 2017; Peck et al., 2011; Granatstein et al., 2010). Rapid growth of rhizomatous perennial dicot weeds in our split treatment Control, regardless of main treatment, could be one reason we did not observe the statistically significant benefits of stacked weed management we anticipated. Regardless of main treatment, there was 44 such a disparity in weed biomass between the split treatment Control and the other three split treatments that more subtle differences among these treatments or main × split treatment combinations were not revealed through our analyses. Perennial weeds have underground storage reserves which are diminished by shoot growth but replenished by photosynthate from the aboveground growth (Mohler et al., 2021). The increasing weed biomass over the course of the study suggests our intervals for management were not short enough to exhaust the belowground storage reserves sufficiently to reduce biomass of perennial weeds in the Cultivation treatment. However, our results demonstrated the efficacy of organic herbicides for control of perennial weeds and may justify the associated expense of these products for use during the establishment period of an orchard. Weed control in the Cultivation treatment may have been improved with more frequent cultivation events, but this has associated time, labor, and equipment costs (Merwin et al., 1995; Peck et al., 2010), as well as potential negative consequences for soil health (Merwin et al., 1994). It is well understood that dead organic mulches, as a single strategy, are insufficient to control rhizomatous perennial weeds (Hammermeister, 2016). Effects on soil health and soil mineral content By the final sampling date in 2019, the Mulch treatment improved nearly all aspects of soil health relative to the main treatment Control, regardless of split treatment. Similar improvements to orchard soil health following wood chip mulch application have been widely reported in other similar studies (Glover et al., 2000; Haynes, 1980; Peck et al., 2011; St. Laurent et al., 2008; Yao et al., 2005) and hay mulches also tend to improve soil organic matter in orchards (Merwin et al., 1994; Zoppolo et al., 2011). In the sandier textured soils of Washington, however, application of wood chip mulch was not sufficient to increase organic matter after 3 45 years (Granatstein and Mullinix, 2008; Hoagland et al. 2008). In a low organic matter silt loam soil of Arkansas, Choi et al. (2011) reported no increase in soil organic matter three years after applying wood chip mulch. Although soil organic matter (OM) declined in the Cultivation treatment, it was not significantly different from the main treatment Control, regardless of split treatment, at the end of our study. The Wonder Weeder was also utilized by Peck et al. (2011) who reported an increase in soil organic matter after four years of cultivation (3x per season), which they attributed to incorporation of weed biomass and chicken manure compost. The ground-driven action of the Wonder Weeder causes less soil disturbance than PTO-driven tillage implements – one possible reason for the slow loss of soil OM observed with this implement (Magdoff and van Es, 2009). Zoppolo et al. (2011) also used ground-driven implements including a tooth arrow tiller and notch disk tiller to control weeds alongside the tree row of newly established apple trees. After four years, soil OM in the tilled area of the tree row was similar to the undisturbed, vegetated area at both 0-10 cm and 0-30 cm depths. For comparison, Merwin et al. (1994) reported an 18% decline in soil organic matter after 5 years of rototilling for weed control despite following a similar tillage schedule and being on a similar soil as in our experiment. Soil respiration was the one aspect of soil health that significantly differentiated all three main treatments after four years of management. By 2019, soil respiration was 29% greater in the Mulch treatment than the main treatment Control, while respiration was 14% lower than the Control in the Cultivation treatment. Soil biological activity is generally stimulated by the addition of carbon rich amendments like wood chip mulch (Peck et al., 2011; Yao et al., 2005) and hindered by frequent cultivation (Magdoff and van Es, 2009). 46 Regardless of split treatment, the Mulch treatment improved soil fertility, as it increased soil content of P and K relative to the main treatment Control and this has been reported in other studies on similar soils (Atucha et al., 2011; Merwin and Stiles, 1994; Peck et al., 2011). The Mulch treatment also increased soil Mn relative to Control. While Atucha et al. (2011) reported a similar finding for the duration of fifteen years when a 15 cm layer of mulch was applied every 2-3 years, Peck et al. (2011) observed a 29% decline in soil Mn following mulch application. Three years after mulch application, Merwin et al. (1995) found no difference in soil fertility between areas managed with mulch or herbicide for weed control. Soil availability of nutrients following mulch application will be influenced by mulch composition and age, time, soil type, and weather conditions. Availability and uptake of nitrogen There were no significant differences in total soil nitrogen among our main or split treatments, yet apple leaf N in the Cultivation treatment was significantly greater than the main treatment Control in 2016-2018, and significantly greater than the Mulch treatment in 2017 and 2019. Hoagland et al. (2008) also reported leaf N was higher where the Wonder Weeder was implemented relative to where wood chip mulch was applied, regardless of compost application to supply fertility in both treatments. Comparatively, Zoppolo et al. (2011) found alfalfa mulch significantly improved soil nitrate and ammonium content compared to the tilled, vegetated, and flame-weeded areas of the tree row. Likewise, Neilsen et al. (2014) found total soil nitrogen was greater in the alfalfa and bark mulch treatments compared to the tilled and black plastic treatments. Microbial immobilization of otherwise plant-available N is one contributing factor to reduced leaf N following application of carbon-rich wood chip mulch, a phenomenon Hoagland 47 et al. (2008) documented using isotopic analyses. However, application of mulches with a lower C:N ratio – such as alfalfa – prevents soil N immobilization (Magdoff and van Es, 2009) and can enhance N uptake (Forge et al., 2013). Total soil nitrogen measures both organic and inorganic forms of nitrogen, which vary in their short and long-term availability for crop growth (Moebius-Clune et al., 2017). Our soil analyses suggest that organic soil nitrogen, which gradually becomes available through mineralization (Glover et al., 2000), was greater in the Mulch treatment than the other main treatments. The Mulch treatment resulted in greater soil respiration and ACE soil protein index; two characteristics of soil health that suggest greater microbial biomass N than the other main treatments (Moebius-Clune et al., 2017). Although we did not measure short-term benefits of mulch application for soil N fertility, soil organic matter accumulated following wood chip mulch application can be a source of nitrate-N in the longer-term (Atucha et al., 2011). Soil availability is not the only factor affecting root uptake of soil nutrients; soil conditions must also permit uptake when nutrients are in demand. Although apple trees rely on internal N reserves rather than soil N for initial spring growth (Cheng and Fuchigami, 2002; TerAvest et al., 2010), there is a period of rapid N uptake from bloom through the end of shoot growth each year (Cheng and Raba, 2009; TerAvest et al., 2010). Soil VWC was 6-10% higher in the Mulch treatment than the main treatment Control in June 2019, a critical period for nutrient uptake due to overlapping and high growth rate of shoots, roots, and fruit (Psarras et al., 2000). Soil moisture was measured in 2019 only but, if increased soil moisture in this treatment persisted in the earlier study years, overall N accumulation may have been inhibited by poor root growth (Merwin et al., 1992) or soil N loss via denitrification or leaching (Hoagland et al., 2008). Increased soil VWC where wood chips or other organic mulches were applied has been 48 reported in other studies across a range of soil types (Agnello et al., 2017; Hoagland et al., 2008; Merwin et al, 1992; Mullinix, 2008; Walsh et al., 1996). Leaf N content was often greater in the herbicide treatments, especially the Capric acid treatment, compared to the split treatment Control. In 2016, leaf N was 2.5 and 3.0 mg‧g-1 greater in the Mulch × Ammoniated soap or Capric acid treatments, respectively, relative to Mulch × split treatment Control. Similarly, in 2017, leaf N in the main treatment Control × Ammoniated soap or Capric acid treatments were 2.8 and 3.7 mg‧g-1 greater, respectively, than the Control × split treatment Control. Although other differences were not statistically significant, the Capric acid split treatment tended to increase leaf N 0.5 to 2.0 mg‧g-1 relative to the split treatment Control, regardless of main treatment or year. The effect of the Ammoniated soap treatment on leaf N was less consistent across years. Enhanced weed control in the herbicide treatments may have contributed to greater availability of plant-available soil nitrogen which resulted in greater apple leaf N content. Measuring N content of harvested weed biomass would have been one way to determine differences in competition for N among treatments based on weed pressure. Variability in leaf N is likely the result of variable availability or uptake of soil N, as N was otherwise applied as uniform foliar applications to the entire study area (Table A3). Apple Leaf Nutrient Content There were statistically significant differences in leaf Ca, K, Mg, and P in at least two study years. With the exception of phosphorus, though, leaf content of these macronutrients was deficient among every main × split treatment combination in 2 (Ca), 3 (K), or 4 (Mg) years as well (Stiles and Reid, 1991). Leaf P was at least optimum for all treatments in 2017-2019 and was significantly affected by both main and split treatment during the study. Leaf P was 49 significantly lower in the Cultivation treatment relative to the other main treatments in 2016 and 2018-19. Although soil P was similar between the main treatment Control and Cultivation treatments, soil disturbance in the Cultivation treatment may inhibited the growth of soil mycorrhizae – a group of fungal organisms known to improve availability and uptake of soil P (Magdoff and van Es, 2009). Additionally, leaf P was greater in the Mowing treatment and split treatment Control compared to Ammoniated soap and Capric acid in 2016 and 2018. The prevalence of macronutrient deficiency in this study highlights the challenge of nutrient management in organic cropping systems, which is commonly referenced in other similar studies (Hoagland et al., 2008). Tree growth and yield response While weed control is a critical factor for initial and long-term profitability of an organic orchard (Hammermeister, 2016), it is not the only factor influencing tree growth and productivity. In the Cultivation treatment, regardless of disparities for weed control among the split treatments, TCSA increased significantly more than the other main treatments – nearly 400% between spring 2016 and fall 2019. Trees in the Cultivation treatment had the largest TCSA by the fall 2019 but TCSA in the main treatment Control × Capric acid and Ammoniated soap combinations were statistically similar to those in the Cultivation treatment. The similarity in TCSA among these treatments follows a similar pattern as leaf N. Although leaf N content was always deficient for this establishing an apple orchard (Stiles and Reid, 1991), the previously discussed increase in leaf N content in the Cultivation main treatment or the Capric acid and Ammoniated soap split treatments could have been biologically significant to cause a boost in tree growth. 50 Despite improvements to soil health and reduced weed pressure in the Mulch treatment, this treatment resulted in poor tree growth which was similar to the main treatment Control × split treatment Control, regardless of split treatment. Increased soil moisture where wood chip mulch was applied resulted in satisfactory to excellent tree growth in similar studies in Washington state (Hoagland et al., 2008; TerAvest et al., 2010), but mulches may offer benefits for tree growth such as reduced incidence of root lesion nematodes (Forge et al., 2013). Unfortunately, studies in New York have regularly reported increased tree mortality related to increased soil moisture following wood chip mulch (Agnello et al., 2017) or hay-straw (Merwin et al., 1992). In British Columbia, though, Neilsen et al. (2014) observed both improvements to soil health and tree growth after applying bark mulch in the tree row. In spite of reduced leaf N, bark mulch in that study resulted in the largest TCSA during the third, fourth, fifth years after planting ‘Ambrosia’/‘B.9’ trees. At that site, where soil conditions were favorable for tree growth and N was not limiting due to regular compost application and fertigation, there were few consistent differences in tree size or fruit yield and quality by the sixth growing season. Fruit yield and fruit quality were not commercially acceptable in the bearing years of 2018 and 2019 in this experiment – the fourth and fifth years after planting, respectively. Robinson et al. (2011) demonstrated cumulative yield of ‘Honeycrisp’/‘B.9’ in the first 7 years after planting to be 40 kg‧tree-1 in a tall spindle production system. In our study, only 59% of trees bore fruit in 2018 and only 21% bore fruit in 2019. Of those that did bear fruit, the average yield was 1.1 kg‧tree-1 in 2018 and 0.6 kg‧tree-1 in 2019. Insufficient leaf N fertility (Stiles and Reid, 1991) coupled with weed competition and excess soil moisture are all likely contributors to poor tree growth and yield in this study. 