Long Term Ecological Research (LTER)
A Member of the LTER Network
| home page > Research > Andrews Forest > Biodiversity |
Summary of Research in LTER5
March 2008
A number of the studies described above will continue as part of our basic LTER research program. In addition, we will begin to develop a more integrated view of biological diversity by examining variation in the abundance and diversity of multiple trophic at the same locations, as they vary annually in response to climate and other factors. Given the importance of competition, trophic interactions, and habitat structure in shaping biotic communities, we feel this integration should lead to greater mechanistic understanding of the controls on biodiversity in our forests. In the sections that follow, we provide an overview of some of our planned studies.
Overstory-understory relationships, diversity-stability relationships, and plant species interactions during succession (C. B. Halpern, M. Dovciak, J. A. Lutz, L. R. Rozzell). Understanding the factors that determine understory plant composition and diversity through forest succession has been a long-term interest at Andrews Forest. Permanent plot studies in WS01 and WS03 have examined the importance of initial composition, disturbance intensity, and species' life histories in shaping early successional patterns (Halpern 1988, 1989; Halpern and Franklin 1990; Halpern and Spies 1995). Measurements now span more than 40 yr and many portions of the watersheds have experienced significant canopy closure. The current emphases of our studies are on (a) changes in forest structure and composition early in stand development and how these are influenced by rates and causes of tree mortality (Lutz 2005, Lutz and Halpern 2006); (b) changes in understory composition, diversity, and biomass, and how these are shaped by the dynamics of the overstory; (c) relationships between plant diversity and population stability (Dovciak and Halpern, in preparation); and (d) plant species interactions during early succession (Rozzell 2003).
Early stand development. Long-term permanent-plot studies in WS01 and WS03 provide new insights into the structural development of Douglas-fir forests (Lutz 2005, Lutz and Halpern 2006). Early forest development is typically viewed as a simple, unidirectional process characterized by rapid closure of the canopy by Douglas-fir, concomitant loss of early successional hardwoods, and intense competition resulting in significant density-dependent mortality. It is generally assumed that hardwoods are short-lived, suppression is the dominant form of mortality, and shade-tolerant conifers are uncommon until later in succession. Our long-term data suggest that this is an overly simplistic model of succession. Stem density and biomass were highly variable among the nearly 200 permanent plots in these watersheds. Although mortality rates were greatest for hardwoods, hardwood biomass increased through canopy closure reflecting continued growth of large, dominant stems (particularly bigleaf maple). Shade-tolerant conifers (primarily western hemlock) accounted for a surprising large proportion of stems: >25% by year 38. Mortality due to suppression was 2.5 times as frequent as that due to mechanical damage (wind and snow-loading), however, biomass lost to mechanical damage was nearly four times greater, a result of periodic storms that created large windthrow patches. Variation in establishment and periodic gap-forming disturbance can contribute to structural complexity early in stand development. These processes may have consequences for the structural development of older forests.
Diversity-stability relationships during succession. Diversity-stability relationships have been studied in aquatic, grassland, and annual plant communities over relatively short time frames (typically <10 yr). Similar studies are lacking from forests and none has explicitly examined whether relationships change during succession as composition shifts from dominance by ruderal to competitive species. Dov?iak and Halpern (in preparation) have explored relationships between species diversity and population stability in forest understories using 40 yr of data from permanent plots in WS1 and 3. This work contributes in several important ways to the broader debate on how diversity affects the stability of biological systems.
Patterns of population stability among forest herbs over 40 yr of succession do not support the general theoretical expectation that stability is negatively related to diversity. Mean population stability was positively related to richness early in succession, but unrelated to richness later, when stability was consistently higher. Similarly, for species with different life histories (early successional colonists vs. forest residuals), stability-diversity relationships were positive and parallel, although stability was generally greater for residual than for colonizing species. Such positive relationships are in variance with the negative associations predicted by competition theory. This suggests that other factors (e.g., facilitation or environmental heterogeneity) may play more important roles than does competition in the stability of plant populations.
Plant species interactions during early succession. It is commonly assumed that competition is responsible for species' replacements early in succession, however few studies have explicitly examined the frequency, strength, or nature of interactions among early successional species. Based a decade of permanent-plot measurements following logging and burning in an old-growth western Cascade forest, Rozzell (2003) used correlation analysis and null modeling to quantify pairwise associations among 33 plant species. Positive associations were more common than negative associations at all sampling times although the proportions of positive associations decreased and negative associations increased over time. A large proportion (>40%) of associations could be attributed to shared, positive correlations with surrounding vegetation, suggesting that diffuse facilitation is of primary importance in this stressful, early successional environment. The increase in proportion of negative associations suggests that that microsite availability and interspecific competition may increasingly limit persistence of early successional species.
