Hemp Growing

Our research builds upon this emergent body of work that employs urban agroecology as an entry point into broader policy discussions that can enable transitions to more sustainable and equitable city and regional food systems in the U.S. . This transition in UAE policy making is already well underway in many European cities . As noted, there are many dimensions of agroecology and ways in which it is conceptualized and applied. We employ the 10 elements of agroecology recently developed by the UN FAO in our discussion of urban agroecology1 . These 10 elements characterize the key constituents of agroecology including the social, ecological, cultural, and political elements. Despite the emancipatory goals of agroecology, a recent review of the literature by Palomo-Campesino et al. found that few papers mention the non-ecological elements of agroecology and fewer than 1/3 of the papers directly considered more than 3 of the 10 FAO-defined elements. In an effort to help guide the transition to more just and sustainable food and agricultural systems in cities across the U.S., we propose that food system scholars and activists consider using the 10 elements as an analytical tool to both operationalize agroecology, and to systematically assess and communicate not only the ecological, but also the social, cultural and political values of urban agroecology. “By identifying important properties of agroecological systems and approaches, as well as key considerations in developing an enabling environment for agroecology, the 10 Elements [can be] a guide for policymakers, practitioners and stakeholders in planning, managing and evaluating agroecological transitions 2 .n San Francisco’s East Bay region, urban food production proliferates in schoolyards, vertical growing racks in half-acre lots converted to urban farms, on rooftops, and in backyards reflecting a diversity of participants, goals, impacts and challenges .

The San Francisco East Bay region is also experiencing rapid gentrification and a worsening affordable housing crisis coupled with high rates of income inequality and food insecurity. The challenge of urban soil contamination creates tradeoffs for aspiring growers between vacant lot availability and siting on the most heavily polluted plots . Specific city policies vary in the degree to which they support or discourage urban agricultural activities, and availability of arable land across the East Bay is uneven. Our case study focuses on urban farmers in the East Bay spanning over 28 miles from El Sobrante in the northeastern edge of the bay, to Hayward in the southern East Bay as shown in Figure 1. We include both for-profit and non-profit farms ranging from educational school gardens to roof-top farms marketing microgreens.We employed a participatory and collaborative mixed methods approach, involving diverse stakeholders from the East Bay Agroecosystem. We held two stakeholder input sessions involving over 40 urban farmers and food advocates to co-create the research questions, advise on the data collection process, interpret the results, and prioritize workshop topics for the community. We administered an online Qualtrics survey to 120 urban farms in the East Bay that had been previously identified by the University of California Cooperative Extension Urban Agriculture working group and additional outreach. The survey launched in Summer 2018, which is a particularly busy time for farmers, and in response to farmer feedback was kept open until November 2018. 35 farmers responded in total, representing a 30% response rate.

While there are limitations in our ability to generalize findings to the East Bay urban farming landscape as a whole due to the relatively small sample size, we obtained a fairly representative sample of the diversity of farm types in the East Bay based on our typology of the original 120 farm types . Survey questions fell into nine categories: 1) Background Info, 2) Farm Description, 3) Operating Expenses and Revenues, 4) Land Access and Tenure, 5) Production and Soil Health, 6) Distribution, 7) “Waste” and Compost, 8) Food Access, and 9) Training, Communications, and Follow Up. There were a few open-ended questions allowing farmers to express what they saw as the three largest challenges facing urban agriculture operations in the area, and policy-relevant suggestions for securing spaces for urban farms and increasing community food security. In addition, we interviewed five urban farmers to deepen our understanding of the social, political, economic, and ecological constraints under which their farms operate. These farmers are particularly involved in networking efforts to strengthen urban farm viability in the East Bay. Four out of five represent locally prominent non-profit farms and one subject represents an alternative cooperatively-run urban farm; three interview subjects are women and two are men. Our study complied with UC Berkeley’s Institutional Review Board protocol for the protection of human subjects and all participants gave consent for participation.Most farms including the UC Oxford Tract and Gill Tract Farms, distribute food to a diverse array of community organizations. The two aforementioned farms together distribute food to over 50 community organizations, ranging from food pantries to community health groups to native land trusts seeking to feed and reclaim land for those of indigenous heritage. 52% of respondents distribute all food within 5 miles of their farm, while 70% distribute within 10 miles.

Produce from each farm site reaches approximately 250 people per week on average during the peak growing season, or approximately 7,000 people from all surveyed farms. Customers reached is moderately correlated with total revenue suggesting a growing impact on CFS as farms access additional income. Farmers reported diversified distribution methods including volunteers harvesting and taking food home , on-site consumption , on-site farm stand distribution, CSA boxes at pick up sites, and volunteers delivering produce directly to distribution sites . Some gleaning and second harvesting occurs at urban farms and gardens with potential for growth given reported “unharvested” and “wasted” food percentages. Backyard produce is also exchanged through crop swaps and neighborhood food boxes . Eight operations reported having access to a refrigerated truck for food deliveries, and two are willing to share their truck with other farmers. There is no universally used or city-organized process for distributing produce off of urban farms and into the community, yet there exists great interest in aggregating produce or distribution channels , an unrealized goal of urban farmers in the East Bay. All of the food system stakeholders involved in our study are working towards transformative food system change, focused on increasing equity, food security, and access to healthy, locally sourced food. See Box 1 for a description of one of the non-farmer stakeholders engaged in the food recovery and distribution system, who has recently established an aggregation hub to serve as a network for reducing food waste and channeling excess food in the urban community to those who are food insecure.Farmers in our study stressed the importance of producing non-food related values on their farms, including education and community building. One farmer in particular emphasized their organization’s mission of “growing urban farmers growing food,” or teaching other people how to grow a portion of their food basket, vertical farming racks thus unlocking food sovereignty and food literacy while increasing healthy food access. Another respondent reported that their farm is “highly desirable for adults with special needs that need a safe place to be outside,” echoing respondents who point out the intimate connection between food and health . Farms frequently reported hosting educational and community-building workshops, cooking and food processing demonstrations, harvest festivals, and other open-to-the-public community events enhancing the resilience and connectivity of people, communities and ecosystems. Social networks emerged as an important theme for enabling the establishment of urban farms , and sustaining operations through social connections between urban farmers and other food justice and health advocates.Farmers identified three primary challenges: revenue, land, and labor inputs. Half of all respondents reported farm earnings of $1,500 annually or less, and all four operations receiving over $250,000 in annual revenue are well-funded non-profit operations . Regardless of for-profit or non-profit status, most farms reported multiple sources of revenue as important to their continued operation , with an average of 3 revenue streams per farm. All non-profit farms reported multiple revenue streams except for three, who were sustained entirely by either board donations, membership fees , and grants. The most important revenue sources for non-profits include grants, grassroots fundraising, and unsolicited donations rather than sales. In addition to these monetary sources, all farms reported receiving substantial non-monetary support , which adds to the precarity of operations when these informal support channels disappear.

Land tenure arrangements range from land accessed without payment through contracts with City or School District officials, to arrangements where a token fee is paid , to more formal leasing arrangements at the utility-owned Sunol Ag Park, where land tenants pay $1000/acre/year for their plots, ranging from 1-3 acres. Only five of the respondents owned their land , representing a mix of for-profit and nonprofit operations . Challenges around land access, security, and tenure were the most frequently occurring theme in the survey long response and interview analysis process, including consensus that land access is the largest barrier to scaling UA in the East Bay. The cost of labor, and relatedly, access to capital and grant funding to pay living wage salaries, were also extremely significant challenges identified by survey respondents. The majority of respondents stated that most of their labor is volunteer rather than paid, with nonprofit respondents reporting this more frequently than for profit enterprises . The maximum number of paid staff at any operation is 20 , while the average is 4. Many farms reported the desire to be able to hire and pay workers more, but not having sufficient revenue to accomplish that goal. Annual volunteer labor participants on farms ranged from 0 to 1542 with an average of 97 volunteers, representing a significant public interest in participating in local food production. Not surprisingly, amount of paid labor and total farm income are positively correlated . However, volunteer labor is also positively but more moderately correlated with total farm income .The farmers in our study acknowledged many challenges facing urban agriculture, stemming both from the high economic costs of production and land rents, and insufficient monetary returns from produce sales. They also framed these challenges through a food justice lens, arguing that the current political economy does not fully compensate farmers for the social-ecological services provided from their farms. Farmers articulated many solutions that could improve the viability of their farm operations including: conversion of city parks into food producing gardens with paid staff, government and institutional procurement goals for urban produced foods, municipal investment in cooperatives or other community based food production , and establishment of aggregation hubs and distribution infrastructure.Our survey results describe a highly diversified East Bay Agroecosystem comprising urban farmers and other food system stakeholders that are growing food as well as food literacy, civic engagement, connectivity, and community. Applying an agroecological lens to interpret our findings of East Bay urban agriculture operations reveals the many agoecological practices farms have long been engaged in, as well as the important distinctions of UAE that still need to be explored, and specific threats to agroecology in urban areas. Pimbert suggests that “agroecology’s focus on whole food systems invites urban producers to think beyond their garden plots and consider broader issues such as citizens’ access to food within urban municipalities and the governance of food systems.” We argue that applying an agroecological lens to the urban context also invites researchers and urban planners and policymakers to think beyond garden plots and singular benefits of food production, to consider these sites as part of a larger agro-ecosystem with synergistic social, cultural and ecological dimensions. We reference the 10 elements of agroecology to illustrate the dynamics of how these elements manifest in practice in this urban context.All of the farms in our survey follow agroecological production practices which include a focus on building soil health through, most commonly, cover cropping, compost application, and no-till practices. These practices produce synergistic effects of adding fertility to the soil through organic matter amendments and boosting water holding capacity. Soil building practices are a response to the impetus to remediate toxins present in urban soils , a prerequisite to intensive cultivation and unique consideration of the urban farm environment. Overall, production practices on our urban farms seek to conserve, protect and enhance natural resources. Our survey respondents described numerous strategies for enabling diversified, intensive production of fruits, vegetables, and other agricultural products.

The inventories, however, must then be structured according to the formats usable by openLCA. This modeling process also requires an abrupt switch from a very high level diagram — the flow diagram — to a detailed inventory requiring low-level information about her farm. There is no process connectivity between the two steps and information does not flow between the models. Two: Time and effort overhead. Models must be manually inspected to ensure completeness and correctness. This may result in a suite of errors, such as models that misrepresent the system, models that are missing portions of the real world system, or models that have incorrect connections between processes. Further, the modeling effort expended to create the flow diagram is not utilized to reduce effort in the LCI creation process. The majority of LCI databases are focused on industrial production systems. Those that are concerned with food are oriented toward the production of processed foods as a resultof partnerships with food processing stakeholders. Data required to conduct an LCA of any alternative agricultural systems is unavailable and puts the onus on the system owner to collect primary data. Three: Lack of flexibility. Only one type of granularity, the unit process, can be modeled. There is no support for creating logical groupings of unit processes in the form of components and no capacity to create hierarchies of such components. Alice is locked into two disconnected levels of abstraction: the high-level flow diagram, vertical grow and the low-level inventory. Even once an LCA model is created, current mechanisms only support the sharing of unit process data is shared.

No other reusable portions of the model are explicitly supported. This also means that as things on the farm change, an entirely new LCA model would need to be created.Life cycle assessment has enabled people to investigate, quantify, and understand the environmental impacts of agricultural and industrial systems. On the whole, LCA is useful, but it is not without its limitations. LCA requires meticulous and tedious data collection for every single process within a system. A good LCA is comprehensive and results in detailed models, however it can be a time consuming and cost-prohibitive process, depending on the size, complexity, and novelty of the system under analysis. The development of software also involves modeling, and there has been substantial research into improving the software modeling workflow. In the following analysis, I call on software engineering research and practice to tease apart some of the modeling challenges faced in modeling the environmental impacts of agricultural systems, as well as to propose opportunities for future work. Fred Brooks discriminates between essential difficulties with software — those relating to intrinsic characteristics — and the accidental difficulties — those relating to temporary or circumstantial characteristics. He goes on to state that to address accidental difficulties, one must promote incremental improvements, but to solve essential difficulties, one must promote revolutionary improvements. The challenges faced in the modeling of the environmental impacts of agricultural systems involve addressing both the essential and accidental difficulties. LCA is one of many environmental impact assessment methods that aim to address these difficulties. LCA methods, tools, and data have undergone immense incremental improvements . However, several essential difficulties remain.