51 2.5 Conclusions The substantial investment cost of establishing a high-density apple orchard demands early yield of high-quality fruit to increase the rate of return on investment. In the present study, no combination of weed management tactics yielded commercially acceptable quantity or quality of fruit. Improvements to soil health and reduced weed biomass in the Mulch treatment failed to promote tree growth and productivity. Cultivation increased tree growth relative to the weedy check, but yield was not optimized. The organic herbicides and Mowing split treatments reduced weed biomass, regardless of the main treatment, yet only the herbicide split treatments improved tree growth relative to the weedy check. There was no demonstrated benefit for tree growth or productivity with stacking weed management strategies. These results highlight the critical importance of appropriate site selection and choice of scion/rootstock combination. Substantial response to management was expected with the choice of ‘Honeycrisp’ and ‘Budagovsky.9’; this dwarfing combination is likely among the most susceptible to variations in weed control. At this wet site and under the limitations of organic fertility amendments, a more vigorous scion/rootstock combination could have adapted more successfully. In addition to the importance of site selection and rootstock/scion selection, weed management decisions also need to be made based on maximizing tree growth and productivity during the establishment period of the orchard. For example, wood chip mulch is a natural choice for weed control in organic orchards, particularly when it is a readily available resource. In heavier textured soils like those in upstate New York; however, the present study supports other similar studies that find wood chip mulch may not be a suitable weed management tactic during the establishment period of an apple orchard. 52 Based on the range of results from this study and others, the success of in-row weed management strategies depends on soil texture, the vigor of the scion/rootstock, ability to maintain adequate tree fertility, and the existing weed community. Perennial fruit trees are less susceptible to weed competition as they mature, so managers of organic orchards must recognize the opportunity to implement more sustainable practices that improve soil health and provide other ecosystems services exists later in the life of the orchard. Weed management strategies must be tailored to the site and age of the orchard with an emphasis on providing a suitable soil environment for root growth and nutrient uptake for establishing orchards. This is especially important in an organic orchard where there is a higher cost of production, cost-effective weed management strategies are limited, and readily plant-available sources of nitrogen are scarce. 53 References Agnello, A., K. Cox, J. Lordan, P. Francescatto, and T. Robinson. 2017. Comparative programs for arthropod, disease and weed management in New York organic apples. Insects 8(3):96. https://doi.org/10.3390/insects8030096. Atucha, A., I.A. Merwin, and M.G. Brown. 2011. Long-term effects of four groundcover management systems in an apple orchard. Hortic. Sci. 46(8):1176-1183. Bradshaw, T. 2017. Non-chemical weed management in apple orchards. University of Vermont Extension, Burlington, Vermont. Breth, D. 2015. Critical weed control requirements in young high density apple orchards. New York Fruit Quarterly. 23(4):11-15. Cheng, L. and L.H. Fuchigami. 2002. Growth of young apple trees in relation to reserve nitrogen and carbohydrates. Tree Physiology. 22(18):1297-1303. https://doi- org.proxy.library.cornell.edu/10.1093/treephys/22.18.1297. Choi, H., C.R. Rom, and M. Gu. 2011. Effects of different organic apple production systems on seasonal nutrient variations of soil and leaf. Scientia Horticulturae. 129(1):9-17. Forge, T., G. Neilsen, D. Neilsen, E. Hogue, and D. Faubion. 2013. Composted dairy manure and alfalfa hay mulch affect soil ecology and early production of ‘Braeburn’ apple on M.9 rootstock. HortScience. 48(5):645-651. https://doi.org/10.21273/HORTSCI.48.5.645. Glover, J.D., J.P. Reganold, and P.K. Andrews. 2000. Systematic method for rating soil quality of conventional, organic, and integrated apple orchards in Washington State. Agriculture, Ecosystems & Environment. 80(1-2):29-45. https://doi- org.proxy.library.cornell.edu/10.1016/S0167-8809(00)00131-6. Granatstein, D. and E. Sánchez. 2009. Research knowledge and needs for orchard floor management in organic tree fruit systems. International Journal of Fruit Science. 9(3):257-281. https://doi.org/10.1080/15538360903245212. Granatstein, D. and K. Mullinix. 2008. Mulching options for Northwest organic and conventional orchards. HortScience. 43(1):45-50. https://doi.org/10.21273/HORTSCI.43.1.45. Granatstein, D., M. Wiman, E. Kirby, and K. Mullinix. 2010. Sustainability trade-offs in organic orchard floor management. Acta Horticulturae. 873:115-122. Granatstein, D., P. Andrews, and A. Groff. 2014. Productivity, economics, and fruit and soil quality of weed management systems in commercial organic orchards in Washington State, USA. Org. Agr. 4(3):197-207. https://doi.org/10.1007/s13165-014-0068-0. Hammermeister, A.M. 2016. Organic weed management in perennial fruits. Scientia Horticulturae. 208:28-42. https://doi.org/10.1016/j.scienta.2016.02.004. Haynes, R.J. 1980. Influence of soil management practice on the orchard agro-ecosystem. Agro- Ecosystems. 6(1):3-32. https://doi-org.proxy.library.cornell.edu/10.1016/0304- 3746(80)90003-7. Hoagland, L., L. Carpenter-Briggs, D. Granatstein, M. Mazzola, J. Smith, F. Peryea, and J.P. Reganold. 2008. Orchard floor management effects on nitrogen fertility and soil biological activity in a newly established organic apple orchard. Biol Fertil Soils. 45(1):11-18. https://doi.org/10.1007/s00374-008-0304-4. 54 Leinfelder, M.M., I.A. Merwin, and M.G. Brown. 2012. Soil health indicators, apple tree growth, and carbon sequestration differ among orchard groundcover management systems. Acta Hort. 938:333-340. Magdoff, F. and H. van Es. 2009. Building soils for better crops: sustainable soil management. 3rd ed. Sustainable Agriculture and Research Education (SARE) program, Handbook series book 10. College Park, Maryland. Merwin, I.A. and J.A. Ray. 1997. Spatial and temporal factors in weed interference with newly planted apple trees. HortScience. 32(4):633-637. https://doi.org/10.21273/HORTSCI.32.4.633. Merwin, I.A. and W.C. Stiles. 1994. Orchard groundcover management impacts on apple tree growth and yield, and nutrient availability and uptake. J. Amer. Soc. Hort. Sci. 119(2):209-215. https://doi.org/10.21273/JASHS.119.2.209. Merwin, I.A., D.A. Rosenberger, C.A. Engle, D.L. Rist and M. Fargione. 1995. Comparing mulches, herbicides, and cultivation as orchard groundcover management systems. HortTech. 5(2):151-158. https://doi.org/10.21273/HORTTECH.5.2.151. Merwin, I.A., W. Stiles, and H. van Es. 1994. Orchard groundcover management impacts on soil physical properties. J. Amer. Soc. Hort. Sci. 119(2):216-222. https://doi.org/10.21273/JASHS.119.2.216. Merwin, I.A., W.F. Wilcox, and W.C. Stiles. 1992. Influence of orchard ground cover management on the development of Phytophthora crown and root rots of apple. Plant Dis. 76(2):199-205. Mia, M.J., F. Massetani, G. Murri, J. Facchi, E. Monaci, L. Amadio, and D. Neri. 2020. Integrated weed management in high density fruit orchards. Agronomy. 10:1492. https://doi.org/10.3390/agronomy10101492. Moebius-Clune, B.N., D.J. Moebius-Clune, B.K. Gugino, O.J. Idowu, R.R. Schindelbeck, A.J. Ristow, H.M. van Es, J.E. Thies, H.A. Shayler, M.B. McBride, K.S.M. Kurtz, D.W. Wolfe, and G.S. Abawi. 2017. Comprehensive assessment of soil health – the Cornell framework. 3rd ed. Cornell University, Geneva, NY. Mohler, C.L., J.R. Teasdale, and A. DiTommaso. 2021. Manage weeds on your farm: a guide to ecological strategies, SARE handbook series. Sustainable Agriculture Research & Education (SARE), College Park, Maryland. Neilsen, G.H. and E.J. Hogue. 1985. Effect of orchard soil management on the growth and leaf nutrient concentration of young dwarf red delicious apple trees. Can. J. Soil. Sci. 65(2):309-315. https://doi-org.proxy.library.cornell.edu/10.4141/cjss85-034. Neilsen, G., T. Forge, D. Angers, D. Neilsen, and E. Hogue. 2014. Suitable orchard floor management strategies in organic apple orchards augment soil organic matter and maintain tree performance. Plant Soil. 378:325-335. https://doi.org/10.1007/s11104-014- 2034-8. Peck, G. and I.A. Merwin. 2009. A grower’s guide to organic apples. New York State IPM. Publication Number 223. Peck, G.M., I.A. Merwin, J.E. Thies, R.R. Schindelbeck, and M.G. Brown. 2011. Soil properties change during the transition to integrated and organic apple production in a New York orchard. Applied Soil Ecology. 48(1):18-30. https://doi.org/10.1016/j.apsoil.2011.02.008. Peck, G.M., I.A. Merwin, M.G. Brown, and A.M. Agnello. 2010. Integrated and organic fruit production systems for ‘Liberty’ apple in the Northeast United States: a systems-based 55 evaluation. Hort. Sci. 45(7):1038-1048. https://doi- org.proxy.library.cornell.edu/10.1016/j.apsoil.2011.02.008. Psarras, G., I.A. Merwin, A.N. Lakso, and J.A. Ray. 2000. Growth phenology, root longevity, and rhizosphere respiration of field grown ‘Mutsu’ apple trees on ‘Malling 9’ rootstock. HortScience. 125(5):596-602. https://doi.org/10.21273/JASHS.125.5.596. Robinson, T.L., S. Hoying, M.M. Sazo, A. DeMaree, and L. Dominguez. A vison for apple orchard systems of the future. New York Fruit Quarterly. 21(3)11-16. Robinson, T.L., A.M. DeMarree and S.A. Hoying. 2007. An economic comparison of five high density apple planting systems. Acta Hortic. 732:481-489. Robinson, T.L., A.N. Lakso and Z. Ren. 1991. Modifying apple tree canopies for improved production efficiency. HortScience. 26(8):1005-1012. Robinson, T.L., S.A. Hoying, and G.H. Reginato. 2006. The tall spindle apple production system. New York Fruit Quarterly. 14(2):21-28. Soil Science Division Staff. 2017. Soil survey manual. C. Ditzler, K. Scheffe, and H.C. Monger (eds.). USDA Handbook 18. Government Printing Office, Washington, D.C. St. Laurent, A., Merwin, I.A., and Thies, J.E. 2008. Long-term orchard groundcover management systems affect soil microbial communities and apple replant disease severity. Plant and Soil. 304(1):209-225. https://doi.org/10.1007/s11104-008-9541-4. Stefanelli, D., R.J. Zoppolo, R.L., Perry, and F. Weibel. 2009. Organic orchard floor management systems for apple effect on rootstock performance in the midwestern United States. HortScience. 44(2):263-267. https://doi.org/10.21273/HORTSCI.44.2.263. Stiles, W.C. and W.S. Reid. 1991. Orchard Nutrition Management. Cornell Cooperative Extension. Information Bulletin 2019. TerAvest, D., J.L. Smith, L. Carpenter-Boggs, L. Hoagland, D. Granatstein, and J.P. Reganold. 2010. Influence of orchard floor management and compost application timing on nitrogen partitioning in apple trees. HortScience. 45(4):637-642. https://doi.org/10.21273/HORTSCI.45.4.637. Walsh, B.D., A.F. MacKenzie, S. Salmins and D.J. Buszard. 1996. Impact of soil management systems on organic dwarf apple orchards and soil aggregate stability, bulk density, temperature and water content. Canadian Journal of Soil Science. 76(2):203-209. https://doi-org.proxy.library.cornell.edu/10.4141/cjss96-028. Williams, M.A., J.G. Strang, R.T. Bessin, D. Law, D. Scott, N. Wilson, S. Witt, and D.D. Archbold. 2015. An assessment of organic apple production in Kentucky. HortTech. 25(2):154-161. https://doi.org/10.21273/HORTTECH.25.2.154. Yao, S., I.A. Merwin, G.W. Bird, G.S. Abawi and J.E. Thies. 2005. Orchard floor management practices that maintain vegetative or biomass groundcover stimulate soil microbial activity and alter soil microbial community composition. Plant and Soil. 271(1):377-389. https://doi.org/10.1007/s11104-004-3610-0. Zoppolo, R.J., D. Stefanelli, G.W. Bird, and R.L. Perry. 2011. Soil properties under different orchard floor management systems for organic apple production. Organic Agriculture. 1(4):231-246. https://doi.org/10.1007/s13165-011-0018-z. 56 Chapter 3: Weed community changes in response to in-row management tactics across three growing seasons in an organic, high density apple orchard in New York State Abstract Weed management remains a major barrier to the success of certified organic apple production in New York, and other regions with humid climates, where weed competition in the first 3-5 years after planting can diminish long term yield potential and return on investment. We sought to evaluate the efficacy of in-row weed management strategies and document changes in the weed community during the establishment period of a high-density apple orchard. In 2015, eight rows of ‘Honeycrisp (Firestorm)’/‘Budagovsky.9’ apple trees were planted (0.9 m × 3.7 m) in a NOFA-NY certified organic orchard block at the Cornell University Agricultural Experiment Station Orchards in Ithaca, NY. A split-plot experimental design was implemented in 2016 with three main treatments including an untreated Control (main treatment Control), cultivation with a Wonder Weeder® (Cultivation), and surface-applied wood chip mulch (Mulch) and four split treatments including an untreated Control (split treatment Control), mowing with a string trimmer (Mowing), and organic herbicides Suppress® (Capric acid) and Final-San-O® (Ammoniated soap). All treatment combinations were implemented four times per season between 2016 and 2019, approximately monthly, from May to August. Mulch was only applied once, in spring 2016. The Mulch treatment maintained significantly less biomass than the main treatment Control through 2019, regardless of split treatment or sample period. Among the split treatments, an abundance of creeping herbaceous perennial weeds such as, Solidago spp. and Symphyotrichum lanceolatum resulted in significantly more weed biomass in the split treatment Control compared to the three other split treatments, regardless of main treatment. Monocot weed biomass increased significantly in the two organic herbicide split treatments 57 compared to the split treatment Control, except within the Mulch treatment. Alternatively, the two organic herbicides improved control of simple herbaceous perennials in the Cultivation treatment and main treatment Control. In the absence of secondary management, biomass from creeping perennials was statistically similar among all main treatments. Regardless of main treatment, biomass from creeping perennials was significantly reduced when the Mowing, Ammoniated soap, or Capric acid split treatments were implemented. Therefore, the effects of stacked weed management tactics varied based on the type or life cycle of weeds present. This highlights the need for implementing weed management strategies that consider the existing weed community to maximize weed control while minimizing costs. 