<p dir="ltr">Non-native biological invasions are one of the leading concerns for global biodiversity. The establishment of invasive species reduces local biodiversity, shifts species composition, changes successional trajectories, and alters ecosystem functions. This thesis examines two aspects of invasive plants: (1) the distribution and the most important climatic and anthropogenic drivers of invasive trees across the eastern United States, and (2) an evaluation of invasive plant removal and herbaceous recovery from a government cost-share program that provides financial support for invasive plant management by private landowners.</p><p><br></p><p dir="ltr">Our first study focused on identifying the distribution of invasive trees, and the factors associated with their distribution. This is essential to predicting spread and planning subsequent management. Using USDA Forest Inventory Analysis (FIA) data and random forest modeling, we examined the distribution, and variables associated with the distribution, of invasive tree species. Invasive trees were found in 10,511 out of 299,387 FIA plots. Invasive species basal area and density (trees per ha; TPH) were highest within the central and southern Appalachian Mountains, Michigan, the Northeast, and the southern Coastal Plain of the United States. A random forest model of invasive species basal area (R²= 0.47, RMSE = 0.47) and density (R²=0.46, RMSE=0.50) vs. environmental variables found that both invasive basal area and density were most strongly associated with human footprint, followed by various climatic variables. An equivalent model of native tree basal (R²=0.53, RMSE=9.25) and TPH (R²=0.47, RMSE=8.64) found that native tree basal area and density were most strongly associated with aridity followed by various climatic variables. As human footprint increased, invasive tree basal area and density increased. These results suggest that the distribution of invasive trees is reliant on human alterations to forests.</p><p><br></p><p dir="ltr">Our second study focused on Environmental Quality Incentives Program (EQIP), a federal cost-share program that has provided $25 billion of financial assistance to farmers and non-industrial private forest owners. Few studies have examined whether this program facilitates the recovery of the herbaceous layer while decreasing the dominance of invasive plant species. We surveyed the herbaceous layer of EQIP-treated and untreated (reference) forests across three physiographic regions of Indiana. Using non-metric multidimensional scaling (NMDS) ordination and linear mixed effects models, we evaluated the species composition, richness, diversity, evenness, floristic quality index, and herbaceous-layer cover of EQIP and reference sites. We also used linear mixed models to evaluate how EQIP site treatment affected the diversity of native plant species. Sites treated with EQIP contracts typically had significantly higher native species richness, Shannon’s diversity, and floristic quality than reference sites. There were significant separations in species composition between EQIP treated and reference forests state-wide and in the southern non-glaciated region of Indiana, although composition overlapped between EQIP and reference forests. Our study suggests that EQIP-funded treatments promote increased species richness and diversity. However, the persistent overlap in species composition we observed may signify biotic homogenization due to a long-shared history of anthropogenic disturbances between EQIP and reference sites. Therefore, active restoration of the herbaceous layer might be needed to allow a full recovery after invasive removal.</p>

<p dir="ltr">To reduce the establishment and spread of invasive plant species, the Environmental Quality Incentive Program (EQIP) was created in 1996 to provide financial and technical assistance to private landowners to aid in conservation practices and address environmental concerns. From 2014-2022, approximately $90 million dollars was obligated to the EQIP for completed contracts of over 240,000 hectares in Indiana. However, to date, there has been no examination of whether participation in conversation cost sharing programs has resulted in the recovery of native tree diversity, growth, and reproduction following the treatment of invasive plants. Furthermore, there is a gap in our understanding of the effectiveness of EQIP and its success in achieving and maintaining management goals. This thesis quantifies the composition, diversity, regeneration density and growth of tree species in response to invasive plant treatments at sites that participated in the EQIP. In particular, I investigated how the species composition of woody seedlings (stems < 2 cm dbh) and woody saplings (2 – 5 cm dbh) differed across EQIP-treated and untreated reference plots. I then examined how native species richness and diversity values differed between EQIP-treated and untreated plots. Lastly, I used dendroecological methods to determine how treatment of invasive shrubs affected overstory tree growth. Across the state, I found that native species richness of seedlings and saplings was greater in EQIP-treated plots then within reference plots. Reference plots were associated with invasive species such as <i>Lonicera maackii</i> (Amur honeysuckle), <i>Rosa multiflora</i> (multiflora rose), <i>Elaeagnus umbellata</i> (autumn olive) and <i>L. japonica</i> (Japanese honeysuckle), while EQIP-treated plots were associated with native tree species, such as <i>Carya ovata</i> (shagbark hickory),<i> Ulmus americana </i>(American elm), <i>Fraxinus americana</i> (white ash), <i>Liriodendron tulipifera</i> (yellow-poplar), <i>Quercus alba</i> (white oak), <i>Q. velutina</i> (black oak), and shrubs, such as <i>Rubus allegheniensis</i> (Allegheny blackberry) and <i>Lindera benzoin</i> (spicebush). I observed that trees generally had greater basal area increment growth following invasive shrub treatments; therefore, the reduction of invasive shrub dominance on EQIP-enrolled lands has led to a small, but significant, increase in overstory tree growth. I observed high within-group variability in growth for EQIP-treated and reference plots, likely due to differing management strategies across privately owned forests. Overall, my results demonstrate that participation in the EQIP has positively benefited species richness, and native species composition, and tree growth.</p>

