California ranchers with substantial private oak woodlands sometimes use public lands as an important component of their production cycle. Yet allowed public grazing has declined and is likely to continue to decline. This, combined with intensifying development pressure and land use change, dramatically affects the resource base for ranch operations, which in turn influences landowner management decisions and practices. Such individual decisions have a significant role in shaping California's future landscape. As private rangelands are developed and taken out of production, public lands become a greatly sought after forage resource for livestock operators. Public agencies could dramatically expand their spheres of conservation influence by leveraging these important connections between private oak woodland ranches and their public leases. This scenario is examined in two key areas in California: the Central Sierra s oak woodland foothills, and the San Francisco Bay Area’s East Bay grassland and hardwood rangelands. In both of these areas the public manages large and crucial swaths of grazed lands surrounded by privately held oak woodland ranch lands. After conducting more than 50 in-depth interviews over the past 6 years we find differences between the two groups of operators in reactions to public forage reductions. In the Sierra foothills seeking replacement forage is a common response, whereas in the Bay Area reducing the herd is the most common reaction to a reduction in allowed public grazing. Nevertheless, in response to a total loss of public grazing over 1/3 of both groups feel it is likely they would sell their ranches. When the private acreage attached to these public leases is taken into consideration, this study reveals how public land management decisions affect the risks to California s remaining hardwood range.

Approximately 70 percent of the land mass of the contiguous United States is in private lands with 50% of this total in row crop or rangelands. Audubon California’s Landowner Stewardship Program works with farmers and ranchers on conservation and restoration projects in a manner compatible with existing agricultural operations. Audubon’s program has worked with over 50 farmers and ranchers in the counties of Yolo, Solano, and Imperial Counties particularly on riparian, oak woodland and native perennial grassland restoration. To assess the success of these efforts Audubon, UC Davis, USDA Agricultural Research Service and Michigan State University conducted individual studies and a meta-analysis of the range and oak woodlands of Yolo County. Results from this multi-disciplinary research as well as lessons learned will be presented.

A Central Coast ranch recently made state and national news with its conservation easement which allowed several forest management methods (e.g., thinning). Questions about the extent of thinning, the sustainability of such forest management practices, and the compatibility with other uses were raised by the media, agency personnel, and environmental groups. In response to this issue, UC Cooperative Extension and Cal Poly Natural Resources Management Department are collaborating to help educate oak woodland owners on the Central Coast. The main purpose of the project is to give the best available knowledge to landowners and managers regarding the management of oak woodlands in the Central Coast, so they can manage their lands without further regulation. A workshop is being planned which will inform the public about the latest scientific findings as well as policy regarding oak woodland management. The workshop will rely on a literature review of publications pertaining to oak woodlands, and a survey of landowners in Santa Barbara, San Luis Obispo, and Monterey counties. The most recent statewide landowner surveys were conducted by Integrated Hardwood Range Management Program (IHRMP) in 1985 and 1992, and this is the first survey to focus exclusively on the Central Coast. Just over 2,900 surveys were mailed and 450 were returned. The workshop scheduled for August 2006 will present information on topics identified in the survey. This paper describes the results of the survey, the results of the workshop, and future actions. We expect this project to lead to an  on the ground demonstration  to show the best practices relating to oak woodland management.

For about 3,000 years, California Indians have used the acorns of several oak species, as well as tanoak, for food. Through lecture, slides, demonstrations, and handouts, this paper will survey the contemporary utilization of acorns for food by California Indians across the state, including an examination of: (1) ancestral burning techniques used to manage oaks; (2) the context in which these techniques are being applied today on certain reservations, rancherias, public lands, and private lands; (3) new and old acorn processing techniques that are still in use today, including cracking stones, winnowing baskets, and hammers, mortars and pestles, electric blenders, and coffee grinders, soaproot brushes, sand leaching beds, porous containers lined with flour sacks, stone boiling, and stovetop boiling in stainless steel pots; (4) the varied foods that result from these processing techniques soups, breads, mushes, and biscuits; (5) familial, community and cultural contexts in which the food is shared and eaten; and (6) the symbolic content of the food in terms of ecological and social relationships that connect people to place, each other, cultural heritage, and sacred time. This paper will also examine the devastating impact that Phytophthora ramorum has had in disrupting, limiting, and sometimes severing these eons-old relationships, by weakening and killing the most important acorn-producing species for food statewide, Quercus agrifolia, Q. kelloggii, and Lithocarpus densiflous, as well as infecting numerous other species found in association with oaks and tanoaks. The paper will conclude with an overview of how California Indians are responding to this threat to cultural continuance.

Results of a 1985 survey of California hardwood rangeland landowners were used to develop a multi-agency research and extension program known as the Integrated Hardwood Range Management Program. In 1992, the same properties were re-surveyed, revealing significant changes in landowner behavior and land use, including reduced cutting of oaks and a decline in livestock production. Landowners, regardless of whether they were ranchers or second-home owners, all valued the amenities of the woodlands: scenery and quality of life. However, the livestock producers that own most of the 10 million acres of rangeland are under stress from urbanization and escalating real estate values. Here we report the results of a re-survey in 2004. As California undergoes rapid demographic and land use change, the changing pattern of hardwood rangeland land ownership will have an important impact on landscape and livelihoods in the oak woodlands.

Long-Term Growth and Economic Viability of Coast Live Oak in Three California Counties - Seventeen-Year Results. Lawrence E. Bonner, Norman H. Pillsbury and Richard P. Thompson A long-term thinning study was established in 10 stands of coast live oak (Quercus agrifolia Née) in the central coast of California in 1984. Information about diameter, number of stems per acre, basal area, and volume growth and yield has been obtained from unthinned control plots and from plots thinned to 50 and 100 square feet of basal area per acre measured in 1984, 1989, 1996, and 2001. Both basal area and total volume growth percentages were significantly greater in the thinned plots compared to the control plots. Average basal area per acre growth rates were 22 percent for the treatment plots and one percent for control plots. In general, total volume growth rates were nearly three times as great in the treatment plots compared to the growth plots. Mortality in unmanaged stands can substantially reduce the growth potential. Proper care and management of the woodland forest can reduce losses from fire and, in general, protect the value of the resource. Wildland-urban interface lands in California are at high risk to loss from fire associated with development pressure. Investments in forest management can allow people to live and work with fire while reducing the overall costs of fire suppression efforts to taxpayers.

Water quality was investigated on the Santa Margarita Ranch, San Luis Obispo County, where drainages carry water from chaparral, oak woodland, and a new vineyard. Three drainages were instrumented with gauges above and below the vineyard for stage height, temperature, and turbidity to assess the effectiveness of water quality protection measures at the Cuesta Ridge Vineyard. Water samples were taken during storm events and analyzed for pH, EC, nitrogen, phosphorus, potassium, sulfur and sediment concentrations. Erratic flow conditions made it difficult to identify differences above and below the vineyard. Water quality was good at all sites and data collected serves as baseline information for future study. Landform, geology, soil type, proximity to wetlands, and land use practices in the surrounding areas should be investigated to consider possible relationships with turbidity and nutrient concentrations.

California's oak-woodlands are a complex, often multi-layered mosaic of grassland, shrubland and woodland patches. While soil, geology and topography influence the distribution of these patches, disturbance and biological interactions are also important determinants of the patchy distribution of these plant communities. Fire intensity and frequency can change the structure of these patches. Different re-establishment strategies of woody species interacting with prevailing weather following disturbance can also produce changes at the patch level which are often smaller than a soil mapping unit or an ecological site. While our knowledge of vegetation dynamics in the oak-woodlands is not great, what we do know, or can reasonably hypothesize, needs to be organized into a format that is sensible and accessible to natural resource managers. State and transition models have been proposed as a format for organizing the complex body of knowledge and concepts describing vegetation dynamics in rangeland ecosystems. Natural and man-caused vegetation change can be incorporated into these models. In this review paper we will describe oak-woodland vegetation dynamics using a state and transition format. These models will describe vegetation change of the long-lived tree and shrub layers as well as the short-lived annual understory and adjacent grasslands. The dynamics will include changes between the mosaic of grass, tree-grass, tree-shrub-grass and shrub-grass patches that are common in California's oak-woodlands. Transitions between states address fire, grazing, drought, woody plant harvesting, vegetation management, and protection. Restoration and apparent irreversible transitions will also be illustrated.