58 3.1 Introduction Adoption of certified organic apple production in New York State and the Northeastern U.S. is constrained by high weed growth potential related to frequent rainfall and high humidity during the growing season. At present, area devoted to certified organic apple production comprises less than 1% of the total planted apple area in New York State (USDA, 2019). Cultivation, organic mulches, and mowing are commonly used for weed control in the tree row of certified organic orchards (Bradshaw, 2017; Merwin and Peck, 2009); however, these practices can be less effective (Granatstein et al., 2010) and more costly (Granatstein and Sanchez, 2009; Merwin et al., 1995; Peck, et al., 2010; Rowley et al., 2011) than synthetic herbicides that effectively manage weeds in conventional apple production systems. Nonetheless, cumulative yield potential and economic return are maximized when early yields are achieved in modern, high-density orchards (Robinson et al., 2006; Robinson et al., 2007) and research in New York has repeatedly demonstrated the yield benefits of maintaining a weed free area in the tree row during May, June, and July (Breth and Tee, 2017; Merwin and Ray, 1997). Widespread adoption of organic apple production in the region has been limited in part because no weed control or soil fertility practices have been developed that are without consequences for tree growth (Agnello et al., 2017). Increasing the number of selective pressures on weeds in the tree row by combining weed management tactics could be one way to maintain weed competition below an economic threshold (Swanton and Weise, 1991). This concept was coined the ‘many little hammers’ approach to weed management by Liebman and Gallandt (1997), who explain the potential for synergistic effects of complementary weed management strategies. 59 The efficacy of any management strategy varies across weed communities, as different species have characteristics that make them more or less vulnerable to different management strategies (Hammermeister, 2016; Mohler et al., 2021; Swanton and Weise, 1991). For example, Granatstein et al. (2014) demonstrated that a 10 cm layer of wood chip mulch in the tree row provided effective weed control for only one year when Elytrigia repens – a creeping perennial weed – was present. The same treatment resulted in three years of effective weed control at a different orchard site where E. repens was not present (Granatstein and Mullinix, 2008). Correct identification of the weed community and an understanding of weed ecology can help orchardists select suitable weed management strategies and manage adaptively as weed communities in the orchard shift over time (Clements et al., 1994). In the present study, 12 combinations of weed management strategies were implemented in a high-density, certified organic apple orchard in New York and repeated monthly between May and August of three consecutive growing seasons. The main objective was to determine the efficacy of these treatment combinations for weed control during the establishment period of an orchard. Some weeds are more difficult to manage with the tools available for organic production, so our secondary objective was to document the effects of these management strategies on the weed community over time. Although this chapter includes information about specific weed species observed in this study, weeds were grouped by cotyledon type and life cycle for most statistical comparisons. This higher level of comparison allows for observations in this study to be applicable to orchard systems as a whole, regardless of the specific weed community. Our hypotheses were as follows: (1) Weed communities would diverge over time based on differences in management 60 (2) Increasing the number of weed management strategies would decrease weed biomass and reduce species richness and diversity due to the increased number of selective filters (3) Organic herbicides and mowing would favor monocot weeds due to their growing point being located at or below the soil surface (4) Treatments without physical disturbance of the soil would favor establishment of perennial weeds 3.2 Materials and Methods Study Site The 0.16-ha site was located on NOFA-NY certified organic land at the Cornell University Agricultural Experiment Station orchards in Ithaca, NY, USA (lat. 42.44519º, long. - 76.45912º). The soils at the site were 78% Hudson and Collamar silt loams and 22% Hudson silty clay loam, moderately well drained, with 2-6% slopes (Soil Science Division Staff, 2017). Semi-dwarf apple trees were grown at this site from 1981-2006 and the site was fallowed following tree removal. In fall 2014, the field was plowed and disked after one line of drainage tile was installed through a wet area on the north end of the site. No cover crop was seeded, and no lime was applied as the soil pH was optimal at 6.8 (E. Shatt, personal communication). The eight tree rows associated with this study were located on the easternmost end of a block containing a total of 24 rows. All trees were planted in the spring of 2015 and the rows were planted to run in a North-South direction. From East to West, the entire block included 15 NOFA-NY certified organic rows, then two organically managed buffer/transition rows, then seven conventionally managed rows. European black alder (Alnus glutinosa L.) flanked the north 61 and east sides of the block, and a 10-m wide grass buffer strip separated the block from a conventionally managed apple block on the south side. A fallow field sat to the West of the block for the duration of the study. Experimental Design Eight rows of ‘Honeycrisp (Firestorm)’/‘Budagovsky.9’ trees comprised the easternmost rows of this certified organic block and were used for this study. The trees were trained following the tall spindle system and spaced 0.9 m between trees and 3.7 m between rows (Robinson et al., 2006). A randomized, split-plot design with four complete blocks was implemented in 2016. Each of the four blocks in this split-plot design consisted of two tree rows which were divided into three main plots, one for each main treatment. Main plots were subsequently divided into four split-plots, one for each split treatment. Each of these split plots (n = 48) represented one experimental unit (Figure A1). Main plots were separated by six buffer trees while split plots were separated by three buffer trees. Main treatments included Cultivation (Wonder Weeder, Harris Manufacturing, Burbank, WA), Mulch (surface-applied mix of hard and soft woods, 15 cm depth, sourced from the Cornell University Grounds Department), and the main treatment Control (untreated control/weedy check). Split treatments included Mowing (hand-held string trimmer), Capric acid [Suppress® Herbicide EC, 6% solution (57 L a.i. ha-1), Westbridge Agricultural Products, Vista, CA], Ammoniated soap [Final-San-O®, (163 L a.i. ha-1), Certis USA, LLC, Columbia, MD], and the split treatment Control (untreated control). Both Suppress® and Final-San-O® are contact, post-emergent, non-selective herbicides approved by the Organic Materials Review Institute for 62 use in certified organic production. These products are most effective on newly emerged weeds less than 15 cm tall. There was a total of twelve treatment combinations (3 main  4 split). The main treatment Control × split treatment Control, which will hereafter be referred to as the ‘weedy check’. Wood chip mulch was applied only once, in May 2016, while all other treatments were repeated four times per season within 1-3 days following each aboveground weed biomass sampling date (Table A1). Within the Cultivation treatment, cultivation preceded herbicide application and cultivation came after Mowing. Treatments were applied with the intent to control weeds within 0.5 m on either side of the tree row (Figure A2). After leaf drop each autumn, a string trimmer was used to mow all weeds in the tree row to reduce habitat for meadow voles (Microtus pennsylvanicus) and prevent subsequent damage over the winter. A diverse stand of monocot and dicot weeds in the inter-row space was mowed approximately four times per season. Weed Biomass Aboveground weed biomass was harvested on four sample dates each year from 2017 to 2019. The specific sampling date varied each year (Table A1) but these dates are grouped into sampling periods for comparison of biomass across years. Sample period 1 represents early to mid-May, period 2 represents early June, period 3 represents late June to early July, and period 4 represents late July to early August. On each sampling date, a 0.5 m2 quadrat was placed directly in the tree row between two randomly selected trees (Figure A2) and all biomass rooted within the quadrat was harvested at the ground level using scissors and/or hand sheers. Locations were 63 unique to each sampling date, and no location was sampled more than once per season. One sample was collected per plot (n = 48) on each sampling date. Harvested biomass was separated into monocot and dicot weeds then the dicot weeds were separated by species. Monocots were identified to the species level in 2019. Aboveground plant material of each species was placed in a separate and smaller paper bag. The bags were oven-dried at 70 ºC for seven days before the dry weight was determined. Analyses in this chapter are based on cotyledon type and weed life cycle. Simple and creeping herbaceous perennials are hereafter referred to as simple perennials and creeping perennials, respectively. A detailed analysis of total weed biomass and percent weed cover is presented in Chapter 2. This chapter covers a more in-depth analysis of weed community assembly in this certified organic apple orchard. Statistical Analysis Data analysis was performed in R (version 1.3.1056). Data were analyzed with linear mixed models which included main treatment (Cultivation, Mulch, Control), split treatment (Mowing, Capric acid, Ammoniated soap, Control) and their interactions as fixed effects. Block was a random effect. Sample date was included as a fixed effect in the model. Therefore, an additional random effect term was Block × Sample Date. Years were analyzed separately. Each model was subjected to an ANOVA to determine the significance of treatment and interaction effects, which were considered significant at the α = 0.05 level according to Tukey’s HSD test. Weed community assembly was analyzed using the vegan package in R and a permutational multivariate analysis of variance (PERMANOVA) was conducted using the ‘adonis’ function in this package. The subject of these analyses was weed species data collected 64 in 2019. Each sampling date was analyzed separately, and weeds present in only one plot per sample period were removed prior to analysis. These data were also fitted to nonmetric multidimensional scaling (NMDS) ordination plots to aid in visualization. 3.3 Results Total Weed Biomass The Mulch treatment reduced total weed biomass relative to the main treatment Control for the duration of the study, but the effect of the main treatment became less significant over time (Table 1). From 2017 to 2019, weed biomass increased annually across all main treatments. Averaged across the sampling periods, biomass in the Mulch treatment more than doubled from 83 to 191 g‧m-2 while biomass in the main treatment Control and Cultivation treatments increased 71% and 67%, respectively (Table A2). Despite this more rapid rate of weed biomass accumulation in the Mulch treatment α = 0.05 * compared with the other main treatments, * the Mulch treatment * * maintained * * * significantly less weed * biomass than the main treatment Control Figure 1. Aboveground weed biomass (g‧m-2) in a certified organic orchard of ‘Honeycrisp’/‘Budagovsky.9’ apple trees grown in Ithaca, NY. These data represent through 2019 and less the aboveground weed biomass per each split treatment in each sample period in 2017- 2019. Values were averaged across the main treatments. 