All measures except open-ended surveys were repeated in 2015. Subsequently, in proceeding research years six additional sites were added for a total twenty-nine farms that participated in this research . Due to variability in support, turnover of garden managers, and a variety of other factors, some farms only participated in portions of the research . During the duration of the urban agroecology survey, six of our community partner sites were abandoned and/or developed. To determine on-farm composition of research sites, we physically measured urban farm size, area of production, non-crop areas, and areas used for infrastructure. Total farm size was measured using Google Earth Pro and ground-proofed during site visits. Farm production space was measured by hand and included all space in the gardens used for producing crops . Not all production occurred in-ground, therefore the overall estimate of area used for production included both raised garden beds and in-ground production. Non-crop areas are defined as managed areas not primarily used for food production and were often set aside as pollinator or natural enemy habitat. These spaces included a variety of perennials and annuals, flowers, and other non-crop features. Infrastructure was defined as area not being used for production, or conserved for non-crop habitat and cangenerally be considered areas utilized for other on-farm uses. These measurements were translated into proportions of total farm area for analysis . Information on common land management and farming practices such as crop rotations, cover cropping,dutch buckets system use of mulch, intercropping, on-site composting, soil management practices, pest control strategies, weed management were collected during the open-ended survey .

Confirmation of practices implemented on the farm were ground truthed over several visits to the farm. In some cases, community farms were managed individually rather than collectively. In most cases we observed common practices among plots and generalized these as commonly used on the site, however, not all participants can be expected to use uniform management practices, and not all practices are visually observable, especially in the context of soil amendments and pest management practices . For analysis, practices including crop rotations, cover-crops, intercropping, mulching, application of soil amendments including compost, manure, and fish emulsion, composting on-site, and no-till practices were aggregated to create an overall management-intensity index.For analysis we questioned how on-farm land use categories and overall size influenced the proportion of production, natural habitat, and infrastructure. We used classification and regression trees to look at overall predictors of the three land use categories. CART analysis indicated that the proportion of production was best predicted by overall proportion of on-farm infrastructure. Eighty-five percent of farms had over 40% of their overall area committed to farm infrastructure, and 58% of those farms had infrastructure in excess of 62% of overall farm size. Infrastructure was by far the largest on-farm land use category, accounting for an average of in all farms measured . When accounting for other non-production land use, an average of 68% of on-farm area was not utilized for food production. Overall size of farms was a poor predictor of any other land use type.Survey results and ground-proofing indicate that agroecological management practices have been widely adopted throughout East Bay urban farms and gardens. Almost all farms assessed incorporated inter-cropping as well as cover cropping and applied compost. Indexed management practices when compared with mean estimated productivity, weed density, and overall crop biodiversity did not have significant effect. Developing a better understanding of the agroecological elements of urban farms will be an important topic in an increasingly urbanized world. Previous analysis suggests that worldwide urban food production can significantly impact global food requirements .

However, as urban populations grow, urban land becomes increasingly valuable, and the “highest and best use” of vacant urban land may limit the implementation of UA. Production capability, impacts on local food security, and the overall economic efficacy of UA will be crucial in promoting and prioritizing it in future and current urban food systems . Further, developing a better understanding of the multi-functionality of UA, including the social, economic, and ecological benefits these systems provide, can better help policymakers and urban planners bolster UA, acknowledging its utility and benefit in the built environment. Understanding spatial composition trends, management practices, and production potential are important and understudied topics that contribute to our understanding of urban farms form and function. This research provides data and context that may influence future discussions regarding the viability and efficacy of UA. A deeper understanding of UA production capabilities, especially in the context of on farm land use, is an important topic when questioning the efficacy of urban food production on high-value urban land. Urban land cycles are largely dependent on rent-seeking and attempt to exploit rent-gaps for profit by landowners and developers . Developing underused urban land is often very profitable, counter to UA operations. High land values consistently challenge urban agriculture systems in the context of “highest and best” use – the concept that land-use should always create the most profit. Urban farms are consistently put in a dilemma; they must justify their existence in the context of production. However, if they are not generating substantial profits, their implementation on high-value urban land will always be questioned. Urban farms also suffer from a fundamental misalignment with “highest and best use” objectives; previously published survey data indicate that urban farm goals are often focused on social goods and food security . Generating profits is often a tertiary goal at best. Despite this misalignment and lack of financial support, estimated yields per unit area are high, with approximately 7.14kg/square meter of fresh vegetables being grown. Urban farms also significantly impact local food security, with ~69% of on-farm production going to the local community.

