Determinants of species composition, functional leaf traits, and floristic quality in restored wetlands

Charles, Brian

Permalink

https://hdl.handle.net/2142/113849 Copy

Description

Title

Determinants of species composition, functional leaf traits, and floristic quality in restored wetlands

Author(s)

Charles, Brian

Issue Date

2021-11-24

Director of Research (if dissertation) or Advisor (if thesis)

Matthews, Jeffrey W

Department of Study

Natural Res & Env Sci

Discipline

Natural Res & Env Sciences

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• Wetland restoration
• invasive plants
• functional traits

Abstract

Wetlands are exceptionally important ecosystems, as they provide ecosystem services like flood abatement and excess nutrient removal. Restoration is a way to enhance these services, and restoration practitioners typically use measures of vegetation, such as plant species composition, to assess the quality of restored sites. Determining what influences plant species composition, how species composition is changing in restored wetlands, and how best to monitor wetlands is important for better understanding restored wetlands and subsequently improving the wetland restoration process. To assess these components of wetland restoration, I surveyed 66 100-m2 plots in 22 wetlands in Illinois that were restored between 1991 and 2010. I collected data on herbaceous plant cover and canopy cover at each plot in 2020 and also collected leaves from herbaceous species to determine specific leaf area (SLA) and leaf dry matter content (LDMC) at each plot. I used previously collected hydrologic well monitoring data to obtain 6 hydrological variables related to the frequency, depth, and duration of inundation. Soil samples were collected to determine soil pH, organic matter content, and nitrogen and phosphorous content at each plot. I performed canonical correspondence analysis to relate predictor variables (hydrological variables, soil variables, canopy cover, time since restoration, and latitude) to herbaceous species composition. I used structural equation modeling to understand how the predictor variables influence each other and ultimately how they influence standard vegetation monitoring metrics (mean coefficient of conservatism, species richness, percent non-native species cover) as well as SLA and LDMC. I also performed non-metric multidimensional scaling to assess changes in species composition between 2015 and 2020, and then used a generalized random effect linear model to determine which predictor variables influenced plot-specific change between 2015 and 2020. Latitude primarily accounted for differences in species composition among wetlands, likely because plots in northern Illinois were often dominated by Phalaris arundinacea, an invasive species that led to stasis in species composition from 2015 to 2020. Plots with greater canopy cover were associated with greater compositional change from 2015 to 2020, likely due to the lack of P. arundinacea in forests, combined with species turnover caused by succession from open plots to forested plots. LDMC and SLA had weak relationships with abiotic predictor variables and low variance explained in the structural equation models, whereas standard vegetation monitoring metrics had stronger relationships with predictor variables and moderate variance explained. The results indicate that P. arundinacea has halted succession where it is present and offer insight into patterns of change in restored wetlands. The results also indicate that SLA and LDMC may not be well suited for use as wetland restoration monitoring metrics, and that standard vegetation monitoring metrics may be more closely related to abiotic plot characteristics.

Graduation Semester

2021-12

Type of Resource

Thesis

Permalink

http://hdl.handle.net/2142/113849

Copyright and License Information

Copyright 2021 Brian Charles

Owning Collections