Curtis H. Flather

Global Change in Forests: Responses of Species, Communities, and Biomes

G change is often perceived as human-induced modifications in climate. Indeed, human activities have undeniably altered the atmosphere, and probably the climate as well (Watson et al. 1998). At the same time, most of the world’s forests have also been extensively modified by human use of the land (Houghton 1994). Thus, climate and land use are two prongs of human-induced global change. The effect of these forces on forests is mediated by the organisms within forests. Consideration of climate, land use, and biological diversity is key to understanding forest response to global change. Biological diversity refers to the variety of life at organizational levels from genotypes through biomes (Franklin 1993). The responses of ecological systems to global change reflect the organisms that are within them. While ecologists have sometimes not seen the forest for the trees, so to speak, it is also true that forests cannot be understood without knowledge of the trees and other component species. It is the responses of individual organisms that begin the cascade of ecological processes that are manifest as changes in system properties, some of which feed back to influence climate and land use (Figure 1). Beyond its role in ecosystems, biodiversity is invaluable to humans for foods, medicines, genetic information, recreation, and spiritual renewal (Pimentel et al. 1997). Thus, global changes that affect the distribution and abundance of organisms will affect future human well-being and land use, as well as, possibly, the climate. This article serves as a primer on forest biodiversity as a key component of global change. We first synthesize current knowledge of interactions among climate, land use, and biodiversity. We then summarize the results of new analyses on the potential effects of human-induced climate change on forest biodiversity. Our models project how possible future climates may modify the distributions of environments required by various species, communities, and biomes. Current knowledge, models, and funding did not allow these analyses to examine the population processes (e.g., dispersal, regeneration) that would mediate the responses of organisms to environmental change. It was also not possible to model the important effects of land use, natural disturbance, and other factors on the response of biodiversity to climate change. Despite these limitations, the analyses discussed herein are among the most comprehensive projections of climate change effects on forest biodiversity yet conducted. We conclude with discussions of limitations, research needs, and strategies for coping with potential future global change.

Patchy Reaction‐Diffusion and Population Abundance: The Relative Importance of Habitat Amount and Arrangement

A discrete reaction‐diffusion model was used to estimate long‐term equilibrium populations of a hypothetical species inhabiting patchy landscapes to examine the relative importance of habitat amount and arrangement in explaining population size. When examined over a broad range of habitat amounts and arrangements, population size was largely determined by a pure amount effect (proportion of habitat in the landscape accounted for >96% of the total variation compared to <1% for the arrangement main effect). However, population response deviated from a pure amount effect as coverage was reduced below 30%–50%. That deviation coincided with a persistence threshold as indicated by a rapid decline in the probability of landscapes supporting viable populations. When we partitioned experimental landscapes into sets of “above” and “below” persistence threshold, habitat arrangement became an important factor in explaining population size below threshold conditions. Regression analysis on below‐threshold landscapes using explicit measures of landscape structure (after removing the covariation with habitat amount) indicated that arrangement variables accounted for 33%–39% of the variation in population size, compared to 27%–49% for habitat amount. Thus, habitat arrangement effects became important when species persistence became uncertain due to dispersal mortality.

Using Landscape Ecology to Test Hypotheses About Large‐Scale Abundance Patterns in Migratory Birds

