As climate change gives rise to water shortages and unstable growing conditions in California and across the United States, agricultural systems must be able to adapt to increasingly extreme environmental stressors. Diversified farming systems, which incorporate biodiversity across multiple temporal and spatial scales to support ecosystem services, offer an alternative to the fragility of the current industrialized regime that dominates US agriculture. When small-scale, thought-intensive, diversified farming systems are supported by research and socio-political movements that defend them and advocate for their wider adoption, food production will transition towards a science, practice, and movement known as agroecology. While many argue that agroecological transitions are necessary to achieve stable food production and climate, economic, and political justice in the US agricultural system, state and federal policies do not reliably support diversified farming systems. In order to craft effective policy interventions, we must have an intimate knowledge of how and why diversified farming practices work to properly support their success and spread. In my dissertation I explore two regional examples of diversified farming practices and their potential for wider adoption given current and possible future policy landscapes. In corn-based crop rotations in the US Midwest and tomato dry farming on California’s Central Coast, climate shocks have sparked a need for dramatic change, opening an opportunity for policy to guide agriculture towards an agroecological future. With the help of farmer collaborators, I ask how each of these systems functions, how policy has failed them, and where it may yet succeed.

The first chapter of my dissertation examines the political and physical landscapes in which farmers grow corn-based rotations in the US Midwest, asking what factors lead farmers towards complex vs simplified rotations. I used publicly available, remotely sensed datasets to look at relationships between rotational complexity, and biophysical (land capability, precipitation) and policy-driven (distance to the nearest biofuel plant) factors on 1.5 million fields in the region, using bootstrapped linear mixed models to account for spatial autocorrelation. I found that policy and economic incentives continue to lead farmers towards simplified crop rotations, such as corn-soy and even corn monoculture. In particular, amidst already elevated corn prices from crop insurance structures and livestock feed, I saw that proximity to biofuel plants–where corn prices tend to be higher due to the federal biofuel mandate–encourages farmers to grow corn in as many years as possible. These policy and economic factors then play out in biophysical landscapes as well, where fields with the highest quality soils and precipitation–which can tolerate degradative soil practices without compromising yields in the near term–tend towards the most simplified rotations.

The second and third chapters of my dissertation explore diversified farm management on California’s Central Coast in dry farm tomato systems, which rely on diversified farming practices (cover cropping, compost application, organic management, etc.) to build soil water holding capacity and fertility. Dry farming allows farmers to grow produce with little to no irrigation water, relying instead on water held in soils from winter rains to support crops through rain-free summers. As climate change increases water scarcity in California, farmers, advocacy groups and policymakers have begun looking to dry farming as a potential solution to the state’s overextended water budget. However, little research has been done on vegetable dry farming in the state, and no coordinated effort has been made to understand the policy conditions that would allow vegetable dry farming to thrive.

In my second chapter, I collaborated with six dry farm operations in a participatory process, coming up with research questions that the farmers who most intimately understand tomato dry farming were eager to answer. After a season of intensive soil and harvest sampling on seven dry farm fields on California’s Central Coast, we were able to come to a better understanding of how the system functions, and develop concrete management suggestions for farmers to consider. We found that, due to quickly drying surface soils, harvest outcomes were only impacted by nutrients below 30-60cm in the soil profile, upending soil fertility management paradigms on irrigated fields, where focus is almost entirely on the top 30cm of soil. We were also able to caution against arbuscular mycorrhizal fungal (AMF) inoculants, which have been marketed to farmers with increasing intensity but, if anything, harm harvest outcomes in dry farm systems (as opposed to resident AMF communities, which are fostered through diversified management and typically improve harvest outcomes). Lastly we showed that dry farm soils develop a signature in their fungal communities that supports fruit quality, suggesting that farmers would likely benefit from developing full dry farm rotations where soils are kept irrigation-free for multiple years.

In my third chapter, I conducted semi-structured interviews with the same farmers who participated in the field study to better understand the full context in which tomato dry farming operates on the Central Coast. As farmers, researchers, and policymakers consider an expansion of dry farm vegetable production in California, I wanted to ask how farmers understand the practice and what its environmental and economic constraints are. I emerged with a synthesis of farmer-stated environmental constraints that I used to create a map of California cropland that could be suitable for future dry farm production. As I considered an expansion of dry farming onto these new lands, I drew on farmers’ experience to explore how to maintain dry farming’s history as an agroecological alternative to the industrial style of farming that dominates the region. Farmers had a clear message about the small operations and direct-to-consumer marketing styles that have been the foundation of dry farming success, and that must serve alongside soil health practices as a model for an agroecological transition towards water savings in California. I identified policies such as publicly funded demonstration farms, participatory breeding programs, and public procurement that could promote dry farm expansion while preserving its fuller context and identity, rather than stoking a shift toward an industrial cooptation of the practice that could edge small growers out of dry farm markets.

Taken together, this dissertation follows on-farm diversification practices–especially those that can help farms guard against water scarcity–through an arc of past and possible future policy. I ask how policy has discouraged crop rotational complexity, and how policy might foster an environment in which dry farming could thrive as a model for an agroecological transition towards water resiliency. I strive to understand dry farming through the eyes of those who practice it to make management and policy recommendations that are grounded in deep knowledge of the system. Through these examples of how complex ecological and policy interactions play out in two separate regions, I highlight the resilience and importance of diversified farm systems, and the possibilities and pitfalls in policy interventions that attempt to prepare our agricultural system for changing climates.