65 biomass than the Cultivation treatment through 2018 (Table A10). Regardless of the main treatment, there was an interaction between split treatment and sample period in each year with respect to weed biomass (Table A10). Weed biomass in the split treatment Control was significantly greater than the other split treatments by sample period 3 in 2017, and sample period 2 in 2018 and 2019 (Figure 1). Within each of the split treatments, aboveground weed biomass was statistically similar across all sample periods, regardless of year. The split treatment Control was the exception, where biomass tended to increase with each passing sample period; in 2019, the increase in biomass between sample periods was statistically significant and increased from 116 to 909 g‧m-2 during the season (Figure 1). Monocot and Dicot Weed Biomass Monocot weed biomass represented 18% of the total weed biomass in 2017 and increased annually to comprise 30% of the total weed biomass harvested in 2019. Correspondingly, dicot weed biomass represented 89% of the total in 2017 and declined to 70% of the total by 2019 (Table A13). Monocot weeds were not identified to the species level in 2017 or 2018 but, in 2019, the most common monocot weeds were Elytrigia repens, Poa annua, and Poa trivialis (Table A11). Dicot weeds were identified to the species level in each year and the most abundant species included Solidago spp., Symphyotrichum lanceolatum, Ranunculus bulbosus, Cirsium arvense, Plantago lanceolata, Plantago major, and Taraxacum officinale (Table A11). Treatment effects on dicot weed biomass were similar to those for total weed biomass due to the high percentage of dicots in the total. Dicot weeds were less prevalent in the Mulch treatment than the other main treatments in 2017 and 2018. On average, the Mulch treatment 66 yielded 37% less biomass from dicot weeds than the Control and Cultivation treatments in 2017 and 30% less in 2018 (Table 1). Biomass from monocot weeds was also 50% lower in the Mulch treatment than the other main treatments through 2019 (Table A12). Among the split treatments, Table 1. Aboveground biomass of all dicotyledon weeds (g‧m-2) in each main treatment (Cultivation, Mulch, main Control), split total weed biomass harvested at treatment (Ammoniated soap, Mowing, Capric acid, split Control), and sample period (1-4) from 2017-2019. Significant interactions are indicated by letters. These values are based on the weed community each sampling period was in a certified organic orchard of 'Honeycrisp (Firestorm)’/ ‘Budagovsky.9’ apple trees in Ithaca, NY. statistically similar in the Mowing, Aboveground biomass of dicotyledon weeds (g‧m -2) 2017 2018 2019 Main treatment Capric acid, and Ammoniated soap Control 120.8 A 123.3 A 171.0 Cultivation 109.9 A 127.9 A 145.0 split treatments (Figure 1), yet the Mulch 73.3 B 87.8 B 158.0 Split treatment Control 153.0 A 263.2 A 437.3 A cotyledon type of weeds tended to Mowing 101.0 B 106.0 B 106.7 B Capric acid 85.4 B 29.7 C 24.5 C Ammoniated soap 65.9 B 53.0 C 64.0 BC differ significantly among these Sample period 1 123.7 78.4 B 85.1 C treatments. Monocot weed biomass 2 84.8 96.0 AB 136.1 BC 3 87.5 127.5 AB 164.5 B 4 109.4 150.0 A 246.8 A was significantly greater in the Statistical significance Main treatment *** ** ns Split treatment *** *** *** main treatment Control × Sample period * * *** Main × Split ns ns ns Ammoniated soap and Capric acid Main × Sample period ns ns ns Split × Sample period *** *** *** Main × Split × Sample period ns ns ns combinations than in the weedy Different letters, ‘A,’ ‘B,’ ‘C,’ indicate mean separation at p ≤ 0.05 using Tukey’s honestly significant difference test. Letters indicating mean separation are specific to either the main treatments, the split treatments, or check and main treatment Control the sample periods and letters should not be compared across columns or across sections in a single column. × Mowing treatment (Figure 2). ns: nonsignificant, *: significant at p ≤ 0.05, **: significant at p ≤ 0.01, ***: significant at p ≤ 0.001. The organic herbicide treatment combinations yielded 75 and 144 g‧m-2 of monocot biomass in 2018 and 2019, respectively, while the weedy check and Mowing treatment combination yielded an average of 4.8 and 36 g‧m-2 in these years, respectively. Similarly, the Cultivation × Capric acid treatment yielded 67 and 139 g‧m-2 monocot weed biomass in 2018 and 2019, respectively, which was at least double 67 the biomass recorded in the Cultivation × split treatment Control. Split treatment did not affect monocot weed biomass within the Mulch treatment in any year. Differences in dicot weed biomass among the split treatments were significantly affected by sample period in each year (Table 1). Dicot weed biomass in the split treatment Control gradually increased throughout each season. In 2018 and 2019, dicot weed biomass in the split treatment Control was more than double that of the other split treatments by sample period 2. By sample period 4 in 2019, biomass from dicot weeds in the split treatment Control was 811 g‧m-2 compared to an average of 59 g‧m-2 in the other split treatments, regardless of main treatment. Figure 2. Total weed biomass by cotyledon type and life cycle harvested during sample period four (9 August 2017, 30 July 2018, 29 July 2019) in a certified organic orchard of 'Honeycrisp'/'Budagovsky.9' apple trees in Ithaca, NY. Total weed biomass and biomass from creeping perennial dicots was significantly greater in the split treatment Control than all other split treatments in every year. In 2018 and 2019, biomass from monocotyledon weeds was signfiicantly greater in the Capric acid (× main treatment Control and Cultivation) and Ammoniated soap (× main treatment Control only) split treatments compared to combinations with the split treatment Control. 68 Dicot Weeds by Life Cycle Table 2. Aboveground biomass of dicotyledon weeds (g‧m -2) with a creeping perennial life cycle in each main treatment (Cultivation, Mulch, main Control), split treatment (Ammoniated soap, Mowing, Creeping perennials Capric acid, split Control), and sample period (1-4) from 2017-2019. Significant interactions are indicated by letters. These values are became the dominant type of based on the weed community in a certified organic orchard of 'Honeycrisp (Firestorm)’/ ‘Budagovsky.9’ apple trees in Ithaca, NY. Aboveground biomass of spreading perennial dicot weeds (g‧m-2) dicotyledon weed by 2019. While 2017 2018 2019 Main treatment biomass from creeping perennial Control 52.8 59.4 138.0 Cultivation 49.5 73.0 121.0 Mulch 38.1 64.9 139.0 dicots increased from 46% to 84% Split treatment Control 75.8 A 165.7 A 389.3 A Mowing 57.8 AB 40.6 B 57.3 B between 2017 and 2019, simple Capric acid 17.0 C 20.6 B 20.6 B Ammoniated soap 36.5 BC 36.2 B 54.4 B perennial dicot weed biomass Sample period 1 39.7 33.1 B 61.3 B 2 44.4 58.5 AB 109.2 B declined from 50% to 15% of the 3 46.2 66.5 AB 132.2 B 4 56.8 105.0 A 227.9 A total dicot weed biomass over this Statistical significance Main treatment ns ns ns Split treatment *** *** *** same period (Table A13, Figure 2). Sample period ns *** *** Main × Split * ns ns Main × Sample period ns ns ns Annual and biennial dicot weeds Split × Sample period ** *** *** Main × Split × Sample period ns ns ns never represented more than 4% of Different letters, ‘A,’ ‘B,’ ‘C,’ indicate mean separation at p ≤ 0.05 using Tukey’s honestly significant difference test. Letters indicating mean separation are specific to either the main treatments, the split treatments, or the total weed biomass harvested the sample periods and letters should not be compared across columns or across sections in a single column. ns: nonsignificant, *: significant at p ≤ 0.05, **: significant at p ≤ in any year (Table A13) and are 0.01, ***: significant at p ≤ 0.001 not discussed further. Main treatment had no impact on biomass from creeping perennials. Instead, there was a significant interaction between split treatment and sample period in each year with respect to creeping perennial dicots (Table 2). There were no differences among the split treatments during sample period 1 but, by sample period 3 in 2017 and sample period 2 in 2018 and 2019, the split treatment Control yielded significantly more biomass than the other split treatments, regardless of main treatment. By sample period 4 in 2019, biomass from creeping perennials was 793 g‧m-2 69 in the split treatment Control which was more than 100-times greater than biomass harvested from the Capric acid split treatment and about 14-times greater than the Ammoniated soap and Mowing split treatments. Relative to the main treatment Control, the Mulch treatment significantly reduced biomass from simple perennial dicots through 2018. In 2019; however, there was an interaction between main and split treatment (Table 3). Within the main treatment Control, the absence of Plantago lanceolata in the two organic herbicide treatments resulted in significantly less biomass from simple perennials there than the Table 3. Aboveground biomass of dicotyledon weeds (g‧m-2) with a simple perennial life cycle in each main treatment Mowing split treatment and weedy (Cultivation, Mulch, main Control), split treatment (Ammoniated soap, Mowing, Capric acid, split Control), and sample period (1- check. Within the Cultivation 4) from 2017-2019. Significant interactions are indicated by letters. These values are based on the weed community in a certified organic orchard of 'Honeycrisp (Firestorm)’/ treatment, only the Cultivation × ‘Budagovsky.9’ apple trees in Ithaca, NY. Aboveground biomass of simple perennial dicot weeds (g‧m-2) Capric acid treatment resulted in 2017 2018 2019 Main treatment Control 62.9 A 60.0 A 31.3 significantly less biomass from Cultivation 55.3 AB 45.1 AB 22.7 Mulch 34.9 B 26.8 B 16.0 Split treatment simple perennials than Cultivation × Control 69.6 A 85.1 A 38.0 A Mowing 39.4 B 61.7 A 44.0 A split treatment Control. Split Capric acid 47.7 AB 12.3 B 3.4 B Ammoniated soap 47.3 AB 16.7 B 7.8 B Sample period treatment had no significant effect on 1 81.1 A 40.4 23.0 2 39.4 B 35.6 26.2 biomass from simple perennials in 3 38.2 B 58.6 26.4 4 45.3 B 41.2 17.6 Statistical significance the Mulch treatment. Main treatment ** ** ns Split treatment * *** *** There was a significant Sample period *** ns ns Main × Split ns ns * Main × Sample period ns ns ns interaction between split treatment Split × Sample period *** * ** Main × Split × Sample period ns ns ns Different letters, ‘A,’ ‘B,’ ‘C,’ indicate mean separation at p ≤ 0.05 using and sample period in each year with Tukey’s honestly significant difference test. Letters indicating mean separation are specific to either the main treatments, the split treatments, respect to biomass from simple or the sample periods and letters should not be compared across columns or across sections in a single column. ns: nonsignificant, *: significant at p ≤ 0.05, **: significant at p ≤ perennial dicots (Table 3). Sample 0.01, ***: significant at p ≤ 0.001. 70 period 1 in 2017 was an anomaly wherein biomass from simple perennials was twice as high in the Ammoniated soap and Capric acid split treatments compared with the split treatment Control. By sample period 4 in 2017, biomass from simple perennials was significantly greater in the split treatment Control than all the other split treatments. In 2018 and 2019, biomass from simple perennials was similar between the two organic herbicide treatments and the split treatment Control during sample period 1. Within the split treatment Control, biomass from simple perennials peaked in sample period 3 in 2018 and 2019 when biomass was 132 and 72 g‧m-2, respectively, and significantly greater than the herbicide treatments. Biomass in the split treatment Control in sample period 4 remained significantly higher than the two herbicide treatments in 2018 only. Biomass in the Mowing split treatment was always similar to the split treatment Control with the exception of sample period 1 in 2019 when biomass was significantly greater in the Mowing treatment than all other treatments. Species Richness, Evenness, And Shannon Diversity Index. Species evenness was only affected by sample period and was significantly greater in sample periods 1 and 4 than sample period 3 (Table 4). There was a split treatment by sample period interaction for species richness. While there were no differences in sample period 1, the Capric acid split treatment reduced species richness relative to the split treatment Control for sample periods 2-4, regardless of main treatment. Species richness in the Mowing and Ammoniated soap split treatments was similar to the split treatment Control at all sample periods (Table 4). There was also an interaction between split treatment and sample period with respect to Shannon Diversity Index (Table 4). However, the diversity index rating for all split treatments 71 was similar to the split treatment Control in most cases. The exceptions were sample period 3 Table 4. Species evenness, species richness, and Shannon Diversity Index in when the diversity index each main treatment (Control, Cultivation, Mulch), split treatment (Control, Mowing, Capric Acid, and Ammoniated Soap), and sample period (1-4) in Capric acid was 0.38 20191. These values are based on the weed community in a certified organic orchard of 'Honeycrisp (Firestorm)’/ ‘Budagovsky.9’ apple trees in Ithaca, NY. which was significantly Species Species Shannon Evenness Richness Diversity Index lower than the index Main treatment Control 0.51 5.47 0.85 Cultivation 0.54 5.25 0.89 value of 1.04 in the split Mulch 0.58 5.41 0.96 Split treatment treatment Control and Control 0.55 6.04 A 0.95 A Mowing 0.58 6.58 A 1.08 A Capric acid 0.48 3.88 C 0.65 B sample period 4 when the Ammoniated soap 0.58 5.00 B 0.91 A Sample period 1 0.59 A 4.77 0.89 AB diversity index of 1.39 in 2 0.56 AB 5.67 0.95 AB 3 0.44 B 5.12 0.74 B the Mowing treatment was 4 0.58 A 5.94 1.01 A Statistical significance Main treatment ns ns ns significantly greater than Split treatment ns *** *** Sample period ** ** * Main × Split ns ns ns the index value of 0.91 in Main × Sample period ns ns ns Split × Sample period ns ** * the split Main × Split × Sample period ns ns ns 1 Means in the same column followed by the same letter are not significantly different according to Tukey’s HSD test (P≤0.05). Uppercase letters A, B, and C designate treatment Control. significant differences among main treatments, split treatments, or sample periods. Multivariate Analysis A permutational multivariate analysis of variance (PERMANOVA) was conducted using the complete species level data collected in 2019. The PERMANOVA was run separately for each of the four sampling periods to determine changes in treatment effect on weed community across the growing season. The PERMANOVA output (Table 5) indicated whether main treatment or split treatment resulted in dissimilar weed communities, and non-metric multidimensional scaling (NMDS) plots were created to visualize these data. The NMDS plots (Figure 3) also provided more specific information regarding the dissimilarity of weed 72 communities among the main and split treatments, as the ellipses and species scores gave context to the PERMANOVA output. The NMDS plots (Figure 3) guided further review of weed communities in Table A10 to determine which species contributed to dissimilarity in community assembly among the treatments. The effect of main treatment on weed community was significant in sample periods 1-3, but these differences were generally more subtle than the differences among the split treatments which were observed in all four sample periods (Table 5, Figure 3). Throughout the sampling periods in 2019, the presence and absence of Solidago spp. and Symphytotrichum lanceolatum was a major contributor to the differences in the weed community among the treatments. Differential growth of grass weeds and sedges also became an important contributor to the treatment differences in sample periods 3 and 4 (Table A10). 