The hypothesis that Neotropical migrant birds may be undergoing widespread declines due to land use activities on the breeding grounds has been examined primarily by synthesizing results from local studies. Growing concern for the cumulative influence of land use activities on ecological systems has heightened the need for large-scale studies to complement what has been observed at local scales. We investigated possible landscape effects on Neotropical migrant bird populations for the eastern United States by linking two large-scale inventories designed to monitor breeding-bird abundances and land use patterns. The null hypothesis of no relation between landscape structure and Neotropical migrant abundance was tested by correlating measures of landscape structure with bird abundance, while controlling for the geographic distance among samples. Neotropical migrants as a group were more sensitive to landscape structure than either temperate migrants or permanent residents. Neotropical migrants tended to be more abundant in landscapes with a greater proportion of forest and wetland habitats, fewer edge habitats, larger forest patches, and with forest habitats well dispersed throughout the scene. Permanent residents showed few correlations with landscape structure and temperate migrants were associated with habitat diversity and edge attributes rather than with the amount, size, and dispersion of forest habitats. The association between Neotropical migrant abundance and forest fragmentation differed among physiographic strata, suggesting that landscape context affects observed relations between bird abundance and landscape structure. Finally, associations between landscape structure and temporal trends in Neotropical migrant abundance were counter to those observed in space. Trends in Neotropical migrant abundance were negatively correlated with forest habitats. These results suggest that extrapolation of patterns observed in some landscapes is not likely to hold regionally, and that conservation policies must consider the variation in landscape structure associations observed among different types of bird species and in physiographic strata with varying land use histories.

Housing growth in and near United States protected areas limits their conservation value

Protected areas are crucial for biodiversity conservation because they provide safe havens for species threatened by land-use change and resulting habitat loss. However, protected areas are only effective when they stop habitat loss within their boundaries, and are connected via corridors to other wild areas. The effectiveness of protected areas is threatened by development; however, the extent of this threat is unknown. We compiled spatially-detailed housing growth data from 1940 to 2030, and quantified growth for each wilderness area, national park, and national forest in the conterminous United States. Our findings show that housing development in the United States may severely limit the ability of protected areas to function as a modern “Noah’s Ark.” Between 1940 and 2000, 28 million housing units were built within 50 km of protected areas, and 940,000 were built within national forests. Housing growth rates during the 1990s within 1 km of protected areas (20% per decade) outpaced the national average (13%). If long-term trends continue, another 17 million housing units will be built within 50 km of protected areas by 2030 (1 million within 1 km), greatly diminishing their conservation value. US protected areas are increasingly isolated, housing development in their surroundings is decreasing their effective size, and national forests are even threatened by habitat loss within their administrative boundaries. Protected areas in the United States are thus threatened similarly to those in developing countries. However, housing growth poses the main threat to protected areas in the United States whereas deforestation is the main threat in developing countries.

IDENTIFYING GAPS IN CONSERVATION NETWORKS: OF INDICATORS AND UNCERTAINTY IN GEOGRAPHIC‐BASED ANALYSES

Mapping of biodiversity elements to expose gaps in conservation networks has become a common strategy in nature-reserve design. We review a set of critical assumptions and issues that influence the interpretation and implementation of gap analysis, including: (1) the assumption that a subset of taxa can be used to indicate overall diversity patterns, and (2) the impact of uncertainty and error propagation in reserve design. We focus our review on species diversity patterns and use data from peer-reviewed literature or extant state-level databases to test specific predictions implied by these assumptions. Support for the biodiversity indicator assumption was varied. Patterns of diversity as reflected in species counts, coincidence of hot spots, and representativeness were not generally concordant among different taxa, with the degree of concordance depending on the measure of diversity used, the taxa examined, and the scale of analysis. Simulated errors in predicting the occurrence of individual species indicated that substantial differences in reserve-boundary recommendations could occur when uncertainty is incorporated into the analysis. Furthermore, focusing exclusively on vegetation and species distribution patterns in conservation planning will contribute to reserve-design uncertainty unless the processes behind the patterns are understood. To deal with these issues, reserve planners should base reserve design on the best available, albeit incomplete, data; should attempt to define those ecological circumstances when the indicator assumption is defensible; should incorporate uncertainty explicitly in mapped displays of biodiversity elements; and should simultaneously consider pattern and process in reserve-design problems.