In Santa Barbara County we have replicated large-scale planting experiments from 1997 to 2001 to determine the effects of cattle and other factors on seedling establishment of valley oak (Quercus lobata) and coast live oak (Q. agrifolia). We manipulated cattle grazing (ungrazed vs. rotational winter-spring grazing) and protection from small and large mammals. Results we have previously reported indicate that rainfall, and seed predation and herbivory by small mammals—most likely gophers and ground squirrels—significantly impacted rates of seedling recruitment. Using the established cohorts of seedlings, we are now examining the factors controlling the transition from seedlings to saplings. Here we present results on growth and survival of both oak species in four different age classes. Mortality rates after year three have been low, but relatively constant in nearly all treatments. For seedlings protected from direct herbivory (i.e., caged individuals), livestock grazing improved survivorship suggesting that cattle have indirect positive effects on oak establishment, perhaps by reducing herbaceous biomass. Contrary to expectations based on the relative abundance of natural oak seedling recruits, establishment and survival of coast live oak has been significantly and consistently lower than that of valley oak. Results from large-scale and long-term experiments will provide information useful for construction of oak population models and a better understanding of the patterns of regeneration in oak woodlands in California and elsewhere.

In California’s blue oak woodlands, plant survival and growth are tightly linked to belowground processes. Microbial biodiversity controls nutrient and water uptake, maintaining oak woodlands. We studied linkages between belowground biodiversity and belowground processes. How do mycorrhizal fungi alter nitrogen acquisition? Our research has altered our understanding of oak woodland function in 5 ways. First, we expected nitrogen to move between plants within  biological pipes  or a common mycorrhizal network (CMN) with little leakage into soil. We discovered that there are no exclusive pathways; roots are  leaky, nitrogen is released into the soil. Second, since oaks form ectomycorrhizal (ECM) networks and annuals form endomycorrhizal (AM) networks, we expected these networks to be independent, resulting in competition for nitrogen. Instead, oak and annual networks appear to interact, cooperating to transfer nitrogen among both networks. Nitrogen transfers between oaks was greater when annuals were present, reduced when annuals were removed. Third, we expected nitrogen to transfer among plants slowly, due to the large size of plant nitrogen pools. However, transfers are rapid and occur into and out of oak roots. This 2-way nitrogen movement is sensitive to nitrogen pools in adjacent oaks; oaks may  communicate , sharing nitrogen. Fourth, we expected ECM biodiversity to be similar in oak woodlands and conifer forests, since ECM occur in both forest types. However, certain fungal groups (hypogeous fungi that fruit belowground) are more diverse in oak woodlands. Fifth, we expected EM species that are more abundant on oak roots would also dominate nitrogen cycling. In 15N tracer field studies, we learned that less abundant fungal species were more active in nitrogen cycling. Perhaps less abundant mycorrhizal morphotypes are keystone species, playing dominant roles in nitrogen cycling. Our studies demonstrate the surprising 2-way nitrogen transfers between plants and the sensitivity of nutrient cycling to mycorrhizal biodiversity.

Oaks depend on hidden diversity belowground. Quercus garryana forms ectomycorrhizas with over 40 species of fungi at a single site. Several of the most common oak mycorrhizal fungi form hypogeous fruiting bodies or truffles in the upper layer of mineral soil. We collected 21 species of truffles associated with Q. garryana. Truffles do not release spores directly into the air, but remain closed belowground. In conifer ecosystems, animals eat truffles and disperse the spores, providing mycorrhizal inoculum for new roots. We hypothesized that in oak woodlands small mammals eat hypogeous fungi and defecate the fungal spores. We trapped small mammals near Q. garryana and examined fecal pellets for hypogeous fungal spores. Three species of rodents, Microtus californicus, Peromyscus maniculatus, and Reithrodontomys megalotis, had 12 species of fungal spores in their fecal pellets. The most common truffles found as spores in fecal pellets were Tuber candidum/T. quercicola, Hydnotryopsis setchellii, and Cazia flexiascus, all Ascomycota. All oak seedlings require mycorrhizal inoculum. Seedlings growing in the root zone of mature oaks have access to the mycorrhizal network of parent trees, but seedlings outside the root zone lack mycorrhizal sources. If the mycorrhizal community on saplings located away from mature oaks includes hypogeous fungi, then small mammals may be dispersing fungal spores into shrublands where saplings are located. We examined roots of oak saplings at distances up to 72 meters from mature oaks and found mycorrhizas of Tuber candidum and Peziza infossa, both hypogeous species, suggesting that small mammals disperse spores for mycorrhizal inoculum. Thus we demonstrate small mammal mycophagy of hypogeous fungi in Quercus garryana woodlands and provide evidence that mycorrhizal inoculum is distributed in fecal pellets. Regeneration of oak woodlands may depend on the dispersal of mycorrhizal fungal spores by small mammals.

Recently mycorrhizal fungi of the Oak Woodlands in the Northern California Sierra Nevada Foothills are receiving more attention. To supplement ongoing mycorrhizal work we investigated the mycorrhizal fungi associated with several oak woodland mature tree species in this dry and nutrient poor area. Our research focused on characterizing and quantifying the fungal hyphal biomass below blue oaks, interior live oaks, foothill pine and Ponderosa pine. The field site was located at the Sierra Foothills Research and extension center outside of Marysville, California. Root restrictive bags made of 25 ¼m nylon were filled with sand, and supplemented with apatite. To discourage saprotrophic fungi growth, no carbon source was added to the media. The bags were placed in the ground at canopy edge beneath each tree species and harvested after one year. The harvested bags were, freeze dried, and the hyphae were removed for analysis. Hyphal tissue was quantified using several techniques. A method currently being assessed is the development of a rapid technique to determine hyphal length using microscope pictures and WinRhizo 2002 software. This new method is being compared to the more common method,  gridline intersect method,  developed by D. Tennant (1976), which is a modified version of Newman’s technique (1966) for determining hyphal length. Enough hyphal biomass had accumulated so that weights of the hyphae could also be measured directly and correlated with hyphal length. The hyphal in growth bags buried around blue oaks containing the highest hyphal biomass. We hope this method will allow faster quantification of fungal biomass for projects in many different forested ecosystems. We plan to utilize this technique to evaluate the influence of fungal biomass and mycorrhizal status may have on the survival and growth of sapling oaks in California’s oak woodlands.

West Nile Virus created an unprecedented rate of infection in North America, sweeping from New York to California in 4 years. It attacked an exceptionally broad range of hosts/vectors over an exceptionally broad geographic area. In 2004, the State Department of Health received approximately 98,000 reports of dead birds, translating into tens of millions of birds killed by West Nile Virus in California. The rates of dead bird reports in California Hardwood Rangelands varied from 54 + 11 birds/km2 in the upper Sacramento River Valley to 3.7 + 0.8 birds/km2 in the southern coastal foothills. The best predictor of bird death was temperature, specifically the number of days exceeding 90o F. Overall, about 80% of hardwood rangelands were in areas of moderate to high rates of West Nile Outbreak. Areas suffering the highest rates of mortality had primarily Blue oak and Valley Oak woodlands, and bird species in blue oak woodlands suffered the greatest potential exposure to West Nile Virus.

Acre for acre, oak woodlands provide habitats for a greater range of wildlife species than the grasslands and irrigated agricultural lands that are often closer to existing metropolitan areas. The richer habitat diversity drives both residential interest in living in or adjacent to oak woodlands as well as preservation interest in maintaining oak woodlands without an overlay of residential land uses. Numerous state and local regulations have been designed to balance these two competing goals. By standardizing residential densities at the Census block group level (containing approximately 400 housing units) we were able to map residential densities within oak woodlands, and other vegetation types, for 1990 and 2000 across the state of California. This provides a geographically detailed analysis of affected acres and housing units at different density levels across the state. The comparison of 1990 and 2000 data provides insights into both patterns of infill within existing census block groups as well as the development of new oak woodland areas with a significant number of housing units. Not surprisingly, the current balance between residential and non-residential oak woodland is not uniform across the state. In addition to the role of local zoning restrictions, we analyze the roles of some of the key economic drivers behind residential conversion. The relative affluence of future residents as well as the local real estate cost of oak woodlands (both per parcel and per acre) can also be mapped geographically and used as explanatory variables for observed rates of residential conversion. In some areas, the same levels of wealth that allows for extensive residential conversion can also be critical in financing the development of permanent areas of oak woodlands without residential conversion.

People value proximity and easy access to protected areas in urban landscapes, including state and regional parks, wildlife refuges, and open space preserves. The popularity of outdoor recreation activities such as hiking and birdwatching has more than doubled in the past 20 years, and surveys indicate that proximity to natural areas is an important factor determining where residents choose to live in California. However, public use of protected areas may limit their effectiveness for conservation. Modeling the variation of recreational use in space could help us better understand its impacts on wildlife and biodiversity. In this study, we used geographic information system (GIS) analysis to examine how site attributes and landscape variables correlate with the relative intensity of recreational use in oak woodland preserves. We collected records of infractions issued by park rangers to recreational users in several Marin County Open Space Preserves between 2001 and 2005 as an index of recreational use intensity. In a multiple regression model, we compared the frequency of infractions to several site- and landscape-level variables related to site accessibility, site amenities, and land uses surrounding each preserve. The results of our model can help park managers anticipate the relative intensity of recreational use and mitigate potential impacts to wildlife and biodiversity.

