Frank W. Davis

NATURE RESERVES: DO THEY CAPTURE THE FULL RANGE OF AMERICA'S BIOLOGICAL DIVERSITY?

Less than 6% of the coterminous United States is in nature reserves. As- sessment of the occurrence of nature reserves across ranges of elevation and soil productivity classes indicates that nature reserves are most frequently found at higher elevations and on less productive soils. The distribution of plants and animals suggests that the greatest number of species is found at lower elevations. A preliminary assessment of the occurrence of mapped land cover types indicates that ;60% of mapped cover types have ,10% of their area in nature reserves. Land ownership patterns show that areas of lower elevation and more productive soils are most often privately owned and already extensively converted to urban and agricultural uses. Thus any effort to establish a system of nature reserves that captures the full geographical and ecological range of cover types and species must fully engage the private sector.

Reserve selection as a maximal covering location problem

Many alternative approaches have been proposed for setting conservation priorities from a database of species (or communities) by site. We present a model based on the premise that reserve selection or site prioritization can be structured as a classic covering problem commonly used in many location problems. Specifically, we utilize a form of the maximal covering location model to identify sets of sites which represent the maximum possible representation of specific species. An example application is given for vertebrate data of Southwestern California, which is then compared to an iterative solution process used in previous studies. It is shown that the maximal covering model can quickly meet or exceed iterative models in terms of the coverage objective and automatically satisfies a complementarity objective. Refinements to the basic model are also proposed to address additional objectives such as irreplaceability and flexibility.

Pollen movement in declining populations of California Valley oak, Quercus lobata: where have all the fathers gone?

The fragmented populations and reduced population densities that result from human disturbance are issues of growing importance in evolutionary and conservation biology. A key issue is whether remnant individuals become reproductively isolated. California Valley oak (Quercus lobata) is a widely distributed, endemic species in California, increasingly jeopardized by anthropogenic changes in biota and land use. We studied pollen movement in a savannah population of Valley oak at Sedgwick Reserve, Santa Barbara County, to estimate effective number of pollen donors (Nep) and average distance of effective pollen movement (δ). Using twogener, our recently developed hybrid model of paternity and genetic structure treatments that analyses maternal and progeny multilocus genotypes, we found that current Nep = 3.68 individuals. Based on an average adult density of d= 1.19 stems/ha, we assumed a bivariate normal distribution to model current average pollen dispersal distance (δ) and estimated δ= 64.8 m. We then deployed our parameter estimates in spatially explicit models of the Sedgwick population to evaluate the extent to which Nep may have changed, as a consequence of progressive stand thinning between 1944 and 1999. Assuming that pollen dispersal distance has not changed, we estimate Nep was 4.57 individuals in 1944, when stand density was 1.48. Both estimates indicate fewer effective fathers than one might expect for wind‐pollinated species and fewer than observed elsewhere. The results presented here provide a basis for further refinements on modelling pollen movement. If the trends continue, then ongoing demographic attrition could further reduce neighbourhood size in Valley oak resulting in increased risk of reproductive failure and genetic isolation.

Regression Tree Analysis of satellite and terrain data to guide vegetation sampling and surveys

. Monitoring of regional vegetation and surface biophysical properties is tightly constrained by both the quantity and quality of ground data. Stratified sampling is often used to increase sampling efficiency, but its effectiveness hinges on appropriate classification of the land surface. A good classification must be sufficiently detailed to include the important sources of spatial variability, but at the same time it should be as parsimonious as possible to conserve scarce and expensive degrees of freedom in ground data. As part of the First ISLSCP (International Satellite Land Surface Climatology Program) Field Experiment (FIFE), we used Regression Tree Analysis to derive an ecological classification of a tall grass prairie landscape. The classification is derived from digital terrain, land use, and land cover data and is based on their association with spectral vegetation indices calculated from single-date and multi-temporal satellite imagery. The regression tree analysis produced a site stratification that is similar to the a priori scheme actually used in FIFE, but is simpler and considerably more effective in reducing sample variance in surface measurements of variables such as biomass, soil moisture and Bowen Ratio. More generally, regression tree analysis is a useful technique for identifying and estimating complex hierarchical relationships in multivariate data sets.