73 Table 5. Adonis (PERMANOVA) output for weed communities at four sampling sample periods in 2019 based on Bray-Curtis distances. These values are based on the weed community in a certified organic orchard of 'Honeycrisp (Firestorm)’/ ‘Budagovsky.9’ apple trees in Ithaca, NY in 2019. Sample d.f. SS MS pseudoF R2 Pr(>F)ab period 1 Main effects Main treatment 2 0.924 0.462 2.007 0.064 0.015* Split treatment 3 3.255 1.084 4.712 0.226 0.005** Interaction terms Main × Split treatment 6 1.924 0.321 1.393 0.134 0.050* Residuals 36 8.288 0.230 0.576 Total 47 14.391 1.000 2 Main effects Main treatment 2 1.005 0.502 1.825 0.066 0.005** Split treatment 3 2.890 0.963 3.499 0.189 0.005** Interaction terms Main × Split treatment 6 1.522 0.254 0.921 0.099 0.620 Residuals 36 9.913 0.275 0.647 Total 47 15.330 1.000 3 Main effects Main treatment 2 1.070 0.535 1.807 0.063 0.030* Split treatment 3 3.417 1.139 3.848 0.202 0.005** Interaction terms Main × Split treatment 6 1.745 0.291 0.983 0.103 0.395 Residuals 36 10.656 0.296 0.631 Total 47 16.888 1.000 4 Main effects Main treatment 2 0.746 0.373 1.268 0.0454 0.230 Split treatment 3 3.372 1.124 3.822 0.205 0.005** Interaction terms Main × Split treatment 6 1.173 0.288 0.980 0.105 0.545 Residuals 36 10.587 0.294 0.644 Total 47 16.435 1.000 Abbreviations: MS, mean sum of squares; SS, sum of squares. Significance values based on 999 permutations. Significance levels: * <0.05, ** <0.01, *** <0.001. 74 Figure 3. Non-metric multi-dimensional scaling (NMDS) ordination of the Bray-Curtis distances for weed community assembly in a certified organic orchard of ‘Honeycrisp’/‘Budagovsky.9’ apple trees in Ithaca, NY. These figures represent weed community assembly in 2019 at sampling sample periods 1-4. Weed species are represented by six letter abbreviations. Major weeds include Solidago spp. and Symphytotrichum lanceolatum (SOLCAN), Poa annua (POAANN), Elytrigia repens (ELYREP), Ranunculus bulbosus (RANBUL), Cirsium arvense (CIRARV), Calystegia sepium (CALSEP), Poa trivialis (POATRI), Plantago spp. (PLALAN, PLAMAJ), and Cyperus esculentus (CYPESC). From top to bottom, sample period 1-4. Split treatment abbreviations: FSO = Ammoniated soap, SPS = Capric acid, MOW = Mowing, UTC2 = split treatment Control. 75 Sample Period 1 and 2 The community of weeds in the split treatment Control was generally the most uniform, regardless of main treatment. Solidago spp. and S. lanceolatum, Ranunculus bulbosus, Plantago lanceolata, and Cirsium arvense were present in all main treatment combinations with the split treatment Control. The weed communities in the other main and split treatment combinations were less uniform (Table A11). The weed community in the split treatment Control was also the most dissimilar from the other split treatments (Figure 3). Sample period 3 E. repens was the main contributor to differences among split treatments in the Cultivation main treatment in this sample period. In the Cultivation × Ammoniated soap or Capric acid treatments, E. repens represented 76% and 98% of the total weed biomass in those treatments, respectively. However, E. repens contributed only 15% of the total weed biomass in the Cultivation × Mowing treatment, and only 7% of the total weed biomass in the Cultivation × Control treatment. A similar phenomenon was observed with Poa trivialis, another grass species, in the main treatment Control. In the Control × Ammoniated soap or Capric acid treatments, P. trivialis represented 61% and 57% of the total weed biomass, respectively. In the Control × Mowing and weedy check, though, P. trivialis represented only 28% and 5% of the total weed biomass, respectively. In the main treatment Control, E. repens was only present when the split treatment was Ammoniated soap or Capric acid (Table A11). Differences were not as pronounced among the Mulch treatment combinations. Regardless of main treatment, Plantago lanceolata differentiated the split treatment Control and Mowing treatments from the two organic herbicide split treatments. During this 76 sample period, P. lanceolata was present in every main treatment combination with Mowing or the split treatment Control, representing 3-52% of the total weed biomass, yet never appeared in the Ammoniated soap and Capric acid treatments (Table A11). Sample period 4 The main treatment effect was no longer significant with respect to community composition during this sample period. Among the split treatments, significant differences in the weed community existed between the split treatment Control and Capric acid treatments only. Again, grass species (particularly E. repens) were abundant in the Capric acid treatment while Solidago spp. and S. lanceolatum contributed the vast majority of biomass in the split treatment Control. In the Capric acid split treatment, grasses and sedges such as E. repens, Poa spp., Agrostis spp., Setaria pumila, Festuca rubra, and Cyperus esculentus represented at least 84% of the total weed biomass. Comparatively, in the split treatment Control, these same species represented 10% or less of the total weed biomass while creeping perennials like Solidago spp., S. lanceolatum, C. arvense, and R. bulbosus represented at least 76% of the total weed biomass in the Control treatment. These creeping perennials contributed less than 8% of the total weed biomass in the Capric acid split treatment (Table A11). 3.4. Discussion The effects of stacked weed management strategies on total weed biomass Stacking weed management strategies reduced total weed biomass in the Mulch and Cultivation treatments where biomass harvested was significantly lower when these treatments 77 were combined with the split treatments of Mowing, Ammoniated soap, or Capric acid compared to combinations with the split treatment Control. Rowley et al. (2011) conducted a similar 2-year study in a tart cherry orchard in Utah and also demonstrated that application of organic herbicides 3-4 times per season significantly improved weed control compared to where paper, straw, or wood chip mulch were used alone. Although stacking improved weed control in the Mulch and Cultivation treatments in the present study, all four split treatments performed similarly regardless of the main treatment; weed biomass in the Mowing, Ammoniated soap, and Capric acid split treatments was significantly lower than the split treatment Control by sample period 3 in 2017, and sample period 2 in 2018 and 2019. Therefore, our findings partially support the hypothesis that increasing the number of weed management strategies would decrease weed biomass. These findings also suggest that the Mowing, Ammoniated soap, and Capric acid treatments are similarly effective when used independently as when used in combination with the Mulch or Cultivation main treatments. While the combination of treatments improved weed control, the use of multiple weed management tactics was not necessary to reduce weed biomass relative to the weedy check. The benefits of stacking were clearly demonstrated by Rowley et al. (2011), but these authors also point out that pelargonic acid (7% v/v) resulted in statistically similar weed control as glyphosate (5.67 g ai/L) even in the absence of mulches. The use of multiple weed management tactics results in increased costs for weed management (Merwin et al., 1995; Rowley et al., 2011) which may not be offset by an increase in yield or quality. There has been little evaluation of the efficacy of Capric acid, Ammoniated soap, and Mowing treatments as individual strategies for weed management in the tree row of perennial fruit crops. Agnello et al. (2017) evaluated both of the herbicide formulations in an organic apple 78 orchard and reported 50-70% weed free area in the tree row in August/September following 5 applications at 3-week intervals. In this 3-year experiment, though, the weed free area was statistically similar between the herbicide treatments and the wood chip mulch treatment. Poor efficacy of the Mulch and Cultivation treatments as individual management strategies in the present study could be related to the predominance of creeping perennial weeds like Solidago spp. and S. lanceolatum. Unlike small-seeded annual weeds, creeping perennials have underground storage organs that enable them to emerge through mulches more readily (Mohler et al., 2021). Presence of creeping perennials has led to mulch failures in other similar studies as well. Agnello et al. (2017) reported a 20 cm layer of bark chip mulch did not prevent growth of C. arvense in the tree row even in the months immediately following application. Similarly, Granatstein et al. (2014) found a 10 cm layer of wood chip mulch suppressed growth of E. repens for only one year after application while, in another orchard without E. repens, the same treatment provided three years of effective weed control (Granatstein and Mullinix, 2008). Hammermeister (2016) recommends tillage low fertility sites where rhizomatous perennials are present, such as the site of this present study. However, increased frequency of tillage to exhaust storage organs of rhizomatous perennials may have been necessary in this study (Mohler et al., 2021). Treatment effects on weed community diversity and assembly Treatment combinations with either of the two organic herbicides resulted in weed communities that were predominantly monocot weeds by 2019. This was most apparent after sample period 1, and within the main treatment Control and the Cultivation treatment (× Capric acid only) where biomass from monocot weeds was significantly greater than treatment 79 combinations with the split treatment Control. E. repens and P. trivialis were the main contributors to treatment differences in monocot weed biomass. Poor efficacy of non-selective, contact herbicides on grass weeds is widely recognized (Webber et al., 2017) and has been demonstrated in other orchard studies (Granatstein et al., 2014; Miñarro, 2012). There were no differences in monocot weed biomass among the Mulch treatment combinations which suggests the initial 15 cm mulch layer was sufficient to suppress growth of monocot weeds for four growing seasons at this site, regardless of secondary management. In the absence of either of the two organic herbicide or Mowing split treatments, creeping perennial weeds like S. lanceolatum, Solidago spp., R. bulbosus, and C. arvense were predominant. Abundance of creeping perennials was anticipated in the main treatment Control and Mulch treatments, as a lack of soil disturbance tends to favor perennial weeds (Ciaccia et al., 2022; Hammermeister, 2016; Miñarro, 2012; Mohler et al., 2021). Disturbance in the Cultivation treatment was expected to increase the abundance of annual weeds as was demonstrated by Miñarro (2012), but annuals remained an insignificant portion of the overall weed community for the duration of this study. The lack of annual weeds in the Cultivation treatment suggests cultivation events were too infrequent to disrupt the life cycle of the existing perennial weeds (Mohler et al., 2021). Alternatively, our observations in the Cultivation treatment support the findings of Ciaccia et al. (2022), who demonstrated that the spontaneous flora that comprise an area before apples are planted are persistent and, in spite of management, may persist for several years following orchard establishment. Before trees were planted at the orchard site for the present study in New York State, the site was maintained as a mowed fallow before being plowed and disked in preparation for planting. 