Characterizing the Network Structure of Headwater Riparian Wetlands in yhe Ridge and Valley Region, Pennsylvania
Open Access
- Author:
- Tyrna, Abbey Anne
- Graduate Program:
- Geography
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- December 12, 2014
- Committee Members:
- Robert P Brooks, Dissertation Advisor/Co-Advisor
Denice Heller Wardrop, Committee Member
Erica A H Smithwick, Committee Member
Elizabeth Weeks Boyer, Committee Member - Keywords:
- wetlands
hydrology
land use disturbance
graph theory - Abstract:
- This dissertation aims to illuminate the cumulative impact of anthropogenic disturbances on the collective capacity of wetland functions. Changes to wetland hydrologic regime and subsequent connectivity in response to landscape disturbances are poorly understood. However, studies have shown that anthropogenic disturbances have decreased ecosystem services in regional watersheds leading to lower water quality conditions. Considering the multi-scale impacts to wetland hydrology in the context of connectivity can further the understanding of important and fundamental interactions of the terrestrial and aquatic realms that contribute to water quality. This work provides a framework for an integrated ecological assessment that can be used to better manage wetland-derived functions and ecosystem services. Chapter 1 provides a conceptual framework for considering wetlands and their aquatic and terrestrial linkages, and outlines the overall research objectives. Hydrologic responses of 10 wetlands classified by disturbance level (highly disturbed and least disturbed or reference standard condition) were examined in Chapter 2. Hydrologic regime variation across the two disturbance groups was much greater than the variation within each group. This suggests that disturbance overrides hydrologic characterization by hydrogeomorphic subclass for riverine upper perennial wetlands. For all sites, percent saturation ranged from 0 - 100% during the growing season. For reference standard sites, percent saturation ranged from 58 - 100%. Highly disturbed sites, which each had a unique package of perturbations, had a larger variation in percent saturation both over the entire study period (0 - 78%) and within a single growing season (3 - 79%). All wetland hydrologic regime characteristics were strongly correlated with disturbance score. However, when examined individually, no one landscape, buffer zone, or site level feature could explain the variability in all wetland hydrologic regime characteristics. Hydrologic regime was used to infer hydrologic flow paths and resultant hydrologic connections between wetlands, streams and terrestrial upland. Wetland and stream inventories in central Pennsylvania were enhanced and wetland condition was documented in Chapter 3. Wetland inventories were enhanced by field observations and condition was assessed using Level 1 and Level 2 assessment tools. Fifty-seven new wetlands totaling 23 ha (57 ac) and three National Wetland Inventory (NWI) wetland extensions equaling 5 ha (12 ac) were added to the Upper Shaver Creek wetland inventory. The new inventory nearly doubled the existing wetland area for headwater riparian wetlands within Upper Shaver Creek watershed. The stream length for the newly inventoried segments equaled 35 km adding nearly one-third more of stream length to the watershed inventory. Level 1 assessment revealed that one-third of the wetland resource in Upper Shaver Creek was rated in the highest or high category of ecological condition and 17% were in lowest category. Wetland buffers, which are known for maintaining quality and reducing landscape scale impact, were discovered to be narrower in the riparian zones that were most impacted by agricultural uses. The importance of a strong buffer was evidenced by the fact that wetlands with < 5% invasive species cover had the highest range of buffer scores. Furthermore, median buffer score decreased as invasive species cover increased. Finally, Level 2 assessment methods revealed wetland condition scores that ranged from severely disturbed (condition score = 7.5) to high ecological integrity (condition score = 99.9), with a mean score in the lower range of the high condition category (M = 59.0, SD = 27.4). The median rapid assessment score for wetlands with 0-5% invasive species cover was much higher than for all other invasive species cover categories. This study revealed that land use in the immediate surrounding area (buffer zone) was more important to reducing the factors contributing to the spread of invasive species than land use at the landscape scale. Finally, a stream-riparian wetland-terrestrial network is assembled using a graph theoretical approach in Chapter 4. This approach presents a framework for connectivity assessments that integrate terrestrial, aquatic and riparian wetland landscapes to rank riparian wetlands by positional importance in network flow. Nodes and edges create the network structure. Stream reaches that extend out to the riparian landscape became nodes connected by confluences as edges. Nodes were weighted by the amount of wetland contained, the quality of those wetlands, and the amount of agricultural land use within its contributing area. Local cumulative wetland influence was calculated by weighted in-degree. Weighted betweenness centrality was calculated and illustrated the places in the landscape that were critical to network flow. Node importance metrics such as the attribute value of the focal node showed that 15 out of the 214 nodes in the network carried 40% of all the attribute weight. Another node importance metric, integral index of connectivity, was used along with component analysis to quantify the capacity and opportunity of the stream landscape to perform biogeochemical cycling functions. The graph theoretical approach to model the positional influence of riparian wetland functions on watershed water quality and to assess the cumulative impact of riparian wetland loss has potential to become a productive way to gain further understanding of these complex systems.