Essential difficulties faced when modeling the environmental impacts of agricultural systems involve the representation, or capture, of complexity, change, and context of such systems. Interestingly, some of these mirror the essential difficulties or characteristics of software as defined by Brooks: complexity, conformity, and changeability.Only some of these attributes are represented in current LCA models: the flow diagram informally represents the scope and boundaries of the system and data documentation formats provide structure when representing unit processes. Dependencies, subsystems, and other potentially relevant attributes are linguistically represented, i.e., they are often described in reports that accompany LCA models. Through the structuring of unit processes as prescribed by DDFs and the list-of-unit processes LCI inventory, LCA results in a powerful declarative model: one where within the system boundary, unit processes represent all major flows of material and energy withinthe system, and are connected through the functional unit.Current LCA modeling is similar, in that it is effective at retroactively assessing static systems. Agricultural systems have a symbiotic relationship with the environment, and are closely tied to the health of the environment due to the interconnectedness with many natural systems. These assessment of agricultural systems is unique as not only is one assessing a system that is directly dependent on the natural environment, but that also has a large component of it that are not of human construction, i.e., they are a part of nature: plants, animals, and the land. The quality of the food produced on a farm is related to the quality of the land, air, and water used. These are seasonal systems that are constantly producing, consuming, and evolving over time.

The global warming potential of a farm in Northern California over the Spring of 2014, when they were growing peas and corn, will not be the same five years from now. For this reason, LCA models are time, location, and system sensitive. Dynamic models of agricultural systems that represent changing resource flows, however, are not available in the LCA modeling paradigm to date. Existing LCA models and tools do not capture change effectively. They provide functionality to increment a model version number, but updating and maintaining models to keep up with changes in the real world is difficult. Excel has the “track changes” functionality,which is a helpful collaboration aid that provides different authors with awareness of how a document is changing, but is not explicitly geared toward capturing changes in LCA models. GaBi does something similar. The tool logs all changes to each object at every save . Once again, the goal is to keep track of which user changed the model to support collaboration and not necessarily to augment the capacity of the model to reflect changes in the world. openLCA simply provides a version number for each object in the file that can be incremented as changes are made.State of the art LCA modeling tools do not provide the ability to revert to previous models, compare models over time, or maintain a model history that reflects the changing world. If one is to update and maintain a model over time, it would involve a manual process of backing up and continuing from where one left off. While this is a valid approach, grow vertical given the scale to which some LCA models can grow and the increasing complexity of agricultural systems, more advanced tool support is required.We are constantly producing and trying to connect different types of information in the real world, as well as in our models. A model has two aspects: the representation of a system, and perspective on the system. When these models are treated in isolation, they provide representation without context. Such models cannot be connected with each other to produce meta-models, nor can existing models be reused. When a model is created of an agricultural system, an artificially created system boundary sharply closes off the rest of the world from the system under analysis. The flow diagram tries to capture some contextual information by showing materials and energy flowing into and out of the system as a whole, but, as discussed earlier, the flow diagram is quite disconnected from the eventual LCA model. Trying to capture the context within which a system exists is a difficult problem. After all, one must scope a modeling problem appropriately lest they end up trying to simulate the world. The question remains, however, how does one capture the context of a system? Standalone LCAs are still common as they are conducted when stakeholders of a particular system are interested in self-evaluation and improvement. Comparative LCAs are conducted when stakeholders in an industry are interested in pitting one set of production techniques against another, to analyze whether or not a new or alternative concept is actually better for the environment than an older or standard approach, or any of the other reason discussed previously in this paper. Conducting an LCA that is wide in scope is difficult, and so, when people are interested in understanding wider or more far reaching impacts, existing LCA studies are used to conduct retrospective meta-analyses. Environmental consulting agencies also tend to conduct these analyses, as part of their business model is LCA as a service , where they charge for their expertise, time, and effort. Depending on the intellectual property issues at play, they may own the model details and are not beholden to reveal the raw data to anyone else.

LCA models provide perspective on a specific system without any context. They are difficult to connect with each other to provide more holistic environmental impact assessments, and it is difficult to reuse partial models. People all over the world are producing extremely detailed models, with a lot of time, money, and effort going into inventorying complex systems. A vast amount of data and a large number of models are produced through LCA. We are unfortunately, producing, but not connecting, environmental impact models. No explicit interfaces or connectors are available to connect entire LCA models.LCA models are intended to provide efficient and convenient access to information about the environmental performance of production systems, such as agricultural systems. However, due to the mismatch between the current LCA modeling language, workflow, tool, and the systems represented, I believe that an important opportunity has been missed to capture the complexity, change, and context of agricultural systems. In this section, I describe a potential avenue through which to address these modeling difficulties. In software engineering, the Object-Oriented paradigm came about as an attempt at capturing the essential complexity of software. Booch describes two kinds of decomposition: Algorithmic and Object-Oriented decomposition. The algorithmic results in a top down structured model, declarative and directed. On the other hand, in the OO approach, one decomposes the systems using the key abstractions. Booch notes: “[In object oriented modeling,] we view the world as a set of autonomous agents that collaborate to perform some higher level behavior”. Current LCA modeling utilizes a hybrid of the two, where the flow diagram is created using an OO approach, but is then quickly abandoned for a mostly algorithmic decomposition in subsequent models. Given the modeling challenges described in this chapter, I propose that a potential means to capturing the inherent complexity of agricultural systems is through a new modeling language that allows for a more object-oriented approach.A new modeling language would not be a silver bullet. It would not necessarily capture all of the complexities, changes, or context of agricultural systems, nor would it remove all accidental misrepresentation of systems. What it would hopefully do, is make the process of representing the environmental performance of such systems easier and less tedious, providing clear and distinct ways to represent and connect modular system models. They are prone to change over time due to rising environmental issues , advances in agricultural practices and technology, changing needs of an ever-growing human society, and a dynamic economic context. Our farms are part of the highly interconnected web of industrial civilization, with dependencies on many other systems. Environmental impact assessments are conducted, regulations written to govern production, tools built to guide the flows of materials and products through supply chains, and food labels created to assist in consumer decision making. At the core of these issues is a mismatch between existing environmental assessment tools and the needs of small- to medium-scale farmers , who do not engage in large-scale industrialized agricultural practices. In Chapter 3, I described the essential difficulties faced when modeling agricultural systems.The Grounded Theory Method is methodology used to develop theory about phenomena through iterative interrogation of data. GTM offers a means to explore new territory, particularly when there is a lack of dominant theory. While the goal of this study was domain understanding, I used GTM as a means of structuring the design of this study and utilize GTM techniques for subsequent data analysis. Through use of GTM, my broader goal was to develop theories of design for sustainable agriculture. Muller writes that one must “remain faithful to the data, and to draw conclusions that are firmly grounded in the data” . I do this by iterating between recruitment, data collection, and analysis, exploring the different concepts at play, while gradually developing a theory of how to design a consistent mechanism for modeling sustainable farms and their interactions with the environment.

Agricultural systems are a critical component of this infrastructure, providing crucial food products as part of the human food system, but also providing vital agricultural by-products as inputs for other industrial systems. They are deeply coupled with the natural environment and undergird human civilization. As this web of systems — industrial, agricultural, environmental, human — grows more intricate, it becomes increasingly difficult to understand and map out the consequences of our actions. Agricultural systems are, thereby, commonly assessed to improve economic performance, reduce environmental impacts, and improve social sustainability. System assessment begins with the representation of a system through some modeling process, and a subsequent evaluation of the model with respect to certain attributes. This process involves the collection, structuring, manipulation, and storage of a variety of data. System assessment often involves multiple modeling processes as each model is created to represent a particular aspect of a system and its performance. While instrumental in im-proving our understanding of these complex systems, these efforts have led to a fragmented field with tensions and pulls in different directions, and have resulted in a duplication of effort, and moreover, data and processes that are disconnected. Over the past decade, small- to medium-scale sustainable farms have been growing in popularity. There is an increasing demand for food that is grown sustainably, raised humanely, and produced with fair and just labor practices. This has spawned many efforts to curtail environmental impacts: including, numerous regulations, eco-labels, indoor cannabis grow system and certifications. Such efforts often require farmers to engage in additional record keeping that can be both tedious and time consuming.

Farmers, in essence, pay for the privilege of growing food sustainably. Despite growth in this sector, there is a lack of appropriate technological tools, arising in part from a mismatch between existing tools — which are typically designed for large-scale industrialized agricultural practices — and the needs of farmers who work at smaller scales. With this comes a growing need for systemic mechanisms to understand, analyze, manage, and further improve, sustainable agricultural systems. There is no dearth of analyses of the environmental impacts of our food systems, but there is a lack of connectivity across the plethora of models created and a lag with the changing real world that is represented.Agricultural systems are composed of heterogeneous subsystems with varied environmental impacts. For example, Figure 1.1 shows Alegria Fresh Farms, an urban farm in Orange County, California. It is a small-scale farm that produces fruits, vegetables and a variety of agricultural byproducts such as organic soil. It is not a single product system. A number of subsystems are present within the farm for irrigation, solar energy production, hydroponic cultivation, hydroponic vertical cultivation, certified organic cultivation, vermicomposting, and a nursery, among others. To understand which subsystems are more responsible for certain environmental impacts, one would need to be able to attribute impacts to particular subsystems and processes.This would show the farmer which cultivation method, for example, is more environmentally friendly. The hydroponic vertical cultivation subsystem shown in Figure 1.2 exemplifies the complexity of modern agriculture, even at this small scale. Many of the materials described are sourced externally from other industrial systems. Many external data are therefore required to accurately assess the environmental impact of the farm.

Availability and access to the environmental impact data of these external components is not guaranteed. Agricultural systems are also constantly evolving: equipment is upgraded, cultivation methods are refined to optimize certain metrics , subsystems grow and shrink, and food types grown change according to supply and demand, in addition to season. Figure 1.3 shows two satellite images of Alegria Fresh Farms: the image on the left is the farm shortly after it was set up. Alegria Fresh Farms was one of a set of small experimental and demonstration farms that were introduced to the Orange County Great Park in around 2009. The image on the right was taken after several years of farm activity. The farm has since been relocated. As changes are made in an agricultural system, any models that were initially created become outdated. When models are constructed for agricultural systems, they are a static representation of a dynamic system. Once data have been collected and things change, the model is no longer an accurate representation of the system. Over the last six years, the Alegria Fresh Farms has grown, systems have changed, and as a result, the relationship of the farm with the environment has been affected. However, many models, from satellite images to layout diagrams, cannot necessarily capture the change in the system and its en-vironmental impacts. While there are many techniques that allow for detailed analyses of an agricultural system’s environmental performance, they are primarily expert-driven technique, which means that once a model is created, it is difficult for the farmer to maintain or update it for continued use.Software systems for agricultural modeling are primarily designed for activities such as farm simulation and yield maximization.

These can be both computationally expensive. Such work, while interesting, relevant, and timely, primarily addresses the modeling and simulation of agricultural systems to improve crop yields and system management, with only a marginal focus on environmental assessment of these systems. Agricultural Ontologies: The Food and Agriculture Organization of the United Nations has maintained the AGROVOC project since the 1980s. It is a controlled vocabulary, designed for, and used by, information management professionals . It consists of over 32,000 agricultural concepts, gleaned from publications, research artifacts, and external thesauri. Similarly, the United States Department of Agriculture’s National Agricultural Library has provided a more America-centric glossary and thesaurus service since 2002. These initiatives are part of a broader goal within the agricultural information management community to standardize agricultural terms, concepts, and data. There are also several efforts — in early stages of development — to extend tradition agricultural thesauri into fully developed ontologies for use at the farm level to inform crop production, “foods-for-health” knowledge systems in the nutrition space, and to augment precision agriculture with plant-driven decision making capa-bilities. The AGROVOC team is also developing an ontology service in anticipation of semantic web requirements. Crop Modeling: Work in crop modeling tends to focus on operational and yield optimization. For example, Honda et al. present a service platform that uses a network of field sensors to obtain real-time field data to plan large-scale field operations e.g. fertilizer application. Meanwhile, Ponti et al. used statistical models to assess the yield gap between organic and conventional agriculture to obtain a deeper understanding of the range of performance in agriculture. Agricultural Simulation: Agriculture researchers often use simulations to understand and make predictions about the performance of various agricultural systems, while computation researchers apply their expertise to improving the performance of simulations in agriculture. For example, the Agricultural Production Systems sIMulator project is a modular simulation tool that allows for the investigation of relationships between plants, animals, soil, climate, and management involved in agricultural systems. Papajorgji et al. present different ways in which model-driven architecture, in particular the use of the Unified Modeling Language, can be leveraged to improve crop simulation models. Miralles & Libeourel study how Geographic Information System models can be brought to bear on crop simulation to allow for better integration of weather data. Life Cycle Assessment: LCA is a modeling technique used to assess the environmental impacts of products and the processes by which they are constructed. Farmers, along with environmental analysts, can conduct LCAs to quantify the environmental impacts of resource flows in a system, vertical grow racks and subsequently make improvements in the farming processes to reduce undesired impacts. Many software systems and databases exist to support LCAs, most of which are domain agnostic . Models of agricultural products and production systems are commonly created in the academic community using LCA. Examples of systems that have been successfully analyzed using LCA include: a large American feedlot where beef is produced; a soybean meal production chain that spans from Argentina to Denmark; and an Australian corn chips production chain that begins at the corn farm, and ends at the consumer-ready packet. While LCA models are intended to provide efficient and convenient access to information about the environmental performance of production systems, such as agricultural systems, there is a disconnect between the current LCA modeling process, the needs of sustainable farmers, and the systems represented.

