Donald L. Hagan

Improving methods to evaluate the impacts of plant invasions: lessons from 40 years of research.

Here the authors review the research methods used to measure the ecological effects of non-native plant invasions. In their synthesis they find that although the number of studies on invasion impacts has increased markedly in recent years, there is a lack of experimental studies, a bias among invader functional groups, and relatively few studies on ecosystem effects of invasions. They recommend utilization of longer-term studies that combine broad-scale observations, experimental manipulations, and predictive modelling across diverse invader functional groups and affected ecosystems to provide more comprehensive insight into the impacts of plant invasions.

Impacts of repeated wildfire on long-unburned plant communities of the southern Appalachian Mountains

The infrequent occurrence of large wildfires in the southern Appalachian Mountains over the last several decades has offered few opportunities to study their impacts. From 2000 to 2008, five wildfires burned a large portion of the area in and surrounding the Linville Gorge Wilderness in North Carolina. Areas were burned either once or twice. The response of acid cove and thermic oak plant communities (structure, cover, richness, diversity) was measured in 78 vegetation monitoring plots, established in 1992 and remeasured in 2010–11. Fire altered forest structure in both communities, resulting in the mortality of larger trees and increases in the abundance of smaller (<5 cm diameter at breast height (DBH)) stems. Burning twice decreased stem counts for mountain laurel (Kalmia latifolia) in both communities, whereas oaks (Quercus spp.) responded positively to burning twice in the thermic oak community. Table Mountain pine stem counts increased in acid cove and thermic oak communities burned once. Fire appears to promote princesstree (Paulownia tomentosa) invasion. Herbaceous species cover responded positively to fire (once or twice; both communities), with concurrent increases in woody species richness and diversity. Tree species composition in acid cove plots was not affected by burning, although some slight changes occurred in thermic oak plots burned twice.

Cogongrass (Imperata cylindrica) Invasion and Eradication: Implications for Soil Nutrient Dynamics in a Longleaf Pine Sandhill Ecosystem

Abstract We assessed pre- and posteradication nitrogen and phosphorus dynamics in longleaf pine sandhill stands severely affected by cogongrass. Across a 7-yr posteradication (glyphosate + imazapyr) “recovery chronosequence,” which included untreated cogongrass, uninvaded reference, and treated plots, we analyzed soils for total N, potentially available P (Mehlich-1 [M1]), pH, and organic matter content. We also used resin bags to assess fluxes of plant available N and P in the soil solution. Additionally, we used litterbags to monitor the decomposition and nutrient mineralization patterns of dead rhizome and foliage tissue. Our results indicate similar total N and M1-P contents in both cogongrass-invaded and uninvaded reference plots, with levels of M1-P being lower than in cogongrass plots for 5 yr after eradication. Soil organic matter did not differ between treatments. Resin bag analyses suggest that cogongrass invasion did not affect soil nitrate availability, although a pulse of NO2 + NO3 occurred in the first 3 yr after eradication. No such trends were observed for ammonium. Resin-adsorbed PO4 was lowest 3 yr after eradication, and pH was highest 5 yr after eradication. Our litterbag study showed that approximately 55% of foliar biomass and 23% of rhizome tissue biomass remained 18 mo after herbicide treatment. Substantial N immobilization was observed in rhizomes for the first 12 mo, with slow mineralization occurring thereafter. Rapid P mineralization occurred, with 15.4 and 20.5% of initial P remaining after 18 mo in rhizomes and foliage, respectively. Overall, our findings indicate that cogongrass invasion has little to no effect on soil nutrient cycling processes, although some significant—but ephemeral—alterations develop after eradication. Nomenclature: Glyphosate, imazapyr, cogongrass, Imperata cylindrica (L.) Beauv. IMCY, longleaf pine, Pinus palustris Mill Management Implications: An improved understanding of how invasive alien plant species alter soil properties and of how novel soil properties persist after eradication is essential for practitioners to develop effective restoration strategies for invaded natural communities. Toward these ends, we undertook this study to assess nutrient dynamics in forest stands affected by cogongrass—an invasive alien plant that affects tropical and subtropical ecosystems on six continents. We chose a longleaf pine sandhill ecosystem as the study site because these forests are frequently targeted in restoration efforts and are commonly invaded by cogongrass in the southeastern United States. Because nitrogen (N) and/or phosphorus (P) availability often affect plant establishment after disturbance, we focused on processes that affect N/P availability and uptake. Our results indicate that cogongrass invasion does not substantially alter soil properties. This is intriguing, given that this species is known to be a superior competitor for belowground resources and often forms dense, single-species stands in longleaf pine ecosystems. The decomposition of cogongrass biomass after herbicide treatment resulted in a rapid pulse of P into the soil and the temporary immobilization of soil N. All measured soil properties (N, P, pH, organic matter), however, returned to a preinvasion state within 5 yr. These results suggest that long-term posteradication soil “legacy” effects, which have been observed with other invasive alien plant species (e.g., alterations to the nitrogen cycle and other biogeochemical cycles), might not occur in cogongrass-affected pine ecosystems—at least those that have been invaded for a relatively short period of time. In light of these findings, we see little evidence to indicate that altered soil properties would impede groundcover restoration efforts in sites that have been invaded by cogongrass for 1–2 yr, as long as practitioners give the soil adequate time to recover after eradication.

