David A. Kinner

Ecological Determinism in Plant Community Structure Across a Tropical Forest Landscape

The ecological mechanisms hypothesized to structure species-rich communities range from strict local determinism to neutral ecological drift. We assessed the degree of ecological determinism in tropical plant community structure by analyses of published demographic data; a broad range of spatial, historical, and environmental variables; and the distributions of 33 herbaceous species (plot size = 0.02 ha) and 61 woody species (plot size = 0.09 ha) among 350 plots in a 16-km2 forest landscape (Barro Colorado Island, Panama). We found a strong degree of cross-landscape dominance by a subset of species whose identities were predictable from sapling survivorship rates under shade. Using canonical ordination we found that spatial and environmental–historical factors were of comparable importance for controlling within-landscape variability in species composition. Past land use had a strong impact on species composition despite ceasing 100–200 years ago. Furthermore, edaphic–hydrological factors, treefall gaps, and an edge effect all had unique impacts on species composition. Hence, ecological determinism was evident in terms of both cross-landscape dominance and within-landscape variability in species composition. However, at the latter scale, the large portion of the explained variance in species composition among plots uniquely attributed to spatial location pointed to an equally important role for neutral processes.

The relative roles of environment, history and local dispersal in controlling the distributions of common tree and shrub species in a tropical forest landscape, Panama

We used regression models and information-theoretic model selection to assess the relative importance of environment, local dispersal and historical contingency as controls of the distributions of 26 common plant species in tropical forest on Barro Colorado Island (BCI), Panama. We censused eighty-eight 0.09-ha plots scattered across the landscape. Environmental control, local dispersal and historical contingency were represented by environmental variables (soil moisture, slope, soil type, distance to shore, old-forest presence), a spatial autoregressive parameter (ρ), and four spatial trend variables, respectively. We built regression models, representing all combinations of the three hypotheses, for each species. The probability that the best model included the environmental variables, spatial trend variables and ρ averaged 33%, 64% and 50% across the study species, respectively. The environmental variables, spatial trend variables, ρ, and a simple intercept model received the strongest support for 4, 15, 5 and 2 species, respectively. Comparing the model results to information on species traits showed that species with strong spatial trends produced few and heavy diaspores, while species with strong soil moisture relationships were particularly drought-sensitive. In conclusion, history and local dispersal appeared to be the dominant controls of the distributions of common plant species on BCI.

Nutrient and Physical Soil Characteristics of River Cane Stands, Western North Carolina

Abstract Arundinaria gigantea (Walt.) Muhl., commonly called river cane, is a member of the grass family (Poaceae). The primary purpose of this research is to characterize the physical and chemical properties of the soils of existing stands of A. gigantea in western North Carolina and to provide guidance for the restoration of river cane to the stream valleys of the Southern Appalachian Mountains. We analyzed soils at 20 sites in Cherokee, Jackson, and Macon Counties in North Carolina and collected data on soil characterization, nutrient levels, bulk density, particle size, pH, and hydraulic conductivity. River cane soils varied significantly for carbon, nitrogen, phosphorus, and sand levels, even within the same watershed (ANOVA, p < 0.05). Typical soils are very sandy, mineral soils with low carbon levels and low nutrient levels. Soils are well drained, have very low bulk densities, and low pH. Despite low nutrient conditions, the plant does not appear to be nutrient limited. Restoration potential for the species is high, but more research is required to determine specific limitations on growth.

GIS-Based Stream Network Analysis for the Upper Río Chagres Basin, Panama

To support a number of projects focused on diverse biological and physical science aspects of the upper Rio Chagres basin, a detailed stream network was extracted from digital elevation data obtained by interferometric radar survey. The elevation data represented the bald earth surface plus a forest canopy of varying height. Therefore, different algorithms for stream network extraction were qualitatively evaluated in terms of their capability to extract accurate stream locations from this challenging type of elevation data. The programs based on a shortest path algorithm and an imposed gradients constraint provided stream locations that were closer to on-ground GPS measurements than the tools based on depressions filling and iterative linking. The influence of different spatial resolutions on network structure and orientation was also explored.