The Bruntland report presents an umbrella definition of sustainable development as, “the ability of humanity to ensure that it meets the needs of the present without compromising the ability of future generations to meet their own needs”. While this definition is commonly used in sustainability research, it is too broad to allow us to engage in grounded, sustainable, human computer interaction work, or other forms of sustainable design activities. We often come away from such discussions with difficulty envisioning specific applications to design and actionable interventions to pursue. There have been many explorations of sustainability in the context of HCI, ICT for Development, and the coordination literature. However, just as most “sustainability-oriented work takes place outside HCI” , much design work for sustainable agriculture also occurs at the periphery of applied computing research. In this section, some of the more promising explorations within this periphery are reviewed.HCI and Agriculture: Interest in the intersection of HCI, design, and agriculture is growing. For example: Raghavan et al. recently suggested use of computation to design better agro-ecological systems, and Frawley & Dyson created non-human animal personas to enhance welfare in animal agriculture. For the most part, design for agriculture tends to be for specific subsets of agriculture. For example, Chinthammit et al. ran a Software Interest Group meeting, looking at “HCI in Food Product Innovation”. Whether it is design ideations specifically for urban residential gardeners or the design of platforms to assist in the creation of backyard permaculture systems, Di Salvo et al. point out that “there is a significant gap between the professional fields of industrial and interaction design and design research in sustainable HCI” . In an attempt to bridge this gap, my collaborators and I ran a workshop on at the ACM CHI conference in 2017 on “Designing Sustainable Food Systems”. We aimed to bring together HCI researchers, designers, and practitioners to explore, design for, and reflect on, opportunities for the HCI community to engage in creating more a sustainable food system. Coordination and Collaboration in Agriculture: Food production is an inherently collaborative process, with many stakeholders involved and varying organizational configurations across system types. Examples include: a qualitative study looking at coordination challenges in organic farm families, and an early warning management technique to enable collaboration among rice farmers participating in small-scale precision agriculture. Such work provides valuable context for understanding how the interplay between stakeholders on farms affects tool design. Farm Management: At the time of this writing, the Information Technology startup community began focusing on the design of tools for agriculture. Earnest examples include: precision farming tools, precision agriculture tools requiring specialized hard-ware ; daily farm management tools aimed at agribusiness; inventory management tools; and tools that provide analytics.Agricultural Environmental Policy: Environmental regulations have serious implications on farm-level assessment, data collection, and record keeping that farmers in California must engage in [38]. Regulations, both state-imposed, and federal, typically require some form of record keeping and form filling, and in some cases, the presentation of these documents during site inspections. Environmental policies that California farmers are subject to include: water, pesticides, and emissions regulations. In addition, farmers may participate in voluntary programs to demonstrate commitment to environmental protection, such as the environmental stewardship program. Environmental Labeling and Certification: The primary goal of environmental labeling and certification agencies is to provide farm quality assurance. These labels and certificates provide the purchaser, whether broker, retailer, or consumer, with the validation that a farm meets a particular standard set out by the certifying agency. The efficacy and usefulness of labels has long been debated from the early days of dolphin-safe labels to the present-day quandary of GMO-labeling. Nevertheless, many farms actively pursue labels and certifications, necessitating an additional layer of record keeping and data collection. There are four main types of labeling and certification in sustainable agriculture: government regulated National Organic Program; non-profit regulated ; retailer specific ; and product-specific .

By meeting others and forming new relationships, community members also increased their social support, an objective of the social and community context domain . When asked to describe their proudest success, one UA leader responded, “seeing friendships form that wouldn’t have happened otherwise.” They believed that the garden was a space for people to “meet people from different cultures, hear different languages, [and] try different cuisines.” Farmers described several types of outdoor community events, such as “Farm Friday,” for people to “have a drink and hang out during the summer.” One gardener said that they “run into good people” at their community garden, and another replied that “garden people are always fun to talk to.” Acknowledging Long Beach’s diversity, one UA site created promotional materials in English, Spanish, and Khmer, demonstrating linguistic capital, or skills from communicating in more than one language or style . During field visits to community gardens, I observed multiple occasions when people would walk by and ask questions about the garden. Gardeners welcomed outsiders by sharing their harvest and providing information on how to join.All interviewees described how UA had changed the land in some way. Community members worked together to enhance their neighborhood and built environment, which influences safety, health behaviors, and risk of disease . The UA sites in this study were built on former oil properties, dumping sites, vacant lots, and in some cases, previous UA sites. Social capital was crucial for building new gardens and farms. UA leaders collaborated with volunteers, the Conservation Corps of Long Beach, pipp racking local troops from Boy Scouts of America, and the UC Master Gardener Program to construct raised beds, sheds, and other amenities.

Ground Education typically constructed raised garden beds and nature paths on existing green space, such as playground fields. In other cases, UA leaders had to build on land that was previously unintended for planting. This passage illustrates how community members used social capital to collectively build a UA site. The garden director explained that they thoroughly tested the soil to ensure that it was safe, and that the process of organic gardening would further improve the soil over time. Gardeners’ determination to turn the former oil property into a green space demonstrates resistant capital, knowledge and skills fostered through oppositional behavior. By repurposing the once “lifeless” soil to grow food, gardeners also demonstrated aspirational capital, the ability to maintain hopes and dreams for the future . UA leaders emphasized the importance of protecting the environment for future generations, a further example of aspirational capital. Though most sites were located near freeways, UA sites planted trees to provide shade and oxygen for fresh air. Trees also served as a noise barrier and gave a “park-like feel” to UA sites. Gardeners and farmers used natural pest control methods and practiced composting to repurpose their food waste. One UA leader harvested seaweed from the beach to make kelp extract, which they used as a fertilizer.UA leaders shared how their sites directly benefitted community health by increasing access to fruits and vegetables, promoting physical activity, and supporting mental health. One UA leader was inspired to donate produce to the local university food pantry after reading articles on food insecurity among college students. Gardeners also relied on growing food for their own food security. For example, a gardener who lived alone and did not own a car relied heavily on their community garden plot. They said growing food helped them eat more fruits and vegetables “loaded with fiber.” Another mentioned their grandparents “save some more money” by eating the food their family grows. In this way, UA supported the SDOH domain of economic stability by allowing community members to reduce food expenses.

UA leaders believed that “if you grow your own vegetables, you’re more likely to eat vegetables” and that food grown from a garden or farm was “fresher,” “tastier,” and more “flavorful” than produce from grocery stores. During my observations, people at UA sites were consistently engaged in outdoor physical activity, usually with their hands, shovels, hoses, wheelbarrows, and other gardening equipment. When I asked an interviewee if they used any power tools for gardening, they replied, “We like to burn calories, not fossil fuels.” They added that “some of the younger folks really get a workout with the pickaxes and stuff,” while gardening for “older folks” is “slow, gentle exercise that [they] can do for prolonged period.” Others agreed that gardening helped them stay “healthy and active.” A gardener who suffered with joint pain from fibromyalgia said, “[Gardening] gets me out of my pain.” Another gardener had to stop gardening at one UA site because of chronic lower back pain but was able to begin gardening again at a different site with raised beds. He sat on his walker to pull weeds and said the raised beds help with accessibility. Participating in UA helped community members cope with physical discomfort, as well as negative feelings and emotions. The phrase “mental health” was repeated by multiple interviewees, such as one who stated, “I think that the biggest crisis that we’re facing in health care other than nutrition, is mental health.” UA sites provided a space for the community to “de-stress,” “heal,” have “peace of mind,” and “manage depression and anxiety.” One UA leader described their community garden as “therapeutic” and a “safe space” that benefitted their family’s physical health and mental health. Another said, “People told me it’s their lifeline… A lot of people told me that they went through a rough time, maybe they lost a spouse, or they had broken up, or they lost their job, and the garden was their focal point.” From my interviews and observations, many others seemed to share this sentiment. During this study, I met three gardeners who had experienced homelessness. They all expressed how gardening helped them stay calm during stressful times. A UA leader who previously lived in their car and struggled with addiction issues said that gardening helped them with sobriety. They said, “The garden is important to me because it gave me purpose.” The act of gardening allowed community members to shift their attention away from negative thoughts and focus on the present moment.UA presented a multitude of learning and teaching opportunities for people of all ages. Ground Education’s lessons throughout the LBUSD are one example that ties directly to the SDOH goal of increasing educational opportunities and helping children and adolescents succeed in school . In October 2023, I had the opportunity to assist third grade students with an herb-picking activity. A lively group of about 30 students walked from their classroom to the school garden. Since it was also Picture Day, the students wore their best clothing. In their suits and dresses, the third graders cheerfully ran up to the Garden Educator and greeted her with hugs. Once the students took a seat on the outdoor benches, the educator began the lesson with a mindfulness exercise. Students placed a hand on their heart and stomach, so they could feel their body move as they inhaled, held their breath, exhaled, and repeated. After the exercise, the energetic class of third graders appeared calmer. The educator asked the students to name plants that were herbs. Then, she discussed how herbs were used as medicine by native peoples, and how herbs are still used today in teas and topical treatments. The educator used a whiteboard with a diagram of the limbic system to teach students about the olfactory bulb. As the educator explained to students, the olfactory bulb in the brain perceives smells and sends signals to the body, which is why smelling different herbs can make people feel more focused, calm, or alert. After the lecture, I helped distribute sachets for the students to collect herbs. We walked from the main school garden to a “Secret Garden,” where the next part of the lesson would take place.Students explored the garden and collected calendula, lavender, mint, rosemary, vertical grow racks and thyme to add to their sachet. Though it was not part of the lesson plan, students also gathered yellow roses, nasturtium, and other flowers, and one student picked up a tiny lizard.

After returning to the benches for a closing discussion, students walked back to their classroom. However, the garden was not empty for long. Many returned during recess to water the plants, or simply enjoy the space while sitting, eating, and chatting with others. This lesson was unique because it focused on biology and the medicinal qualities of herbs. Other lessons I observed related to nutrition and tasting fruits and vegetables, connecting to the previous themes of “Improving Health.” Students also learned about food supply chains, the structure of plants, and using fractions to plant seeds in a garden bed evenly. School garden lessons are examples of organized educational curriculum at UA sites. A few sites offered youth education programs, such as LBO’s Gateway to Gardening program taught by a horticultural therapist. The Martin Luther King Jr. Peace Garden offered garden-based education classes. There were also informal opportunities for teaching and learning through hands-on experience. It is likely that students may share their garden experiences at home with their parents and siblings. Many adult gardeners and farmers had learned agricultural skills from their families. This is an example of familial capital, cultural knowledge nurtured among kin . A UA leader admitted that when they started, they “literally knew nothing about gardening” but now “[are] trying to learn as much as they can” while gardening with their family. During field visits, I saw that gardeners were often working together with children, parents, grandparents, and siblings. UA leaders recalled learning how to grow food with their family or a significant other. One UA leader, who had learned agricultural techniques from their mother, felt it was important for their children to learn that “the growing experience is normal and natural.” They believed that if “kids are more attuned to things growing and dying, and then going into the compost, and then creating life again, it makes the idea of death, less onerous.” For this interviewee and others, educating children was crucial for preserving UA skills and knowledge.UA leaders identified conflicts with landowners, lack of funding, engaging local community members, and language barriers as major challenges of maintaining gardens and farms. In addition to social capital, UA leaders exercised navigational capital, the skill of maneuvering through social institutions , to use land owned by the city or private owners. An interviewee said that ideally, this was a “win-win situation” because “[landowners] don’t have to take care of these odd properties and they save money and time, and [community members] get a garden.” However, there was often “red tape” obtaining permits from the city, which delayed UA leaders from starting construction. Even after UA sites were constructed, their operation was not guaranteed. The City of Long Beach could temporarily lease land to UA leaders, then choose to not renew the lease in the future. Private landowners had the power to reclaim their land or sell it to a buyer. An interviewee reflected, “We’re all on borrowed land.” UA leaders also felt the constant pressure to acquire funding through grants or fundraisers to keep their sites operational. Due to inadequate funding to hire full-time staff, UA leaders were tasked with “doing everything,” from fundraising and coordinating volunteers, to cleaning bathrooms and dealing with animal pests. Although UA sites generated income from produce sales, community garden memberships, donations, and grants, revenue was unpredictable. Oftentimes, UA sites were spearheaded by one main person, with either a few staff, volunteers, and/or interns. Ground Education was the exception to this, with a team of over 20 employees; however, the two co-founders worked for seven years without pay to grow their nonprofit organization. They gained financial support through the LBUSD, parent teacher associations, and Long Beach Gives, a citywide fundraiser. A common challenge that community garden leaders identified was “a lack of people who can consistently come and maintain the garden.” They would repeatedly contact gardeners who neglected their plots, despite policies requiring regular upkeep. Schools similarly struggled to maintain their gardens, which was why Ground Education was founded. Before they had the resources to construct new gardens, Ground Education cared for existing school gardens that were untended during the summer or completely abandoned.

