The existing conservation planning program Zonation , with its predefined benefit functions, could be particularly useful for utility maximization problems. Further similarities between the utility maximization approach and Zonation include the ability to incorporate several important data inputs including benefit, threat, and cost data; an iterative heuristic similar to our greedy heuristic; and a built in routine to dynamically update all sites’ marginal values after conservation action. Continued research into the application of these methods and different software solutions is warranted due to the increasing opportunity for use in applied conservation and natural resource management problems in ecosystem services, restoration planning, or multifunctional agriculture conservation prioritization, in addition to biodiversity conservation. California has been the top agricultural producer in the United States for more than 60 years; production in 2008 was 11.2 percent of the total U.S. value of agricultural crops and commodities . California supplies nearly half of the nation’s fruits and vegetables. The value of gross agricultural cash receipts was $36.2 billion in 2008, of which exports were 16 percent. Thus, agricultural vulnerabilities and adaptation to climate change in California are important to millions of people, many of whom know little or nothing about the state, its resources, or its agricultural sector. To build public support for understanding agricultural adaptation to climate change and the need to mitigate greenhouse gas emissions,led grow lights it will be necessary to develop greater awareness of a broad set of biophysical and socioeconomic factors that influence agricultural sustainability .

Previous studies on impacts and adaptation to climate change in California have mainly focused on responses to abiotic factors such as water supply and increases in temperature . Social issues such as labor, markets, and policy for land use change need more attention in the context of climate change . To address these issues, this project took several approaches to studying agricultural vulnerability and adaptation to climate change in California. We explored a wider conceptual framework for climate change responses than has been addressed for California agriculture in the past. Each approach is a stand‐alone study that utilizes different types of methods, develops different types of adaptive capacity, and is relevant to different stakeholder groups. Rather than provide an integrated analysis, the main outcome shows the versatility of new formats for developing and synthesizing interdisciplinary information at multiple scales that can be useful in design of strategies for adaptation to climate change.Global environmental changes tend to have a disproportionate impact on agriculture compared to other parts of the economy. Since agriculture relies directly on natural resources, those who work in agriculture are inherently vulnerable to changes in climate, water availability, and land use . Volatility in agricultural markets and the cost of energy, fertilizers, and other inputs are also major sources of concern among farmers . Such changes can have a multitude of biophysical and social consequences that are often difficult to predict. While some farmers will anticipate changes and reap benefits, others will face increasing vulnerability unless efforts are made to strengthen their adaptive capacity and enhance the resilience of agricultural ecosystems . Vulnerability, defined here as “the potential for loss,” is often assessed by examining biophysical and social indicators that reflect aspects of exposure, sensitivity and adaptive capacity, which vary over time and space . Given the orientation of vulnerability towards negative outcomes, it is also necessary to understand the factors that ensure resilience within social‐ecological systems .

Since vulnerability and resilience vary spatially, a number of recent studies have developed methods for mapping dimensions of vulnerability using geographic information systems. The Social Vulnerability Index is one approach that has been used to link social indicators with biophysical data and explore vulnerability to environmental hazards . The body of work which uses the SOVI has compared changes in social vulnerability among U.S. counties over the last 40 years, and integrated social vulnerability with exposure to flood risks in the Sacramento‐San Joaquin Delta . From a theoretical perspective, the SOVI has helped establish the “hazards of place” concept and provides a model for indentifying vulnerable regions and communities that merit closer examination through contextualized and place‐based approaches . In the context of climate change vulnerability, O’Brien et al. use an indexing approach to highlight the “double exposure” of agricultural populations to the impacts of climate change and economic globalization. In their work, socioeconomic indices are superimposed on top of mapped climate data to illustrate spatial differences in vulnerability. They then use case studies, surveys and interviews to help interpret impacts on agricultural livelihoods in vulnerable locations. These studies and others, illustrate the need to balance large‐scale spatial analysis of climate change and socioeconomic impacts with localized, often community‐level, assessments of vulnerability and adaptive capacity .Here we develop an Agricultural Vulnerability Index for California that aims to integrate a broad set of biophysical and social indicators that are relevant to state and local efforts to adapt to changes in climate, land use and economic forces. Given its geographic heterogeneity and diverse agricultural economy, California offers a prime opportunity to examine spatial differences in agricultural vulnerability, as well as the responses that will be needed to adapt successfully. A second objective of this study is to identify regions of concern, which may require a more careful assessment of local impacts and adaptive responses by stakeholders in the agricultural community.

