Influence of habitat amount,arrangement, and use on population trend estimates of male Kirtland’s warblers

Kirtland’s warblers (Dendroica kirtlandii) persist in a naturally patchy environment of young, regenerating jack pine forests (i.e., 5–23 years old) created after wildfires and human logging activities. We examined how changing landscape structure from 26 years of forest management and wildfire disturbances influenced population size and spatial dispersion of male Kirtland’s warblers within their restricted breeding range in northern Lower Michigan, USA. The male Kirtland’s warbler population was six times larger in 2004 (1,322) compared to 1979 (205); the change was nonlinear with 1987 and 1994 identified as significant points of change. In 1987, the population trend began increasing after a slowly declining trend prior to 1987, and the rate of increase appeared to slow after 1994. Total amount of suitable habitat and the relative area of wildfire-regenerated habitat were the most important factors explaining population trend. Suitable habitat increased 149% primarily due to increasing plantations from forest management. The relative amount and location of wildfire-regenerated habitat modified the distribution of males among various habitat types, and the spatial variation in their abundance across the primary breeding range. These findings indicate that the Kirtland’s warbler male population shifted its use of habitat types temporally and spatially as the population increased and as the relative availability of habitats changed through time. We demonstrate that researchers and managers need to consider not only habitat quality, but the temporal and the spatial context of habitat availability and population levels when making habitat restoration decisions.     

Pattern of Kirtland's warbler occurrence in relation to the landscape structure of its summer habitat in northern Lower Michigan

Studies of the endangered Kirtland's warbler in relation to landscape ecosystems were conducted from 1986–1988 on a large wildfire-burn surrounding Mack Lake in southeastern Oscoda County, Michigan. A landscape ecosystem approach was used to distinguish low- and high-elevation segments of the landscape, as well as 11 local ecosystem types. The ecosystems were distinguished by physiography, microclimate, soil, and vegetation. The early occurrence of the warblers was strongly related to landscape structure, i.e., to the broad low- and high-elevation areas and the local ecosystem types within them. Territories of male warblers were observed in 5 of the 11 ecosystems. The five ecosystem types where warblers were observed were characterized by (1) a physiography of level or rolling terrain; (2) soil series of Grayling, Graycalm, Montcalm, or Rubicon; (3) uplands with relatively warm temperature during the breeding season; (4) vegetation dominated by low sweet blueberry, bearberry, wintergreen, northern pin oak, blue stem grasses, and hair cap moss; and (5) canopy of relatively tall, dense, and patchy jack pine and oak. Landscape structure appears to be an important factor affecting the occurrence of the warbler in its summer habitat in northern Lower Michigan.     

Long-term effects of managing for an endangered songbird on the heterogeneity of a fire-prone landscape

Context

Jack pine (Pinus banksiana)-dominated ecosystems of northern Lower Michigan are the primary breeding habitat for the federally endangered Kirtland’s warbler (Setophaga kirtlandii, KW). Historically, young stands used by KW were produced by stand-replacing wildfires, but fire suppression has necessitated the management of jack pine plantations for KW habitat since the 1970s. Effects of this long-term management on landscape age heterogeneity have previously not been quantified.

Objectives

We hypothesized that forest management has altered the spatial and temporal distribution of jack pine-dominated ecosystems beyond their historic range of variability.

Methods

By developing a diameter-age relationship for jack pine, we estimated ages of pre-European settlement trees found in General Land Office survey notes. We compared pre-European and current landscapes using geostatistical modeling of survey notes, and landscape metrics to quantify changes in pattern.

Results

Three KW management-based age classes (<20, 21–50, >50 years) are now more evenly distributed (31, 39, and 30 %, respectively) compared to the pre-European distribution (5, 19, 76 %) with little variability over time. Landscape metrics suggest the current landscape is younger and more fragmented than the pre-European landscape. These changes indicate restriction of the historic range of age variability, largely due to conversion of older jack pine stands to young KW habitat plantations.

Conclusions

Management has met KW population objectives, but has altered the temporal variability of the landscape’s age structure. Pre-European settlement patterns of stand-ages may provide a foundation for an ecosystem-based management plan for the region that supports both KW and the ecosystems upon which they depend.