In 2004 the California State Legislature passed Senate Bill1334 (Bill), titled The Oak Woodlands Conservation Act. This Bill outlined oak woodland mitigation options for counties to achieve feasible and proportional habitat mitigation under the California Environmental Quality Act. The Bill required that if a county approved project may have a significant effect to oak woodlands then the county must require one or more mitigation alternatives. Faced with implementing this Bill, many counties began asking the University of California (UC), the California Department of Fish and Game, the California Department of Forestry and Fire Protection and the Wildlife Conservation Board (WCB) questions including what types of projects fell under the purview of the Bill, what constituted  significant impacts , and how to determine suitable, appropriate mitigation. To help answer these questions, the UC Integrated Hardwood Range Management Program (IHRMP) submitted a grant to the WCB that was funded in early 2006. This grant proposed activities and products to specifically help planners implement SB 1334. In addition, the different organizations being asked questions about SB 1334 formed a Committee to develop and disseminate SB 1334 information, including compliance standards and effective strategies to conserve oak woodlands. Specifically this Committee charged itself with developing guidelines that planners could use to analyze proposed project effects on oak woodlands. These will include web-based decision keys, PowerPoint presentations and visual comparison standards for assessing oak woodland impacts. Although these efforts are ongoing, the net products will be important tools to help counties comply with SB 1334 and conserve their important oak woodlands.

Woodland ecosystems often consist of a mosaic of interacting dominant woody species that vary in density and abundance. Local variation in dominant species abundance may influence spread of plant pathogens across this heterogeneous landscape. We investigated this possibility in a 275 km2 study area in eastern Sonoma County, which is being invaded by the pathogen that causes Sudden Oak Death, Phytophthora ramorum. We assessed symptoms of infection by P. ramorum of foliar and canker hosts in 2004 and 2005 in 202 randomly located plots. Our results showed that the widespread foliar host, bay laurel (Umbellaria californica), exhibited symptoms of disease throughout the study area, while canker hosts showed no signs of infection in most plots. This suggests that SOD spread among foliar hosts much more rapidly than from foliar hosts to canker hosts. With the high precipitation over the past two rainy seasons, we expect the disease to progress into canker hosts over the next several years. We also found that density of bay stems and degree of infection on bay laurel depended partly on presence of oak species. For example, the proportion of bay stems with symptoms of P. ramorum was 18% greater when coast live oak (Quercus agrifolia) was present than when it was absent. Coast live oak presence was also related to number of symptomatic bay leaves per tree. We suspect that our findings arise from the fact that the pathogen spreads most rapidly in stands where few canker host individuals inhibit dispersal among foliar host trees. Although previous studies have shown that abundance of bay laurel is related to infection levels in oaks, this is the first study that we know of suggesting that the presence of canker hosts affects infection levels on bay laurel, the main foliar host of P. ramorum.

Phytophthora ramorum has a broad host range in coastal California forests, but it disproportionately impacts coast live oaks (Quercus agrifolia), black oaks (Q. kelloggii), Shreve oaks (Q. parvula var shrevei), and tanoaks (Lithocarpous densiflorus). These overstory trees provide the principal high quality plant proteins used by multiple animal species, serve as primary food sources for animals and fungi, and support numerous other species through their architecture. From continuously monitored plots in Marin County, we can describe a broad pattern of tree responses to P. ramorum. The sequence of disease progression in infected coast live oaks and black oaks is consistent: bleeding; ambrosia and bark beetle attacks; development of decay fungi, most notably Hypoxylon thouarsianum; followed by death. Weibull regression model estimates of mean survival associated with these disease stages strongly indicate that the beetle attacks significantly shorten the time to mortality, probably by overwhelming tree defenses. Coast live oaks and black oaks with larger stem diameters have significantly increased probability of P. ramorum infection. Once infected, larger diameter coast live oaks are more likely to be attacked by beetles. Stem breakage of living infected coast live oaks is consistently associated with ambrosia beetle tunneling in the sapwood. Although the rate of new infections has slowed in coast live oaks since 2000, the mean infection level in our study plots and in landscape-scale evaluations is little changed, at 22%. The percentage of infected living tanoaks has increased from 39 to 73% since 2000. We can project that a shift in the age and size structure of these dominant trees is continuing, leading to a number of ecologically significant changes. Loss of oaks and tanoaks and their ecological roles, plus the introduction of weedy invasive plants has begun in these coastal oak woodlands and forests.

To study impacts of sudden oak death (SOD), a lethal bark canker disease caused by Phytophthora ramorum, we established permanent plots in Sonoma County forest types at risk due to SOD. Baseline stand and tree health data were collected in 2001 and the plots were resurveyed in 2004. The 250 plots (0.02 ha each) are located at 11 study locations in stands containing Quercus agrifolia, Q. kelloggii, or Lithocarpus densiflorus as the dominant hardwood species. By 2004, P. ramorum was present at two locations that lacked P. ramorum symptoms in 2001 and caused tree mortality at one of these locations. Between 2001 and 2004, P. ramorum canker incidence increased from 29% in 2001 to 40% in plots containing L. densiflorus and from 2% to 10% in plots containing Q. kelloggii. Plots with Q. agrifolia showed a slight drop P. ramorum canker incidence (from 9 to 7%) due to apparent symptom remission in trees at one location. Between 2001 and 2004, the percentage of trees with P. ramorum canker symptoms increased at three of four locations with symptomatic SOD canker hosts. Mortality due to both P. ramorum and other agents increased at 9 of 11 study locations between 2001 and 2004. P. ramorum accounted for 27% of the mortality increase in Q. kelloggii, 49% in Q. agrifolia, and 30% in L. densiflorus. For most locations, annualized background mortality unrelated to P. ramorum was less than 1% per year between 2001 and 2004. Over this period, mortality rates associated with P. ramorum exceeded background mortality at three locations and was equal to it at a fourth. At one L. densiflorus location, mortality associated with an unidentified bark canker was comparable to levels of mortality associated with P. ramorum at other locations. Only P. nemorosa and P. pseudosyringae were isolated at this location.

Sudden Oak Death, caused by the water mold Phytophthora ramorum, is a plant disease responsible for the death of 100s of thousands of oak and tan oak trees. California bay laurel (Umbellularia californica) has been shown to play a key role in P. ramorum inoculum build-up and subsequent spread to oaks. While bay laurel appears to vary in susceptibility to P. ramorum, little is known about the causes or extent of this variability. An understanding of such dynamics helps in identifying the limits of infection, dispersal, and impact of plant pathogens. In this research, we examine how bay laurel susceptibility to P. ramorum is related to bay laurel phenotype, genotype, and environment. Based on data from lab susceptibility trials and field infection data collected from 97 trees from 12 populations in Northern California, we found that lab lesion size and field infection levels varied significantly among both bay laurel trees and populations. The phenotypic trait of leaf area was significantly related to lab lesion size, where bigger leaves produced bigger lesions. Variability in lesion size produced in the lab and infection levels in the field were significantly related to AFLP markers, suggesting a genetic basis to resistance. We also identified markers associated with phenotypic traits putatively involved in conferring susceptibility, including leaf toughness and leaf water content. At the population level, environmental variability significantly explains susceptibility to P. ramorum. This work demonstrates how susceptibility to P. ramorum in bay laurel depends on genetic, phenotypic, and environmental traits, providing useful information for predicting risk of P. ramorum spread from bay laurel to oak and tan oak trees, an important consideration for controlling the spread of Sudden Oak Death.

Over the past century, plant diseases have had major impacts on forested ecosystems worldwide. An important emerging forest disease is Sudden Oak Death, the causal agent of which, Phytophthora ramorum, was first identified in 2000. Currently confined to several counties in coastal California and a small area in Oregon, the disease has the potential to become more widespread. In 2002, we began an investigation into the impacts of the disease on mixed evergreen forest communities dominated by coast live oak (Quercus agrifolia) and bay laurel (Umbellularia californica). We measured understory herbaceous plant and shrub parameters and woody plant seedling densities, over a gradient of disease severity to assess whether newly downed trees serve as refugia for woody seedling recruitment. We found diverse understory assemblages in infected forests, with understory species richness ranging from 19.1 to 32.2 species per hectare. Percent cover of herbaceous plants in these understories ranged from 14.3 to 51.7%, woody seedlings are present at densities of 0.37 to 2.08 seedlings/m2, and course woody debris varies from 1.20 to 46.86 m3/ha. Initial results suggest that newly downed trees do not serve as refugia for woody seedlings, but correlations do exist between forest infection gradients and other understory parameters. Through direct lethality to infected trees, this novel plant-pathogen combination has the potential to indirectly affect the composition and diversity of non-target understory species.