FIRE, SOIL HEATING, AND THE FORMATION OF VEGETATION PATTERNS IN CHAPARRAL

We documented patterns of surface heating associated with chaparral fire to characterize funda- mental scale variation in the intensity of this stand-replacing disturbance. To test how this variation may influence community structure, we studied its effect on the soil seed bank and the distribution of seedlings and resprouts that emerged after fire. To evaluate the long-term significance of initial patterns, we monitored vegetation development for 4-5 yr, thereby encompassing the dynamic portion of the chaparral fire cycle. We studied two stands on level uniform terrain before, during, and after fall fires. Stands were dominated by chamise (Adenostoma fasciculatum), a postfire seeder/sprouter. Nonsprouting Arctostaphylos and Ceanothus spp. were also present. Preburn vegetation, seed populations, soil heating, and postburn plant growth were analyzed along transects of contiguous 1-m 2 plots, so that we could block them together incrementally to identify scale dependence of patterns. In addition, we directly compared heating effects under the fuel array with those just outside by establishing plots in canopy gaps, under the adjacent canopy, and in gaps created and eliminated by reciprocally translocating fuel. Pre- and postburn seed populations were estimated in soil samples collected from all plots. The proportion of seed that survived above and below 2.5 cm in the soil was determined in a subset of plots. The amount and distribution of canopy fuel that collapsed during fire and smoldered on the ground caused pronounced spatial variation in total surface heating. The strength of relationships among patterns of soil heating, preburn canopy, surviving seeds, and seedlings and herbaceous resprouts was consistently most pronounced in blocks 3-5 m long. At this scale, postburn patterns were strongly negatively associated with the amount of preburn canopy and the pattern of soil heating this fuel created. Seedlings or herbaceous resprouts of numerous species were abundant where soil heating was relatively low, most notably in natural and created canopy gaps. Conversely, areas where dense canopy occurred before fire, especially gaps displaced by fuel addition, were barren except for occasional Arctostaphylos and Ceanothus seedlings. These obligate postfire seeders, along with the subshrub Helianthemum scoparium , had more deeply buried seeds, and some of them were able to survive where soil heating was prolonged. However, Helianthemum did not emerge from depth. Seedlings of Arctostaphylos and Ceanothus nearest Adenostoma burls survived significantly better when Adenostoma failed to resprout. This was common in one burn where heating was relatively high and burl size was small. Seed mortality prevented Adenostoma seedling emergence from occurring where its seeds were most abundant prior to fire, which was in proximity to its burls. Adenostoma seedlings did emerge in areas of lower soil heating, but their survival was inversely related to the density of Helianthemum seedlings. No shrub seedlings emerged after the first year following fire because their seed banks were exhausted by fire-induced mortality and/or germination. After 4-5 yr, few young Adenostoma remained. The combination of seedling and resprout regen- eration allowed this shrub to maintain dominance, but to a lesser extent in the older stand. Our results support a vegetation pattern-process model in which local species distributions after fire in Adenostoma chaparral are antecedently linked to the physical and chemical properties of the canopy. These control the nature of combustion, the soil heating that results, and the distribution of seeds and resprout tissues that survive. The vegetation develops entirely from these sources, so fire-induced patterns are manifest in the long-term structure of this vegetation.

Modeling vegetation pattern using digital terrain data

Using a geographic information system (GIS), digital maps of environmental variables including geology, topography and calculated clear-sky solar radiation, were weighted and overlaid to predict the distribution of coast live oak (Ouercus agrifolia) forest in a 72 km2 region near Lompoc, California. The predicted distribution of oak forest was overlaid on a map of actual oak forest distribution produced from remotely sensed data, and residuals were analyzed to distinguish prediction errors due to alteration of the vegetation cover from those due to defects of the statistical predictive model and due to cartographic errors.Vegetation pattern in the study area was associated most strongly with geologic substrate. Vegetation pattern was also significantly associated with slope, exposure and calculated monthlysolar radiation. The proportion of observed oak forest occurring on predicted oak forest sites was 40% overall, but varied substantially between substrates and also depended strongly on forest patch size, with a much higher rate of success for larger forest patches. Only 21% of predicted oak forest sites supported oak forest, and proportions of observed vegetation on predicted oak forest sites varied significantly between substrates. The non-random patterns of disagreement between maps of predicted and observed forest indicated additional variables that could be included to improve the predictive model, as well as the possible magnitude of forest loss due to disturbances in different parts of the landscape.

Interactions of Factors Affecting Seedling Recruitment of Blue Oak (Quercus Douglasii) in California

Acorn germination and seedling recruitment of blue oak (Quercus douglasii) were studied in relation to postdispersal predators, planting depth, oak canopy cover, slope angle and aspect, and herb layer. Removal of acorns and seedling recruitment from surface—sown and buried acorns were measured for 2 yr at two sites, a north—slope forest and a ridgetop savanna. Nested exclosures were used to measure acorn predation by birds, mice, gophers, deer, and cattle. The rate of seedling recruitment (Ps) was related to treatment variables using cluster analysis combined with stepwise logistic regression. Average acorn fall ranged from 3.5 to 58.7 acorns/m2, with germination rates varying from 28 to 85% in different years. 8172 sown acorns yielded 2922 (35.76%) seedlings, but Ps varied from 0.09 to 0.71 among the different treatments. Year, site, and rodents interacted strongly to affect seedling recruitment. Ps averaged 0.48 during a cool wet year compared to 0.23 in drier years. Low correlations between seedling recruitment rates for the same plot in different years were due to annual changes in the distribution of favorable microsites and patchy, unpredictable acorn predation. Recruitment rates for buried acorns were twice those of surface—sown acorns due to improved germination and reduced predation. Pocket gophers (Thomomys bottae) were the major predator of buried acorns. Surface acorns suffered high mortality from dying and overheating, as well as from predation by mice, gophers, and cattle. Seedling recruitment was positively associated with increasing canopy cover and more mesic microsites at the low elevation site, but was negatively associated with these factors at the cooler, high elevation site. Hierarchichal classification combined with stratum—specific logistic regression models was important in revealing strong interactions among seedling recruitment factors.