     

Advance regeneration and seed banking of woody plants in Ohio pine plantations: Implications for landscape change

Silviculturally-managed pine plantations within southern Ohio are chronically disturbed patches of introduced vegetation distinct from the surrounding matrix of hardwood forest. To determine the successional pathways by which such pine stands might blend back into the hardwood forest matrix under different types of silvicultural management, we determined the current status of hardwood regeneration under 24 pine stands. Stands of Pinus virginiana (Virginia pine) had the highest density of hardwood seedlings and saplings (20,560 stems ha^-1) whereas P. strobus (white pine) stands averaged only 7090 hardwood stems ha^-1; P. resinosa (red pine) stands were intermediate. The most abundant hardwood seedling and sapling species under pine canopies were Acer rubrum and Cornus florida. DCA ordination of the seedling + sapling assemblages clustered most of the P. resinosa and P. strobus stands in the center of the ordination along with a group of species which are common in second-growth forests of the area. P. virginiana stands, in contrast, were scattered throughout the ordination space. Most of the woody species common in second-growth forests of the region were also common in the pine understory. Multiple regression indicated that large plantations with deeper litter, higher soil pH and lower total hardwood density had the greatest abundance of mesic-site species in the understory. This relationship did not hold for P. resinosa stands, however, due to more frequent and intense silvicultural intervention. The seed bank was not an important source of woody seedlings to the understory assemblage under intact pine plantations. The vegetation of 1–4 yr old clear-cut sites was dominated by wind and bird dispersed species which were generally absent from the understory of intact plantations. We conclude these chronically disturbed planted patches will revert to matrix vegetation faster if the disturbance is allowed to end in a gradual manner through stand senescence than if it is abruptly ended by clear-cutting.     

Aspen regeneration in the Colorado Front Range: differences at local and landscape scales

Elk (Cervus elaphus) populations in Rocky Mountain National Park are higher than at any time in the past century, and heavy browsing by elk may interfere with aspen (Populus tremuloides Michx.) regneration. We used aerial photographs to identify all aspen stands within Rocky Mountain National Park, and all aspen stands within the elk winter range range (defined as 2400 to 2800 m elevation) in three portions of the adjacent Roosevelt National Forest. From this population of aspen stands, we randomly selected 57 stands for evaluation of aspen regeneration. Stands that contained stems younger than 30 years and taller than 2.5 m tall were classified as regenerating successfully. Only 20% of the aspen stands in Estes Valley contained a cohort of regenerating aspen stems, whereas 45-to-75% of aspen stands across the larger landscape of the Front Range had regenerating cohorts of aspen. Within the elk winter range of the Roosevelt National Forest, 13 of 17 aspen stands were regenerating. In the elk winter range on the east side of the Park but outside of Estes Valley, 11 of 15 aspen stands were regenerating successfully. Only a few aspen stands exist in the elk winter range on the western side of the Park, and none of the five aspen stands sampled in Kawuneeche Valley had a regenerating cohort. The lack of regeneration in Kawuneeche Valley may result from locally heavy elk use in both winter and summer. In the summer elk range at higher elevations in the Park (2800 to 3200 m), 16 of 23 stands had regenerated. At landscape scales, all locations outside of the heavily impacted Estes Valley averaged about two cohorts/stand that regenerated after the mid-1960s. All stands that lacked a regenerating cohort showed evidence of moderate-to-severe damage from elk browsing of stems. No regenerating stands showed evidence of severe browsing. We conclude that at landscape scales, regeneration within aspen stands is very common across the Front Range, except in local areas of the highest elk use where little regeneration has occurred in the past 30 years.     

Fire severity and seed source influence lodgepole pine (<Emphasis Type="Italic">Pinus contorta</Emphasis> var. <Emphasis Type="Italic">murrayana</Emphasis>) regeneration in the southern cascades,Lassen volcanic National Park,California