Forest fragmentation and bird community dynamics: inference at regional scales

With increasing fragmentation of natural areas and a dramatic reduction of forest cover in several parts of the world, quantifying the impact of such changes on species richness and community dynamics has been a subject of much concern. Here, we tested whether in more fragmented landscapes there was a lower number of area-sensitive species and higher local extinction and turnover rates, which could explain higher temporal vari- ability in species richness. To investigate such potential landscape effects at a regional scale, we merged two independent, large-scale monitoring efforts: the North American Breeding Bird Survey (BBS) and the Land Use and Land Cover Classification data from the U.S. Geological Survey. We used methods that accounted for heterogeneity in the probability of detecting species to estimate species richness and temporal changes in the bird communities for BBS routes in three mid-Atlantic U.S. states. Forest breeding bird species were grouped prior to the analyses into area-sensitive and non-area-sensitive species according to previous studies. We tested predictions relating measures of forest structure at one point in time (1974) to species richness at that time and to parameters of forest bird community change over the following 22-yr-period (1975-1996). We used the mean size of forest patches to characterize landscape structure, as high correlations among landscape variables did not allow us to disentangle the relative roles of habitat fragmentation per se and habitat loss. As predicted, together with lower species richness for area-sensitive species on routes surrounded by landscapes with lower mean forest-patch size, we found higher mean year- to-year rates of local extinction. Moreover, the mean year-to-year rates of local turnover (proportion of locally new species) for area-sensitive species were also higher in landscapes with lower mean forest-patch size. These associations were not observed for the non-area- sensitive species group. These results suggest that landscape structure may influence forest bird communities at regional scales through its effects on the total number of species but also on the temporal rates of change in community composition. Evidence for higher rates of local extinction and turnover in more fragmented landscapes suggests that bird communities function as metapopulations at a regional scale, and points out the importance of colonizations and recolonizations from surrounding landscapes to local community dynamics. Further, our results illustrate that the methods used to estimate the community parameters can be a powerful statistical tool in addressing questions relative to the dynamics of communities.

RELATIVE SPECIES RICHNESS AND COMMUNITY COMPLETENESS: BIRDS AND URBANIZATION IN THE MID‐ATLANTIC STATES

The idea that local factors govern local richness has been dominant for years, but recent theoretical and empirical studies have stressed the influence of regional factors on local richness. Fewer species at a site could reflect not only the influence of local factors, but also a smaller regional pool. The possible dependency of local richness on the regional pool should be taken into account when addressing the influence of local factors on local richness. It is possible to account for this potential dependency by comparing relative species richness among sites, rather than species richness per se. We consider estimation of a metric permitting assessment of relative species richness in a typical situation in which not all species are detected during sampling sessions. In this situation, estimates of absolute or relative species richness need to account for variation in species detection probability if they are to be unbiased. We present a method to estimate relative species richness based on capture–recapture models. This approach involves definition of a species list from regional data, and estimation of the number of species in that list that are present at a site–year of interest. We use this approach to address the influence of urbanization on relative richness of avian communities in the Mid-Atlantic region of the United States. There is a negative relationship between relative richness and landscape variables describing the level of urban development. We believe that this metric should prove very useful for conservation and management purposes because it is based on an estimator of species richness that both accounts for potential variation in species detection probability and allows flexibility in the specification of a “reference community.” This metric can be used to assess ecological integrity, the richness of the community of interest relative to that of the “original” community, or to assess change since some previous time in a community.