Interspecific competition enhances nitrogen fixation in an actinorhizal shrub

In forest understory restoration, the establishment of reintroduced species may be strongly linked to their ability to compete for belowground resources. In this study, we provide isotopic and morphological evidence for competition-induced increases in nitrogen fixation by Morella cerifera (L.) Small (wax myrtle) when planted with Pinus palustris Mill (longleaf pine). Compared to a competition-free treatment, we found no significant differences in tissue N concentrations for M. cerifera. However, 15N enrichment in leaves, stems and roots, as well as whole-plant values for nitrogen derived from fertilizer were significantly lower when the plants were subject to interspecific competition from P. palustris. Plants in the competition treatment also allocated a significantly greater percentage of belowground biomass to root nodules than those in the competition-free treatment (0.65 vs. 0.41%). This strongly suggests that M. cerifera is capable of upregulating nitrogen fixation in response to interspecific competition. This may help explain why M. cerifera outperformed non-nitrogen-fixing species reintroduced on the same site.

Partitioning of applied 15N fertilizer in a longleaf pine and native woody ornamental intercropping system

The cultivation of ornamentals to produce woody floral products—the fresh or dried stems that are used for decorative purposes—may be an attractive option for southeastern landowners looking to generate income from small landholdings. Since many shrubs native to the understory of the longleaf pine (Pinus palustris Mill.) ecosystem have market potential, one possibility is the intercropping of select species in the between-row spacing of young longleaf pine plantations. The objective of this study was to evaluate how interspecific competition affects the fate of 15N fertilizer when American beautyberry (Callicarpa americana L.), wax myrtle (Morella cerifera (L.) Small) and inkberry (Ilex glabra (L.) A.Gray) are intercropped with longleaf pine. Nitrogen derived from fertilizer (NDF), utilization of fertilizer N (UFN) and recovery of fertilizer N (RFNsoil) were compared between agroforestry and monoculture (treeless) treatments to assess the effects of competition. Results varied by species, with NDF being higher for C. americana foliage and lower for all M. cerifera tissues in the agroforestry treatment. No effect was observed for I. glabra. UFN was lower for all species in the agroforestry treatment. RFNsoil was higher in the agroforestry treatment for I. glabra, but no treatment effects were observed for C. americana or M. cerifera. Overall, while it is clear that interspecific competition was present in the agroforestry treatment, the inefficiency of fertilizer use suggests that nitrogen was not the most limiting resource. Management interventions, particularly those that address competition for water, will likely be critical to the success of this system.