While UA can increase access to food, it also has the potential to bolster CCW through cross-cultural social interactions and educational and employment opportunities . Communities can potentially use UA to address health inequities through social and environmental changes .While SDOH may be broadly defined as “the conditions in which people are born, grow, live, work, and age,” Hahn clarifies that SDOH encompasses social resources and health hazards controlled by societal systems which, in turn, have consequences on health outcomes and risks . SDOH can include health-related knowledge, attitudes, beliefs, and behaviors, but these factors are often the result of social factors that are uncontrollable by individuals, such as discrimination . Populations that experience wide disparities in SDOH are affected by health inequities, such as higher risk of disease and earlier mortality . Research suggests that low income and education levels are strong predictors of physical and mental health problems . Socioeconomic status , which can be broadly defined as the combined total measure of a person’s social and economic position in relation to others, is also positively correlated with health outcomes . The following sections will describe SDOH in the context of Long Beach, based on the domains described by The Healthy People 2030 initiative: 1) social and community context, 2) economic stability, 3) education access and quality, 4) neighborhood and built environment, and 5) health care access and quality.The social and community context domain refers to the relationships people have with others. Factors like social support, self-esteem, pipp horticulture and self-efficacy may protect against health risks from adverse social conditions . These protective factors may be fostered through the CCW model . This is especially important for marginalized groups targeted by discrimination .

Those who have a low SES due to social disadvantages, such as discrimination due to race/ethnicity, gender, sexual orientation, and disability, are more likely to suffer from health inequities . Children born into low SES have greater risk of experiencing sudden infant death, infectious diseases, exposure to lead poisoning, household smoke, accidents, and child abuse, which may explain why low SES children have higher rates of asthma, developmental delay, and avoidable hospitalizations . Children from low SES neighborhoods face greater barriers to health-promoting behaviors and often experience stressors from family conflict and economic instability . Additionally, they are at greater risk of being exposed to intimate-partner and community violence. Low SES adolescents report worse health than their peers, experience higher rates of obesity, pregnancy, sexually transmitted disease, depression, and suicide, and more likely to be sexually abused, drop out of high school, or be killed. Compared to those who are more economically advantaged, low SES adults experience higher rates of mental illness, food insecurity, coronary heart disease, and other chronic health conditions, and experience earlier mortality . When discussing SES, it is important to note that in the United States, “race, socioeconomic status, and health have historically been inextricably intertwined” . Federal agencies collect data primarily by race due to Statistical Directive No. 15 of the Office of Management and Budget , originally adopted in 1977 . As of 1997, the directive requires federal agencies in the United States to report statistics for one ethnic category, Hispanic, and five racial groups: American Indian and Alaskan Native, Asian, Black, Native Hawaiian or Other Pacific Islander, and White . United States census data is based on how individuals self-identify to one or more groups, and reflects a general, social definition of race, independent of biological, anthropological, or genetic factors .

For clarification, when this dissertation describes “minorities,” “people of color,” or “communities of color,” this typically refers to non-White racial groups. There is no internationally agreed definition for minority , but in the United States, racial and ethnic minorities are groups of non-European descent: American Indian/Alaska Native, Asian, Black or African American, Native Hawaiian/other Pacific Islander, and Hispanic/Latino . As of 2020, most United States residents identify as White . The U.S. Census Bureau predicts that the nation’s population will become more racially and ethnically diverse, as immigration is projected to surpass birth as the primary driver of population growth. People who identify as more than one race are projected to be the fastest growing racial or ethnic group over the next several decades, followed by Asian Americans and Hispanic/Latino Americans . Over 70% of Long Beach’s population identifies as a racial/ethnic minority, or people of color . About 44.1% of the population is ethnically Hispanic or Latino2 .A stable income is necessary to afford food, housing, and health care. Steady employment can prevent poverty, which is experienced disproportionately among most non-White populations. Compared to 8.2% of non-Hispanic White persons, poverty rates are over twice as high for Black and Hispanic people . Unemployment is strongly associated with worse health and higher mortality, and those who live in poverty are unable to afford health-promoting living conditions . Higher education often leads to employment in jobs with higher compensation, better health care benefits, and safer working conditions. Conversely, those with a lower education are at greater risk of being injured and exposed to hazardous chemicals while working . As mentioned previously, historical redlining practices shaped the neighborhood demographics of Long Beach. For example, housing lenders imposed deed restrictions to prohibit non-White residents to purchase, lease, or occupy property .

Such restrictions were common in East Long Beach and Bixby Knolls, which, decades later, still report lower minority populations than other parts of Long Beach . The National Association for the Advancement of Colored People advocated against such discriminatory housing policies, and worked toward local policy reform in education, employment, economic development, and law enforcement. Their work resulted in the end of racial deed restrictions during the 1960s, though a 1975 study found that unfair housing practices still continued . That same year, Cambodians escaping civil war found refuge in Central Long Beach, which lenders considered risky for investment. Despite this, Cambodians created a commercial district there, and Long Beach became home to the largest Cambodia diaspora. In 2006, the Long Beach City Council officially designated a portion of the city as “Cambodia Town” . Recent data shows that communities of color are still concentrated in North, Central, and West Long Beach. Figure 9 shows a map of Long Beach . The dark purple areas represent where communities of color are the most concentrated.Education increases access to economic opportunities and resources which can influence health . Persons with less than a high school education are expected to live six years less compared to those with a college education . Almost half of all deaths among working-age adults in the United States can be attributed to potentially avoidable factors associated with lower educational attainment, including discrimination in health care settings . In addition to racial/ethnic groups and women, lesbian, gay, bisexual, and transexual groups and persons with disabilities also experience health inequities, which can be attributed to reduced education and employment . Higher education is associated with increased social support, which may benefit physical and mental health by buffering the effects of stress, enhancing health knowledge, and encouraging healthy behaviors . Education level is also highly correlated with health literacy, the ability to comprehend and use information to manage medical care and make informed health decisions . According to Long Beach Unified School District enrollment data from 2021-22, 37,952 students were socioeconomically disadvantaged. The California Department of Education states that socioeconomically disadvantaged students meet at least one of the following: neither parent received a high school diploma, eligible for the Free or Reduced Price Meals Program, are a migrant, homeless, or foster youth, or were enrolled in a Juvenile Court School. In Long Beach, over 40% of residents are Hispanic or Latino, and 12% are Black. Hispanic/Latino and Black families are more likely to have lower educational attainment and quality, due to living in neighborhoods with under-resourced schools . LBUSD enrollment data from the California Department of Education shows that most socioeconomically disadvantaged youth identified as Hispanic or Latino , African American , or Asian/Pacific Islander .Residential segregation forces communities of color into hazardous areas, resulting in detrimental effects on mental and physical health . Neighborhoods with increased social disorder may heighten anxiety and depression . Particularly for poor and Black neighborhoods, health risks are exacerbated by increased exposure to polluted air and contaminated water, because toxic waste facilities, industrial plants, best way to dry cannabis and landfills are often intentionally sited in low SES neighborhoods . Pollution also increases risk of COVID-19 , which, after its discovery in December 2019, became the nation’s third leading cause of death after heart disease and cancer. Compared to non-Hispanic Whites, those who identified as American Indian or Alaska Native, Black, and Hispanic or Latino were about twice as likely to become hospitalized and die from COVID-19 . According to Sprainer , this is because “low-income communities and communities of color across the country are exposed to higher long-term concentrations of an air pollutant that makes COVID-19 more deadly.”

Low SES communities are historically exposed to pollutants due to perceived lack of political power to control zoning, which is controlled by largely white governance structures for industrial development . In urban areas, low SES communities also have less access to green space . Additionally, not all green spaces are equally healthy and well maintained. Data from Su et al. suggested that low-income and minority residents with access to city parks have greater exposure to air pollutants such as nitrogen dioxide , fine particulate , and ozone . Therefore, low SES populations have less opportunity to experience green spaces and their associated benefits, such as improved physical and mental health, and even safer neighborhoods . These patterns of pollution exposure and access to green space can be clearly seen throughout Long Beach. The Long Beach Airport was ranked by the Environmental Protection Agency as having the second highest lead emissions of airports nationwide . Air pollution is further exacerbated by emissions from cars and trucks, particularly those transporting goods from the Port of Long Beach. As seen in Figure 10, areas closest to the port and 710 Freeway, toward the west, have pollution levels of 70- 100%. In comparison, areas of East Long Beach, where the minority population is lower and household incomes are higher, pollution levels are less than 40% .The areas with the highest concentration of minorities, air pollution, asthma, and diabetes are furthest from the largest parks in Long Beach . West Long Beach only has one acre of green space per 1,000 residents . Low SES communities in Long Beach are also susceptible to contamination from industries. For example, an empty lot of upper West Long Beach was used as a sludge dumping ground for oil companies in Long Beach and Signal Hill . This has resulted in years of toxic waste build up, including lead and arsenic. In 2021, local residents opposed developer plans to build a storage facility and requested officials to conduct an environmental impact report. They also argued that the empty lot, which became an area for homeless encampments, should be used to develop a park .The multitude of health risks associated with low SES cannot be addressed by health care alone and may even be exacerbated by discrimination in health care settings . Discrimination perpetuates health inequities by increasing health risks, lowering health care quality, and disrupting the economic opportunities available for low SES populations . There is evidence that people of color in Long Beach, who live in areas with higher air pollution and lower access to green space, have higher hospitalization rates. Long Beach’s hospitalization rates for asthma are higher than that of LA County and California . Data from the Long Beach Department of Health and Human Services also reveals that age-adjusted emergency room rates due to adult and pediatric asthma are over eight times higher for Black residents , than White residents . Asthma rates for the Asian and Pacific Islander population Hispanic/Latinx population were also higher . Figure 11 displays adult and pediatric rates of asthma by ZIP Code.Similar to asthma, hospitalization rates for diabetes are also higher in Long Beach compared to the county and state . In 2016, 9.7% of adults in Long Beach reported being diagnosed with diabetes.

A major unanswered question is whether expansion began with hunter-gatherer groups, perhaps as a result of the invention of particular technologies or behavioral innovations, or much more recently with the advent of agriculture. Early mtDNA studies suggested that humans experienced a burst of population growth between 30 and 130 thousand years ago —well before the start of agriculture. More recent results have extended the time frame for sub-Saharan African growth to 213–12 kya, depending in part on mtDNA haplogroup. However, it is populations—not haplogroups— that are subject to growth, and many present-day hunter-gatherer groups, including those in Africa, do not exhibit any mtDNA signal of demographic expansion at all. On the other hand, Y chromosome sequence data are compatible with a model of constant size for both hunter-gatherer and farming populations in Africa. Autosomal microsatellites tend to indicate an early start to population growth, but there is disagreement among studies on the time of expansion and whether or not the expansions involved African populations. Zhivotovsky et al. examined a large autosomal microsatellite dataset in 52 worldwide populations and concluded that African farmers, but not hunter-gatherers, exhibit the signal of population growth. Unfortunately, inferences of demographic parameters based on the above mentioned loci may be unreliable due to the possible confounding effects of natural selection or evolutionary stochasticity , best way to dry bud or uncertainty in our understanding of mutation rates or the underlying mutation process. A more reliable source of information regarding past population size change comes from multilocus nuclear sequence studies.