80 Implications for weed management in organic orchards in the Northeastern U.S. The Mulch and Cultivation treatments × split treatment Control were not sufficient to reduce weed biomass relative to the weedy check during the critical weed free period for apple trees (Breth and Tee, 2017; Merwin and Ray, 1997). Increasing the frequency of cultivation with the Wonder Weeder may be one strategy to improve control of creeping perennials. Cultivation to suppress these weeds depends on timeliness; following tillage that results in fragmentation of creeping perennials, these weeds must be tilled under again when the undergrounds storage organ reaches its minimum weight (Mohler et al., 2021) and our calendar-based timing may not have been adequate or best timed to manage the specific weed species at the site. More frequent or intense cultivation may have negative effects on soil health (Merwin et al., 1994) and associated time, equipment, and labor costs (Peck et al., 2010). However, minimizing crop-weed competition during the establishment years of an orchard is critical for long-term cumulative yield (Hammermeister, 2016). Similarly, a second application of wood chip mulch may have improved control of perennial weeds in the Mulch treatment by increasing the depth weeds must break through for emergence and establishment. However, wood chip mulch application may be cost-prohibitive, particularly when mulch is sourced off-farm (Merwin et al., 1995; Rowley et al., 2011). Wood chip mulch may not be a good option in poorly drained soils due to tree mortality from excessive moisture, as Agnello et al (2017) suggested, and we saw no benefit to tree growth or fruit yield in Chapter 2. Split treatments of Mowing and either of the two organic herbicides were similarly effective when used alone as when used in combination with the Mulch or Cultivation main treatments. Although organic-approved herbicides may be cost-prohibitive to apply for the lifetime of an orchard, our results demonstrate the efficacy of Suppress (Capric acid) and Final- 81 San-O (Ammoniated soap) and the corresponding benefits for tree growth (Chapter 2). Early use of these products may be cost-effective as the benefits of maintaining a weed free area during the establishment years of an orchard are well-documented (Breth and Tee, 2017; Merwin and Ray, 1997). Other studies also suggest that, once trees are established, they are less susceptible to competition from in-row vegetation so management strategies can change as the orchard matures (Choi et al., 2011; Hoagland et al., 2008; Stefanelli et al., 2009). Orchards with higher fertility may also be less susceptible to weed competition during the establishment years (Hammermeister, 2016; Mia et al., 2020). 3.5 Conclusions The results of this study demonstrate the impact of weed management tactics on the composition of weed communities, and the bearing that the weed community holds on the efficacy of those tactics. At this orchard site, an abundance of creeping herbaceous perennial weeds resulted in the Mulch and Cultivation treatments alone yielding weed biomass similar to the weedy check by 2019. Alternatively, the Mowing, Capric acid, and Ammoniated soap split treatments significantly reduced weed biomass compared to the weedy check, regardless of the main treatment. Although the Capric acid and Ammoniated soap treatments alone did not reduce monocot weed biomass, harvested weed biomass was still significantly reduced in these treatments compared to the weedy check. While the merit of an integrated approach to weed management cannot be refuted, we observed no benefit of using multiple weed management strategies with regard to weed control. Likewise, these results do not justify the additional expense of using multiple tactics for weed control in the tree row of establishing orchards for the goal of increasing tree growth and productivity (as described in Chapter 2). Examining the 82 legacy effect (persistence) of weeds present before planting an apple orchard is one way to determine ideal management strategies for different orchard sites. While there is an international interest in increasing the biodiversity and ecosystem services of agricultural systems, the primary goal during the establishment period of an orchard needs to be maximizing early yield and return on investment, as farm viability is a critical component of sustainable agriculture. Efficacious weed management tactics depend on matching the approach with the weed community to be controlled; approaches that minimize weed competition, optimize nutrient availability, and promote tree growth and productivity are still needed to support the adoption of certified organic apple production in the Northeastern U.S. 83 REFERENCES Agnello, A., K. Cox, J. Lordan, P. Francescatto, and T. Robinson. 2017. Comparative programs for arthropod, disease and weed management in New York organic apples. Insects 8(3):96. https://doi.org/10.3390/insects8030096. Bradshaw, T. 2017. Non-chemical weed management in apple orchards. University of Vermont Extension, Burlington, Vermont. Choi, H., C.R. Rom, and M. Gu. 2011. Effects of different organic apple production systems on seasonal nutrient variations of soil and leaf. Scientia Horticulturae. 129(1):9-17. https://doi-org.proxy.library.cornell.edu/10.1016/j.scienta.2011.02.009. Ciaccia, C., E. Testani, T. Amoriello, and D. Ceccarelli. 2022. Weed community evolution under diversification managements in a new planted organic apricot orchard. Agriculture, Ecosystems & Environment. 335(2022):108014. https://doi- org.proxy.library.cornell.edu/10.1016/j.agee.2022.108014. Granatstein, D. and E. Sánchez. 2009. Research Knowledge and Needs for Orchard Floor Management in Organic Tree Fruit Systems. International Journal of Fruit Science. 9(3):257-281. https://doi.org/10.1080/15538360903245212. Granatstein, D. and K. Mullinix. 2008. Mulching options for Northwest organic and conventional orchards. HortScience. 43(1):45-50. https://doi.org/10.21273/HORTSCI.43.1.45. Granatstein, D., M. Wiman, E. Kirby, and K. Mullinix. 2010. Sustainability trade-offs in organic orchard floor management. Acta Horticulturae. 873:115-122. Granatstein, D., P. Andrews, and A. Groff. 2014. Productivity, economics, and fruit and soil quality of weed management systems in commercial organic orchards in Washington State, USA. Org. Agr. 4(3):197-207. https://doi.org/10.1007/s13165-014-0068-0. Hammermeister, A.M. 2016. Organic weed management in perennial fruits. Scientia Horticulturae. 208:28-42. https://doi.org/10.1016/j.scienta.2016.02.004. Hoagland, L., L. Carpenter-Briggs, D. Granatstein, M. Mazzola, J. Smith, F. Peryea, and J.P. Reganold. 2008. Orchard floor management effects on nitrogen fertility and soil biological activity in a newly established organic apple orchard. Biol Fertil Soils. 45(1):11-18. https://doi.org/10.1007/s00374-008-0304-4. Liebman, M. and E. Gallandt. 1997. Many little hammers: Ecological management of crop-weed interactions. In L.E. Jackson (ed.). Ecology in Agriculture. Academic Press. San Diego, California. Merwin, I.A. and J.A. Ray. 1997. Spatial and temporal factors in weed interference with newly planted apple trees. HortScience. 32(4):633-637. https://doi.org/10.21273/HORTSCI.32.4.633. Merwin, I.A. and W.C. Stiles. 1994. Orchard groundcover management impacts on apple tree growth and yield, and nutrient availability and uptake. J. Amer. Soc. Hort. Sci. 119(2):209-215. https://doi.org/10.21273/JASHS.119.2.209. Merwin, I.A., D.A. Rosenberger, C.A. Engle, D.L. Rist and M. Fargione. 1995. Comparing mulches, herbicides, and cultivation as orchard groundcover management systems. HortTech. 5(2):151-158. https://doi.org/10.21273/HORTTECH.5.2.151. Merwin, I.A., W. Stiles, and H. van Es. 1994. Orchard Groundcover Management Impacts on Soil Physical Properties. J. Amer. Soc. Hort. Sci. 119(2):216-222. https://doi.org/10.21273/JASHS.119.2.216. 84 Mia, M.J., F. Massetani, G. Murri, J. Facchi, E. Monaci, L. Amadio, and D. Neri. 2020. Integrated weed management in high density fruit orchards. Agronomy. 10(10):1492. https://doi.org/10.3390/agronomy10101492. Miñarro, M. 2012. Weed communities in apple orchards under organic and conventional fertilization and tree-row management. Crop Protection. 39:89-96. https://doi- org.proxy.library.cornell.edu/10.1016/j.cropro.2012.04.002. Mohler, C.L., J.R. Teasdale and A. DiTommaso. 2021. Manage weeds on your farm: a guide to ecological strategies, SARE handbook series. Sustainable Agriculture Research & Education (SARE), College Park, Maryland. Peck, G. and I.A. Merwin. 2009. A grower’s guide to organic apples. New York State IPM. Publication Number 223. Peck, G.M., I.A. Merwin, M.G. Brown, and A.M. Agnello. 2010. Integrated and organic fruit production systems for ‘Liberty’ apple in the Northeast United States: a systems-based evaluation. Hort. Sci. 45(7):1038-1048. https://doi- org.proxy.library.cornell.edu/10.1016/j.apsoil.2011.02.008. Robinson, T.L., A.M. DeMarree, and S.A. Hoying. 2007. An economic comparison of five high density apple planting systems. Acta Hortic. 732:481–489. Robinson, T.L., S.A. Hoying, and G.H. Reginato. 2006. The tall spindle apple production system. New York Fruit Quarterly. 14(2):21-28. Rowley, M.A., C.V. Ransom, J.R. Reeve, and B.L. Black. 2011. Mulch and organic herbicide combinations for in-row orchard weed suppression. International Journal of Fruit Science 11(4):316-331. https://doi-org.proxy.library.cornell.edu/10.1080/15538362.2011.630295. Soil Science Division Staff. 2017. Soil survey manual. C. Ditzler, K. Scheffe, and H.C. Monger (eds.). USDA Handbook 18. Government Printing Office, Washington, D.C. Stefanelli, D., R.J. Zoppolo, R.L., Perry, and F. Weibel. 2009. Organic orchard floor management systems for apple effect on rootstock performance in the midwestern United States. HortScience. 44(2):263-267. https://doi.org/10.21273/HORTSCI.44.2.263. Swanton, C.J. and S.F. Weise. 1991. Integrated weed management: The rationale and approach. Weed Technology. 5(3):657-663. https://doi.org/10.1017/S0890037X00027512. Webber III, C.L., J. W. Shrefler, and L.P. Brandenberger. 2012. Organic weed control. In R. Alvarez-Fernandez (ed.). Herbicides - Environmental Impact Studies and Management Approaches. IntechOpen. https://doi.org/10.5772/1206. 85 CHAPTER 4 Conclusions The experiment I described in this thesis was based on the idea that changing one management practice – weed management – can have implications on the apple production system as a whole. Specifically, weed competition and weed management tactics have implications for soil nutrient and water availability, soil health, tree nutrient uptake, and tree growth and productivity. Additionally, different methods of weed management act as filters on the weed community, so diversity and abundance of the weed community were measured. These horticultural response variables were evaluated during the four years of this experiment to help develop a clearer understanding of the suitability of prospective weed management tactics for organic apple production systems where weed control and nutrient management have been identified as major barriers to adoption. This experiment was unique in that few previous studies have evaluated stacked, or integrated, approaches to weed control in the tree row of perennial tree fruit. The design of the experiment created an opportunity to evaluate weed management strategies both individually, and in combination. The premise of this was that increasing the number of weed control tactics would reduce weed abundance and diversity. While the merit of an integrated approach to weed management cannot be refuted, we observed no benefit of using multiple weed management tactics in this particular study, as all the weed control treatments reduced weed biomass relative to the weedy check. With regard to weed control, the Mowing, Ammoniated soap, and Capric acid split treatments resulted in similar weed biomass regardless of the main treatment. Similarly, within the Mulch and Cultivation treatments, split treatment had no impact on tree growth. Although TCSA was significantly greater within the main treatment 86 Control × Ammoniated soap and Capric acid split treatments, these organic herbicides provided no benefit for tree growth in the Cultivation and Mulch main treatments. Main and split treatments did have statistically significant impacts on several of the other response variables, yet significant correlations between the response variables were rarely identified. For example, the soil health parameters were almost always positively correlated with one another, but neither soil health nor weed control nor leaf nutrient content were significantly correlated with tree growth as we originally anticipated. The trees in this experiment did not progress at a rate that would be acceptable on a commercial farm, and there are several possible reasons for poor growth of the trees in all our treatments. Soil moisture was very high in the orchard and especially for prolonged periods after rainfall. We were not able to meet the recommended N fertility requirements for establishing apple trees. Cultivation did promote N uptake, with N being greater than Control in 2016-2018 and Mulch in 2017 and 2019, but these treatments did not reach recommended leaf N levels for commercial apple production. We attempted to address this through foliar-applied fish fertilizer, but this was not sufficient. Soil applied fertilizer could have made a difference. Trees were deficient in several macronutrients, so soil fertility or root uptake were low. The scion and rootstock combination of ‘Honeycrisp’ and ‘Budagovsky.9’, respectively, was among the most dwarfing possible. This combination was selected because we expected it to respond more dramatically to differences in weed competition, but these trees never grew at an acceptable rate. More vigorous trees better suited to individual sites should be chosen for future work or, where possible, a range of weak and strong trees could be compared in the same orchard. It is difficult to state whether weed competition alone was a main contributor to poor tree growth. Differences in weed biomass were not reflected in tree growth or productivity. A weed 87 free treatment would have permitted us to evaluate the influence that competition was having on