Although outdoor recreation is often viewed as having minimal impacts on protected wildlands, a growing number of studies have shown that hiking trails can serve as corridors for the spread of exotic plant species. Very little research has focused on recreation as a means of spread for plant pathogens but here, we present research that links two forms of recreation (hiking and mountain biking) with the dispersal of an exotic forest pathogen, Phytophthora ramorum, which causes the disease known as Sudden Oak Death. By sampling soil collected from the shoes and tires of visitors entering and exiting protected wildlands, we found that 7% of visitors entered the site with viable pathogen on their shoes and tires, and 23% carried it out with them. Although hikers and bikers did not differ significantly in their capacity to transport P. ramorum, there was a trend indicating that during dryer conditions, the further a person traveled along a trail, the more likely they were to pick up the pathogen. Lastly, although our data show that humans can disperse P. ramorum, the temporal window for doing so is constrained, as the pathogen could not be cultured from soil on hikers  shoes after 24 hr, although this time was extended to at least 72 hr if the soil on shoes remained moist. Collectively, our results suggest that recreation in protected areas can lead to dispersal of the pathogen that causes Sudden Oak Death. However, because P. ramorum is sensitive to desiccation, human dispersal may be limited to situations where the pathogen remains moist. These include further spread of the pathogen during individual visits to already infected areas and movement of the pathogen to previously uninfected areas when visitors move rapidly from one region to another, especially when shoes and tires are stored in moist conditions.

A striking consequence of globalization is the tremendous influx of infectious diseases and invasive, non-natives species worldwide. One invader of great concern is the fungus-like pathogen, Phytophthora ramorum, which causes a devastating forest disease known as Sudden Oak Death (SOD) in many coastal regions of California and Oregon. In addition, P. ramorum has been found in nurseries and managed landscapes throughout Europe (11 countries so far) and recent laboratory studies have indicated that numerous oak species in the eastern United States are extremely susceptible to attack by the pathogen, should it reach these areas in the future. Here, we evaluate the influence of human activity on the distribution of this pathogen at three spatial scales in California. At the local scale, we found that P. ramorum was significantly more common in soil found on hiking trails at Fairfield Osborn Preserve than from adjacent areas off trail. At the landscape scale, forests on public lands in eastern Sonoma County open to recreation had significantly higher proportions of diseased host trees than those on private lands. And at the regional scale, forested areas in northern and central California surrounded by large human populations were significantly more likely to have infected host trees. Collectively, these findings suggest that humans are important dispersal agents of a destructive pathogen and promote its spread. Efforts to address this epidemic may thus require aggressive management of human activity, which could be logistically and politically difficult to achieve.

The number of emerging infectious diseases are thought to be increasing worldwide   many of which are non-native, invasive plant diseases in forest ecosystems. A predictive understanding of invasion processes is necessary to manage and prevent further spread, especially in complex spatially heterogeneous landscapes. Mathematical modeling of susceptible-infectious transitions in plant epidemics often incorporate spatial dynamics, but are rarely applied in a GIS to real-world wildland landscapes. In this paper, we present and evaluate a GIS-based epidemiological model of the spreading forest pathogen Phytophthora ramorum, which is causing the devastating forest disease known as Sudden Oak Death. First, we describe a generic mathematical model for a susceptible-infectious epidemic that simulates spatial and temporal patterns of disease spread on a weekly time step for application at large spatial scales. Next, we describe how data from field and lab studies were used to parameterize the driving system variables, including daily rainfall and temperature, host abundance and susceptibility, human density, and dispersal characteristics. The parameterized model was implemented (1990-2005) in a GIS to simulate disease spread across California at a spatial resolution of 250 x 250 meters. We evaluated model performance in the GIS by examining the correspondence between predicted patterns of disease spread and over 1000 geo-located field observations of disease presence. Additionally, we examined the nature of prediction errors by ecoregion, vegetation composition, and climate. The model predicts almost 80% of the spatial variability in current patterns of disease spread and identifies numerous oak woodland systems at high risk of infection. We ve found the application of epidemiological models to realistic landscapes in a GIS can allow for a rigorous validation of model performance using geo-located field data of disease presence and can be used as an effective management tool to identify actual landscapes at high risk to disease spread.

Land-use practices to sustain expanding human populations have resulted in dramatic alterations of forested ecosystems worldwide. One potential consequence of land-use change is the spread of infectious disease. In this research, we investigated patterns of human-related land-cover change between 1942 and 2000 and examined whether or not changes have led to conditions that promote the establishment and spread of the invasive pathogen Phytophthora ramorum, causal agent of the destructive forest disease Sudden Oak Death. We established 102 field sites randomly located within a 275 km2 region of Sonoma County, CA comprised of several land cover types. Within a 150 m radius area around each site, we mapped fine-scale changes in land cover in a GIS using 1942 and 2000 imagery. We found that woodland area increased by 25% from 1942 to 2000, while grassland and chaparral decreased in area by 34% and 49% respectively, and development increased by 1105%; mean size of woodland patches increased by 51% and number of woodland patches decreased by 41%. To examine if these changes influenced the establishment of P. ramorum, we sampled P. ramorum infection level on host species and measured woody species abundance within each 150 radius area in spring 2005. Multiple regression analyses showed that density of the host bay laurel (an important carrier of the disease) was positively related to woodland cover change. Also, there were a greater number of bay trees showing symptoms of P. ramorum in areas that have increased in woodland. We hypothesize that these increases in woodland density and expansion into grasslands and shrublands facilitates spread of P. ramorum by increasing contagion of host and increasing shading and moisture levels. Understanding the influence of land cover change on disease spread is critical for informing land management and preventing intensification and spread of destructive pathogens.

Sudden oak death (SOD) is a disease of epidemic proportions, sweeping through many coastal hardwood forests in California. Since 1999, SOD has produced overstory mortality in China Camp State Park (CCSP), creating a number of gaps which effectively alter the structure of the forest. In following year, and the three thereafter (2000-2003), ADAR high-resolution (1 m) multispectral imagery for CCSP was acquired. We classified this existing imagery to identify gaps within the forest mosaic. Once the gaps were identified, they were vectorized and measured for perimeter and area. We also characterized the gaps within their spatial context through closeness to edge and closeness to nearest gap. We then compared these spatial measures of variation to temporal ones, creating a time-sequence of gap formation and the resulting gaps  closure or persistence. In addition to multispectral imagery, we sampled basal area measures within CCSP and correlated them with gap size, structure, and configuration, testing the accuracy of remote sensing techniques against field based information in determining forest structure changes. We then scaled our understanding of these structural changes up to a larger landscape scale not covered by the field plots.

We conducted surveys that documented animal and plant species diversity, abundance, and distribution across a landscape that included areas transformed from annual grass rangeland to agricultural production. Surveys were conducted quarterly over four years. Trends in the distribution and abundance of species were examined within the agricultural area (vineyard) and in undisturbed grassland. Vineyard habitat and annual grassland habitat were compared with respect to statistical differences in species number, population number, distribution of species, and net diversity as an index of species and abundance. Contrary to expectation the effect of agricultural conversion to a well-managed vineyard is positively correlated over time to an increase in species diversity and abundance. Further, compared to undisturbed grassland, the vineyard areas provide a more complex habitat supporting more species and larger populations of insects and birds.

Wild pigs were first established around coastal Spanish settlements in California in the 1800s and expanded over the last century by hunting introductions, domestic releases, and natural dispersal. The current distribution of wild pigs is closely associated with oak woodlands where foraging and rooting by the species may impinge on many native plants and animals. Rooting disturbance by wild pigs may be especially problematic for regeneration of oak woodlands by physical damage to small tree seedlings. In 1998 we initiated a long term study of the ecological effects of wild pigs in California oak woodlands using multiple exclosures paired with nearby control plots that were established in two state parks in the north and central coast region of California. Mesh size and height of exclosure fences were designed to repel wild pigs but allow access by all other organisms, and functioned as intended. We were especially interested in rooting effects on oak seedling number and size. The focus period for the 8 year study recently ended, during which we monitored seedlings in (1) large 50-m X 50-m exclosure and control  habitat plot  pairs centered in oak woodland, and (2) smaller 3-m X 3-m exclosure and control  canopy plot  pairs established beneath the canopies of individual trees producing large crops of acorns in fall 1998 and 1999. We previously reported that wild pigs significantly reduced acorn survival in control compared to exclosure plots, thereby reducing forage availability for native organisms. In this paper we will summarize the longer time series of data on changes in number and size of oak seedlings in habitat and canopy plot exclosures by summer 2005. Although mean size of seedlings was significantly larger in exclosures in general, the final differential for seedling number between exclosure and control plots was nearly 400% at one site.