Once polymorphism data from multiple X-linked and autosomal loci began to appear, clear discrepancies with inferences based on both mtDNA and microsatellites emerged. For example, most non-African populations tend to have positive Tajima’s D values— reflecting possible contractions in Ne—while most African populations tend to have only slightly negative values. Indeed, the largest re-sequencing study to date that targets unlinked autosomal non-coding regions finds that patterns of neutral polymorphism in non-African populations reject the standard constant size model, and are most compatible with a range of bottleneck models invoking a large reduction in effective population size some time after the appearance of modern humans in Africa. In contrast, data from the sole African population examined, the Hausa of Cameroon, were compatible with demographic equilibrium, as well as with a set of recent population expansion models. In this paper, we expand upon the work of Voight et al. by analyzing a re-sequencing dataset comprised of 20 independentlyevolving autosomal non-coding regions in 7 human populations. Our sub-Saharan African populations include the San from Namibia, Biaka from the Central African Republic, Mandenka from Senegal, and Yorubans from Nigeria. Our multilocus analysis, which focuses on two summary statistics with power to detect population growth , follows a two-step approach. We employ a simulation-based method to test the hypothesis that populations experienced exponential growth after a period of constant size. When the hypothesis cannot be rejected, we then fit parameters of this two-phase growth model to our data using approximate Bayesian computation. As in previous studies, we find that the non-African data are not consistent with a simple growth model. On the other hand all four sub-Saharan African samples fit the two-phase growth model, and we are able to infer a range of onset times and growth rates for each population.

We sample sub-Saharan African populations that practice different subsistence strategies and then ask whether the inferred signals of population growth are shared between, or specific to, food-gathering or food-producing groups.Our understanding of population size changes in human prehistory has improved as our genetic datasets and analysis methods have become more sophisticated. Early studies of the pairwise mismatch distribution in mitochondrial DNA suggested dramatic increases in population size between 110 and 70 kya in sub-Saharan Africa. More recent coalescent studies have also favored 50- to 100-fold growth occurring between 213 and 12 kya. Conversely, modern surveys of nuclear sequence variation at unlinked loci have not provided clear evidence for rapid population growth from small ancestral size. For example, African populations usually exhibit slightly negative Tajima’s D values, while non-African populations tend to have positive Tajima’s D values. Different patterns of polymorphism in African and non-African populations have been interpreted as reflecting a history of bottleneck in the ancestry of non-Africans. Therefore, the question of when anatomically modern human populations began to expand in size is better addressed in sub-Saharan African populations because more recent demographic events likely obscure signals of population growth in the ancestors of nonAfrican groups. Bottlenecks, in particular, can mask the effects of earlier, as well as later, population growth.However, thus far, very few surveys of nuclear DNA sequence variation have been performed in sub-Saharan African populations, and interpretations drawn by existing studies have been complicated by the different populations and loci analyzed, the kinds of analyses performed, and the different growth models assumed.

The earliest studies considered only the few existing nuclear sequence data available in the literature at the time, and explored only a small set of growth model parameters. Later studies adopted a more explicit hypothesis-testing framework, but focused on only a single African population. For instance, Pluzhnikov et al. analyzed a large resequence dataset of noncoding autosomal regions for the Hausa of Cameroon . They determined that while observed summaries of the site frequency spectrum did not statistically reject a null model of constant size, they were consistent with a range of alternative growth models. Consequently, Voight et al. turned to a goodness-of-fit approach to determine better estimates of the time of onset of growth and the growth rate in the Hausa. By generating approximate likelihoods for the mean of observed summary statistics over a grid of parameter values, they determined that the Hausa best fit a growth model beginning ,1,000 generations ago with a per-generation growth rate a of 0.7561023 . Assuming a generation time of 25 years, this corresponds to an overall ,2-fold growth rate from ancestral to modern size beginning ,25 kya. Here, we extend these sorts of analyses to a greater range of African populations: two hunter-gathers, the San of Namibia and the Biaka of the Central African Republic; and two food producers, the Mandenka of Senegal and the Yorubans of Nigeria. All four groups show depressed values of Tajima’s D and Rozas’ R2 coupled with a high proportion of singleton mutations . These patterns of sequence polymorphism are suggestive of population growth. We therefore tested our multilocus African dataset to determine whether we could reject models of population growth, and adopted the best aspects of previous hypothesis-testing and inference approaches. We first employed hypothesis-testing to determine, by coalescent simulation, whether a range of growth models could be rejected in favor of constant size using the method pioneered by Pluzhnikov et al.. When growth could not be rejected, we fitted parameters of the two phase growth model to our data using approximate Bayesian computation . Thus, we conditioned simulations on each locus individually , cannabis grow setup and explored a continuous range of parameter values rather than restricting our search to a set of predetermined grid coordinates. All of our African populations best fit models with relatively low population growth beginning in the late Pleistocene . Even with ,112-kb of sequence data per individual, a large range of growth models are consistent with our 95% credible regions for t and a. We cannot, for instance, statistically distinguish different rates and times of growth among our four sub-Saharan African samples. However, our hunter-gather populations show a tendency towards slightly older and stronger growth than our food-producing populations . Furthermore, we detect a strongly negative, non-linear association between t and a . This effect, which has been identified previously, implies that sequence data from our four African populations are consistent either with weaker growth beginning earlier in the Late Pleistocene, or with stronger growth commencing more recently. Interestingly, we can reject an onset of population growth for the San during the Holocene , and therefore, growth in this population is not linked to the development of agriculture. Although we cannot reject an onset of growth associated with agriculture for the Biaka, Mandenka and Yorubans, our best fitting models do not favor this interpretation. Indeed, the limited size of our dataset gives us more power to infer older rather than more recent growth. We see little effect from the increased size of the dataset obtained for Yorubans. Even though we increased both the number of samples and the number of loci , estimates of the rate and timing of growth are comparable to those inferred for the Mandenka, and our 95% credible region is not appreciably smaller. This is interesting given that, under a model of population growth, expected values of Tajima’s D depend to some extent on sample size. With regard to the small increase in the number of loci in our Yoruban dataset, recent power analyses by Adams and Hudson suggest that orders of magnitude more data may be necessary to obtain growth model parameters with substantially greater accuracy, especially in models involving recent growth. Furthermore, the modern effective sizes we infer – on the order of 105 – are much smaller than regional census sizes.

This discrepancy partly reflects the fact that effective size is not a simple proxy for census size. However, another explanation also seems likely: under a model of exponential growth, the bulk of the population increase is weighted towards the present, and for the aforementioned reasons [28], we are not likely to capture the effects of substantial increases population size in modern times. Although population growth seems like a reasonable demographic model for human groups on non-genetic grounds [1,2,34], humans have likely experienced both population growth and population structure at some time in the past. The question is whether and to what extent either or both of these aspects of population history left a signature on patterns of variation. To explore the effects of alternate models of population structure on patterns of genetic variation, we use a coalescent simulation approach. In particular, we examine how Tajima’s D and Rozas’ R2 respond under models incorporating low-frequency gene flow in a structured population, recent admixture, and cryptic population structure . We assume a two-deme splitting model with i) a constant low level of gene flow, ii) a single admixture event occurring ,3 kya , and iii) population structure collapsing ,150 years ago . All of these processes produce very slight reductions in Tajima’s D and Rozas’ R2, but the mean deviations never exceed 0.27 and 0.011, respectively. To put these values in perspective, such deviations represent no more than 10% and 12% of the variance naturally observed for Tajima’s D and Rozas’ R2 under the corresponding standard neutral models with no gene flow, admixture, or cryptic population structure. Although these confounding factors may have caused our growth estimates to appear slightly older or stronger than they actually are, their effects are minor. Similarly, biases in our estimates of per-locus mutation and recombination rates are unlikely to have major effects on our inferences. For instance, elevated recombination would lead to a lower variance of Tajima’s D and Rozas’ R2, which would return growth estimates with less uncertainty, while elevated mutation rates would shorten our time frames, and hence return younger growth estimates. Estimates of growth rates under the isolation-with-migration model, which simultaneously accounts for population structure and gene flow, are consistent with our inference of an increase in the effective size of sub-Saharan African populations. Although growth rates are lower than suggested by ABC, we still infer that African populations experienced ,5-fold growth from ancestral sizes. While a simple two-phase growth model is too simplistic to fully describe African population history, it is interesting to note that a more complex model incorporating an ancient bottleneck does not fit African resequencing data. This is in marked contrast to the large reduction in population size that the same studies inferred for non-Africans. We therefore suggest that our growth estimates genuinely reflect a substantial increase in effective size among sub-Saharan African populations beginning in the Late Pleistocene. However, we note that these inferences could be complicated by other forms of population structure not accounted for in our models. While some authors have speculated that human populations underwent sudden expansions in population size in response to dramatic climatic events, technological inventions, or behavioral changes that took place earlier than 50 kya, our data are more consistent with a model of exponential growth beginning after 50 kya, but certainly before the Holocene.

Our results emphasize that lytic phages are likely to be an important component of the microbiome and are capable of influencing both bacterial abundance and diversity over short timescales.In our first of multiple experiments , we conducted a proof of concept experiment. We used ddPCR to measure quantities of known phage and bacterial host in size fractions of our mock community , and we determined that our fractionation method effectively concentrates phages from the leaf wash, allowing us to deplete them from both the “bacteria only” and 100K MWCO filtrate fractions of the leaf microbiome . FRS and SHL bacteriophages were effectively depleted, although the ddPCR signal was not entirely eliminated in the 0.22- µm filter bacterial recovery fraction . Phage levels were concentrated from the 0.22- µm flow-through fractions in the 100K MWCO centrifugation unit, representing bacteria plus phage treatment. Lastly, we also measured decreased levels of phage in the 100K MWCO flow-through fractions, representing the additional phage-depleted inoculum: bacteria plus filtrate. FRS and SHL phages are approximately 60 and 80 nm in size, respectively, and we thus presume that most phages in the environmental samples that are that size or larger should be retained in the upper portion of the 100K MWCO centrifugation unit. Membrane pore size for the unit we used is 10 nm; therefore, curing and drying weed smaller phage particles should have been retained in the upper fraction as well. Overall, we therefore consider both the bacterial/fungal fraction and the 100K MWCO flow-through fraction phage-depleted, but not necessarily absent of all phage.

Lastly, levels of P. syringae pv. tomato abundance was measured in all fractions , and signal was also detected in the non-bacterial fractions. However, this is likely due to the detection of DNA and not the presence of live cells, as bacteria could not be cultured from those filtered fractions . As seen in Figure 4-1d,infectious phage particles were present in the initial leaf wash, and they were also sufficiently high in concentration to completely lyse the bacterial lawn in the 0.22 µm flow through and 100K MWCO concentrate fractions, as little to no bacterial growth is observed. By comparison, a solid bacterial lawn is seen in the 0.22- µm filter recovery sample, where most phages appear to be depleted. As evidenced by a small number of plaques, a few bacteriophages are present in the 100K MWCO filtrate. This further supports the possibility that the third treatment, bacteria plus filtrate, was phage-depleted, but not completely free of phages, in our subsequent field experiments.After rarefaction and filtering, there were a total of 200 OTUs present in the spray inoculum from field experiment 2 representing taxa from the four top phyla commonly found in the phyllosphere: Proteobacteria, Actinobacteria, Bacteroidetes, and Firmicutes. As expected, the bacterial composition of inoculum from the three different treatments, sampled after resuspension with/without phage but before growth, has similar rank order of relative abundance for the top OTUs . Observed differences in relative abundance of specific taxa may be due in part to concentrated free bacterial DNA in the 100K MWCO fraction. Given the way in which inocula was prepared , it is unlikely that the bacterial communities differed substantially between treatments at inoculation.Using a community-level phage depletion approach, we found that the phage fraction of the phyllosphere microbiome from field-grown tomato plants impacted bacterial abundance and composition during microbiome establishment on a new host.

When microbial communities were sprayed onto juvenile tomato plants after either phage depletion or resuspension with the depleted phage-fraction, we observed decreased abundance in the latter treatment after 24 hours across three different experiments : first with six independent leaf wash sources , then with one leaf wash source and six plant replicates per treatment , and finally with a constructed bacterial community and natural phage fraction . Using 16S rRNA Illumina MiSeq data from field experiment 2, we were able to further show that the phage-fraction of the phyllosphere affects microbiome composition, including relative abundance of specific OTUs . We observed an effect of phage depletion treatment on community dissimilarity between treatments after 24 hours, but not after 7 days . We also found some evidence for differences in both alpha and beta diversity between phage depleted and phage re-suspended communities after 7 days . Overall, these results support the idea that lytic phages can mediate bacterial dynamics within host-associated bacterial communities, as they have been found to do in free-living communities. Across these experiments we observed a decrease in overall bacterial abundance 24 hours after inoculation, suggesting that phages affected growth of the most common and/or fastest growing bacterial strains during colonization of a new plant host. However, it is important to note that decreased overall bacterial abundance is not necessarily an expected outcome of lytic phage action within a microbiome. This is both because phage-mediated lysis has been shown in some cases to increase population growth due to release of nutrients but also because other strains that are not being targeted by phages should be able to offset any decreased growth of susceptible bacteria.