these trees and more clearly determine what level of weed control is considered acceptable and able to produce commercially acceptable yields. Specifically, a weed free treatment could have enabled direct comparison of tree growth, yield and nutrient uptake in the absence of weed pressure. Measuring the N content of the harvested weed biomass may have also given insight to the level of competition occurring within each treatment and how that relates to the corresponding weed community. Split treatments became most important for differences in weed control over time, indicating that the given weed community was abundant with rhizomatous perennials from a long-time fallow, which were not well controlled by the Mulch or Cultivation treatments when used independently. Although the organic herbicide and Mowing treatments did improve weed control in these main treatments, the organic herbicides and Mowing treatments performed similarly with or without the Mulch and Cultivation treatments as the base treatment. So, in this case, the organic herbicides and Mowing treatments were apparently sufficient as individual strategies. However, only the organic herbicides significantly improved tree size compared to the weedy check, while the main treatment Control × Mowing treatment resulted in tree size similar to the weedy check in 2019. At this orchard site, creeping herbaceous perennials that proliferated during the pre-plant fallow period seemed to overwhelm treatments that did not control these weeds. Therefore, these results do demonstrate the need for pre-plant weed control measures that target the most troublesome weeds. An economic comparison of these individual and combined weed management practices is one analysis that was not conducted as part of this thesis that would have added greatly to the value of this work for orchard managers. Future work should strive to include an economic 88 analysis to aid in on farm decision making with regard to weed management decisions for the establishment period of orchards. Lastly, conducting these types of experiments at multiple locations would lend more insight into the efficacy of certain weed management tactics as they relate to weed community change in different environmental conditions. Given the site and weed community dependent nature of these weed control tactics, a similar experimentation at multiple sites would be valuable in determining the most efficacious approaches to managing weeds, promoting tree growth, and optimizing nutrient availability under different farm conditions. 89 APPENDIX Table A1. Sampling dates for aboveground weed biomass data collection in a certified organic orchard of ‘Honeycrisp’/‘Budagovsky.9’ apple trees grown in Ithaca, NY. Weed biomass sampling dates 2017 2018 2019 Sample Period 1 18 May 10 May 8 May Sample Period 2 6 June 31 May 4 June Sample Period 3 7 July 2 July 26 June Sample Period 4 9 August 30 July 29 July 90 Ammoniated Capric Capric Ammoniated Mowing Control Row 1 soap acid acid soap Block 1 Ammoniated Control Mowing Capric acid Control Mowing Row 2 soap Color Main treatment Control Mulch Cultivation Figure A1. Experimental design of block 1, one of four complete blocks established in a certified organic orchard of ‘Honeycrisp’/‘Budagovsky.9’ apple trees grown in Ithaca, NY in 2015. 91 Table A2. Pesticides applied to manage insect pests and disease in a certified organic orchard of ‘Honeycrisp’/‘Budagovsky.9’ apple trees grown in Ithaca, NY between 2015 and 2019. EPA Registration Product Name Active ingredient Manufacturer Number Aza-direct® Azadirachtin 71908-1-10163 Gowan Company, Yuma, AZ Azera Azadirachtin, Pyrethrins 1021-1872 MGK®, Minneapolis, MN Loveland Products Inc., Greeley, BioCover™ MLT Petroleum oil 34704-805 CO Cueva® Copper octanoate 67702-2-70051 Certis USA LLC, Columbia, MD Valent Biosciences LLC. DiPel® Bacillus thuringiensis 73049-39 Libertyville, IL Bacillus Certis USA LLC, Columbia, MD Double Nickel 55® 70051-107 amyloliquefaciens Corteva™ Agriscience, Entrust® SC Naturalyte ® Spinosad 62719-282 Wilmington, DE Chromobacterium Marrone® Bio Innovations, Grandevo® 84059-17 subtsugae Davis, CA Dodecene pheromone Pacific BioControl Corporation, Isomate® CM/OFM TT 53575-30 analogs Vancouver, WA JMS Flower Farms Inc., Vero JMS Stylet-Oil® paraffinic oil 65564-1 Beach, FL Cydia Certis USA LLC, Columbia, MD Madex® HP 69553-1 pomonella granulovirus Microthiol® Disperss® Sulfur 70506-187 UPL, King of Prussia, PA Pyganic® Crop Protection EC 1.4 II Pyrethrins 1021-1771 MGK®, Minneapolis, MN Marrone® Bio Innovations, Regalia® Reynoutria sachalinensis 84059-3 Davis, CA 92 Table A3. Fertilizers foliar applied to ‘Honeycrisp’/’Budagovsky.9’ apple trees, as needed based on annual leaf tissue analysis, in a certified organic orchard in Ithaca, NY between 2015 and 2019. Rate of Product Name Active Ingredient Manufacturer Dates of Application Application 21-June-2017, 6-July-2017, 25-May-2018, 30-May- Aqua Power™ -1 JH Biotech Inc., 2018, 7-June-2018, 20-Hydrolyzed fish protein 23.4 L ha 5-1-1 Ventura, CA June-2018, 7-June-2019, 19-June-2019, 28-June- 2019, 12-July-2019 1.8 kg per 31-May-2017, 30-May- Epsom salts Magnesium sulfate heptahydrate 100L 2018, 19-June-2019 0.12 kg per U.S. Borax Inc., 31-May-2017, 21-June- Solubor® Disodium octaborate tetrahydrate 100 L Valencia, CA 2017, 30-May-2018 8-June-2017, 7-June-2018, -1 G.P. Solutions, Zinc Chelate 8% Chelated zinc 4.68 L ha 20-June-2018, 7-June-2019, LLC., LaBelle, FL 28-June-2019 93 Swath of cultivation Weed biomass sampling area Tree row ( 0.5m x 0.5m) Swath of cultivation 0.9 m Figure A2. Diagram of weed biomass sampling location in relation to the swath of cultivation, the tree row, and the in-row space between trees in a certified organic orchard of ‘Honeycrisp’/‘Budagovsky.9’ apple trees grown in Ithaca, NY. 94 0.5 m 0.5 m Table A4. Average maximum, minimum, and average daily temperature (ºC) during study compared to historic daily temperatures for Ithaca, NY. Month 20171 20181 20191 Historic2 Daily maximum temperature May 17.63 23.2 18.8 19.7 Jun 24.5 24.2 23.6 24.4 Jul 26.0 27.9 27.8 26.6 Aug 25.0 26.13 25.6 25.9 Sep 23.2 23.2 22.7 21.7 Daily minimum temperature May 8.44 11.2 8.4 6.4 Jun 13.2 12.8 12.0 11.8 Jul 16.0 15.9 17.0 14.3 Aug 13.9 17.44 14.2 13.5 Sep 11.4 14.0 10.9 9.4 Average daily temperature May 12.93 17.2 13.5 13.0 Jun 18.8 18.5 18.2 18.1 Jul 20.9 22.0 22.2 20.4 Aug 19.2 20.63 19.7 19.7 Sep 16.9 18.4 16.9 15.6 1Source: NEWA for Ithaca (CUAES: Cornell Orchards), NY weather station – latitude: 42.44, longitude: -76.46, elevation: 920 ft. 2Source: Northeast Regional Climate Center https://www.nrcc.cornell.edu/wxstation/ithaca/normal.html. Based on data for the period 1981-2010. 3Data missing from NEWA (CUAES: Cornell Orchards) for 16 May and 17 May 2017 plus 7 August through 15 August 2018; backfilled with data from the Northeast Regional Climate Center for Ithaca, NY 95 Table A5. Total monthly precipitation (cm) during study compared to historic monthly precipitation amounts for Ithaca, NY. Month 20171 20181 20191 Historic2 May 11.73 9.9 11.4 8.1 Jun 8.5 5.0 12.3 10.1 Jul 14.3 11.0 8.6 9.7 Aug 5.2 9.93 12.2 9.2 Sep 5.2 5.8 4.5 9.4 Total 44.9 41.7 48.9 46.6 1Source: NEWA for Ithaca (CUAES: Cornell Orchards), NY weather station – latitude: 42.44, longitude: -76.46, elevation: 920 ft. 2Source: Northeast Regional Climate Center https://www.nrcc.cornell.edu/wxstation/ithaca/normal.html. Based on data for the period 1981-2010. 3Data missing for two days in 2017: 16 May and 17 May 4Data missing for nine days in 2018: 7 August through 15 August 96 Table A6. Average daily temperatures (ºC) and total precipitation (cm) between sampling dates. Source: NEWA for Ithaca (CUAES: Cornell Orchards), NY weather station – latitude: 42.44, longitude: -76.46, elevation: 920 ft. 2017 2018 2019 Time Period Temperature (ºC) Budbreak (~5/5) to Sample Period 1 10.51 16.0 11.5 Sample period 1-2 15.4 17.1 14.1 Sample period 2-3 20.3 19.0 18.2 Sample period 3-4 20.6 21.7 22.2 Sample period 4 to Harvest (~9/10) 17.2 20.52 19.2 Precipitation (cm) Budbreak (~5/5) to Sample Period 1 4.81 1.1 1.1 Sample period 1-2 4.8 8.1 10.1 Sample period 2-3 8.6 5.7 11.9 Sample period 3-4 14.0 10.3 7.4 Sample period 4 to Harvest (~9/10) 9.0 10.7 15.7 Seasonal Total Precipitation 41.2 35.9 46.2 1Data missing for 16 May and 17 May 2017; backfilled with data from Northeast Regional Climate Center for Ithaca, NY 2Data missing for 7 August through 15 August 2018; backfilled with data from Northeast Regional Climate Center for Ithaca, NY 97 Table A7. Aboveground weed biomass (g‧m-2) in a certified organic orchard of ‘Honeycrisp’/‘Budagovsky.9’ apple trees grown in Ithaca, NY. These data represent the average value for each main treatment, split treatment, and sample period in a given year. Aboveground weed biomass (g‧m-2) 2017 2018 2019 Main treatment Cultivation 134.9 A 177.0 A 225.0 AB Mulch 82.6 B 112.0 B 191.0 B Control 151.8 A 163.0 A 260.0 A Split treatment Ammoniated soap 104.1 B 99.0 B 153.0 B Mowing 110.1 B 112.0 B 127.0 B Capric acid 92.5 B 89.1 B 134.0 B Control 185.8 A 302.4 A 490.0 A Sample Period 1 133.0 95.7 B 119.0 C 2 102.0 112.6 B 192.0 BC 3 112.0 183.9 A 247.0 B 4 146.0 200.3 A 345.0 A Statistical significance Main treatment *** *** ** Split treatment *** *** *** Sample period * *** *** Main × Split ns ns ns Main × Sample period ns ns ns Split × Sample period *** *** *** Main × Split × Sample period ns ns ns Different letters, ‘A,’ ‘B,’ ‘C,’ indicate mean separation at p ≤ 0.05 using Tukey’s honestly significant difference test. Letters indicating mean separation are specific to either the main treatments, the split treatments, or the time points and letters. ns: nonsignificant, *: significant at p ≤ 0.05, **: significant at p ≤ 0.01, ***: significant at p ≤ 0.001. 98 Table A8. In-row soil volumetric water content collected on 10 sampling dates in 2019 from within the tree row of a certified organic orchard of ‘Honeycrisp’/‘Budagovsky.9’ apple trees in Ithaca, NY. 6-Jun 13-Jun 25-Jun 8-Jul 16-Jul 23-Jul 8-Aug 19-Aug 26-Aug 6-Sep Volumetric water content (m3‧m-3) Main treatment Cultivation 43.4 B 33.5 B 38.0 B 28.5 20.8 28.5 27 26.2 B 23.3 B 27.9 B Mulch 48.7 A 36.9 A 42.5 A 29.8 19.9 30.8 24.4 32.0 A 27.7 A 33.8 A Main Control 44.3 B 34.7 AB 39.2 B 29.3 18.6 30.8 25.5 33.0 A 28.6 A 33.8 A Split treatment Ammoniated soap 45.9 35 41 31.6 A 22.3 A 32.0 A 27.2 AB 31.6 A 26.8 A 33.0 A Mowing 46.1 35.8 40.6 30.3 A 20.2 A 31.7 A 24.7 BC 32.0 A 28.6 A 33.6 A Capric acid 44.8 36.4 40.2 31.7 A 22.8 A 30.9 A 29.5 A 32.2 A 29.2 A 33.3 A Split Control 45 33 37.7 23.2 B 14.0 B 25.4 B 21.1 C 25.8 B 21.6 B 27.5 B Statistical significance of main and split treatments Main treatment *** * *** ns ns ns ns *** *** *** Split treatment ns ns ns *** *** *** *** *** *** *** Main × Split ns ns ns ns ns ns ns * ns ns treatment 1 Means in the same column followed by the same letter are not significantly different according to Tukey’s HSD test (P≤0.05). Uppercase letters A, B, C designate significant differences among main treatments or split treatments. Letters for mean separation should not be compared across columns or between main and split treatments. 99 Table A9. Trunk cross-sectional area of each main (Cultivation, Mulch, main Control) and split (Ammoniated soap, Mowing, Capric acid, split Control) treatment at five sampling times between 2016 and 2019. Spring Winter Winter Winter Winter 2016 2016 2017 2018 2019 Main treatment Control 1.22 1.54 2.27 B 3.19 3.85 Cultivation 1.18 1.64 2.58 A 3.98 4.79 Mulch 1.21 1.56 2.21 B 3.12 3.78 Split treatment Control 1.20 1.51 bc 2.22 bc 3.01 3.64 Mowing 1.18 1.44 c 2.10 c 3.06 3.73 Capric acid 1.22 1.70 a 2.55 a 3.87 4.64 Ammoniated soap 1.21 1.63 ab 2.47 ab 3.62 4.36 Statistical significance Main treatment ns ns *** *** *** Split treatment ns *** *** *** *** Main × Split ns ns ns ** ** 100 Table A10. Aboveground weed biomass (g‧m-2) in a certified organic orchard of ‘Honeycrisp’/‘Budagovsky.9’ apple trees grown in Ithaca, NY. Main treatment values were averaged across the respective split treatments and split treatment values were averaged across all main treatments. Sample period values are averaged across all main and split treatment combinations at a given sample period (sampling date) within a year. Aboveground weed biomass (gm-2) 2017 2018 2019 Main treatment Cultivation 134.9 A 177.0 A 225.0 AB Mulch 82.6 B 112.0 B 191.0 B Control 151.8 A 163.0 A 260.0 A Split treatment Ammoniated soap 104.1 B 99.0 B 153.0 B Mowing 110.1 B 112.0 B 127.0 B Capric acid 92.5 B 89.1 B 134.0 B Control 185.8 A 302.4 A 490.0 A Sample period 1 133.0 95.7 B 119.0 C 2 102.0 112.6 B 192.0 BC 3 112.0 183.9 A 247.0 B 4 146.0 200.3 A 345.0 A Statistical significance Main treatment *** *** ** Split treatment *** *** *** Sample period * *** *** Main × Split ns ns ns Main × Sample period ns ns ns Split × Sample period *** *** *** Main × Split × Sample period ns ns ns Different letters, ‘A,’ ‘B,’ ‘C,’ indicate mean separation at p ≤ 0.05 using Tukey’s honestly significant difference test. Letters indicating mean separation are specific to either the main treatments, the split treatments, or the sample periods and letters should not be compared across columns or across sections in a single column. ns: nonsignificant, *: significant at p ≤ 0.05, **: significant at p ≤ 0.01, ***: significant at p ≤ 0.001. 