All the ecological components of a farm are tightly interlinked – influencing each other, negatively orpositively, through multiple divergent pathways. Belowground, roots can form a well-developed network with arbuscular mycorrhizal fungi (AMF) for greater nutrient uptake. Aboveground, floral resources (pollen and nectar) influence bee visitation needed for crop pollination. In this way, AMF could positively influence crop pollination through improving nutrient uptake. However, agricultural intensification, characterized by high nutrient inputs, low crop diversity, and high tillage frequency, have been linked to biodiversity loss, negatively impacting multiple species interactions. While diversifying crops could help to promote these beneficial ecological interactions, it is poorly understood how changes in crop diversity impact bees that specialize on limited floral resources or mycorrhizal fungi that exhibit preferential host associations. By partnering with small-scale farmers of color, my dissertation focused on understanding the mechanisms in which crop diversity influences below- and above-ground interactions, and their connections (AMF to pollination).

The first part of my dissertation examined how crop diversity (monoculture versus polycultures)affected AMF and pollinator communities in the intensively managed agricultural region in California’s San Joaquin Valley. I examined how crop diversity influences several aspects of AMF communities: (1) richness and diversity, (2) composition, and (3) colonization. I demonstrate that crop diversity enriches AMF communities, counteracting the negative effects of agricultural intensification, providing the potential to increase agroecosystem functioning and sustainability. Next, to investigate the impact of on-farm diversification on pollinator communities, I focused on how specialist squash bees (Peponapis genera) responded to crop diversity (monoculture versus polyculture) using squash (Cucurbita pepo) as the focal crop. I hypothesized that squash bees would be more abundant on monocultures, due to greater numbers of squash flowers, than polycultures. Despite this, I found that increasing non-squash floral resource diversity in polycultures supports a greater abundance of squash bees. On-farm diversification may be an important refuge for specialist bees that are vulnerable in landscapes dominated by agriculture.

The second part of my dissertation sought to more closely characterize mechanisms that enhancepollination services critical to promoting food security. While aboveground floral visitors provide pollination services to plants, belowground AMF could enhance the quality of floral resources (pollen and nectar). Yet, the effect of the AMF community composition on pollination is still unclear because existing studies consist of a single AMF species or disregard the interacting effect of soil nutrient availability. Therefore, in a controlled greenhouse and field experiment, I investigated how different levels of AMF diversity and nutrient availability impacted floral resources and pollinator visitation. Our results demonstrate that greater AMF diversity, regardless of nutrient availability, has the largest positive effect on floral resource quality and pollinator visitation, an indicator of pollination services.

Altogether, this dissertation points to the effects of crop diversity on belowground and abovegroundbiodiversity that could bolster ecosystem functions. By working directly with small-scale farmers of color, this research also outlines potential pathways that farmers can take towards harnessing beneficial ecosystem functions. Importantly, the results demonstrate that crop diversification could help to recover biodiversity from below- to above-ground even in the intensively managed region in California’s San Joaquin Valley.

The relentless global pursuit of food and fiber production, often at the expense of natural ecosystems, has resulted in agricultural systems that degrade local biodiversity and environmental quality. While farmers have traditionally possessed the knowledge to maintain soil health and ecosystem balance, modern industrial agriculture has shifted towards simplistic, input-dependent practices. Organic farming offers a step towards sustainability by reducing synthetic inputs, yet it often falls short, still relying on organic substitutes and failing to address deeper ecological concerns. In contrast, soil health management practices, rooted in principles from the Natural Resource Conservation Service (NRCS), offer holistic strategies beyond mere input substitution. These practices prioritize maximizing living roots and soil cover, fostering biodiversity, and minimizing soil disturbance. However, practices that put these principles into action such as cover crops and reduced tillage remain underutilized, even among organic farmers. Understanding the barriers to the adoption of soil health management is crucial for transitioning to a more balanced agricultural paradigm that sustains productivity and environmental integrity.Most of our current understanding of the impacts of soil health management comes from research-station trials that isolate 1-2 practices at a time, within a single edaphic and climatic context. Recent on-farm research endeavors to bridge this gap by examining how management practices impact soil health metrics in real-world settings. Rebuilding soil organic matter (SOM) is a core goal of soil health management, given its manifold benefits, including supporting soil life and enhancing structure and nutrient availability. Increasing SOM levels aligns with short- and long-term goals of bolstering soil health and combating climate change through carbon sequestration. However, it isn’t clear how implementation of soil health practices, as utilized on actively managed farms, impacts these different carbon goals. Beyond carbon, soils provide many critical ecosystem services, and the impacts of soil health management on various soil-mediated ecosystem services and general multifunctionality is not well resolved. Managing farms for multiple ecosystem services beyond crop production can also present challenges, as trade-offs and conflicting priorities often arise. Understanding the intricate interplay between local soil characteristics, management practices, and various stakeholder objectives is essential for crafting effective policies. My dissertation addresses these challenges by delving into three key areas of agricultural development: identifying barriers to adopting soil health practices, exploring the impacts of management on soil carbon, and assessing the relationship between management practices and multiple ecosystem services. By integrating social, economic, and environmental perspectives, my dissertation informs policies that facilitate the transition to sustainable soil management practices, fostering a more resilient and environmentally sound food system. Briefly, we find that different farming models face unique and varied challenges in adopting soil health practices. On farms with higher levels of continuous living cover, reduced disturbance, and crop diversity, we observe higher carbon stocks and increased mineral-associated and particulate organic matter. We also find that management tends to be more influential on distinct carbon outcomes relative to inherent soil properties. Lately, we identified eleven beneficial relationships between various soil health practices and soil-mediated ecosystem services including yield, soil fertility, carbon sequestration, nutrient cycling, soil microbial diversity, and mitigating excess end-of-season soil nitrate. Continuous living cover in particular emerges as a key practice to enhance multiple services simultaneously (ecosystem multifunctionality). While not a panacea, improving soil management practices represents a crucial step toward achieving multifunctional and sustainable agricultural landscapes.

Supplementary information