That the impact of phages on abundance in our experiments was short-lived suggests either that phages are particularly impactful during initial colonization, as bacterial population are rapidly growing, or that resistant bacterial strains/species increased in density over time to utilize existing resources. Indeed, the Kill the Winner hypothesis predicts that phages should most commonly prey upon highly competitive bacterial species. Results of our sequencing efforts supports this model, as we found different relative abundances of the two dominant families when the phage fraction was versus was not present in the initial inoculum. After 24 hours, the bacteria plus phage treatment plants were observed to have lower abundances of Pseudomonads, but when the phage-fraction was depleted there was an overabundance of an OTU within the family Enterobacteriaceae. However, after seven days the differences in relative abundance of these two OTUs were no longer observed to differ among treatments. Although only marginally significant, the presence of phage in the inoculum also led to an increase in alpha diversity at seven days post-inoculation. Again, this result may have been driven by a decrease of Pseudomonads after the first 24 hours, perhaps allowing a richer community to develop after the first week. Interestingly, when comparing beta diversity among treatments using averaged Bray-Curtis distances between samples within a treatment, we found an interaction effect between day sampled and inoculum treatment. This suggests that the phage fraction of the microbiome may also be having an effect on among-host microbiome diversity, initially driving divergence among communities as the empty niches are filled, , but eventually leading to more synchronous community structure. It is important to note that the patterns we observed were based on the depletion of lytic phages from the microbiome at the point of inoculation, but there were almost certainly many temperate phages remaining and possibly some lytic phages contained within bacterial cells at the time of collection/filtration. As such, it is possible that differences in treatment effect observed between 24 hours and 7 days were due to the resurgence of phages in the phage-depleted communities rather than loss of phages in the bacteria plus phage treatment. The observed transience of phage-mediated impacts on abundance and diversity is intriguing, and longer-term studies with more time points are needed to better understand temporal effects of phages on bacterial communities. One question we were not able to directly address in this series of experiments is the constituents of the leaf wash filtrate . The molecules and small proteins found in this filtrate had a surprisingly large and variable impact on the phyllosphere microbiome, impacting both abundance and community composition and causing high variation among biological replicates. In future experiments, additional size fractionation of the leaf wash filtrate and/or mass spectrometry analysis of these fractions may help address this question. As observed in our proof of concept experiment, cannabis drying system it is also possible that some bacteriophages made it through the filtration step and were present in this treatment. We decided to eliminate this treatment from many of our analyses due not to the effect of the treatment itself but rather due to the high variances observed across replicate plants. In most cases, plants within this treatment spanned the variation observed in both the bacteria alone treatment and the bacteria plus phage treatment. It was therefore unclear to us how to interpret this treatment and what biological significance it might have, but further study is certainly warranted. Another limitation of this work is that we have not identified the specific phages in the phage-fraction of the experiment. We have taken measures to ensure that the method used for separation of microbiome fractions is effective at separating phage from bacteria, but in order to fully describe the diversity of phage, as we have done for the bacterial community, one would need to take a metagenomics approach.

Furthermore, there may be other entities that are phage-sized in thatfraction of the microbiome, such as extracellular vesicles or spores of bacteria such as Bacillus that impact upon microbiome colonization. However, given that the current estimates of phages largely outnumber bacteria in the environment, we expect non-phage particles to be far less abundant than phages in this size fraction. This was recently shown for outer-membrane vesicles, where they were estimated to represent less than 0.01-1% of SYRB DNA-stained phage-sized particles quantified in seawater. Furthermore, we cannot rule out the possibility that the presence of phage, but not their predation on specific taxa, is causing the effects we are observing. However, by recapitulating the results of decreased abundance in bacteria after 24 hours when a phage fraction was present in our constructed community, we were able to lend some insight to this question. In this case, our detection of a phage capable of lysing a member of the constructed community suggested that the phage fraction was most likely driving the observed decrease in abundance. This is further supported by the fact that the phage was found to lyse Pantoea agglomerans, a member of the family Enterobacteriaceae, which we have found to be in high relative abundance in 16S rRNA community data in both this experiment and other unpublished work. Another important note is that the ddPCR protocol used here relies on lysis of bacteria cells through a hot-start step in the PCR. Because of this, it is possible that our abundance measures do not take into account hard-to-lyse bacteria. Finally, we did not include any analyses of the fungal communities in these microbiomes, as it was outside the scope of the current work. However, it is possible that our filtration methods also impacted any fungal viruses that might have been present in this study. How fungal communities are influenced by viruses within the microbiome is certainly an open question in the field that warrants further study. Given the building evidence that the phyllsophere microbiome is a key component of plant fitness, influencing key functional traits and likely protecting host plants against disease, the idea that lytic phages impact these communities is of direct relevance to plant health. A better understanding of bacteria-phage dynamics within these systems may present opportunities for manipulating the plant microbiome and ultimately increasing plant health. These ideas can be extended to the human microbiome, where the role of phages is proving to be appreciable. With regard to using phages in therapeutics, their role in controlling bacterial community dynamics and local adaptation is an important consideration for both phage-therapy to target specific pathogens and full-microbiome perturbations or replacements via fecal transplants. Overall, our results make a significant contribution towards the empirically demonstration of the role that phages play in shaping bacterial community structure in natural systems. This may be through, but is not limited to, impacts on bacterial abundance, composition, competitive-dynamics, and/or diversity. These effects are ultimately likely to affect the overall stability and function of the microbiome, and consequently, host fitness. In conclusion, it is becoming increasingly clear that phages should be considered when seeking to understand the diversity, evolution, and ecology of any microbiome.With the goal of using a ‘natural’ microbiome for subsequent studies, we sampled tomato leaves from the UC Davis Student Farm between the months of August and October. For field experiment 1, inoculum was generated from each of six different sites from across three different fields . For the subsequent experiment with sequencing data, field experiment 2, leaves were pooled across fields into a single diverse inoculum source .

The first experiment was performed less than one month after collection, and the last experiment was done six months after collection.For each experimental replicate within an experiment , 6 seeds from each field-collected tomato type were placed into sterile 1.5 mL Eppendorf tubes and submerged in 400µl of sterile 10 mM MgCl2 solution and sonicated for 15 mins in a Branson M5800 sonicator. This sonicating water bath is different from laboratory sonicators used to disrupt cells; instead, these baths will dislodge bacteria with minimal disruption of their cell integrity. The liquid was then transferred into new sterile tubes and used as seed microbiome inocula. Prior to inoculation, seeds were surface sterilized using the following procedure: Seeds were first soaked in 2.7% bleach solution for 20 minutes, then washed with sterile ddH2O three times to remove any excess bleach. The last washes were plated on KB agar plates and incubated at 28 °C for 24 hours. After sterilization, 40µl of the original seed wash was pipetted directly on top of each individual seed. We did this so that each seed would receive roughly the same number of microbes that was removed during the sonication step. The removal and re-addition process was done, in general, so that every seed used in the experiment would undergo the exact same procedure, and the only difference would be receiving microbiota or not. Negative control seeds were each inoculated with 40µl of 10 mM MgCl2.In order to culture bacteria from the seeds used in this experiment, seeds were sonicated into sterile buffer, as above. Next, cannabis storage the seed wash was diluted 1:10 in sterile 10 mM MgCl2 solution and plated onto KB agar and Lysogeny Broth agar.

They were incubated for 48 hours at 28 °C. We were only able to culture bacteria from tomato types 4 and 2. On average, we cultured 40 colony- forming units from each TT4 seed. To isolate individual strains from the microbial community, we picked morphologically distinct colonies, based on color and surface, and streaked them on new nutrient agar where they were grown for 24 hours at 28 °C. Liquid cultures were attained by inoculation into liquid KB and grown on an orbital shaker at 28 °C overnight.For consistency amongst tomato plant hosts, Money Maker seeds were used for all further experiments measuring the impact of particular seed-associated microbiota. Seeds were sterilized as described above. In addition to testing our own bacterial isolates ZM1, ZM2, and ZM3, we also included Biological Control strains, kindly provided by Dr. V. Stockwell, Oregon State. These two strains are Pantoea agglomerans strain E325A and Pantoea vagans strain C9-1. Bacterial inoculum was prepared as follows: isolates were grown overnight on an orbital shaker in LB at 28 °C. We measured the optical density at 600 nm of the overnight culture and plated the culture on LB agar, incubated overnight at 28 °C to obtain their CFU counts. The remainder of the liquid culture was stored at 4 °C overnight. The next day, we calculated a CFU to OD ratio, and re-measured the OD to account for any growth that occurred of the liquid culture overnight. We pelleted the bacteria at 4000 X G for 5 minutes, re-suspended in sterile 10 mM MgCl2 solution, and diluted to the appropriate concentration. Each seed was inoculated by pipetting the bacterial culture directly on top of each individual seed. In Figure 2-6, we inoculated seeds with pure cultures at a final inoculum density of 40 CFU/seed to approximately match the observed natural densities.

Each experimental replicate held four seedlings, and we had three experimental replicates per isolate per treatment. Disease severity was monitored for 10 days after plate flooding . For dose response curves, the density of bacteria applied to the seeds ranged from 4×10-1 to 4×106 CFU/seed with control replicates not receiving any, and disease was monitored for nine days. We did not replicate at the plate level for dose response curves.Pseudomonas syringae density was quantified from each experimental replicate using droplet digital PCR using a fluorescent probe targeting the Pseudomonas 16S gene as fully described elsewhere. Briefly, seedling homogenates were diluted 1:10, and 2µl of homogenate was used as template in the BioRad QX200 ddPCR reaction. In analyzing positive droplets, all thresholds were set using negative, no template controls and positive pure Pst DNA controls. As with analysis of AUDPC, Pst densities are a measure of each plate experimental replicate, as described above. Bacterial abundances are normalized to total seedling weight within a plate and reported as copy of 16S rRNA gene per gram of plant material. For negative ddPCR controls, we always attempted to measure Pst in the MgCl2 inoculated plant controls for all experiments as well as Pantoea DNA. Although the Pst probe was designed to be specific to Pseudomonads, we did this to ensure our probe was only amplifying Pst and not Pantoea nor any plant material. The signal amplitudes for sterile plant-only and Pantoea isolates-only controls were the same as those of no-template, sterile ddH20 controls, indicating that indeed, there was no detectable background amplification of Pantoea species when using the Pseudomonas probe, and no Pseudomonas was present in the negative controls.

Using pure DNA of individual bacterial isolates, we amplified and sequenced 16S genes using 27R and 1492R primers . In addition to 16S, we sought to further discriminate against our potential strains, and so we amplified the gyrB gene and rpoB genes with previously published primers and PCR protocols. We performed a BLASTn search of all isolate sequences and recorded the top hits with the highest identity . Phylogenetic tree of isolates and neighbors were built using gyrB sequences. We placed our isolates within a subset of samples previously mapped in a Pantoea phylogenetic tree by Rezzonico et al.. Dr. T. Smits kindly provided the E325 gyrB sequence. The evolutionary history of our isolates and other strains was inferred by using the Maximum Likelihood method based on the Tamura-Nei model. The tree with the highest log likelihood is shown. Initial trees for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the Maximum Composite Likelihood approach, and then selecting the topology with superior log likelihood value. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. The analysis involved 13 nucleotide sequences. All positions containing gaps and missing data were eliminated. There were a total of 316 positions in the final dataset. Evolutionary analyses were conducted in MEGA7.MiSeq sequencing files were demultiplexed by QB3 sequencing facility. Reads were combined into contigs using VSearch, and the remainder of the analysis was carried out in Mothur version 1.41.3 following their MiSeq SOP. Data were quality-filtered by length, ambiguous bases, and homopolymer length using the recommended Mothur parameters. Chimeras were removed using UChime. We used a 99% similarity cut-off for defining OTUs. The Silva reference database was used for sequence alignment and taxonomic assignment. Archaeal, chloroplast, mitochondrial and unknown domain DNA sequences were removed. Once an OTU table was generated in Mothur, the remainder of the analysis was performed in R using the Phyloseq package version 1.19.1 and Vegan package version 2.4-5. To account for reagent contaminants, we also sequenced DNA extraction kit controls and PCR controls along with our samples. Contaminant OTUs from control samples that were at a similar or higher relative abundance in control samples compared to experimental samples were removed from the full OTU table. Data were rarified to 50,000 reads per sample and singletons were removed.The field of microbiome science spans both basic and applied research in human health, agriculture, and environmental change. As our understanding of the ability of the microbiome to influence host health and shape host traits deepens, curing bud there is increasing interest in selecting and/or designing microbiomes for specific traits or functions. Such trait-based selection of microbiomes has the potential to shape the future of agriculture and medicine. In agriculture, below ground microbiota have already proven capable of shifting the flowering time of plant hosts, enhancing drought resistance, and even altering above-ground herbivory. However, long-term, repeatable success of future efforts will rely on a fundamental understanding of the assembly of, selection within, and co-evolution among microbiota within these communities. One of the challenges facing successful, rational microbiome manipulation and assembly is disentangling the forces naturally shaping the communities, including both host characteristics and constant microbial immigration on community stability. For example, in both humans and plants, there is contrasting evidence for the relative importance of the environment versus host genotype in shaping the microbiome, and dispersal has been shown to override host genetics in an experimental zebra fish system. One powerful but under-utilized approach to understand and experimentally control for the factors shaping microbiome composition and diversity is experimental evolution.