Rocky Mountain lodgepole pine, (Pinus contorta var. latifolia) regenerates quickly after high severity fire because seeds from serotinous cones are released immediately post-fire. Sierra lodgepole pine (P. contorta var. murrayana) forests burn with variable intensity resulting in different levels of severity and because this variety of lodgepole pine does not have serotinous cones, little is known about what factors influence post-fire regeneration. This study quantifies tree regeneration in a low, moderate, and high severity burn patch in a Sierra lodgepole forest 24 years after fire. Regeneration was measured in ten plots in each severity type. In each plot, we quantified pre- and post-fire forest structure (basal area, density), counted and aged tree seedlings and saplings of all species, and measured distance to the nearest seed bearing tree. There was no difference in the density of seedlings and saplings among severity classes. Distance and direction to the nearest seed bearing lodgepole pine were the best predictors of lodgepole seedling and sapling density in high severity plots. In contrast to Rocky Mountain lodgepole pine, regeneration of Sierra lodgepole pine appears to rely on in-seeding from surviving trees in low or moderate severity burn patches or live trees next to high severity burn patches. Our data demonstrate that Sierra lodgepole pine follows stand development pathways hypothesized for non-serotinous stands of Rocky Mountain lodgepole pine.     

Aspen persistence near the National Elk Refuge and Gros Ventre Valley elk feedgrounds of Wyoming, USA

We investigated aspen (Populus tremuloides)regeneration in the Gros Ventre River Valley, the National Elk Refuge and a small part of Grand Teton National Park, Wyoming, USA to see if elk (Cervus elaphus) browsing was as damaging as previously thought. We conducted a landscape-scale survey to assess aspen regeneration across gradients of wintering elk concentrations using 68 randomly selected aspen stands in the 1090 km² study area. Forty-four percent of the stands sampled supported some newer regeneration that had reached the canopy. There were no significant differences of regeneration across elk winter range classification (p=0.25) or distance from feedgrounds (p=0.96). However, a multiple linear regression found that the concentration of elk was one of several important predictors of successful aspen regeneration (p=0.005, R ²=0.36). Our results suggest that stand-replacing regeneration occurs across the landscape at a variety of elk densities despite some trends of reduced regeneration under greater elk concentrations. We propose that high spatial and temporal variation and scattered patches of successful aspen regeneration characterize aspen persistence between periods of episodic regeneration and recruitment.     

Incoming! Association of landscape features with dispersing mountain pine beetle populations during a range expansion event in western Canada

Mountain pine beetle, Dendroctonus ponderosae Hopkins (Coleoptera: Curculionidae, Scolytinae), is a forest insect that undergoes intermittent population eruptions, causing landscape-level mortality to mature pines. Currently, an outbreak covers over 16.3 million ha of British Columbia and Alberta in western Canada. Recent incursion into the jack pine (Pinus banksiana Lamb.) of northwestern Alberta threatens further range expansion through the boreal forest to central and eastern Canada. The spread from British Columbia into northwestern Alberta has been facilitated by above-canopy dispersal of the insect by meso-scale atmospheric currents. At these scales, dispersing D. ponderosae may behave like inert particles, causing terrain-induced tropospheric convective and advective currents to influence population dispersal and establishment. We use spatial point process regression models to examine the association of meso-scale variables, including landscape features and their orientations, habitat suitability, elevation and treatment efforts, with occurrence of D. ponderosae infestations in 2004, 2005, and 2006. Infestations of D. ponderosae primarily established in canyons and valleys, before moving into more open-sloped areas. Southwestern slopes of midslope ridges and small hills, southwest facing open slopes, and valleys that run in a northeast–southwest cardinal direction were positively associated with higher intensities of infestation. This study provides insight into the influences of complex terrain on landscape disturbance by a forest insect, and can be used to prioritize areas for potential management.     

Utilization of a habitat mosaic by cotton rats during a population decline

This study describes the demographic features of a population of Sigmodon hispidus utilizing the habitat mosaic provided by a Carolina Bay on the Atlantic coastal plain of South Carolina. A total of 71 cotton rats were captured 160 times on a 4 ha grid during a winter decline from 25/ha to less that 1/ha. Body weights of adults declined until early February and then increased; those of subadults grew very slowly until February followed by a spurt in growth. Weight gain did not differ between survivors and non-survivors for males, but female survivors gained 1.5 g per week more than non-survivors. Female subadults exhibited higher mortality early in the decline and males later.Adult females were randomly distributed across 8 microhabitats, whereas adult males were almost exclusively confined to heavy Rubus cover. Subadult males used wet sites more than any other cohort; subadult females were widely distributed using drier sites most frequently. By the end of the decline, all survivors were localized in Rubus-dominated patches. No statistically significant changes in electromorph genotypes or allele frequencies were detected, but survivors had a higher frequency of the F-allele at the adenylate kinase locus than did non-survivors (42.3% vs. 16.7%).Our findings affirm the importance of a landscape perspective in understanding the population dynamics of cotton rats, and show how a habitat mosaic influences survival differentially among sex-age cohorts.     