We studied the nesting habitats of cavity-nesting birds in oak woodlands at the San Joaquin Experimental Range, Madera County, CA. We measured habitat surrounding nests and examined similarities and differences in nesting habitat used by eight species of secondary cavity-nesting species. Nesting habitat of Bewick’s Wrens (Thryomanes bewickii) was unique compared to other species. Their nests were the lowest and located in the densest habitat on the steepest slopes where interior live oaks (Quercus wislizenii) were abundant. Nesting habitat was similar among European Starlings (Sturnus vulgaris), Western Bluebirds (Sialia mexicana), and Violet-green Swallows (Tachycineta bicolor). These species all nested in open areas with the highest density of blue oaks (Quercus douglasii), the lowest densities of live oaks and snags, and the lowest basal area of trees. Nests of White-breasted Nuthatches (Sitta carolinensis) were similar to the above three species, differing only in being lower compared to nests of European Starlings. Ash-throated Flycatcher (Myiarchus cinerascens) nesting habitat was characterized by dense stands of trees with a low density of live oaks and snags. Oak Titmouse (Baeolophus inornatus) nests were low and located in dense stands with a low density of live oaks on steep slopes. House Wren (Troglodytes aedon) nests were found in dense stands of trees. Habitat features important to nest survival coincided with nest sites used. Successful Ash-throated Flycatcher and Oak Titmouse nests had higher basal area of live trees compared to unsuccessful nests and, compared to other species, habitat surrounding their nests had high basal area. European Starling nests were highest, and nests located high in trees had higher nest survival. Detailed information about the nesting habitats of species, including those needed for successful nesting, can provide a better understanding of the ecological factors that permit coexistence of different species and may aid in conservation efforts. Our results suggest a potential for nest-site competition among European Starlings, Western Bluebirds, Violet-green Swallows, and possibly White-breasted Nuthatches, and are consistent with results found for nest-cavity variables.

California and Israel are both characterized by Mediterranean climates, and the native oaks growing in these two locales occupy similar ecological niches. However, unlike their Israeli counterparts, low regeneration among certain California species appears to be limiting their ability to colonize suitable habitats released from disturbance. For the past three years, a greenhouse study has been conducted in both Israel and California to sort out the role of evolutionary history in influencing species regeneration rates and strategies. Results suggested that potted California oak seedlings might be less tolerant to drought than species from Israel, as evidenced by lower water use efficiency, reduced seedling survival, lower SLA and changes in stable carbon isotopes. This past year an outplanting study was initiated at the UC Sierra Foothill Research and Extension Center in Yuba County to determine how the species respond under normal field conditions. Three California oak species, including Quercus agrifiolia, Q. douglasii and Q. berberidifolia, as well as two Israeli oak species (Q. ithaburensis and Q. calliprinos), have been planted in a common garden. This study is evaluating phenology, growth rates, and how seedlings respond to summer drought in terms of pre-dawn plant moisture stress (PMS). To date, results indicate that the deciduous species from both locations (Q. douglasii and Q. ithaburensis) break bud far earlier than their evergreen counterparts. Growth rates vary greatly by species, with Q. agrifolia from California and Q. ithaburensis from Israel growing far more rapidly that the other species.We hope that this research will shed light on different adaptive strategies of evergreen and deciduous species and how oaks evolving under similar environments, but in different locales, deal with adverse environmental conditions such as drought. Ultimately, this information should benefit land management and oak restoration efforts.

Oak woodlands in northern California are becoming increasingly fragmented due to heightened urban and agricultural development. Much of the change in land use in Napa and Sonoma Counties has been from oak woodland to vineyards. Grape flowers are wind pollinated and therefore such a change in land use could potentially impact the local bee fauna. We studied how Bombus (Apidae) abundance and species richness are affected by oak woodland fragmentation. Adaptations such as polylectic foraging (utilizing multiple floral resources) and long flight distances may decrease the impact of fragmentation on bumblebees. We studied six sites surrounded by a high percentage of vineyard and six sites embedded in oak woodland. Collections were made at each site twice per month between March and September in 2002 -2004. Bumblebees were collected in two ways: netting and using pan traps. Pan traps were placed at 5m apart along two transects in each plot. To determine type of floral resource used (native or exotic), after each bumblebee was netted, identified and released, the flower it was captured on was also identified and recorded. We compared means of bumblebee abundance and species richness between paired sites. We also looked at whether there was a relationship between the type of floral resources. There was no significant difference between high and low vineyards in bumblebee abundance. There was a significant effect of floral use and bumblebee abundance, but no significant interactive (floral x vineyard) effect.

The first (1981-84) and second (1991-94) statewide inventories of California’s oak woodlands found low levels of regeneration for several common oak species. A regeneration stocking algorithm applied to blue oak woodlands in 1991-94, for example, resulted in classifying eighty-four percent of this forest type as nonstocked or lightly stocked with blue oak (Quercus douglasii) seedlings. In 2001, a new statewide inventory of California’s oak woodlands was initiated, with 10 percent of field plots measured each year. We use the first five years of data to re-examine regeneration of common oak woodland types. Blue oak continued to show very low levels of regeneration (seedlings and saplings) in this third decade of monitoring.

In 1987, a study was begun at five sites throughout California to determine how the sprouting of harvested blue oak (Quercus douglassii) is affected by the date the trees are cut down, the height of the residual stumps, and whether stumps are protected by fencing. Results of the study two years post-harvest and 10 years post-harvest were reported in the Proceedings of the Third (1991) and the Fifth (2001) Oak Symposia, respectively. In general the smaller stumps sprouted more, but no difference in sprouting could be attributed to season of harvest. The effect of protecting cut stumps became increasingly evident. Ten years after harvest over 4 times as many protected stumps had living sprouts as unprotected stumps. At the sixth Oak Symposium (October 2006), we will report the results of the blue oak stump sprouting study 19 years after harvest.

Ecological processes occur, synchronously and asynchronously, across many spatial scales. In California, we studied spatial synchrony in acorn production of California black oaks (Quercus kelloggii) among seven study sites between 1994 and 2002. One site was in Placer County, and consisted of four spatially distinct but relatively close stands (n = 144-147), and the other six sites, each with 25 study trees, were farther apart in San Diego, Los Angeles, Tuolumne, Shasta, Mendocino, and Monterey counties. To determine if synchrony exists with acorn production for these seven sites, correlation coefficients were calculated for annual acorn crops from randomly-selected pairs of trees. Distances between trees, based on UTM coordinates, were categorized into distance categories to assess relative synchrony. With the seven sites, there was limited spatial synchrony because r-values decreased with increasing distance. Trees < 0.001) while trees >500 km and 0.05). Separate analysis for the Placer County trees found spatial synchrony for trees up to 25 km apart (P < 0.001), the maximum distance separating these trees. With the other six sites, spatial synchrony (P < 0.001) was found for all distance categories. The greater number of trees with the Placer site introduced more tree-specific variation which increased variation in acorn crops, while trees from the other sites were large diameter individuals where within-population variation was reduced. Our results indicate that acorn production by California black oak is moderately spatially synchronous for trees

Recent mortality in Quercus agrifolia, Q. kelloggii, and Lithocarpus densiflorus exceeds historical levels in forests affected by the pathogen Phytophthora ramorum. We assessed the balance between recent mortality and seedling populations in these species to examine the status of regeneration in stands with varying levels of mortality caused by P. ramorum. Regeneration data was compiled in two sets of plots (all 0.02 ha). In 150 plots distributed across 12 locations primarily in Marin County, seedling counts and mature tree condition were rated annually from 2000 through 2005. In 250 plots distributed across 11 locations in Sonoma County, data were collected in 2001 and 2004. In plots monitored annually, Q. agrifolia seedling populations fluctuated more widely from year to year than tanoak seedling populations. Tanoak seedlings were present in nearly all plots with tanoak trees. Nearly all plots with tanoak mortality had tanoak seedlings which could potentially grow to replace dead trees. Coast live oak plots were less well-stocked with seedlings. Many plots with coast live oak mortality lacked coast live oak seedlings. Mean counts of coast live oak seedlings per plot were significantly lower in 2004 than in 2001 among plots in Sonoma County. Less than half of all plots with California black oak trees had California black oak seedlings. Three-quarters of the plots with California black oak mortality lacked seedlings of this species. In these forests, regeneration of California black oak appears inadequate to maintain stand density even without the additional mortality due to P. ramorum.