<p>Land use change constitutes a primary driving force in shaping social-ecological systems world wide, and its effects reach far beyond the directly impacted areas. Graph based landscape ecological tools have become established as a promising way to efficiently explore and analyze the complex, spatial systems dynamics of ecological networks in physical landscapes. However, little attention has been paid to making these approaches operational within ecological assessments, physical planning, and design. This thesis presents a network based, landscape-ecological tool that can be implemented for effective use by practitioners within physical planning and design, and ecological assessments related to these activities. The tool is based on an ecological profile system, a common generalized network model of the ecological infrastructure, graph theoretic metrics, and a spatially explicit, geographically defined representation, deployable in a GIS. Graph theoretic metrics and analysis techniques are able to capture the spatio-temporal dynamics of complex systems, and the generalized network model places the graph theoretic toolbox in a geographically defined landscape. This provides completely new insights for physical planning, and environmental assessment activities. The design of the model is based on the experience gained through seven real-world cases, commissioned by different governmental organizations within Stockholm County. A participatory approach was used in these case studies, involving stakeholders of different backgrounds, in which the tool proved to be flexible and effective in the communication and negotiation of indicators, targets, and impacts. In addition to successful impact predictions for alternative planning scenarios, the tool was able to highlight critical ecological structures within the landscape, both from a system-centric, and a site-centric perspective. In already being deployed and used in planning, assessments, inventories, and monitoring by several of the involved organizations, the tool has proved to effectively meet some of the challenges of application in a multidisciplinary landscape.</p>

Land use change constitutes a primary driving force in shaping social-ecological systems world wide, and its effects reach far beyond the directly impacted areas. Graph based landscape ecological tools have become established as a promising way to efficiently explore and analyze the complex, spatial systems dynamics of ecological networks in physical landscapes. However, little attention has been paid to making these approaches operational within ecological assessments, physical planning, and design. This thesis presents a network based, landscape-ecological tool that can be implemented for effective use by practitioners within physical planning and design, and ecological assessments related to these activities. The tool is based on an ecological profile system, a common generalized network model of the ecological infrastructure, graph theoretic metrics, and a spatially explicit, geographically defined representation, deployable in a GIS. Graph theoretic metrics and analysis techniques are able to capture the spatio-temporal dynamics of complex systems, and the generalized network model places the graph theoretic toolbox in a geographically defined landscape. This provides completely new insights for physical planning, and environmental assessment activities. The design of the model is based on the experience gained through seven real-world cases, commissioned by different governmental organizations within Stockholm County. A participatory approach was used in these case studies, involving stakeholders of different backgrounds, in which the tool proved to be flexible and effective in the communication and negotiation of indicators, targets, and impacts. In addition to successful impact predictions for alternative planning scenarios, the tool was able to highlight critical ecological structures within the landscape, both from a system-centric, and a site-centric perspective. In already being deployed and used in planning, assessments, inventories, and monitoring by several of the involved organizations, the tool has proved to effectively meet some of the challenges of application in a multidisciplinary landscape.