Vegetation dynamics, fire, and the physical environment in coastal Central California

Current concepts of vegetation dynamics include that of the shifting land- scape mosaic, but evidence for shifting mosaics in disturbed and undisturbed systems is primarily based on negative spatial relationships among adults and recruits, and not on measurements of actual shifts over time. We used aerial photographs to measure transition rates as evidence for mosaic shifts among grassland, coastal sage scrub, chaparral, and oak woodland communities in central coastal California between 1947 and 1989. In unburned plots without livestock, transition from grassland to coastal sage scrub was 0.69% per year, coastal sage scrub to oak woodland was 0.30% per year, and oak woodland to grassland was 0.08% per year. These transition rates, considered together, indicate that vegetation patterns may be dynamic on landscapes dominated by these communities. In burned plots without livestock, and in unburned plots where livestock were not excluded, transition rates were lower, except for the conversion of oak woodland to grassland. In burned plots, a high rate of transition of coastal sage scrub to grassland was measured. Markov chain models predicted much less directional change in community proportions in either grazed or burned conditions than in ungrazed, unburned conditions. Some transition rates varied with substrate and topographical position, indicating that fire, grazing, and the physical environment interacted to determine direction and rate of vegetation change. Variation in transition on different substrates suggests that only portions of the vegetation of these landscapes may be dynamic, with some patches in certain combinations of environment and disturbance that change rapidly, and other patches that remain static as edaphic or topographic climax communities.

Gene movement and genetic association with regional climate gradients in California valley oak (Quercus lobata Née) in the face of climate change

Rapid climate change jeopardizes tree populations by shifting current climate zones. To avoid extinction, tree populations must tolerate, adapt, or migrate. Here we investigate geographic patterns of genetic variation in valley oak, Quercus lobata Née, to assess how underlying genetic structure of populations might influence this species’ ability to survive climate change. First, to understand how genetic lineages shape spatial genetic patterns, we examine historical patterns of colonization. Second, we examine the correlation between multivariate nuclear genetic variation and climatic variation. Third, to illustrate how geographic genetic variation could interact with regional patterns of 21st Century climate change, we produce region‐specific bioclimatic distributions of valley oak using Maximum Entropy (MAXENT) models based on downscaled historical (1971–2000) and future (2070–2100) climate grids. Future climatologies are based on a moderate‐high (A2) carbon emission scenario and two different global climate models. Chloroplast markers indicate historical range‐wide connectivity via colonization, especially in the north. Multivariate nuclear genotypes show a strong association with climate variation that provides opportunity for local adaptation to the conditions within their climatic envelope. Comparison of regional current and projected patterns of climate suitability indicates that valley oaks grow in distinctly different climate conditions in different parts of their range. Our models predict widely different regional outcomes from local displacement of a few kilometres to hundreds of kilometres. We conclude that the relative importance of migration, adaptation, and tolerance are likely to vary widely for populations among regions, and that late 21st Century conditions could lead to regional extinctions.

Sources of variation in radiometric surface temperature over a tallgrass prairie

Abstract Numerous studies have noted a strong negative correlation between radiometric surface temperature and spectral vegetation indices such as the NDVI, and have suggested that this relationship might be exploited in strategies to model land surface energy balance from satellites. These studies have been largely empirical in nature and the relationships among remotely sensed data, land surface properties, and land surface energy balance that produce this phenomenon remain unclear. We studied the relationship between radiometric surface temperature and NDVI over a tallgrass prairie in northeastern Kansas. The study site included a mix of landcovers, with fractional vegetation cover and exposed soil backgrounds over much of the site. We observed a persistent negative correlation between radiometric surface temperature and NDVI, but found that the relationship was highly date- and time-specific. In this context, the relationship between surface temperature and NDVI was observed to depend on landcover type, and a significant proportion of the total variance in both NDVI and radiometric surface temperature was explained by stratifying the data by landcover class. More importantly, our results show the relationship between surface temperature and NDVI to have little association with surface energy balance for data sets acquired from aircraft and helicopters on several dates during the growing seasons of 1987 and 1989. Based on results from a simulation model of the soil-canopy-sensor system, we hypothesize the observed covariance between radiometric surface temperature and NDVI to be largely caused by temperature differences between the soil background and vegetation canopy and by variation in fractional vegetation cover. This hypothesis is supported by evidence showing soil moisture to be an important secondary control on radiometric surface temperature due to its effect on soil thermal inertia, rather than as a limiting control on latent heat flux, as might be expected. These findings indicate that invertible surface energy balance models must account for the effects of landcover, soil background temperatures, and soil moisture before thermal infrared imagery can be effectively used to estimate land surface fluxes.