We examined the potential importance of pollen limitation to variation in acorn production among individuals and between years in three species of California oaks by testing for a relationship between phenology and subsequent acorn production both within and across years. Within years, trees flowering closer to the mean flowering date of the population at large produced significantly more acorns in two of the species. Assuming that more is available when more conspecifics are blooming, this is consistent with pollen limitation affecting individual variation in acorn productivity. Across years, relatively warm, dry conditions during the peak of flowering correlated significantly with larger mean acorn crops the next fall in all three species, explaining as much as 59% of the variance in the mean annual acorn crop. Assuming that such conditions favor either increased pollen movement or increased fertilization, this is consistent with the hypothesis that pollen limitation plays a large and significant role in the highly variable seed crops characteristic of masting in these species. Overall, the proportion of total variance in the acorn production of individual trees explained by these two indices of pollen availability ranged from 29% to 39%. These results support the hypothesis that pollen limitation plays an important role in causing variance in seed production in wind-pollinated oaks.

Over at least the last half century, rates of recruitment of young trees in California's oak woodlands have been found to be extremely uncommon, and lower than rates of mortality of mature trees for some species such as valley oak, Quercus lobata. This phenomenon threatens the integrity of these ecosystems. As part of a larger study of factors limiting seedling and sapling recruitment, we conducted measurements on 4 yr. old seedlings established from experimental plantings to address the questions: What effects does summer drought have on water relations, gas exchange and photosynthetic characteristics of Q. lobata (valley oak) and Q. agrifolia (coast live oak) seedlings, and how are these effects related to a successful transition from seedling to sapling life-stage? Results indicate that seedlings of both species were subject to much lower water potentials (PDXPPs) than nearby adult trees. Seedlings of Q. lobata had lower PDXPPs than Q. agrifolia seedlings. In both species, seedlings had lower maximum rates of gas exchange per unit leaf area than trees. Seedlings and trees of Q. agrifolia had lower gas exchange rates than seedlings and trees of Q. lobata. Chlorophyll fluorescence characteristics indicated leaves of seedlings and trees of both species are well protected from permanent photoinhibitory damage during summers. Variances in these physiological characteristics are being analyzed to determine relationships with seedling growth and survival.

While there has been little research on California grassland birds, their populations are thought to be declining due to habitat loss, fragmentation, and degradation. The aim of this study is to look at the association between grassland birds and their landscape-scale habitat matrix. The habitat is a mosaic of valley grasslands with blue oak and coast live oak woodlands. This study is part of a larger long-term monitoring project in collaboration with the East Bay Regional Park District, investigating the influence of livestock grazing on plant species composition and vegetation structure. In particular, this study analyzes total bird species richness and the presence of Grasshopper Sparrows (Ammodramus savannarum), Horned Larks (Eremophila alpestris), Western Meadowlarks, (Sturnella neglecta), and Savannah Sparrows (Passerculus sandwichensis) in response to patch size, perimeter: area ratio, proportion of different cover types, and distance to urban development. These landscape variables are being analyzed at three spatial scales: 500m, 1km, and 2km radii from the point count center. We expect to see a positive association between total bird species richness and proportion of surrounding oak woodland. We expect a positive association between grassland birds and patch size, distance to urban development and distance to roads and trails. Finally, we expect to see a negative association between grassland birds and perimeter:area ratio.

We quantified bird populations with 100-m radius point counts at two study areas on public lands between late March and mid-May 2004. The study areas were in Sierra Nevada foothill habitats in Yuba and Tehama counties. Eighty-nine species of birds were detected, and population and community indices were derived for Blue Oak (Quercus douglasii) Woodland, Annual Grassland, and Montane Hardwood habitats. Habitat attributes at the study areas were most similar in Annual Grassland and Montane Hardwood habitats, while Blue Oak Woodlands in Yuba County had more tree cover (P < 0.001) and larger diameter and taller trees (P < 0.001) than Tehama County. Herbaceous cover was greater in Yuba County for Annual Grasslands (P < 0.001) and Blue Oak Woodlands (P < 0.001) than Tehama County. Differences in bird communities tracked habitat differences as bird populations were more numerous and there were more species in Yuba County compared to Tehama County for Blue Oak Woodlands (P < 0.001) and Annual Grasslands (P < 0.05). In Blue Oak Woodlands, 16 species were more numerous (P < 0.05) and 14 species had greater detection rates (P < 0.05) in Yuba County than Tehama County. Only three and two species, respectively, were more numerous (P < 0.05) or had greater detection rates (P < 0.05) for Blue Oak Woodlands in Tehama County compared to Yuba County. Abundance estimates and detection rates were significantly greater (P < 0.05) for only three species and one species, respectively, for Annual Grassland and Montane Hardwood habitats in Yuba County than Tehama County. These data, while limited to a single year, provide population indices and wildlife-habitat relationship information to assist efforts to conserve wildlife habitats in the Sierra Nevada foothills.

We examined selection for trees species used for nesting by bird species breeding in ponderosa pine forests in the Sierra National Forest, California. We located nests of all bird species on four sites and examined tree species used for nesting. Overall, hardwood tree species were used as nest substrates more than expected compared to available trees. Forty-five percent of all nests found were in hardwood tree species. Twenty-one species used hardwoods for nesting, with nineteen using California black oaks (Quercus kelloggii) and twelve using canyon live oak (Quercus chrysolepsis). Hardwood snags were seldom used. Of ten species with adequate sample sizes to examine individually, six species used hardwoods more than expected, two species used conifers more than expected, and two species showed no overall preference. Western Wood-Pewees (Contopus sordidulus), Warbling Vireos (Vireo gilvus), and Cassin’s Vireos (Vireo cassinii) nested more often in California black oaks (73, 71, and 43% of nests, respectively). Hutton’s Vireos (Vireo huttoni) and American Robins (Turdus migratorius) used predominantly canyon live oaks (Quercus chrysolepsis) (83 and 36% of nests, respectively). Anna’s Hummingbirds (Calypte anna) showed selection for both oak species. Western Tanagers (Piranga ludoviciana) and Purple Finches (Carpodacus purpureus) nested predominantly in ponderosa pine (79 and 81% of nests, respectively). Although Hammond’s Flycatchers (Empidonax hammondii) showed no significant preference for hardwood vs. conifer species compared to available trees, 81% of nests were in conifers. Stellar’s Jays (Cyanocitta stelleri) also showed no overall preference for conifers or hardwoods, although 31% of nests were found in canyon live oaks. Our results underscore the importance of hardwood trees species, particularly oaks, to birds breeding in ponderosa pine forests. These shade-intolerant species need openings in the understory that were probably created primarily by fire in the past. We recommend managing for retention of mature oaks in ponderosa pine stands and activities that encourage recruitment of oaks, such as prescribed fire.

Developing efficient, robust biological indicators of landscape change is a challenge for conservation assessment, especially in rapidly changing landscapes like California’s oak woodlands. Birds are often used as indicators of environmental change, despite some well-founded concerns, because they are conspicuous, they have a very broad constituency, and standard census protocols have been developed. Simple standardized counts of birds may provide an incomplete picture of the response of bird populations to environmental change. Unless specific efforts are made to assess density or reproductive success, it is very difficult to distinguish viable population  source  habitat from non-viable population  sink  habitat. We evaluated and compared two standard methods for assessing bird populations in oak woodlands of Sonoma and Napa counties, a region of northern California in which habitat conversion and fragmentation are rapidly changing the character of remaining oak woodlands. We collected reproductive success information on over 300 nests of over 30 species of birds nesting in oak woodlands of Sonoma and Napa counties in 2003 and 2004. We compare these data with standard point counts samples from the same sites in 2003 and 2004. Our results show that the two methods yield very different results with point counts sampling many more species than could be efficiently sampled by labor-intensive nest monitoring. Nevertheless, nest monitoring provided more accurate information on several species. For example, European Starlings were detected at only half of our sites through point counts, but were found at all sites where nests were monitored. We recommend that nest monitoring be used in conjunction with point counts to evaluate avifaunal response to environmental change in oak woodlands, and that nest monitoring focus on a small suite of species hypothesized to exhibit changing reproductive success.

Widespread conversion of oak woodland to row-crop agriculture in central coastal California the past two decades has resulted in landscapes where remaining native habitats are highly altered, with potential consequences for wildlife diversity and the long-term sustainability of these communities. In winter 2002, we established study sites in cropland and in undeveloped rangeland in Santa Barbara and San Luis Obispo counties. Between April and June on each of the five study sites, we surveyed 15 50-m-radius point count stations four times each and searched systematically for nesting cavities. We measured the dbh, height, and species of all trees with cavities and checked occupied cavities every three days to determine nest initiation date, clutch size, and nest success. We used Poisson and regression and ANOVA to relate songbird diversity to site characteristics and use of nesting cavities at the cropland and rangeland locations. Results indicate that of the 75 bird species that we detected by point counts, 16 that the literature usually describes as sensitive to fragmentation were recorded more at rangeland sites. We recorded 10 others that may be adapted to cropland. We counted 1.5 times as many cavities in valley oak as in coast live oak trees. For coast live oak trees, those with cavities were significantly larger than those without cavities. Results of this study will assist the scientific, policy, and agricultural communities with guidelines for management and restoration of intensively managed landscapes.