Frequent Prescribed Burning as a Long-term Practice in Longleaf Pine Forests Does Not Affect Detrital Chemical Composition.

The O horizon, or detrital layer, of forest soils is linked to long-term forest productivity and health. Fuel reduction techniques, such as prescribed fire, can alter the thickness and composition of this essential ecosystem component. Developing an understanding of the changes in the chemical composition of forest detritus due to prescribed fire is essential for forest managers and stakeholders seeking sustainable, resilient, and productive ecosystems. In this study, we evaluated fuel quantity, fuel structure, and detrital chemical composition in longleaf pine ( Miller) forests that have been frequently burned for the last 40 yr at the Tom Yawkey Wildlife Center in Georgetown, SC. Our results suggest that frequent prescribed fire reduces forest fuel quantity ( < 0.01) and vertical structure ( = 0.01). Using pyrolysis-gas chromatography/mass spectrometry as a molecular technique to analyze detrital chemical composition, including aromatic compounds and polycyclic aromatic hydrocarbons, we found that the chemical composition of forest detritus was nearly uniform for both unburned and burned detritus. Our burning activities varied in the short term, consisting of annual dormant, annual growing, and biennial dormant season burns. Seasonal distinctions were present for fuel quantity and vertical fuel structure, but these differences were not noted for the benzene/phenol ratio. These results are significant as more managers consider burning existing longleaf stands while determining effective management practices for longleaf stands yet to be established. Managers of such stands can be confident that frequent, low-intensity, low-severity prescribed burns in longleaf pine forests do little to affect the long-term chemical composition of forest detritus.

Thermocouple Probe Orientation Affects Prescribed Fire Behavior Estimation

Understanding the relationship between fire intensity and fuel mass is essential information for scientists and forest managers seeking to manage forests using prescribed fires. Peak burning temperature, duration of heating, and area under the temperature profile are fire behavior metrics obtained from thermocouple-datalogger assemblies used to characterize prescribed burns. Despite their recurrent usage in prescribed burn studies, there is no simple protocol established to guide the orientation of thermocouple installation. Our results from dormant and growing season burns in coastal longleaf pine ( Mill.) forests in South Carolina suggest that thermocouples located horizontally at the litter-soil interface record significantly higher estimates of peak burning temperature, duration of heating, and area under the temperature profile than thermocouples extending 28 cm vertically above the litter-soil interface ( < 0.01). Surprisingly, vertical and horizontal estimates of these measures did not show strong correlation with one another ( ≤ 0.14). The horizontal duration of heating values were greater in growing season burns than in dormant season burns ( < 0.01), but the vertical values did not indicate this difference ( = 0.52). Field measures of fuel mass and depth before and after fire showed promise as significant predictive variables ( ≤ 0.05) for the fire behavior metrics. However, all correlation coefficients were less than or equal to = 0.41. Given these findings, we encourage scientists, researchers, and managers to carefully consider thermocouple orientation when investigating fire behavior metrics, as orientation may affect estimates of fire intensity and the distinction of fire treatment effects, particularly in forests with litter-dominated surface fuels.

Landscape, Soil, and Plant Analysis of Japanese Stiltgrass (Microstegium vimineum) Invasion in the Piedmont Region, SC

