A mid-range projection forecasts up to 59 million residents statewide by 2050, with massive conversion of agricultural to urban land in the Central Valley, and cities such as Fresno doubling in population . Urbanization in California tends to consume lands with high quality soils and relatively abundant water supply due to their proximity to existing towns and cities in the valleys . Given such prospects of population growth, the purpose of this task was to develop future urbanization scenarios for Yolo County, and assess implications for agriculture, greenhouse gas emissions, and other issues related to land use change. Urbanization presents both opportunities and challenges for agriculture. In some regions, it does generate markets for agricultural products, such that farm production increases locally . But urbanization is more typically accompanied by challenges: the loss of agricultural land due to subdivision and development; vandalism at the urban edge ; and conflicts with new suburban residents about noise, odor, and potential spray drift associated with farming operations. If development takes place in a dispersed pattern that fragments agricultural land, farming may become difficult on some remaining agricultural parcels due to difficulties in moving farm machinery from field to field. Also, fragmentation and loss of farmland causes farmers to lose benefits associated with being part of a large farming community, such as sourcing inputs, accessing information, sharing equipment, and supporting processing and shipping operations . Impacts on agriculture from urbanization will then be disproportionate to the land area covered. Suburban or exurban development increases GHG emissions per land area substantially when compared with agricultural land uses .

It is useful to know the extent of these increases,blueberry packing boxes especially since California counties will need to demonstrate ongoing commitment towards reducing GHG emissions in response to state mandates, such as the Climate Action Plan that was adopted in 2011 for the unincorporated areas in Yolo County . In addition, land use planning for climate change can potentially set the stage for greater provision of other ecosystem services at the rural‐urban interface, such as regulation of environmental resources, biodiversity conservation, livelihood options, and business opportunities that build social capital . The A2 and B1 scenarios of the International Panel on Climate Change are based on story lines for higher and lower GHG emissions, respectively, which can be conceptually down scaled at local scales to explore how future local land use patterns will respond to climate change . A2 has higher economic and population growth, and less emphasis on environmental, social, and sustainability priorities than B1. The downs caled story lines can form the basis for spatial modeling of land use change and the challenges that would occur at the rural‐urban interface. In California, UPlan is a simple rule‐ based urban growth model used for regional or county level modeling . The spatial configuration of each land use type is based on demographics, land use designations of the General Plan , and on a set of attractors and detractors for land use change that can be informed by the story lines of climate change scenarios.The majority of California’s new residents will settle in urban areas in coastal counties and in the Central Valley. The Sacramento metropolitan region, where Yolo County is located, will house a significant portion of this growth. Projections prepared for the Sacramento Area Council of Governments Blueprint project in 2005 estimated a population increase from 1,948,700 persons in 2000 to 3,952,098 persons in 2050, i.e., >100 percent increase . The conversion of the region’s undeveloped land into urban, suburban, and exurban development often occurs at the expense of agriculturally productive land. 

Yolo County includes 653,452 acres according to the 2008 California Department of Conservation Farmland Mapping and Monitoring Program . Agricultural land occupied 538,043 acres in 2008. About 87 percent of the acreage was in agricultural use . Land use was classified as 4.6 percent urban in the incorporated cities of Davis, West Sacramento, Woodland, and Winters. Important farmland was 57 percent, and livestock grazing land was 24 percent of the county’s acreage. In 1998, Yolo County alone contained about 43 percent of the prime farmland that existed within the Sacramento region , and it yielded the highest farm market values out of all the counties . Thus Yolo County is an important reservoir of productive farmland within the Sacramento metropolitan region. A net loss of about 30,000 acres of agricultural land occurred between 1992 and 2008, and this includes a net gain of about 16,000 acres of grazing land . New grazing lands were formed by draining parts of the Yolo Bypass along the Sacramento River and by transitioning dry‐farmed grain fields to grassland, such as near the Dunnigan Hills. Overall, only 1 percent of Yolo County’s total prime farmland was lost up until 2000 . Between 1998 and 2008, the rate of agricultural conversion to wetlands, especially along the Sacramento River for wildlife conservation, has increased to approximately 2,000 acres yr‐1 . Urbanization accounted for the loss of about 6,500 acres of agricultural land between 1992 and 2008 , i.e., approximately 406 acres yr‐1 . Most of this was prime farmland and farmland of local importance. Yolo County has been relatively successful at protecting agricultural land from urban conversion through land preservation programs, incentives for farmers, and land use policies that make it difficult to develop land zoned for agriculture. Yolo County’s population grew an average of 2.2 percent per year from 1985 to 2007, from 120,300 to 197,530 residents . But by 2050 the county’s population may reach 320,000 to 394,000 , depending on assumptions used in scenarios for either regional or statewide planning. This would result in an increased urban population and pressures to expand the current urban footprint.