We used behavioral experiments to evaluate competition for nest sites and the extent to which European Starlings (Sturnus vulgaris) are seen as a threat by native bird species at the San Joaquin Experimental Range, Madera County, CA. We quantified the level of aggressive behavior of native cavity-nesting birds to starlings at active nests in trees and nest boxes. In 2000, we presented a life-like model of a starling at active nests of native cavity-nesting species, facing the nest. In 2001, a House Sparrow (Passer domesticus) was used as a control, with both species presented at nests. Models were mounted on a piece of wood and placed above the nest box, facing away from the box entrance. Responses were coded as sporadic scolding, continuous scolding, aggressive flight, and attack. Placement of the model facing into the cavity elicited a relatively more aggressive response than that elicited by a model placed on top of the nest box. Oak Titmouse (Baeolophus inornatus) responses consisted entirely of scolding except for one aggressive flight toward the model, but Western Bluebird (Sialia mexicana), Ash-throated Flycatcher (Myiarchus cinerascens), and Acorn Woodpecker (Melanerpes formicivorus) responses included physical attacks on the starling model. The European Starling model elicited a significantly stronger response from Oak Titmice and Western Bluebirds than the House Sparrow model. Although sample sizes were small, Ash-throated Flycatcher responses to starling models included both aggressive flights and attacks, while neither behavior was seen in response to the House Sparrow model. Acorn Woodpecker response to the starling model at three nests tested was strong, consisting of attacks on the model by up to seven members of the group. Over the two years of the study, eight nests of five species were usurped and/or depredated by starlings. Our results show that Western Bluebirds, Oak Titmice, Ash-throated Flycatchers, and Acorn Woodpeckers recognize starling as potential aggressors. We recommend the use of nest boxes to reduce the potential for nest-site competition between starlings and some native cavity-nesting species.

Quercus garryana, along the west coast of North America, has a similar range of habitats, phenology, and lack of regeneration as Q. pyrenaica in Spain. Both species range from Mediterranean-type climates to temperate, summer-wet climates, but this study was limited to stands in Mediterranean-type climates. Although the degree of genetic similarity and phylogenetic relatedness between these two species have apparently not been studied, the ecological convergence is remarkable. In this paper, we compare the age structure of these oak species and their contribution to stands in terms of density, canopy cover, basal area, frequency, pattern, and associated species. In Spain, the range of undisturbed stands visited included relatively young stands, undisturbed for less than 100 years, to old-growth stands, undisturbed for 200+ years. In California, the range of undisturbed stands was more limited to those >150 years old. Sampling was conducted using a variety of methods, including belt transects and circular plots. All other species were noted and their relative abundances estimated. Tentative conclusions are considered regarding the convergent properties of Q. garryana and Q. pyrenaica and their associated vegetation types.

Coast live oak (Quercus agrifolia née) occurs in the Coast Range mountains from southern Mendocino County, California to the Sierra San Pedro Martir, Baja California, Mexico. In northern and central California, coast live oak is suffering heavy mortality as a result of infection by Phytophthora ramorum. Earlier work indicated biochemical differentiation of central coast populations from northern and southern California, an important hybrid zone in northern California and possible differential response to inoculation between populations from northern and southern California. Here we report results of a phylogeographic study of this species using chloroplast and nuclear microsatellite markers. The chloroplast genome is inherited uniparentally and is particularly useful for identifying the migration of lineages over evolutionary time scales that include the important events of the Pleistocene glaciations. We sampled 41 populations of coast live oak from throughout the geographic range and analyzed nuclear and chloroplast microsatellies for over 500 individuals. Based upon the chloroplast haplotype distributions, at least 4 biogeographic groups should be recognized within the California range of this species: 1. three common haplotypes in the San Francisco Bay area and northwards, of which only one is found outside of this region in San Luis Obispo Co. and at Ojai in Ventura County. 2. three haplotypes in the Monterey-Big Sur region that were unique to this region, 3. two haplotypes unique to the coastal mountains of San Luis Obispo County and, 4. six haplotypes in extreme southern California, of which 5 were unique to the region. Nuclear microsatellite markers revealed reduced levsl of genetic structure as a result of pollen dispersal. We discuss the roles of seed and pollen dispersal in the evolution of populations of coast live oak and how this information may help in developing strategies for studies of resistance to the sudden oak death disease.

Many plant species are of sufficient ecological importance to merit species-specific conservation plans.  For species threatened by land use change, nature reserves will be a key element of their conservation strategy.  It can be particularly challenging to identify which sites to preserve when species are broadly distributed, such as California valley oak (Quercus lobata Neé), and include geographically structured genetic variation that needs to be taken into account.  Here, we investigate the use of molecular genetic markers to identify populations of high genetic diversity or high genetic uniqueness for designing a network of conservation reserves.  In an earlier study, we sampled individuals from 32 valley oak populations distributed throughout the species range and determined their genotypes based on chloroplast haplotypes (six primers) and nuclear genetic markers (7 primers).  Our findings from that study indicated a north-south gradient in genotypes, significant genetic differences between western and eastern populations, and two centers of high allelic richness and genotypic uniqueness.  For this paper, we analyzed the data from a reserve design perspective.  We conclude that a minimum of six of 32 locations would be needed to represent genetic variation as indicated by chloroplast DNA and at least 10 of 55 locations would be required to represent genetic variation as indicated by allelic variation in nuclear DNA. The analysis suggests that an efficient reserve network for protecting genetic variation in the species can be developed by including sites of high allelic diversity that are also complementary in their allelic composition. While a molecular genetic approach will enhance the development of a conservation plan for valley oak, conservation priorities should also take into account the regions that are in jeopardy, conservation costs and opportunities.

The complexity of savanna ecosystems with the shared dominance between woody and herbaceous primary producers and the associated spatial and temporal heterogeneity poses challenges to measuring such fundamental ecosystem characteristics as net primary productivity (NPP). We address these challenges in the blue oak savanna in California by constructing comprehensive estimates of NPP for three adjacent watersheds. We directly assessed annual biomass increment of all plant components in 12 randomly stratified plots (380 m2) per watershed and used Monte Carlo techniques to quantify uncertainty in these estimates. Annual estimates of mean NPP for the three experimental watersheds during 2001-02 ranged from 4.35 to 5.69 Mg ha-1 yr -1 of dry biomass. Confidence intervals (95%) around these means averaged ± 33% of the mean. On average, belowground NPP accounted for 22% of total NPP. In general, there was much greater uncertainty in the belowground estimates. Trees accounted for approximately 50% of aboveground NPP in all three watersheds. Across these watersheds, tree productivity increased in a linear fashion with canopy closure. In contrast, herb productivity was nearly constant for relatively open sites (canopy closure < 40%) and then monotonically declined as canopy closure increased. The result is that total NPP increased gradually from the most open sites in the watersheds to a maximum around 55% canopy closure. Our biometric approach to measuring productivity in the blue oak savanna provided an initial estimate of the resolution of any changes in NPP that we can expect to detect at the scale of a small watershed.

Current theory suggests Mediterranean climates favor evergreen over winter-deciduous vegetation because the former can sustain photosynthetic production during the mild winter when water is not limiting. However, winter-deciduous trees, especially oaks, are not uncommon in some Mediterranean regions, and frequently replace evergreens in the most xeric upland savannas as well as in very mesic riparian forests. To explain this anomalous distribution of deciduous dominance, I have developed a simple mechanistic model of leaf habit for Mediterranean climates. The model is based on the relative photosynthetic C economy of evergreen vs. winter-deciduous leaves during two developmental stages, and predicts a bimodal distribution of deciduous-dominance along a moisture gradient. The higher intrinsic photosynthetic capacity of deciduous relative to evergreen leaves should allow drought-adapted deciduous species to grow roots more rapidly over the first wet season, and thereby recruit in environments too xeric to support otherwise similar evergreens. Where moisture is sufficient for seedlings of both habits to establish, the advantage should go to the strategy that maximizes photosynthetic C gain, and thus growth, over a year. If the summer drought period is very short (e.g., riparian forests), winter-deciduous leaves achieve the larger annual C gain because they can extend their intrinsic photosynthetic advantage over a longer season each year. Where the drought period becomes sufficiently long, however, evergreen leaves gain the advantage because they can continue photosynthetic production through winter when deciduous species are inactive, and thereby more than compensate for their lower spring-time assimilation rates. I am currently testing the predictions of this model for several evergreen and winter-deciduous California oaks. I will briefly discuss these projects as well as the management implications of this work.