In essence, the California AVI is meant to be a starting point for “place‐based” adaptation planning throughout California, perhaps patterned on an early example from Yolo County summarized in Section 3 of this paper . A principal component analysis was conducted on the variables in each of the four sub‐ indices . This was done to examine the covariance structure of the variables in each sub‐index and facilitate subsequent compilation of the overall agricultural vulnerability index. Each variable was standardized to have a mean of zero and a standard deviation of one. Principal components with eigenvalues greater than one were retained, since these satisfied the Kaiser criterion . Each variable with a rotated loading greater than ± 0.5 was assigned to the principal component where it had the highest loading value. Communalities were used to estimate the proportion of variance for each variable explained by the retained components. The variable loadings were examined to ensure that the direction of the components were all consistent, specifically that positive loadings for a variable were indicative of high vulnerability. If the directionality of the component was contrary to this logic, the rotated component scores were multiplied by negative one to reverse the direction but retain the covariance structure . If the variables that loaded on a component axis were ambiguous in relation to vulnerability, the component was not used in the sub‐index calculation . The rotated component scores for each grid cell were then summed to determine the sub‐index value. Finally, the four sub‐index values for each grid cell were added together to generate a value for the overall agricultural vulnerability index. Data for the sub‐index and overall index values are mapped according to seven vulnerability levels based on the standard deviation around the mean.In the climate vulnerability sub‐index, eight initial variables were reduced to two retained components that explained 85.2 percent of the variance among grid cells . Annual precipitation had a high negative loading on PC1, while potential evapotranspiration, climate vulnerability precipitation, days in July above 35oC , and days above 30oC all had high positive loadings. This component effectively characterizes statewide patterns in precipitation and summer temperature. In contrast,nft hydroponic the variables in PC2 reflect patterns in winter temperature with high positive values for both lowest minimum temperature and days in the growing season and high negative values for chill hours. The inverse relationship between chill hours and the other two variables in PC2, while intuitive, made it impossible to assign an unambiguous direction to the component in relation to vulnerability. For example, while warmer winter temperatures may result in inadequate chill hours for many orchard and vineyard crops, they can also reduce the incidence of freezing temperatures and expand the growing season for other crops . Due to the ambiguity of PC2 in relation to vulnerability, only PC1 was used in the sub‐index. The fact that PC1 accounted for 69.3 percent of the cumulative variance, confirmed that dropping PC2 would not reduce the amount of variance explained to levels below what is captured by the other sub‐ indices, which ranged between 84.0 and 67.0 percent . The spatial distribution of the sub‐index values indicates moderately high and high climate vulnerability throughout the southeastern part of the state . The small total amount and high variability of precipitation combined with high summer temperatures and high potential evapotranspiration present more severe challenges to agriculture in southern California than in other parts of the state. 

In particular, parts of San Bernardino, Kern, and Inyo counties tended to have the highest levels of climate vulnerability, though it should be noted that few crops are currently grown in the most vulnerable areas. As such, the primary agricultural regions of Kings, Kern, Riverside, and Imperial counties that have moderately high climate vulnerability merit closer consideration. While a consensus has yet to be reached on how precipitation will change over the next century throughout California, there is broad agreement that temperature and potential evapotranspiration will generally rise . Such changes will have a profound effect on regional hydrology, in many cases reducing the availability of surface and ground water while increasing irrigation demand . Strategies to safeguard supplies and minimize irrigation demand include expanding storage infrastructure, water pricing and markets, conjunctive use, groundwater banking, allocation limits, improved water use efficiency, public and private incentives for irrigation technology, reuse of tail‐water, shifting to less water‐intensive crops, and fallowing . Even when water is not limiting, high summer temperatures can have direct impacts on the yield of many crop species, particularly if extreme temperatures occur at key points during the reproductive phase . Since exposure to high temperatures is difficult avoid in the field, adaptation strategies may require shifting to new crops or varieties with better tolerance to high temperatures . Place‐based adaptation plans at the county or irrigation district scale would provide opportunities to better understand the local risks and uncertainties; improve communication among stakeholders, officials and scientists; and ultimately enhance the community’s capacity to adapt .For the crop vulnerability sub‐index, two retained components cumulatively accounted for 86.3 percent of the variance among grid cells . The crop dominance and crop sensitivity indices had high positive loadings on PC1, while pesticide rate had a very high positive loading on PC2. The Salinas and Santa Maria Valleys, as well as the areas surrounding Fresno and Bakersfield, had very high crop vulnerability due to a combination of high crop sensitivity and high pesticide use . Much of the Central Valley had moderately high vulnerability due to a mix of moderate crop sensitivity and moderate pesticide use. While Napa, Sonoma, Marin, and Mendocino counties had relatively low crop sensitivity due to the widespread cultivation of wine grapes, parts of these counties also had moderately high vulnerability due to high crop dominance . Changes in climate can directly impact crop growth though new temperature regimes and a northward shift in the range of pests and disease. In response to a reduction in chill hours, nut and stonefruit growers may require new low‐chill hour varieties or a shift to new crops . Warmer winter temperatures may extend the growing season for alfalfa or certain cool season crops , and expand the range of subtropical crops like citrus. Warmer summer temperatures may allow for the cultivation of hot‐season crops in regions where they are not currently grown . Longer growing seasons will likely enable pest species to complete more reproductive cycles, which can increase the severity of infestations . Improving agrobiodiversity can limit some of these risks by serving as a repository of germplasm for future plant breeding efforts, and providing specialized knowledge that may help growers shift more easily to new crops .Results of the PCA for the land use vulnerability sub‐index indicate that 67.0 percent of the cumulative variance among grid cells is explained with two principal components .