Given the county’s geography, urban expansion will almost certainly occur at the expense of farmland and open space if growth is not restricted to infill development within existing boundaries. With respect to California’s climate change policies aimed at reducing GHG emissions 16 and Senate Bill 375 17 which connects land use planning with implementation of AB 32, urbanization onto agricultural land raises two important issues for the 2050 time frame: magnitude of the loss of agriculturally productive land that provides ecosystem services such as meeting the food needs of an expanding state and global population, wildlife habitat, and open space for residents; and an increase in GHG emissions from decentralized urbanization when compared with more compact, centralized forms of urban development that leave agricultural lands undeveloped. There is a need to better understand the relationships between urbanization, agriculture, and climate change, and their interrelated effects on ecosystem services. In order to understand the type, extent, and likely locations of urbanization in the county, we used UPlan GIS‐based software, a rule‐based, land use allocation model developed by the Information Center for the Environment at the University of California, Davis . UPlan is an open‐source, relatively simple model that can be run on a sub‐county area, a county, or a group of counties. It is a suitable model for broad‐brush urbanization modeling of large land areas using multiple development scenarios, and has been used by more than 20 counties in California,package of blueberries including a group of rural Blueprint counties in the San Joaquin Valley . In the past it has been employed to assess the impacts of urbanization policies and growth on natural resources , to understand the risk of wildfires in rural woodlands from urban growth , and to evaluate the effect of land use policies on natural land conversion . Households are divided into four residential land‐use types based on density parameters, while employees are assigned to nonresidential land use types , also by density. New development is divided by land use type and allocated across the landscape based on the geographic cells with the highest combined attraction weights and the user‐defined land use order. The model uses a cell size of 50 meters, roughly about half an acre. The final output is a map displaying the location, by land use type, of future urbanization.For the purposes of this project, we modified UPlan in several ways when compared to previous usages. Since our time frame is longer than in many previous applications, we no longer required that the model place growth in areas conforming to the current county General Plan. Land use politics and regulation can change greatly over 40 years, which is equal to at least two General Plan cycles in most California counties. Furthermore, the County Board of Supervisors by majority vote can approve zoning variances four times a year, allowing development that does not conform to a current General Plan and zoning code. Thus, the planning documents and zoning codes that are a short‐term deterrent to development may no longer be relevant in the longer term. The purpose of this project was also to model three significantly different scenarios, and restricting development to the current policy framework would make this difficult.

For these reasons we did not include the countywide General Plan land use designations. We also modified UPlan to allow development within existing urban areas, on the assumption that a significant amount of urban redevelopment is likely within the 2010–2050 time frame. Lack of an infill development option was a significant drawback with previous versions of UPlan. Sharply increased levels of infill are likely within more environmentally oriented future scenarios. Indeed, our AB32+ scenario assumes that 100 percent of development takes place within existing urban areas. This approach is likely to rapidly decrease GHG emissions because lifestyles of urban area dwellers tend to have smaller carbon footprints, such as less energy expenditure for transportation , as long as their economic actions do not increase to the point of significantly outweighing that benefit . Urban development is already increasingly taking the form of infill within the state’s largest urban areas, including Los Angeles and the San Francisco Bay Area, and infill development is a leading goal of the Sacramento region’s 2004 Blueprint vision for the future . Lastly, we established density categories that are relatively high by historical California standards, but fairly close to the density levels of recent development in the more urban portions of the state. Our categories were “Very Low Density Residential,” with an average lot size of one acre; “Low Density Residential,” with an average density of 8 units per acre ; “Medium Density Residential”, with an average density of 20 units per acre; and “High Density Residential,” with an average of 50 units per acre. The latter two categories are similar to densities currently being achieved within many of California’s more urban communities . Within each scenario, we also apportioned development differently between these types. The A2 scenario focuses primarily on Low Density Residential development, while B1 is relatively evenly split between High, Medium, and Low Density types, and AB32+ favors High and Medium densities. In terms of building types, the Medium Density category might consist of two‐ to three‐story apartment or condominium buildings with significant green space around them, while the High Density category might include three‐ to five‐story buildings in a more urban format. It is important to emphasize that none of these categories require high‐rise apartment living, although this development type is not forbidden, and might in fact be desirable for limited locations within the county during the study period.Several types of urbanization attractors are typically used in UPlan, including blocks with growth in the previous census period , freeway ramps, arterial streets, collector streets , and urban spheres of influence. To predict infill development more accurately, we added additional attractors such as existing commercial strips, shopping centers, freeway retail zones , existing neighborhood centers, and rail transit stations. In most cases adding these factors meant creating new GIS data layers with information from publicly available sources or visual analysis of Google aerial imagery. We took into account existing land uses within cities by creating a data layer of current zoning districts, and consolidating these districts into high and low infill potential layers. The first of these includes existing commercial and industrial land, which typically consists of relatively large parcels of land being used for relatively short‐lived purposes , owned by landowners who are likely to be open to profitable redevelopment over a 40‐year time frame.