Use of space by the yellow-footed antechinus, Antechinus flavipes, in a fragmented landscape in South Australia

An understanding of how individual species are able to persist and move within fragmented landscapes is critical for elucidating the effects of fragmentation and aiding in the management of species. Here, we studied movement behaviour of the dasyurid Antechinus flavipes in a heavily fragmented landscape using trapping and radiotracking. We assessed the ability of animals to move within and amongst small (<6 ha) remnants and make use of the matrix, and investigated how females used the available space within remnants. Seventeen between-remnant movements were detected from 428 recaptures, ranging in length from 30 to 720 m and averaging 352 m. Most were by adult males during the breeding season, with 40% more than 500 m. Landscape types traversed would have included exotic pine plantations, open grazed areas and roads. Between-site movements of juveniles were only detected on three occasions. However, few young males were captured as adults, suggesting high dispersal rates and considerable matrix use. Conversely, despite high female recapture rates, again only three between-site movements were recorded. Radiotracking further indicated that females confined foraging to remnants, with occasional forays to isolated trees in paddocks. Female home range areas were similar for remnants and forest (0.04–0.66 ha). A. flavipes is clearly able to persist in very small patches of native vegetation in the landscape studied here. Its long-term persistence appears dependent on the ability of females to maintain a presence in the small remnants, and of unrelated males to move between remnants to breed with resident females. This study illustrates the importance of recognising the occurrence of metapopulations in fragmented landscapes for conservation management.     

Post-storm surveys reveal large-scale spatial patterns and influences of site factors,forest structure and diversity in endemic bark-beetle populations

The storm that struck France on december 26^th and 28^th 1999 felled 140 million m³ of timber and had a high economic, social and landscape impact. This event offered the opportunity to study large-scale patterns in populations of forest insect pests that would benefit from the abundant breeding material. A large-scale survey was carried out in France in 2000 to sample the most frequently observed species developing on spruce (Ips typographus, Pityogene schalcographus) and pine (Tomicus piniperda, Ips sexdentatus) in 898 locations distributed throughout wind-damaged areas. The local abundance of each species scored on a 0 to 5 scale was analysed using geostatistical estimators to explore the extent and intensity of spatial autocorrelation, and was related to site, stand, and neighbourhood landscape metrics of the forest cover (in particular the interconnection with broadleaf forest patches) found within dispersal distance. All species but I. sexdentatus, which was much less abundant, displayed large-scale spatial dependence and regional variations in abundance. Lower infestation levels per tree (windfalls and standing trees) were observed in stands with a high proportion of wind-damaged trees, which was interpreted as the result of beetles distributing themselves among the available breeding material. More infestations were observed in wind-broken trees as compared to wind-felled trees. More importantly, populations showed significant relationships with the structure of coniferous stands (in particular with the number of coniferous patches). T. piniperda population levels were negatively correlated to the amount of coniferous edge shared with broadleaf forest patches, possibly because of the disruptive effect of non-host volatiles on host-finding processes at the landscape-scale. The differences observed between species regarding patterns and relationships to site, stand, and forest cover characteristics are discussed in relation to the ecological characteristics of each species.     

Experimental evidence of the effects of a changed matrix on conserving biodiversity within patches of native forest in an industrial plantation landscape

We implemented a replicated before-after-control-impact (BACI) experiment to quantify vertebrate response in native forest patches to a major change in the surrounding exotic Radiata Pine (Pinus radiata) plantation. We contrasted vertebrate occupancy of patches of native eucalypt forest where the surrounding stands of exotic Radiata Pine (Pinus radiata) were clearfelled (termed “treatment patches”) with matched “control patches” where surrounding pine stands remained unlogged. Different species of arboreal marsupials varied in their response to our experimental treatments. The Common Ringtail Possum was unaffected by cutting of the surrounding pine stands, whereas all sightings of the Mountain Brushtail Possum were in control patches. For birds, species richness was significantly reduced by 4–9 species in treatment patches. Birds with cup and dome nests were those negatively affected by the cutting of the surrounding pine stands. They may be susceptible to altered microclimatic conditions or increasing levels of nest predation when the surrounding pine matrix is clearfelled. Our study emphasized how the biota inhabiting retained patches of native forest within plantation landscapes can be changed when stands of surrounding Radiata Pine are clearfelled. In the case of birds, more species will be maintained within eucalypt patches if logging is scheduled so that not all the surrounding pine plantation is clearfelled at once.     