RELATIONSHIPS AMONG NORTH AMERICAN SONGBIRD TRENDS, HABITAT FRAGMENTATION, AND LANDSCAPE OCCUPANCY

Fragmentation of breeding habitat has been hypothesized as a cause of population declines in forest-nesting migratory birds. Negative correlations between the degree of fragmentation and bird density or fecundity at local or regional scales support the fragmentation hypothesis. Yet, in spite of reduced fecundity and densities in fragmented systems, many forest-nesting passerine species have increased in numbers over time. We hypothesized that range-wide population change in species for which habitat fragmentation negatively affects reproductive success should depend on the proportion of the population that actually occupies fragmented landscapes. We predicted that fragmentation-sensitive species (e.g., species that occur in reduced densities in fragmented landscapes) should increase globally in numbers at a greater rate than species that readily occupy fragmented landscapes, because fragmentation-sensitive distributions place a large proportion of the global population in contiguous landscapes that are superior for breeding. We used Breeding Bird Survey (BBS) data and associated landscape metrics to test this prediction for 10 species of forest-nesting passerines in the United States that experience reproductive dysfunction associated with habitat fragmentation: Ovenbird (Seiurus aurocapillus), Red-eyed Vireo (Vireo olivaceus), Wood Thrush (Hylocichla mustelina), Northern Cardinal (Cardinalis cardinalis), Worm-eating Warbler (Helmitheros vermivorus), Kentucky Warbler (Oporornis formosus), Indigo Bunting (Passerina cyanea), Scarlet Tanager (Piranga olivacea), Acadian Flycatcher (Empidonax virescens), and Hooded Warbler (Wilsonia citrina). Our approach was to: (1) quantify landscape features associated with BBS routes across the eastern United States, (2) classify landscapes around BBS routes as “fragmented” or “contiguous,” (3) estimate the proportion of detected individuals that occurred in fragmented landscapes on a species-by-species basis, and then (4) associate 10-yr trends for each species with the proportion of breeding individuals occupying fragmented landscapes. Regression analysis indicated a significant, negative relationship between the proportion of the breeding population occupying fragmented landscapes and the population trend from 1970 to 1980. Although this result links habitat fragmentation to population change and provides support for the fragmentation hypothesis, other factors (e.g., land use change, weather, varying life history traits, varying winter survivorship, habitat amount thresholds) could generate similar results. More work is needed to partition the relative influence of these factors on regional bird population dynamics if conservationists are to understand more clearly the effects of fragmentation on the distribution and abundance of species across their ranges.

Species richness and patterns of invasion in plants, birds, and fishes in the United States

We quantified broad-scale patterns of species richness and species density (mean # species/km2) for native and non-indigenous plants, birds, and fishes in the continental USA and Hawaii. We hypothesized that the species density of native and non-indigenous taxa would generally decrease in northern latitudes and higher elevations following declines in potential evapotranspiration, mean temperature, and precipitation. County data on plants (n = 3004 counties) and birds (n=3074 counties), and drainage (6 HUC) data on fishes (n = 328 drainages) showed that the densities of native and non-indigenous species were strongly positively correlated for plant species (r = 0.86, P < 0.0001), bird species (r = 0.93, P<0.0001), and fish species (r = 0.41, P<0.0001). Multiple regression models showed that the densities of native plant and bird species could be strongly predicted (adj. R2 = 0.66 in both models) at county levels, but fish species densities were less predictable at drainage levels (adj. R2 = 0.31, P<0.0001). Similarly, non-indigenous plant and bird species densities were strongly predictable (adj. R2 = 0.84 and 0.91 respectively), but non-indigenous fish species density was less predictable (adj. R2 = 0.38). County level hotspots of native and non-indigenous plants, birds, and fishes were located in low elevation areas close to the coast with high precipitation and productivity (vegetation carbon). We show that (1) native species richness can be moderately well predicted with abiotic factors; (2) human populations have tended to settle in areas rich in native species; and (3) the richness and density of non-indigenous plant, bird, and fish species can be accurately predicted from biotic and abiotic factors largely because they are positively correlated to native species densities. We conclude that while humans facilitate the initial establishment, invasions of non-indigenous species, the spread and subsequent distributions of non-indigenous species may be controlled largely by environmental factors.

Minimum viable populations: is there a 'magic number' for conservation practitioners?

Establishing species conservation priorities and recovery goals is often enhanced by extinction risk estimates. The need to set goals, even in data-deficient situations, has prompted researchers to ask whether general guidelines could replace individual estimates of extinction risk. To inform conservation policy, recent studies have revived the concept of the minimum viable population (MVP), the population size required to provide some specified probability of persistence for a given period of time. These studies conclude that long-term persistence requires ≥5000 adult individuals, an MVP threshold that is unaffected by taxonomy, life history or environmental conditions. Here, we re-evaluate this suggestion. We find that neither data nor theory supports its general applicability, raising questions about the utility of MVPs for conservation planning.