Measuring changes of populations or communities over time under controlled settings in response to a known selection pressure has proved a powerful force in gaining fundamental understanding of both host-pathogen evolution and microbial evolution. Here, we harness an experimental evolution approach in order to study how an entire microbial community can be selected upon in a plant host environment that varies across disease resistance-associated genotypes. We employ a microbiome passaging approach using the phyllosphere microbiome of tomato as a model system to determine if the microbial community could become adapted to the plant host environment. The phyllosphere, defined as the aerial surfaces of the plant, is a globally important microbial habitat. Microbial communities in this habitat can shape important plant traits such as protection against foliar disease and growth. Successful traitbased selection on the phyllosphere could therefore allow for enhancement of plant health, but this critically depends on the ability to select for a well-adapted microbial community that is relatively stable against invasion. We collected a diverse phyllosphere microbiome from tomatoes grown in an agricultural setting and transplanted it onto green-house grown plants using a transplantation method previously shown to be effective for lettuce [118]. We serially passaged this diverse microbiome on each of four cohorts of tomato plants of five different genotypes for a total of 30 weeks. We then measured adaptation of the community both computationally by fitting community structure to neutral models, and empirically using community coalescence experiments in which communities from different passaged lines are combined together and re-inoculated onto host plants in a common garden experiment. Overall, we were able to measure and characterize the response of the phyllosphere microbiome to selection in the plant host environment under greenhouse conditions, and select for a stable and well-adapted plant-associated microbiome.A diverse starting inoculum was collected from field grown, mature tomato plants. This field-microbiome was spray inoculated onto 30 tomato plants of 5 different genotypes, with six replicates each. Two-week old tomato plants were spray-inoculated once per week for five weeks, and then sampled in their entirety ten days after the final inoculation . The phyllosphere microbiome of each plant was then individually passaged on these genetically distinct hosts over the course of four eight-week long passages; P1, P2, P3, and P4 . Microbiomes were not pooled across plants within a given plant genotype, resulting in 30 independent selection lines. Control plants were inoculated with an equal volume of either heat killed inoculum or sterile buffer every week. At the end of each passage, bacterial density was measured and normalized to the weight of each plant , and communities were sequenced using 16S rRNA amplicon sequencing. We first measured the impact of host genotype on bacterial community structure . Using Bray-Curtis dissimilarity measures, we performed an ANOSIM test and found that plant genotype could explain 29% of dissimilarity between microbiomes in P1 . In P2, plant genotype similarly explains 28% of the variation in bacterial community dissimilarity . However, genotype becomes an insignificant driver of community composition in both P3 and P4 . The genotype effect observed in P1 was robust to removal of the primary outlying line , and that same line had too low read depth to be analyzed at P2, and thus was excluded from this analysis at the rarefaction step. By P3, this line was included, as it did not fall outside of the 95% confidence intervals for P3 clustering. We also sought to determine if there were more subtle influences of host genotype on the community that were not uncovered through analyzing Bray-Curtis distances alone.

This upstream production facility uses the same method of expression and follows the same schedule as the base case upstream facility.Transient expression in plants is a method of recombinantly producing proteins without stable integration of genes in the nuclear or chloroplast genome. The main advantages of using this method are reducing the extensive amount of time needed to develop a stable transgenic line and overcoming biosafety concerns with growing transgenic food crops in the field expressing heterologous proteins. Transient expression is attainable through several systems including biolistic delivery of naked DNA, agrobacteria, and infection with viral vectors. Notably, the use of viral vectors has been marked suitable for application on a field-scale due to the flexibility of production, and the quick accumulation of target proteins while achieving high yields. A new report has shown efficacy in delivering RNA viral particles using a 1–3 bar pressure, 1–4 mm atomizer nozzles spray devices in the presence of an abrasive to cause mechanical wounding of plant cell wall. GRAS notices GRN 738 and GRN 910 describe production of thaumatin in edible plant species and N. benthamiana, respectively. The expression of thaumatin in leaf tissue of the food crops Beta vulgaris , Spinacia oleracea , or Lactuca sativa  is generally lower than in N. benthamiana. However, despite having lower expression levels, the absence of pyridine alkaloids that are present in Nicotiana species is a major advantage for production in food crops because of the significant downstream resources needed to remove alkaloids in Nicotiana-based products. The ultimate solution may be a high-expressing engineered Nicotiana host devoid of alkaloid biosynthesis, but that option was not modeled in this study.

The transient production facility is designed to produce 50 MT of purified thaumatin in spinach, annually, air racking over 153 batches due to longer turnaround time required for S. oleracea compared to N. tabacum crops. Each batch has a duration of 67.8 days and a recipe cycle time of 1.94 days.The proposed base case upstream field production facility, displayed in Figure 1, consists of a 540 acre block of land divided into 22 plots, each of which is suitable for growing 318,000 kg FW of N. tabacum, carrying 477 kg of thaumatin, accounting for downstream recovery of 66.8%. It is assumed that the facility is located in a suitable climate where the growth of N. tabacum is attainable throughout the year, ignoring variations in production between batches . Each batch starts with direct seeding of transgenic N. tabacum plants in the field . The seeds are left to germinate for two weeks followed by vegetative growth for 3 more weeks post germination . After a total of 35 days post seeding, a tractor sprayer applies 4900 L of a 4% ethanol solution to the plot’s crop, triggering the synthesis and accumulation of thaumatin in plant biomass. The plants are incubated for 7 more days, during which time they continue to uptake nutrients and express thaumatin. After 42 days from seeding, the batch is harvested through two mechanical harvesters andfour hopper trucks at a rate of 17,000 kg/h and transported to downstream processing facility using a conveyer belt . The plot undergoes a turnaround period of three days for which the labor and equipment cost is included. No pesticides, fungicides, or herbicides costs are added due to the assumption that not enough growing degree days are accumulated during the batch cycle duration , for disease-causing organisms to be a concern.

The base case downstream processing facility is designed to purify and formulate 318.5 kg/batch of thaumatin with 98% purity. A DSP batch starts with shredding plant biomass using two industrial shredders , each processing 40,000 kg of plant biomass/h. This step is designed to homogenize the leaves and stems to facilitate the extraction process. Shredded plant material is then mixed with an acetate buffer in a 0.8 L of buffer to 1 kg of biomass ratio. This step leverages stability of thaumatin at low pH to precipitate host plant proteins that aren’t stable under acidic conditions. The extraction buffer consists of 50 mM acetic acid and 150 mM sodium chloride mixture at a pH of 4.0. The resulting plant slurry is then fed into a screw press to separate most of the dry plant material. A screw press is recommended for this step because it minimizes the amount of extraction buffer needed by forcing out more plant sap with the increasing pressure inside the chamber. The crude extract stream obtained from the screw press unit is sent to three parallel P&F filtration units for initial clarification, each having a membrane area of 190 m2 . Furthermore, the model assumes the use of food-grade filter membranes designed to include 10 filter sheets with decreasing particle retention size from 25 to 0.1 µm. The acetate buffer is applied once again as cake wash with a 0.2 L buffer to 1 L extract ratio. Diatomaceous earth is added to this step as a filter aid in a 6:100. The stability of thaumatin at low pH and high temperatures facilitates the precipitation of more host cell proteins as well as other undesired plant-derived compounds. Using seven heating tanks , the plant extract is then heated to 60 ◦C for 60 min. Following heat incubation, the stream is sent to a P&F filtration unit to capture the heat-precipitated proteins. It is assumed that a 90% reduction of N. tabacum total soluble proteins is attainable following the heat incubation and precipitation steps. Concentrating the thaumatin stream prior to the ultrafiltration/diafiltration step is necessary to avoid processing large liquid volumes ~573,000 L further downstream.

It has been reported that thaumatin experiences a loss in sweetness when heated above 70 ◦C at a pH of 7.0; therefore, the product stream undergoes concentration by evaporation prior to neutralizing the solution since the protein can sustain higher temperatures at a low pH.The triple effect evaporation unit is designed to evaporate 90% of the water content in the stream at 109 ◦C, 77 ◦C, and 40 ◦C in the first, second, and third effect, respectively, over 4 h.The exiting stream is then neutralized with 1:1 molar ratio and mixed in V-101 for 30 min and sent to the P&F filtration unit to remove any precipitated materials. An additional 1.5% loss of thaumatin during this step is assumed. Because soluble impurities such as nicotine and other pyridine alkaloids are abundant in N. tabacum plants, a UF/DF step is necessary to eliminate small molecules. The UF/DF unit consists of 4 stacked cassette holders, each containing twenty 3.5 m2 cassettes. Since thaumatin is a 22 kDa protein, a membrane with MWCO of 5 kDa is used per working process knowledge. Assuming a conservative flux of 30 L/, the inlet stream is concentrated using a concentration factor of 5, diafiltered 10 times against reverse osmosis water, then re-concentrated using a CF of 5 over 20.6 h, resulting in a 75% pure thaumatin and nicotine content of 1.08 mg/kg thaumatin. A retention coefficient of 0.9993 was assumed for thaumatin, resulting in 5.8% thaumatin loss in UF/DF . The retentate is then sent to five CEX chromatography columns operating in parallel which was modeled based on unpublished data from Nomad Bioscience GmbH . GE Healthcare Capto S resin with an assumed binding capacity of 150 g/L was used in this analysis. Table S2 shows the downstream losses breakdown per unit operation. Spray drying is used as a final formulation step over other means of industrial drying due to the heat sensitivity of thaumatin.The simulated facility consists of three sections—Virion production laboratory , curing cannabis spinach field growth, and DSP. A list of base case design parameters and assumptions is shown in Table S3. The VPL process is adopted from a recent article entailing the production of RNA viral particles from agrobacteria carrying a PVX construct. The laboratory is sized to produce 7900 L of spray solution per batch for application in the field. Nicotiana benthamiana plants are used as the host to produce the viral particles to inoculate spinach. N. benthamiana seeds are germinated in soilless plant substrate at a density of 94 plants per tray. Seedlings are grown hydroponically , under LEDs, until reaching manufacturing maturity at day 35. Agrobacterium tumefaciens is grown for 24 h, before being left in a 4 L flask overnight, and the A. tumefaciens suspension is added to MES buffer in V-101. N. benthamiana infiltration takes place in a vacuum agroinfiltration chamber for 24 h followed by incubation for 7 days in . N. benthamiana biomass production, agrobacterium growth, agroinfiltration, and incubation parameters are adapted from. After the incubation period, 41.5 kg of N. benthamiana fresh weight are ground and mixed with PBS buffer in a 5:1 buffer:biomass ratio.