Influence of fire regimes on lodgepole pine stand age and density across the Yellowstone National Park (USA) landscape

A probabilistic spatial model was created based on empirical data to examine the influence of different fire regimes on stand structure of lodgepole pine (Pinus contorta var. latifolia) forests across a >500,000-ha landscape in Yellowstone National Park, Wyoming, USA. We asked how variation in the frequency of large fire events affects (1) the mean and annual variability of age and tree density (defined by postfire sapling density and subsequent stand density) of lodgepole pine stands and (2) the spatial pattern of stand age and density across the landscape. The model incorporates spatial and temporal variation in fire and serotiny in predicting postfire sapling densities of lodgepole pine. Empirical self-thinning and in-filling curves alter initital postfire sapling densities over decades to centuries. In response to a six-fold increase in the probability of large fires (0.003 to 0.018 year⁻¹), mean stand age declined from 291 to 121 years. Mean stand density did not increase appreciably at high elevations (1,029 to 1,249 stems ha⁻¹) where serotiny was low and postfire sapling density was relatively low (1,252 to 2,203 stems ha⁻¹). At low elevations, where prefire serotiny and postfire lodgepole pine density are high, mean stand densities increased from 2,807 to 7,664 stems ha⁻¹. Spatially, the patterns of stand age became more simplified across the landscape, yet patterns of stand density became more complex. In response to more frequent stand replacing fires, very high annual variability in postfire sapling density is expected, with higher means and greater variation in stand density across lodgepole pine landscapes, especially in the few decades following large fires.     

Microhabitat conditions and landscape pattern explain nocturnal rodent activity,but not seed removal,in burned and unburned lodgepole pine forests

Context

Disturbances create spatial variation in environments that may influence animal foraging. Granivory by rodents can influence seed supply and thus plant establishment. However, effects of disturbance patterns on rodent seed removal in western North American conifer forests are generally unknown.

Objectives

We conducted a study in lodgepole pine (Pinus contorta var. latifolia) forests of Greater Yellowstone (Wyoming, USA) to answer: (1) How do seed removal and rodent activity vary between recently burned and adjacent unburned forests and with distance from fire perimeter? (2) Which microhabitat conditions explain variability in seed removal and rodent activity?

Methods

One or two years after wildfires, we established transects (n = 23) with four stations each: at 10 and 40 m from the fire perimeter in both burned and unburned forest. At stations, we deployed trays with lodgepole pine seeds and cameras pointed at trays for 28 days and quantified habitat structure and seed abundance.

Results

Seed removal, which averaged 85%, and diurnal rodent activity did not differ between burned and unburned forests or with distance from the fire perimeter; however, nocturnal rodent activity was lower in burned forests. Seed removal and diurnal rodent activity were not associated with any microhabitat conditions we measured. However, nocturnal rodent activity was associated with microhabitat in both burned and unburned forests.

Conclusions

High seed removal rates suggested that rodent foraging was not reduced by high-severity wildfire. If observed seed removal represents natural conditions, post-dispersal seed predation could influence post-fire recruitment of a widespread foundation tree species.

     

The spatial pattern of a northern conifer-hardwood landscape

A geographic information system, fractal analyses, and statistical methods were used to examine the spatial distributions of old growth hemlock, northern hardwood, mixed hardwood/hemlock stands and wetlands with respect to each other and also soils and topography. Greater than 80% of the stands of any covertype were less than 20 ha in area. Nearly pure hemlock and northern hardwood stands were associated with soils having a fragipan, while mixed hardwood/hemlock stands were associated with sandier soils. Hemlock stands were distributed independently of hardwood and mixed hardwood/hemlock stands, but hardwood and mixed hardwood/hemlock stands were usually surrounded by hemlock. Bogs and lakes were usually surrounded by hemlock stands and are distributed independently of hardwood stands. The shapes of all stands vary from extremely simple to extremely complex, with a general tendency for hemlock stands to be more convoluted than hardwoods. The analyses suggest segregation across soil types and a disturbance regime favoring the establishment of hardwoods and mixed hardwood/hemlock stands in a hemlock matrix as reasons for the origin of the observed spatial patterns.     