Island ecosystems commonly harbor a diversity of unique, endemic species that are particularly susceptible to the effects of introduced organisms. Historically, no large mammals existed on Santa Catalina Island, California, but several species of ungulates were introduced after the 1850’s including mule deer (Odocoileus hemionus) in the 1930s. Although mule deer have been present on Santa Catalina Island for 70+ years and are known to consume native trees and shrubs, there are no quantitative data on their diets and foraging behavior useful for assessing impacts to island endemic trees and shrubs, particularly the island scrub oak (Quercus pacifica), a species that is the current focus of an island-wide restoration program. Working together with the Santa Catalina Island Conservancy, a private non-profit organization responsible for the stewardship of 88% of the island, we are investigating browse impacts of introduced mule deer to trees and shrubs in island scrub oak woodland habitats. Our two year study was initiated in January 2005, and research activities include (1) monitoring oak seedling survival in sixteen 30-m X 30-m experimental plots where 1600 nursery grown seedlings (100 seedlings/plot) were planted in spring 2005, (2) assessing oak tree regeneration potential from data on number and size of seedlings detected along 4-m X 40-m seedling transects randomly positioned in woodland habitats, (3) estimating diets of mule deer from microhistological analyses of fecal pellets systematically collected each month of the study, and (4) developing recommendations for managing mule deer consistent with the Conservancy’s mission to maintain and restore native plants and animals of Santa Catalina Island. Our study is one of thirteen complementary projects focused on understanding the ecology of island scrub oak woodland ecosytems on Santa Catalina Island. We will describe and discuss preliminary data from the study in the context of this larger oak woodland restoration effort.

European red foxes (Vulpes vulpes) were introduced into lowland California in the 1880s for fur farming and hunting. The introduced foxes quickly spread throughout much of the state and have been implicated in the decline of several federally threatened and endangered ground-nesting bird species. Red foxes have been present in the East Bay for 25-30 years and they are regularly sighted in coastal wetlands and in the Oakland and Berkeley hills. We were interested in documenting the extent of the invasion into oak woodlands in the East Bay, as foxes may negatively impact both prey populations and native carnivores such as gray foxes and bobcats. We surveyed fire roads and hiking trails in core oak woodland sites in Contra Costa and Alameda counties for carnivore scat. All scat samples were collected, and their locations entered into a GIS. To positively distinguish the scat to species, DNA was extracted, amplified, and identified using PCR and RFLP. Four carnivore species were identified, including coyote, gray fox, red fox, and bobcat. Red foxes were only present in woodlands that were adjacent to urban or suburban development, and were not present in more rural sites. This may be due to larger relative numbers of coyotes in these more remote sites, or to more variable prey and water availability. This research is providing managers essential information about red fox distribution, habitat requirements, and interactions with other carnivores, which can be used to better monitor and eventually control red fox invasions and subsequent impacts to native species.

Water Pollution is defined as the corruption of ecosystems, human health, and local economies by the inappropriate introduction of contaminants into water-ways. Although counter-intuitive, water itself has become a pollutant in California’s hardwood rangeland, with ecological consequences that exceed the impacts predicted by models of global climate change. The greatest manifestations of this problem are year-round flows of surface water in streams that previously dried-out in summer months. The primary cause is excess and inefficient irrigation of suburban landscaping, which is often applied at rates equivalent to 50 inches of annual rainfall. Increases in summer flows at USGS gauging stations can be correlated with housing construction in areas like the Santa Clarita Valley, creating entirely new types of woodland ecosystems along the wildland/urban interface. These dysfunctional wetlands create a link between human and wildland ecosystems, putting both human and wildlife populations at risk. Although many of these woodland ecosystems are less than 20 years old, they played a major role in the spread of West Nile Virus and could play a role in the spread of Bird Flu, Dengue fever, and Malaria.

Newly-restored valley oak (Quercus lobata) woodlands often have serious invasions of nonnative annual grasses and thistles. Although prescribed fire can be an effective control for these exotics, managers may be reluctant to use fire if it causes substantial mortality of recently-planted valley oak trees. We studied the effects of prescribed fires on the survival and subsequent growth of five- and six-year-old valley oaks near Davis, California. One set of trees was burned in summer 2003 to control yellow star thistle, a second set was burned in spring 2004 to control annual grasses, and a third set was left unburned. Prior to burning, measurements were taken on each tree's height, diameter, and understory grass biomass. Fire temperatures were measured using temperature-sensitive paints on ceramic and/or aluminum tags hung on each tree. Survival, coppicing, height, and diameter were measured for each tree several times during the two years following the burns. Both burns showed similar patterns. Although smaller trees suffered more damage (top-kill and coppicing) than larger trees, very few oaks died as a result of either fire (3-4%). Tree size was initially reduced by the fires, and although the burned trees were approaching the size of the control trees two years after the burns, they had not yet returned to the size they would have been had they not been burned. However, the growth rates of burned valley oak trees quickly returned to, and often exceeded, the growth rates of unburned control trees, suggesting the potential for full recovery. Since the trees in our experiment suffered very little mortality and appeared robust two years after the fire, the presence of recently-planted valley oak trees does not appear to preclude the use of a single prescribed burn to control invasive grasses or yellow star thistle.

A major vegetation survey of California was conducted in the 1930s under the direction of Albert Wieslander. His crews mapped one third of the state, creating detailed renditions of vegetation patterns on topographic quadrangles. Recorded in these maps are the dominant species of each vegetation polygon. We developed methods to render the historical maps to GIS, and processed over 25,000 square kilometers of the central Sierra Nevada that form a transect from the Central Valley to the east side. We queried the resulting GIS to develop maps of individual oak species for the region. We present the historical patterns and extents for Quercus douglasii, Q. kelloggii, Q. lobata, Q. wislizeni, and Q. chrysolepis. We compare the historical extents to modern extents, as measured by modern digital vegetation maps, predominantly CalVeg. Preliminary results indicate a loss of over 50% of Blue Oak woodlands, and an increase in Montane hardwoods.

Conservation easements are an increasingly popular strategy for conserving biological diversity on private land. Recent recommendations for increasing easement effectiveness have focused on the need for ecological monitoring on easement properties. Here we describe an approach for long-term ecological monitoring, which can provide critical information on persistence and viability of biological diversity in protected areas. We sampled understory plant composition on five of The Nature Conservancy's conservation easements in the foothills of Mt. Lassen. We established 73 plots of 10 meter radius and performed pin drops as well as a full species composition macroplot for each plot in 2005 and 2006. We compared diversity and relative abundance of native and nonnative species in blue oak woodlands and the interspersed forb-dominated communities. Our analysis of community diversity and relative abundance provides a baseline for future monitoring and guidelines for the design and intensity of future monitoring efforts.

Oaks 2040: The Status and Future of Oaks in California is a research project sponsored by the California Oak Foundation and the Goldman Foundation. This study grew out of the realization that we cannot adequately protect California’s oak woodlands without a clear understanding of oak geographic distribution, botanical diversity, and conservation risks. To that end, we have completed an exhaustive review of available vegetation maps to compile the first comprehensive up-to-date oak forest and woodland map of 10  oak types  within California. Then, by populating the mapped oak types with statewide forest survey data from the Forest Inventory and Analysis (FIA) program, which has established over 800  cluster  plots within these oak woodlands and oak forests since 2001, we compiled type-stratified summary inventories. Finally, by examining past and future human population expansion, land use issues and other trends, we determine which oak woodlands are most likely to face development pressures or be otherwise "at risk" between now and 2040. Throughout this report, findings are reported for each of six regions: North Coast, North Interior, Central Coast, Sacramento, San Joaquin, and Southern. This paper is a summary of a more extensive report soon to be published by the California Oak Foundation.

In the 1920s and 1930s A.E. Wieslander and his crew surveyed much of the California landscape. The data they collected, known as the Wieslander Vegetation Type Mapping collection (VTM), contains vegetation data, detailed vegetation type maps and an extensive photograph collection. Our research will examine VTM plots in the Quercus agrifolia-Umbellularia californica woodlands surrounding San Francisco Bay to determine the role of plant community change and landscape change in the Sudden Oak Death disease complex. The causal agent of this disease, Phytophthora ramorum, has emerged as a significant pathogen within the Quercus agrifolia-Umbellularia californica woodlands. Initially detected in 1994 in Marin County, it has since spread to fourteen counties in California and one county in Oregon. There are still many unanswered questions regarding the ecology of this pathogen including its relationship to historical conditions and its spatial pattern across the landscape. Our research will integrate plot based plant community data with landscape level spatial data to increase our understanding of presence and spread of SOD both spatially and temporally. In the winter of 2005, we selected 15 VTM plots across a gradient of Sudden Oak Death infection and from within the Quercus agrifolia-Umbellularia californica woodlands. We have begun the process of re-sampling each plot location to assess plant community and stand structure to determine if certain plant communities are more prone to Phytophthora ramorum infection. We will a