The extract is then sent to a decanter centrifuge to separate plant dry matter from the liquid phase which is clarified by dead-end filtration , followed by mixing the permeate with 35.9 kg of diatomaceous earth and 7780 L of water to reach a final concentration of 1014 viral particles/L and 4.55 g diatomaceous earth/L. Diatomaceous earth is used as an abrasive to mechanically wound plant cell walls allowing the virions to enter the cytoplasm of the cell. The final spray is stored in for 13 h before field application. Field operation starts at the beginning of each batch with the direct seeding of 28.3 million Spinacia oleracea seeds over 22.6 acres. Spinach is planted over 80-inch beds with an assumed 3 ft spacing between beds, resulting in 14,520 linear bed feet per acre. Seeds are germinated and grown in the field for 44.5 days, during which time a drip irrigation system delivers irrigation water and soluble fertilizer to the soil. It is assumed that 200 acre-inches of irrigation water and 64 tons of fertilizer are needed per batch. A tractor on which multiple high-pressure spray devices are mounted is used to deliver the viral particle solution at a rate of 2 acres/h. This method of delivery has shown high effectiveness. Spinach plants are incubated in the field for 15 days post-infection. During that period, thaumatin starts to accumulate in the crop at an average expressionlevel of 1 g/kg FW after 15 days post-spraying. At day 60, two mechanical harvests collect a total of 344 MT spinach biomass, carrying 344 kg thaumatin, with the aid of four hopper trucks, which is transferred to a 500-m-long conveyor belt that extends from the field collection site to the DSP section of the facility. Harvesting occurs at an average rate of 17,000 kg FW/h, which is estimated based on a harvester speed of 5 km/h and 14,520 linear bed feet per acre. A more simplified downstream processing, enabled by the use of spinach as a host, starts with mixing plant material with 65 ◦C water before extracting the green juice through a screw press . The resulting GJ is heated for 1 h at 65 ◦C in ten jacketed tanks , then concentrated by evaporation to reduce product stream volume for further purification steps. Since thaumatin is not stable at temperatures above 70 ◦C at neutral pH, evaporation is performed at a low temperature of 40 ◦C and 0.074 bar vacuum pressure. Thermally degraded host cell proteins and impurities are eliminated in a P&F filtration unit designed to include 10 filter sheets with decreasing particle retention size from 25 to 0.1 µm. Smaller impurities are removed using a diafiltration unit with 5 kDa molecular weight cut off cassettes in a similar process as described in Section 3.3, the retentate is spray dried in to obtain a final product which has 5% water content, and 348 kg of solid material containing 94% pure thaumatin and 6% spinach impurities. These impurities are expected to be water soluble, heat stable molecules in the range of 5–100 kDa, according to the theoretical design of the filtration scheme.As shown in Figure 3a, field labor is the highest contributor to the upstream field facility followed by consumables. Detailed labor requirement and cost estimation calculations can be found in Tables S7 and S8. Consumables include mechanical harvester and tractor’s fuel, lubrication, and repair costs and other field equipment repair costs. Upstream indoor facility AOC breakdown elucidates a high cost of consumables due to the cost of soilless plant substrate, followed by high energy consumption from the LED lighting system used for plant growth.

Bio-fuel production is driven by mandates for renewable transport fuels, weak land use regulation, production subsidies, and speculation by energy and commodity companies in both developing and industrial countries . Although global estimates of the scale of industrial bio-fuel production are difficult to make, the World Bank calculates that 36 million ha were dedicated to bio-fuel production globally in 2008, doubling the 2004 level. Oil palm production in Indonesia and Malaysia indicates the emerging trajectory: aided by government policies and subsidies, oil palm plantations grew in Indonesia from 3.6 million ha in 1961 to 8.1 million ha by 2009 . The consequences of the expansion of oil palm include ongoing displacement of smallholders, increasing monoculture, and abandonment of food cropping, though the extent to which these effects are occurring remains uncertain . Across the Global South, oil palm and sugarcane plantations may provide only a tenth of the jobs when compared to the livelihoods generated through smallholder farming .Despite expansion of large-scale commercial agriculture, smallholders still make up 85% of circa 525 million farms worldwide . Such farmers span a spectrum from traditional, pipp drying racks indigenous growers using no external inputs to those with heavy dependency on modern seed varieties, fertilizers, and pesticides, but up to 50% of smallholders are thought to utilize resource conserving farming methods .

While they represent the bulk of the agricultural population, estimated at circa 2.6 billion people , due to land inequalities they often do not control the bulk of the arable land . These disparities are largest in South America, and least pronounced in Africa . Another sign of intensifying inequalities is that mean farm size has decreased in many parts of Africa and Asia , increasing the vulnerability of small farmers and exacerbating the poverty in these regions, while large landholdings are increasingly controlled by a small number of people . Despite poverty, the current contribution of small farms to global food production is significant. Herrero et al. estimate that mixed crop and livestock systems supply 50% of the worlds’ cereal, 60% of the world’s meat and 75% of the world’s dairy production. Much of this production is locally produced and consumed, and provides the main source of food for the world’s 1 billion poor . Altieri considers that traditional indigenous agriculture supplies 30 – 50% of the world’s food. suggests that the contribution of smallholders to food production is increasing in some countries because of changing national socioeconomic and political situations and government policies favoring domestic food self-sufficiency . As indicated previously, not all smallholder agriculture would be considered DFS. Perhaps 50% of smallholder farmers use agro-industrial inputs or have not adopted agroecological methods .

Qualitative research suggests that through implementation of “sustainable intensification”, a set of resource conserving practices also used in DFS , such farms could become 60-100% more productive, potentially contributing far more to local and global food security , although rigorous, quantitative comparisons are both lacking and needed . Overall, small-scale diversified farmers face continuous, intensifying pressures from the encroachment of industrial supply chains . However, in parts of the developing world, diversified farming systems are actually expanding, in response to food sovereignty movements, smallholder desires for healthier and more economically independent lives, and some level of civil society and government support. Agroecological techniques are site specific and tend to be transferred from location to location through horizontal communication and social networks, with much adaptation by local communities . Evidence of the rising adoption of agroecological principles in many Latin and Central American countries exist through the many cases of campesino-tocampesino training reported, as well as the increasingly global spread of the La Via Campesina movement . Cuba is a case where the transition to agroecological practices has been particularly rapid ; in this case the expansion was a response to a severe food security crisis and lack of fossil fuel inputs following collapse of the former USSR and associated subsidies to industrialized agriculture . To some degree, DFS are also expanding in industrial countries despite the vastly more inhospitable political and economic conditions that may prevail, particularly in the U.S.

There, as in Australia and many European countries, there is growing demand for organic and locally produced fruits, vegetables, fish, and meat, which is spawning an increase in the number of small-scale, highly diverse farms, often supplying urban markets . In the U.S., certified organic agriculture has grown markedly, rising from less than 1 million acres in 1990 to 4.8 million acres in 2008 and comprises 0.7% of agricultural production with 20,000 producers . Worldwide, organic agriculture has tripled from 11 million ha in 1999 to 37.2 million ha in 160 countries as of 2009 and currently makes up 0.9% of agricultural production , with 1.8 million producers in 2009, predominantly from Asia and Africa. Nonetheless, while organic agriculture tends to support greater biodiversity than conventional farms , not all organic farms are DFS . Much organic agriculture has become increasingly large-scale and homogeneous as producers and food companies strive to maximize profits and meet growing market demand .The series begins by examining what is known about how DFS maintain a range of ecosystem services that provide critical inputs to farming, including soil quality, water use efficiency, control of weeds, diseases and pests, pollination services, carbon sequestration, energy efficiency/greenhouse warming potential, resistance and resilience to climate change, food production, and biodiversity. By comparing DFS to conventional industrial systems, Kremen and Miles find that DFS significantly enhance all the ecosystem services measured with the exception of crop production, although not necessarily to the level required to control pests and diseases or provide sufficient pollination. The authors note that relatively few research dollars have yet been applied to the improvement of DFS compared to conventional systems; redressing this substantial inequality in public and private investment is necessary to close yield gaps while maintaining environmental benefits. The authors recommend that new research should be holistic and integrated across many components of the farming system to identify management systems that can take advantage of potential synergies. Next, Bacon et al. seek to simultaneously deepen our understanding of the social consequences of DFS vs. industrial production and to unpack several key influences affecting continuity, change, and possibilities for transformation of these systems. Case studies from California’s Central Valley, Mesoamerican coffee agroforestry systems, and agricultural parks in the European Union, identify the critical role of government policy in an agricultural system’s emergence and the combination of market demand and multiactor governance that provide continuity. They find that the spread of DFS will generate social benefits, including decreased pesticide exposures, improved food security, longer agricultural working seasons, and healthier diets, but may also generate new costs, such as increased muscular skeletal injuries associated with higher manual labor demands. Social movements can alter governance arrangements and influence both the spread of DFS and the creation of policies that increase environmental benefits and reduce social costs. However, broader changes to the market and political structures and economic policies of agriculture are needed to enable a socially sustainable expansion of DFS. Iles and Marsh consider several examples of obstacles to the adoption and spread of DFS in industrialized agricultural systems. These include the broader political economic context of industrialized agriculture, the erosion of farmer knowledge, and supply chain and marketing conditions that limit farmers’ ability to adopt diversified practices.

To overcome these obstacles and nurture DFS, policy makers can transform agricultural research, develop peer-to-peer learning processes, support recruitment of new farmers, invest in improved agricultural conservation programs, compensate for provision of ecosystem services in working landscapes, pipp horticulture and develop direct links to consumers and institutional markets. In contrast to analyzing a market-led expansion of DFS, Rosset and Martinez-Torrez propose a theoretical framework focused on disputed rural territories and repeasantization to understand how and why rural social movements have increasingly adopted agroecology and diversified farming systems as part of their discourse and practice. Rural spaces are increasingly disputed as agribusiness seeks to “grab land”, control production systems, and remove many rural inhabitants from the land, while small-scale farmers, rural workers, indigenous communities and women are increasingly organized into social movements, such as Via Campesina, that seek to repopulate or maintain these landscapes through the defense of their food, seed, and land sovereignty. For peasants, family farmers and their social movements, agroecology helps both to build autonomy from unfavorable markets and to restore degraded soils. The social process of sharing these practices and values from farmer to farmer , coupled with broader global social movements, help bring alternatives such as DFS to scale. We finish the series with an in-depth analyses of specific farming or social systems. Sayre et al. examine how ranching is the most ecologically sustainable segment of the U.S. meat industry and exemplifies many of the defining characteristics of DFS. Rangelands also provide other ecosystem services, including watershed functioning, wildlife habitat, recreation, and tourism. Innovations in marketing, incentives and easement programs that augment ranch income, creative land tenure arrangements, and collaborations among ranchers can support greater diversification. Taking advantage of rancher knowledge and stewardship can support the sustainability of ranching and its associated public benefits. We have attempted to launch the concept of DFS by encouraging broad based interdisciplinary collaboration and practice from the outset, through combining our analysis of the ecology of food production with complementary questions of food access, distribution, and structure of the agri-food systems. This special feature thus incorporates insights from ecology, economics, political economy, and related social science fields to create a more inclusive analysis of the challenges and opportunities that influence efforts to achieve food security and the multiple dimensions of sustainable agriculture.In 1519, at the time of the arrival of the Spanish invaders to the Basin of Mexico, the people in the region ran a sophisticated system of agriculture that was able to feed its large human population, estimated by different studies between 1 and 3 million . Successful farming in central and western Mesoamerica depended critically on the ability to keep an accurate calendar to predict the seasons. Apart from the wet tropics of the coastal plains of the Gulf of Mexico and the Caribbean, all other regions of Mesoamerica, namely the Mexican Altiplano, the Balsas Basin, and the seasonally dry ecosystems of the Pacific slopes of Mexico, share a highly cyclical precipitation pattern with a dry spring followed by a monsoon-type rainy season in summer and early fall. Precipitation-wise, the most unpredictable time of the year is mid-, and in some parts late, spring; the “arid fore-summer” that precedes the arrival of the Mexican monsoon . Planting too early, following the cue of a first haphazard early rain, can be disastrous if the true rainy season does not continue. Waiting to plant late, after the monsoon season has clearly started, can expose the corn field, or milpa, to an overly short growing season and will also put the crop under competition from weeds that have already germinated. Wild plants in these highly seasonal ecosystems often have traits that allow them to hedge the risk of a false moisture cue. Annual plants often have heteromorphic seeds, some of which germinate with a single rain pulse while others remain dormant and germinate after successive rainfall events . Other plants have lignified seed-retention structures that release seeds gradually into the environment as the dry spring progresses . Finally, woody perennials often flower in the dry early spring in response to photoperiod, independently of moisture availability, and shed their seeds in late spring or early summer when the monsoon season is starting . In this latter group, the physiological ability to detect the season independently of precipitation cues is critically important to avoid premature germination. Accurate timekeeping must have also been strategically critical for pre-Hispanic farmers, who, in order to be successful, had to prepare the milpa fields before the onset of the monsoon rains and plant as early as possible while, at the same time, being able to disregard false early rainfall signals. In the 16th century, Diego Durán noted the importance that the native calendar had for these communities “to know the days in which they had to sow and harvest, till, and cultivate the corn field.” He also noted as “a very remarkable fact” that the Mexica farmers followed strictly the calendrically based instruction of the elders to plant and harvest their fields and would not start their farming activities without their approval.