Regeneration and transformation of the premier UK apple (Malus × pumila Mill.) cultivar Queen Cox

Summary

A number of factors were assessed for their effects on in vitro shoot proliferation and adventitious shoot regeneration. More in vitro leaves of a quality suitable for use in regeneration and transformation experiments were obtained from shoots on DKW proliferation medium compared with MS medium, and also on MS and DKW media containing phloroglucinol. Compared with MS medium, shoot proliferation was greater on MS with halved levels of NH₄NO₃ and KNO₃. Adventitious shoots were hyperhydric on MS-based but not on DKW-based regeneration medium. More adventitious shoots regenerated on media solidified with Sigma Agargel than on media with Sigma Phytagel or Gelcarin (FMC). Viable transformed shoots were recovered on Sigma Agar or Agargel but not Phytagel. Wounding of leaf explants by stabbing with needles, and stabbing combined with scoring with a scalpel, increased the number of calli regenerating, and these methods, as well as solely scoring with a scalpel, increased the number of calli regenerating shoots compared with the control. Combined stabbing and scoring resulted in more calli producing shoots than solely scoring or stabbing. Vortexing leaf explants with silicon carbide whiskers increased the percentage of subsequently formed calli that regenerated shoots compared with the control. Transformed shoots were regenerated following co-cultivation with Agrobacterium tumefaciens EHA101 harbouring the binary vector pSCV1.6 (with selectable marker gene npt II and GUS reporter gene uid A). The number of transformed shoots as a percentage of explants varied from 0.5% to 2.2%. Molecular analysis of the four extant transformed lines confirmed integration of the transgene and indicated that in three lines there was one integration site, and in one line there were four sites of integration.     

Spatial association between forest heterogeneity and breeding territory boundaries of two forest songbirds

Human activities and natural disturbances create spatial heterogeneity within forested landscapes, leading to both sharp and gradual boundaries in vegetation and abiotic attributes, such as rocks. Those boundaries may affect the detailed delineation of avian territories (independently of their general location), but their role is largely unknown. We tested, using a spatial analysis approach, whether spatial heterogeneity of vegetation and abiotic attributes were associated with territory boundaries of ten black-throated blue warblers (Dendroica caerulescens) and 14 ovenbirds (Seiurus aurocapillus). The study was conducted during summer 1999 in a mature deciduous forest near Québec City, Canada. Singing males were mapped from repeated surveys at 756 points, 25 m apart, on a 49 ha grid. Spatial heterogeneity was obtained from 27 attributes measured at each point. Boundaries of bird territories, vegetation, and abiotic attributes were delineated using the lattice-wombling boundary detection algorithm. The spatial association between territory and microhabitat boundaries was computed using the spatial overlap statistics. There was significant spatial overlap between territory boundaries and those of 15 and 17 attributes for black-throated blue warbler and ovenbird, respectively. The attributes most strongly associated with territory boundaries were conifer seedling cover, grass and total vegetation cover between 0-2 m high for black-throated blue warbler and fern cover, vegetation-covered rocks and shrub diversity for ovenbird. Complementary to this, a redundancy analysis (RDA) was used to compare attributes associated with the general occurrence of males to those whose boundaries were associated specifically with territory boundaries. Most attributes whose boundaries were associated with territory boundaries did not correspond to resource attributes, i.e., those where birds were detected most frequently. We conclude that soft boundaries associated with spatial heterogeneity may help shape forest bird territories by providing landmarks not necessarily related to resources used within territories.     