Geolocation of man-made reservoirs across terrains of varying complexity using GIS

The Reservoir Sedimentation Survey Information System (RESIS) is one of the worlds most comprehensive databases of reservoir sedimentation rates, comprising nearly 6000 surveys for 1819 reservoirs across the continental United States. Sediment surveys in the database date from 1904 to 1999, though more than 95% of surveys were entered prior to 1980, making RESIS largely a historical database. The use of this database for large-scale studies has been limited by the lack of precise coordinates for the reservoirs. Many of the reservoirs are relatively small structures and do not appear on current USGS topographic maps. Others have been renamed or have only approximate (i.e. township and range) coordinates. This paper presents a method scripted in ESRIs ARC Macro Language (AML) to locate the reservoirs on digital elevation models using information available in RESIS. The script also delineates the contributing watersheds and compiles several hydrologically important parameters for each reservoir. Evaluation of the method indicates that, for watersheds larger than 5km^2, the correct outlet is identified over 80% of the time. The importance of identifying the watershed outlet correctly depends on the application. Our intent is to collect spatial data for watersheds across the continental United States and describe the land use, soils, and topography for each reservoirs watershed. Because of local landscape similarity in these properties, we show that choosing the incorrect watershed does not necessarily mean that the watershed characteristics will be misrepresented. We present a measure termed terrain complexity and examine its relationship to geolocation success rate and its influence on the similarity of nearby watersheds.

Estimation of Landslide Importance in Hillslope Erosion Within the Panama Canal Watershed

This paper presents an approach for assessing the regional importance of landslides based on conventional daily-discharge and daily-sediment data from sub-basins within the Panama Canal Watershed. In many wet mountainous regions, sediment yields are controlled by both surficial erosion and deep, landslide erosion. Landslides require that rainfall (and by inference runoff) exceed a threshold. Runoff can be used to derive a parameter termed ‘landslide days’, with a suitable correction factor, to account for evapotranspiration, infiltration, and rainfall patchiness. The approach developed then uses runoff as the driver for a simple surficial-erosion model and landslide days as the driver for a landslide model. In a case study of the Panama Canal Watershed, this model describes spatial and temporal patterns of annual yields with a high degree of efficacy, demonstrating that simple daily data can be used to determine whether a river basin, such as the upper Rio Charges basin, might be undergoing substantial landslide-related erosion.

The Hunters Crossing weathered rock slide, Haywood County, North Carolina, USA

Abstract The Hunters Crossing landslide is a slow-moving, weathered rock slide affecting a small community of condominiums in the town of Waynesville in the Blue Ridge Mountains of Haywood County, North Carolina. In November 2005, studies were begun to assess the characteristics of this landslide and the potential for further movement and damage to structures. Work included drilling several boreholes, performing seismic velocity surveys, and surveying benchmarks among other investigations. Data indicate that the potential failure surface is located no more than 11 m below the ground surface, possibly at the contact between saprolite and partially weathered rock. However, inclinometers installed at two locations on the slope have not detected enough movement to corroborate that assessment. Studies continue at this site to determine the location of the failure surface, to identify the mechanisms that accelerate movement, and to relate these findings to a broader understanding of weathered rock slides elsewhere in the southeastern USA.

Presence of Arundinaria gigantea (river cane) on numerous non-wetland sites suggests improper ecological classification of the species

Arundinaria gigantea, or river cane, is a potential riparian buffer, and there is general interest in the restoration of the species to the southern Appalachian mountains. Proper classification of A. gigantea is critical in determining suitable candidate sites for planting. The 1988 National List of Plant Species That Occur in Wetlands lists A. gigantea as facultative wetland (FACW) for the southeast and elsewhere. In this article, we review plant, soil, and hydrologic data collected at 20 A. gigantea sites in western North Carolina, USA, and discuss the species wetland classification in light of our findings. Saturated hydraulic conductivity (Ksat), bulk density (ρb), elemental abundance, soil description, plant identification, and hydrologic data are presented and suggest that river cane is currently classified incorrectly, and that an upland (UPL) or facultative upland (FACU) classification is more appropriate. None of the sites studied meet all 3 criteria common to wetland definitions (hydric soils, hydrophytic vegetation and wetland hydrology), although one of the sites had several wetland characteristics. We suggest that river cane restoration should occur in well drained, sandy soils and that the species should only be used as a riparian buffer in this type of setting.