101 Table A11. Weed community composition across each main (Cultivation, Mulch, main Control) and split (Ammoniated soap, Mowing, Capric acid, split Control) treatment combination at four sampling sample periods in 2019. Represents the average aboveground weed biomass of each species in each main × split treatment combination. Main Cultivation Mulch Main-Control treatment Sample Split Species g/m2 % Species g/m2 % Species g/m2 % period treatment Ammoniated 1 Ranunculus bulbosus 53.9 39.0 Ranunculus bulbosus 37.5 36.1 Poa annua 37.9 34.6 Soap Solidago spp. and Elytrigia repens 43.6 31.5 20.1 19.4 Elytrigia repens 23.1 21.1 Symphytotrichum lanceolatum Taraxacum officinale 32.1 23.2 Elytrigia repens 14.4 13.9 Ranunculus bulbosus 21.8 19.9 Solidago spp. and Poa annua 4.1 3.0 Unidentified Poaceae 3 11.0 10.6 Symphytotrichum 18.1 16.5 lanceolatum Total 138.3 96.7 Rumex crispus 7.4 7.1 Taraxacum officinale 4.9 4.5 Cirscium arvense 3.5 3.4 Total 109.6 96.5 Persicaria amphibia 3.4 3.3 Taraxacum officinale 2.0 1.9 Total 103.8 95.7 1 Mowing Ranunculus bulbosus 69.2 44.6 Ranunculus bulbosus 81.5 58.3 Plantago lanceolata 87.9 65.8 Elytrigia repens 30.0 19.3 Plantago lanceolata 39.5 28.3 Ranunculus bulbosus 15.2 11.4 Plantago lanceolata 23.5 15.2 Taraxacum officinale 7.2 5.2 Unidentified Poaceae 3 6.3 4.7 Solidago spp. and Taraxacum officinale 13.2 8.5 5.2 3.7 Trifolium spp. 6.1 4.6 Symphytotrichum lanceolatum Unidentified Poaceae 2 11.9 7.7 Total 139.7 95.5 Poa annua 6.0 4.5 Total 155.1 95.3 Taraxacum officinale 5.3 4.0 Prunella vulgaris 2.5 1.9 Total 133.6 96.8 1 Capric acid Elytrigia repens 47.0 48.8 Ranunculus bulbosus 47.9 50.4 Elytrigia repens 64.6 58.8 102 Poa annua 27.5 28.5 Elytrigia repens 15.3 16.1 Poa annua 30.9 28.1 Solidago spp. and Taraxacum officinale 6.1 6.3 12.4 13.1 Unidentified Poaceae 2 5.5 5.0 Symphytotrichum lanceolatum Cirscium arvense 4.5 4.7 Taraxacum officinale 4.5 4.7 Ranunculus bulbosus 4.6 4.2 Prunella vulgaris 2.6 2.7 Poa annua 4.2 4.4 Total 109.9 96.1 Ranunculus bulbosus 2.0 2.1 Persicaria amphibia 3.6 3.8 Cerastium vulgatum 1.5 1.6 Cirscium arvense 3.3 3.5 Plantago lanceolata 1.5 1.6 Total 95.0 96.0 Total 96.4 96.2 Solidago spp. and Solidago spp. and Solidago spp. and 1 Split-Control Symphytotrichum 58.2 42.3 46.5 58.2 Symphytotrichum 55.1 42.0 Symphytotrichum lanceolatum lanceolatum lanceolatum Ranunculus bulbosus 53.9 39.2 Plantago lanceolata 10.9 13.6 Ranunculus bulbosus 44.9 34.2 Cirscium arvense 17.7 12.9 Cirscium arvense 8.6 10.8 Plantago lanceolata 17.2 13.1 Plantago lanceolata 2.7 2.0 Ranunculus bulbosus 6.6 8.3 Cirscium arvense 5.9 4.5 Total 137.6 96.3 Elytrigia repens 6.3 7.9 Unidentified Poaceae 3 4.2 3.2 Total 79.9 98.7 Total 131.3 97.0 Ammoniated Solidago spp. and 2 Ranunculus bulbosus 67.3 37.5 29.7 28.8 Elytrigia repens 68.5 33.9 Soap Symphytotrichum lanceolatum Elytrigia repens 60.7 33.8 Ranunculus bulbosus 29.4 28.5 Ranunculus bulbosus 51.3 25.4 Solidago spp. and Symphytotrichum 33.5 18.7 Unidentified Poaceae 2 16.8 16.3 Unidentified Poaceae 2 35.8 17.7 lanceolatum Taraxacum officinale 6.15 3.4 Persicaria amphibia 10.6 10.3 Unidentified Poaceae 3 26.7 13.2 Solidago spp. and Unidentified Poaceae 3 2.8 1.6 Cirscium arvense 6.4 6.2 Symphytotrichum 13.5 6.7 lanceolatum Cerastium vulgatum 2.76 1.5 Trifolium spp. 3.7 3.6 Total 195.8 96.8 Total 170.5 94.9 Taraxacum officinale 2.0 1.9 Total 98.6 95.6 2 Mowing Ranunculus bulbosus 84.4 62.5 Ranunculus bulbosus 63.9 44.7 Plantago lanceolata 68.9 42.6 Elytrigia repens 17.1 12.7 Plantago lanceolata 47.9 33.5 Ranunculus bulbosus 46.1 28.5 Solidago spp. and Taraxacum officinale 10.5 7.8 Unidentified Poaceae 3 9.1 6.4 Symphytotrichum 11.9 7.3 lanceolatum 103 Solidago spp. and Symphytotrichum 5.6 4.2 Elytrigia repens 6.6 4.6 Unidentified Poaceae 2 10.0 6.2 lanceolatum Solidago spp. and Prunella vulgaris 4.8 3.6 5.4 3.8 Prunella vulgaris 7.3 4.5 Symphytotrichum lanceolatum Plantago lanceolata 4.2 3.1 Prunella vulgaris 3.4 2.4 Unidentified Poaceae 3 6.8 4.2 Daucus carota 3.7 2.7 Total 136.3 95.4 Cichorium intybus 2.7 1.7 Total 130.2 96.5 Total 153.7 95.0 2 Capric acid Elytrigia repens 95.7 64.0 Persicaria amphibia Elytrigia repens 68.8 61.1 36.3 42.3 Unidentified Poaceae 3 25.5 17.1 Elytrigia repens Unidentified Poaceae 3 30.0 26.7 18.2 21.2 Ranunculus bulbosus 13.7 9.2 Cirscium arvense Unidentified Poaceae 1 5.8 5.1 8.6 10.0 Cerastium vulgatum 5.4 3.6 Unidentified Poaceae 2 Cerastium vulgatum 4.9 4.3 6.7 7.8 Plantago lanceolata 2.5 1.7 Ranunculus bulbosus Total 109.5 97.2 6.3 7.3 Total 142.8 95.6 Unidentified Poaceae 3 4.7 5.5 Solidago spp. and Symphytotrichum lanceolatum 2.3 2.7 Total 83.0 96.9 Solidago spp. and Solidago spp. and Solidago spp. and 2 Split-Control Symphytotrichum 97.8 32.4 220.8 65.0 Symphytotrichum 136.0 35.2 Symphytotrichum lanceolatum lanceolatum lanceolatum Ranunculus bulbosus 57.4 19.0 Plantago lanceolata 39.0 11.5 Ranunculus bulbosus 120.7 31.2 Unidentified Poaceae 2 54.9 18.2 Ranunculus bulbosus 23.7 7.0 Plantago lanceolata 71.1 18.4 Cirscium arvense 33.6 11.1 Cirscium arvense 22.8 6.7 Cirscium arvense 17.5 4.5 Elytrigia repens 30.2 10.0 Elytrigia repens 18.6 5.5 UNK2 17.5 4.5 Plantago lanceolata 14.4 4.8 Total 324.9 95.7 UNK3 10.2 2.6 Total 288.2 95.4 Total 373.0 96.5 3 Ammoniated Elytrigia repens 125.3 76.1 Elytrigia repens 33.9 40.2 Poa trivialis 139.1 61.4 Soap Solidago spp. and Ranunculus bulbosus 20.0 12.2 18.1 21.5 Elytrigia repens 47.7 21.1 Symphytotrichum lanceolatum Achillea millefolium 7.2 4.4 Poa trivialis 12.7 15.1 Agrostis gigantea 17.3 7.6 Solidago spp. and Symphytotrichum 4.3 2.6 Daucus carota 8.8 10.4 Ranunculus bulbosus 10.2 4.5 lanceolatum Solidago spp. and Total 156.7 95.2 Festuca rubra 6.0 7.1 Symphytotrichum 7.0 3.1 lanceolatum Persicaria amphibia 1.9 2.3 Total 221.3 97.7 Total 81.4 96.6 104 3 Mowing Daucus carota 23.5 24.4 Ranunculus bulbosus 34.9 45.8 Plantago lanceolata 59.9 52.3 Solidago spp. and Ranunculus bulbosus 22.4 23.3 8.5 11.1 Poa trivialis 31.5 27.5 Symphytotrichum lanceolatum Plantago lanceolata 15.0 15.6 Daucus carota 7.5 9.9 Ranunculus bulbosus 6.9 6.0 Elytrigia repens 14.6 15.2 Prunella vulgaris 6.3 8.3 Juncea tenuis 3.0 2.6 Solidago spp. and Solidago spp. and Symphytotrichum 9.8 10.2 Plantago lanceolata 5.5 7.2 Symphytotrichum 2.9 2.5 lanceolatum lanceolatum Poa trivialis 6.1 6.3 Plantago major 4.9 6.4 Plantago major 2.5 2.1 Plantago major 3.1 3.2 Agrostis stolonifera 3.9 5.2 Prunella vulgaris 2.0 1.8 Total 94.5 98.1 Poa trivialis 2.5 3.3 Daucus carota 2.0 1.7 Total 74.0 97.2 Total 110.6 96.5 3 Capric acid Elytrigia repens 168.2 98.4 Persicaria amphibia 60.4 63.3 Poa trivialis 91.6 56.6 Total 168.2 98.4 Poa trivialis 15.1 15.8 Elytrigia repens 66.9 41.3 Elytrigia repens 12.6 13.2 Total 158.5 97.9 Agrostis stolonifera 2.0 2.1 Cyperus esculentus 1.8 1.9 Total 91.8 96.2 Solidago spp. and 3 Solidago spp. and Split-Control Cirscium arvense 167.4 28.7 298.7 58.5 Symphytotrichum 458.9 68.5 Symphytotrichum lanceolatum lanceolatum Solidago spp. and Symphytotrichum 123.3 21.2 Elytrigia repens 69.8 13.7 Plantago lanceolata 67.0 10.0 lanceolatum Alnus glutinosa 91.8 15.8 Cirscium arvense 58.6 11.5 Ranunculus bulbosus 66.8 10.0 Ranunculus bulbosus 78.8 13.5 Plantago lanceolata 20.0 3.9 Poa trivialis 34.8 5.2 Elytrigia repens 39.6 6.8 Calystegia sepium 17.3 3.4 Cirscium arvense 10.8 1.6 Calystegia sepium 29.3 5.0 Daucus carota 16.1 3.1 Total 638.3 95.3 Plantago lanceolata 17.7 3.0 Prunella vulgaris 13.4 2.6 Unidentified Poaceae 5 13.4 2.3 Total 493.9 96.8 Total 561.2 96.4 105 4 Ammoniated Elytrigia repens 86.0 42.7 Festuca rubra 18.4 18.2 Poa trivialis 58.8 27.6 Soap Solidago spp. and Symphytotrichum 48.9 24.3 Poa trivialis 18.0 17.8 Agrostis stolonifera 44.8 21.0 lanceolatum Ranunculus bulbosus 22.3 11.1 Persicaria amphibia 12.1 12.0 Elytrigia repens 41.6 19.5 Taraxacum officinale 12.1 6.0 Achillea millefolium 10.9 10.8 Ranunculus bulbosus 21.7 10.2 Cyperus esculentus 9.4 4.7 Ranunculus bulbosus 8.7 8.6 Setaria pumila 14.4 6.7 Solidago spp. and Poa trivialis 7.6 3.8 Senecio vulgaris 7.2 7.1 Symphytotrichum 11.5 5.4 lanceolatum Daucus carota 4.0 2.0 Elytrigia repens 6.8 6.7 Achillea millefolium 8.0 3.7 Festuca rubra 3.2 1.6 Agrostis stolonifera 5.8 5.7 Prunella vulgaris 3.5 1.6 Total 193.5 96.1 Cirscium arvense 5.0 4.9 Total 204.3 95.7 Taraxacum officinale 3.1 3.1 Trifolium spp. 1.7 1.7 Total 97.7 96.6 4 Solidago spp. and Mowing Ranunculus bulbosus 35.0 28.5 37.4 31.5 Plantago lanceolata 62.5 49.7 Symphytotrichum lanceolatum Plantago lanceolata 19.4 15.8 Ranunculus bulbosus 26.2 22.0 Ranunculus bulbosus 18.4 14.6 Elytrigia repens 13.9 11.3 Plantago lanceolata 14.4 12.1 Poa trivialis 10.6 8.4 Solidago spp. and Symphytotrichum 12.9 10.5 Poa trivialis 10.1 8.5 Setaria pumila 8.0 6.4 lanceolatum Daucus carota 11.0 8.9 Prunella vulgaris 9.9 8.3 Prunella vulgaris 6.4 5.1 Setaria pumila 10.9 8.8 Setaria pumila 5.7 4.8 Daucus carota 5.9 4.7 Solidago spp. and Prunella vulgaris 9.0 7.4 Daucus carota 3.4 2.9 Symphytotrichum 3.9 3.1 lanceolatum Trifolium spp. 4.5 3.6 Elytrigia repens 3.4 2.9 Persicaria pensylvanica 2.5 2.0 Taraxacum officinale 3.1 2.5 Taraxacum officinale 2.9 2.5 Taraxacum officinale 2.2 1.7 Total 119.5 97.3 Total 113.4 95.3 Total 120.4 95.7 106 4 Capric acid Elytrigia repens 90.2 46.7 Festuca rubra 29.5 31.4 Elytrigia repens 112.0 47.3 Poa trivialis 59.1 30.6 Elytrigia repens 21.9 23.3 Agrostis stolonifera 97.6 41.2 Solidago spp. and Cyperus esculentus 17.2 8.9 7.7 8.2 Poa trivialis 18.7 7.9 Symphytotrichum lanceolatum Setaria pumila 15.5 8.0 Poa trivialis 6.8 7.2 Total 228.3 96.5 Agrostis gigantea 4.2 2.2 Setaria pumila 5.7 6.0 Total 186.2 96.5 Unidentified Poaceae 5 5.2 5.5 Cyperus esculentus 5.0 5.3 Cirscium arvense 2.9 3.0 Achillea millefolium 2.6 2.8 Agrostis stolonifera 2.1 2.3 Total 89.3 95.1 Solidago spp. and Solidago spp. and 4 Solidago spp. and Split-Control Symphytotrichum 416.6 52.4 701.5 79.1 Symphytotrichum 774.1 72.4 Symphytotrichum lanceolatum lanceolatum lanceolatum Ranunculus bulbosus 151.3 19.0 Calystegia sepium 48.4 5.5 Cirscium arvense 97.6 9.1 Elytrigia repens 81.8 10.3 Elytrigia repens 39.6 4.5 Poa trivialis 94.6 8.8 Cirscium arvense 37.9 4.8 Cirscium arvense 33.6 3.8 Ranunculus bulbosus 66.7 6.2 Calystegia sepium 23.1 2.9 Ranunculus bulbosus 17.3 2.0 Total 1033.0 96.5 Vitis sp. 22.7 2.9 Plantago lanceolata 16.5 1.9 Festuca rubra 22.2 2.8 Total 856.9 96.6 Total 755.6 95.1 Abundance is presented as percentage and absolute value of biomass. Lists include species representing at least 95% of the total biomass. 107 Table A12. Represents the average aboveground biomass (gm-2) of monocotyledon weeds in each main treatment (Cultivation, Mulch, main Control), split treatment (Ammoniated soap, Mowing, Capric acid, split Control), and sample period (1-4) from 2017-2019. Significant interactions are indicated by letters. Aboveground biomass of monocotyledon weeds (gm-2) 2017 2018 2019 Main treatment Control 30.9 B 40.1 89.9 B Cultivation 25.0 B 48.6 80.0 B Mulch 9.36 A 24.2 32.9 A Split treatment Control 32.8 B 39.2 B 52.2 B Mowing 9.0 A 6.0 A 20.3 A Capric acid 26.6 AB 59.5 B 109.0 C Ammoniated soap 18.7 AB 45.9 B 88.9 C Sample period 1 5.2 A 17.3 A 34.0 A 2 17.6 AB 26.4 AB 55.6 AB 3 24.0 AB 56.4 B 83.0 BC 4 36.2 B 50.3 AB 97.6 C Statistical significance Main treatment ** * *** Split treatment * *** *** Sample period ** *** *** Main × Split ns *** *** Main × Sample period ns ns ns Split × Sample period ns ns ns Main × Split × Sample period ns * ns Different letters, ‘A,’ ‘B,’ ‘C,’ indicate mean separation at p ≤ 0.05 using Tukey’s honestly significant difference test. Letters indicating mean separation are specific to either the main treatments, the split treatments, or the sample periods and letters should not be compared across columns or across sections in a single column. ns: nonsignificant, *: significant at p ≤ 0.05, **: significant 108 Table A13. Biomass of herbaceous dicotyledon weeds harvested from each main (Cultivation, Mulch, main Control) and split (Ammoniated soap, Mowing, Capric acid, split Control) treatment at four annual sampling sample periods between 2017 and 2019. Represents the average aboveground biomass of weed species grouped by life cycle in each main or split treatment. Annuals Biennials Creeping Perennials Simple Perennials Sample Period 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 Main Year treatment/split ------------------------------------------------------------------- g•m-2----------------------------------------------------------------------------- treatment 2017 Cultivation 8.0 0.4 0.2 6.9 0.1 0.8 3.0 1.1 31.1 36.7 21.9 52.3 102.4 46.3 56.8 71.7 Mulch 0.0 0.1 0.1 0.0 0.0 0.0 0.7 0.7 22.2 17.0 28.2 21.8 59.6 48.9 36.1 51.5 Control 0.7 0.8 0.1 0.2 0.0 0.8 5.3 3.3 39.7 47.6 57.5 39.1 107.3 55.0 52.7 73.2 Ammoniated soap 3.7 1.1 0.2 0.1 0.1 1.0 0.1 0.0 37.0 23.8 27.8 12.6 126.4 47.9 25.5 33.9 Mowing 1.6 0.2 0.0 0.2 0.1 0.2 10.5 2.6 46.0 39.0 16.1 18.5 65.9 56.4 62.7 84.2 Capric acid 4.8 0.3 0.0 0.1 0.0 0.0 0.0 0.0 12.2 26.0 9.9 3.4 105.2 37.6 32.8 23.5 Control 1.5 0.1 0.2 9.0 0.0 1.0 1.5 4.2 28.7 46.3 89.7 116.4 61.7 58.1 73.2 120.3 2018 Cultivation 9.3 0.0 0.0 0.4 2.2 3.1 8.3 16.7 21.5 46.1 39.0 81.9 58.2 59.8 76.2 89.9 Mulch 0.3 0.0 0.0 0.0 0.8 1.2 1.9 0.3 21.6 33.4 44.6 76.7 39.2 44.9 63.6 24.9 Control 0.9 0.0 0.0 0.0 1.5 5.0 3.4 5.6 23.7 24.6 48.0 79.4 56.3 71.0 99.6 74.9 Ammoniated soap 3.6 0.0 0.0 0.2 1.0 0.0 0.0 0.0 19.9 19.0 24.2 18.5 40.4 37.7 16.2 31.7 Mowing 2.7 0.0 0.0 0.0 2.3 4.2 6.4 0.7 17.9 7.2 12.9 13.9 91.1 63.7 126.4 76.2 7.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 17.3 1.9 12.7 26.2 28.5 11.3 10.3 3.4 Capric acid Control 0.4 0.0 0.0 0.4 2.7 8.1 11.7 29.4 33.8 110.7 125.7 258.6 45.1 121.6 166.2 141.6 2019 Cultivation 0.6 0.0 0.0 0.0 0.5 1.3 7.7 4.3 23.2 47.6 88.1 140.2 65.7 70.2 65.2 66.8 Mulch 0.5 0.0 0.8 1.8 0.4 0.6 8.1 0.9 26.0 78.1 107.1 216.6 61.3 55.5 19.6 24.2 Control 0.0 0.0 0.0 0.6 0.4 0.3 1.1 2.0 23.2 49.0 124.8 231.3 51.8 94.0 54.9 48.3 Ammoniated soap 0.3 0.0 0.0 2.4 0.0 0.0 3.0 1.3 15.9 30.7 13.4 33.2 53.8 55.2 11.9 25.5 Mowing 0.0 0.0 0.0 0.8 1.2 1.3 11.0 7.3 6.8 13.9 11.2 29.2 115.0 113.2 53.2 62.7 Capric acid 1.1 0.0 0.0 0.0 0.4 0.4 0.0 0.0 9.2 5.2 1.7 7.0 23.5 12.6 0.7 2.2 Control 0.0 0.0 1.0 0.0 0.1 1.1 8.5 1.0 64.7 183.3 400.4 714.8 46.1 111.9 120.5 95.4 109