Spatial patterns of sand pine invasion into longleaf pine forests in the Florida Panhandle

Land use practices since European settlement have had profound effects on the composition and structure of certain forested ecosystems in the southeastern U.S. Coastal Plain. One significant change since the turn of the century has been the invasion of upland pine forests by sand pine in the state of Florida panhandle and peninsula. This study quantified sand pine extent and expansion and examined links between sand pine distribution and environmental factors in the Florida panhandle. Geographic information system analysis using aerial photographs (1949 and 1994) showed sand pine expansion and also increased canopy cover of sand pine over time. There was a high rate of conversion of longleaf pine to sand pine from 1949 to 1994 (44%), and conversion of sparse sand pine stands to dense sand pine stands (69%). Therefore, widespread changes in the Florida landscape were evident during a relatively short time period. Adjacency analyses showed a strong negative association between longleaf pine and dense sand pine and a positive association between riparian vegetation and dense sand pine. Distribution of sand pine across elevation demonstrated that sand pine expanded inland and upland into longleaf-pine forests. In 1949, sand pine was selectively located on sites below 30 m in elevations; by 1994, sparse and moderately dense sand-pine stands were found at all elevations. Thus, any area that may receive an input of sand pine seeds, most notably from riparian areas, is vulnerable to establishment.     

Aspen structure and variability in Rocky Mountain National Park,Colorado, USA

Elk, fire and climate have influenced aspen populations in the Rocky Mountains, but mostly subjective studies have characterized these factors. A broad-scale perspective may shed new light on the status of aspen in the region. We collected field measurements of aspen (Populus tremuloides Michx.) patches encountered within 36 randomly located belt transects in 340 km² of Rocky Mountain National Park, Colorado, to quantify the aspen population. Aspen covered 5.6% of the area in the transects, much more than expected based on previously collected remotely sensed data. The distribution and structure of aspen patches were highly heterogeneous throughout the study area. Of the 123 aspen patches encountered in the 238 ha surveyed, all but one showed signs of elk browsing or had conifer species mixed with the aspen stems. No significant difference occurred in aspen basal area, density, regeneration, browsing of regeneration and patch size, between areas of concentrated elk use (elk winter range) and areas of dispersed elk use (elk summer range). Two-thirds of the aspen patches were mixed with conifer species. We concluded that the population of aspen in our study area is highly variable in structure and that, at a landscape-scale, evidence of elk browsing is widespread but evidence of aspen decline is not. This revised version was published online in July 2006 with corrections to the Cover Date.     

Spatial patterns of primary productivity in the Greater Yellowstone Ecosystem

Landscapes are often heterogeneous in abiotic factors such as topography, climate, and soil, yet little is known about how these factors may influence the spatial distribution of primary productivity. We report estimates of aboveground net primary productivity (ANPP) in 90 sample stands stratified by cover type and elevation class, and use the results to predict ANPP across a portion of the Greater Yellowstone Ecosystem. Tree ANPP was estimated by sampling tree density by species and diameter classes and estimating average annual diameter increment by tree coring. Biomass for current tree diameter and past tree diameter were calculated by species and diameter class for each stand using the dimension analysis software BIOPAK. Shrub ANPP was estimated by calculating current biomass from basal area using BIOPAK and dividing by the assumed average life span of the shrubs. Clipping at the end of the growing season was used to estimate herb ANPP. Differences in ANPP among cover types and elevation classes were examined with analysis of variance. Multiple regression was used to examine relationships between ANPP, and soil parent material, topography, and cover type. The best regression model was used to predict ANPP across the study area.We found ANPP was highest in cottonwood, Douglas-fir, and aspen stands, intermediate in various seral stages of lodgepole pine, and lowest in grassland and sagebrush cover types. Parent material explained significant variation in ANPP in mature and old-growth lodgepole pine stands, with rhyolite ash/loess being the most productive parent material type. ANPP decreased with increasing elevation in most cover types, possibly because low temperatures limit plant growth at higher elevations in the study area. ANPP was not related to elevation in mature and old-growth lodgepole pine stands, due to relatively rapid growth of subalpine fir at higher elevations.A regression model based on cover type and elevation explained 89% of the variation in ANPP among the sample stands. This model was used to generate a spatially continuous surface of predicted ANPP across the study area. The frequency distribution of predicted ANPP was skewed towards lower levels of ANPP, and only 6.3% of the study area had a predicted ANPP level exceeding 4500 kg/ha/yr. Patches high in predicted ANPP were primarily at lower elevations, outside of Yellowstone National Park, and near the national forest/private lands boundary. These patterns of ANPP may influence fire behavior, vertebrate population dynamics, and other ecological processes. The results reinforce the need for coordinated management across ownership boundaries in the Greater Yellowstone Ecosystem.