Insect biodiversity (J. C. Miller). Studies of insect biodiversity and abundance have been (Miller 1995; Miller and Hammond 2000, 2003) and will continue to be, focused on Lepidoptera and Coleoptera. Three reasons justify the emphasis on these taxa: 1) arthropods comprise the majority of species at Andrews (86%) (Parsons et al. 1991), 2) the distribution of species and their abundance provides a statistically powerful dataset to assess patterns relevant to landscape heterogeneity and management practices throughout the HJA (Hammond and Miller 1998; Heyborne et al. 2003), and 3) the prevalence of primary consumers and the foodplant associations regarding angiosperms and gymnosperms provides an avenue for incorporating studies on plant communities and ecosystem processes with biodiversity at a higher trophic level (Miller et. al 2003). As in LTER4 the emphasis in LTER5 will be on Lepidoptera because of the legacy of data we have recorded starting in 1986. Initially, we will compare differences in Lepidoptera species richness, abundance and functional guild composition on a basis of general habitats categorized as riparian versus upland and open canopy (young regrowth forest) versus closed canopy (mature forest). This project will employ UV light traps placed at 20 locations to represent the aforementioned conditions. In the latter half of LTER5 we will sample from three sites representing the range of climates and forest ages present on the Andrews: (1) low elevation (400 m, WS02); (2) mid elevation (1000 m, Mack Creek); and (3) high elevation (1400 m, Frissell Ridge). At each site we will sample both young (25-30 yr) and old forest (>150 yr) habitats. These sites will overlap with the proposed "all-taxa study area" (see below). As in all previous studies, moths at each site will be sampled on one night every other week, from May-September. In concert with the theme of "temporal variability" we will quantify the inter-annual (in)consistency in patterns of species abundance and synchrony in population trends among various species based on guild classification across each sample sites.
Temporal (annual) variability within forest understories (C. B. Halpern, M. Dovciak). Considerable attention has been devoted to the spatial distributions and successional dynamics of forest organisms in the Pacific Northwest. However, we have limited understanding of the patterns and correlates of biotic variability at finer temporal scales. For example, it is often assumed that forest understory communities are fairly stable in old-growth forests, but that populations of ectomycorrhizal fungi, insects, and small mammals exhibit high inter-annual variation. For most organisms, we have limited empirical data to test these assumptions and there is only cursory understanding of the degree to which temporal variability is shaped by changes in climate, local environment, or biotic interactions.
During the second half of LTER5 we initiated studies of temporal variability in forest understory communities. Our objectives were to (1) quantify the nature annual variation in understory communities, (2) identify plant population and/or community traits that are sensitive to environmental changes, and (3) document the spatial coherence of this variation.
Our conceptual model assumes that annual variation in plant performance is mediated by site productivity (moisture availability) and forest age/structure (via moderation of environmental stress). We hypothesized that temporal variation would be lower in more productive (mesic) habitats due to greater resource availability and in older forests where canopy structure more fully moderates microclimatic stress. Our sampling design permits comparisons among habitats of contrasting productivity (lower vs. upper slopes) and microclimatic stress (N- vs. S-facing slopes) in both young (40 yr) and older (>400 yr) forest.
Two summers of data collection provide only a limited view of spatial and temporal variation in the understory. Nevertheless, we have observed the following patterns:
- Species turnover (presence/absence) is fairly low among 1m2 plots at all topographic positions in both young and old forest. By contrast, stem density and foliar cover can be highly variable.
- Temporal variation in stem density and plant cover do not appear to vary with topographic position or forest age as predicted.
- Seasonal trends in temperature (daily means and maxima) are strikingly similar among young and old forests at corresponding topographic positions. As expected, upper slopes are consistently warmer than shaded, lower slopes; however, N- and S-facing slopes show small differences in temperature (more so for means than for maxima).
- All sampling locations show generally parallel seasonal declines in soil moisture (volumetric moisture content at 0-10 cm depth). However, effects of topography appear to differ between young and old forest: Soil moisture was greater on S-facing slopes in young forest (WS1), but greater on N-facing slopes in old forest (WS2). Local spatial variation in soil moisture tended to be greatest at upper-slope positions on S-facing slopes and lowest at lower-slope positions on S-facing slopes.
|
|
|