Our findings indicate that increasing overall production capacity in UA, an important consideration in urban land use, can be linked to on-farm land use. Despite high yields per unit of area, our on-farm spatial analysis found that an average of only ~32% of available area is being utilized for production. With land being such a limiting factor of UA adoption, we found that UA sites may not be maximizing potential production area. These findings indicate that overall urban farm size is not a limiting factor to increased production. Two possible explanations may influence underutilization of production area. Firstly, these farms often exist on volunteer labor and often lack consistent funding to pay farm managers and employees . Investment, infrastructure, and labor may be limiting full production potential. Moreover, these spaces are serving residents more than markets. If local food needs are met there may be less incentive to put additional land into production. Conversely, spatial composition, especially in the context of production area,grow hydroponic may be influenced by management practices. The three farms with the highest proportion of production area all utilized in-ground management practices. Contrariwise, UA sites with the most significant proportion of infrastructure all utilized raised-beds in their production systems. Raised-bed production is often linked to concerns about soil health or security of tenure. Raised-bed production can help mitigate potential soil contamination issues, often found in UA. This production practice is also modular and can be broken down and moved in cases of insecure tenure. In summary, UA production is not limited by yields per unit of area but more explicitly linked to social-ecological factors that prohibit the full implementation of long-term, in-ground production systems. We found that implementation of sustainable farming practices is widespread among urban farmers and practiced across measured sites. Intercropping, cover-cropping, and soil building practices are common and often practiced simultaneously . Adoption of sustainable farming practices may be in response to abiotic and ecological challenges faced by converting impacted urban land into productive farms. Crop rotations, cover-cropping, mulching, and manure and compost application were often cited during interviews with farm managers as strategies to remediate impacted urban soils. Management practices were also frequently cited as strategies used in response to pest and weed pressures. Weeds were prevalent in all measured sites, but broad leaf weeds were most pervasive and were especially abundant for in-ground production systems as opposed to raised beds. Average weed coverage in quadrats was reduced by the implementation of intercropping. These results have important analogs to findings in rural agricultural systems and show that these practices can be implemented at small scales in novel urban agroecosystems. Urban agriculture is defined as agricultural production within urban areas managed by urban residents including home gardens, market farms, orchards, and often, animal rearing . The popularity of UA has expanded in cities around the world . The American Gardening Association reported a 34% increase in new urban farms between 2007–2011 and identified over 8500 operating urban farms and gardens in 38 US cities . The realized and potential benefits of UA are far-reaching; recent estimates claim UA could annually contribute $80–160 billion in food production, nitrogen fixation, energy savings, pollination, climate regulation, soil formation, and biological control of pests . There are innumerable variations of UA worldwide, with various on-farm compositions, each situated in their own agronomic and geopolitical context. This review does not attempt to be inclusive of all variations of UA, but to focus on the ecological management of crop pests and assess the current state of research of biological control in urban agriculture—an ecosystem service with an estimated value of $1.12 billion .Pests in UA are ubiquitous, and characteristics of urban areas can make pests particularly damaging and difficult to control.

Herbivorous insect populations have been reported to decrease in diversity but increase in abundance in urban areas , and pest outbreaks are linked to factors endemic to urbanization—habitat fragmentation and disturbance . Other unique features of urban areas such as vegetation maintained year-round, nutrient-stressed perennials, and higher temperatures from the urban heat island effect can also increase pest density and/or the severity of pest damage . Despite these challenges, many urban farmers choose not to use pesticides for public and environmental health reasons , instead using ecological pest management practices . Moreover, because of re-entry and pre-harvest intervals, many of the more effective pesticides cannot be used on typical urban farms where multiple plant species are adjacent, and the farm is visited daily by UA practitioners. In contrast, conservation biological control uses practices that are commensurate with many UA practices and limitations by employing habitat manipulation to provision resources that can support “natural enemy” arthropods to improve pest suppression . Diverse management practices such as crop rotations, intercropping, increased plant species richness, and incorporation on non-crop habitats contribute to high spatial and temporal diversity in UA systems , but information about how these manipulations affect ecosystem function, especially CBC, is inadequate in comparison to research in rural farms. For example, numerous studies have reported that habitat manipulation and diversification of the surrounding landscape and on-farm biodiversity have been effective at increasing beneficial insect richness, abundance, and biological control on more rural farm scapes . At the local-scale incorporation of non-crop perennials, floral resources, and crop rotations within farms , and at the landscape-scale, greater proportions of natural vegetation, non-crop land, and landscape heterogeneity surrounding rural farms have proven to promote the biological control of pests . Adding floral resource additions, crop rotations, and ground cover management practices including mulching, and soil amendments such as compost additions can provide benefits such as alternative food sources and habitats necessary for maintaining consistently high natural enemy populations and increased rates of biological control over time and space . Further investigating these practices in UA can help provide ecologically based, cost-effective interventions to reduce crop damage from insect and mite pests, thereby increasing local food security. UA practitioners often adopt agroecological practices that include local habitat diversification, but there are few studies that document whether the impacts of diversification on small urban farms are similar to more rural, larger agricultural systems that are not subject to affects unique to UA systems, including urban microclimates, reduced species diversity, and landscape-scale characteristics. A growing field of study in urban biological control has sought to fill this research gap. Some limited work has shown how local- to landscape-scale effects often vary by taxa , and by the type of crop damage, ranging from chewing herbivory to fungal and bacterial disease.