ABSTRACT Japanese stiltgrass (Microstegium vimineum), an exotic invasive plant, is native to Southeast Asia. This study was conducted to determine the chemical composition of Japanese stiltgrass as well as soil and landscape characteristics that correlate with invasion of Japanese stiltgrass around Lake Issaqueena in the upper Piedmont of South Carolina. Geographic Information Systems (GISs) and Light Detection and Ranging (LiDAR) were used to determine the spatial pattern of invasion with respect to the aspect, slope, canopy cover, soils, and distance to roads and trails. Japanese stiltgrass was distributed on both sides of Lake Issaqueena in Pacolet and Madison soil map units (Fine, kaolinitic, thermic Typic Kanhapludults) on the average slopes of 21%, but it was particularly common on the eastern shore of the lake in low-lying wet and shaded areas (mean canopy cover 51%). In addition, invasion by Japanese stiltgrass was correlated with the proximity to roads and trails. Plant tissue analysis revealed many differences in the distribution of macronutrients, macrominerals, and micronutrients in the leaves, stems, and roots of Japanese stiltgrass, although those differences were not always statistically significant. Concentrations of nitrogen (N), phosphorus (P), and calcium (Ca) were the highest in leaves while zinc (Zn) concentrations were the highest in stems and concentrations of magnesium (Mg), copper (Cu), manganese (Mn), iron (Fe), and sodium (Na) tended to be higher in roots. Carbon (C), sulfur (S), and potassium (K) concentrations were generally higher in above-ground tissues versus roots. Soil chemical analysis revealed no statistical differences between control and invaded plots. Our findings suggest that watershed areas surrounding lakes may be particularly susceptible to the invasion of Japanese stiltgrass due to their microclimates, low-lying wet pathways for seed distribution and recreational uses.

Geospatial Soil and Plant Tissue Analysis of Sericea Lespedeza (Lespedeza cuneata) Invasion around Lake Issaqueena, South Carolina

ABSTRACT This study was conducted to determine the chemical composition of sericea lespedeza (Lespedeza cuneata) as well as soil and landscape characteristics that correlate with its invasion. Geographic Information Systems (GIS) and Light Detection and Ranging (LiDAR) were used to determine the pattern of invasion and to quantify landscape metrics. Sericea lespedeza was distributed on both sides of the lake in Pacolet and Madison soil map units (Fine, kaolinitic, thermic Typic Kanhapludults) on average slopes of 17.5%. It was common along roads and trails with a preferred mean canopy cover of <50%. Plant tissue analysis revealed statistically higher concentrations of macronutrients nitrogen (N), phosphorus (P), potassium (K), magnesium (Mg), calcium (Ca) and sulfur (S) and selected micronutrients in leaves compared to stems and roots. Micronutrients copper (Cu), manganese (Mn) and iron (Fe) were significantly higher in stems and roots. Soil analysis results showed no statistical difference between control and invaded plots for measured soil chemistry parameters.

The Role of Soil and Landscape Factors in Chinese Privet (Ligustrum sinense) Invasion in the Appalachian Piedmont

Abstract There is a limited understanding about the ecological mechanisms that enable certain plant species to become successful invaders of natural areas. This study was conducted to determine the soil and landscape characteristics that correlate with invasion of Chinese privet (CHP), and to develop a model to predict the probability of CHP invasion in Piedmont forests. A landscape ecosystem classification (LEC) system—based on the percentage of clay in the B horizon, depth to maximum clay (cm), exposure, terrain shape, and aspect (degrees)—was used to determine the soil moisture characteristics of invaded and uninvaded plots. Additional measurements included the cover classes of CHP and other species, litter depth (cm), slope (degrees), overstory basal area (m2 ha−1), and soil chemical properties. CHP invasion was negatively correlated with overstory basal area and slope and positively with litter depth and pH. A stepwise logistic regression model containing these four variables was highly sensitive, with an overall accuracy of 78%. Given the accuracy of this model, we propose that it can be used to calculate the probability of invasion in a given area, provided that some basic, readily obtainable site characteristics are known. Nomenclature: Chinese privet, Ligustrum sinense Lour. Management Implications: Chinese privet (CHP) (Ligustrum sinense Lour) is one of the most common woody invasive alien plant species in the Appalachian Piedmont. An improved understanding of the factors that correlate with CHP invasion will benefit land managers in the region, as this species can reduce native diversity, alter forest structure, and be costly to control. We found CHP invasion to be negatively correlated with overstory basal area and slope and positively correlated with litter depth and pH. A model containing these four variables was highly sensitive, being able to predict CHP invasion 78% of the time. By identifying the areas that are most likely to be invaded, this model could facilitate early detection and control of CHP, thereby slowing its spread and helping to conserve native flora and fauna.