Untangling the confounding effects of urbanization and high water level on the cover of emergent vegetation in Cootes Paradise Marsh, a degraded coastal wetland of Lake Ontario Journal Articles uri icon

  •  
  • Overview
  •  
  • Research
  •  
  • Identity
  •  
  • Additional Document Info
  •  
  • View All
  •  

abstract

  • An approach based on a digital elevation model (DEM) was used to untangle the confounding effects of long-term water-level fluctuations and increasing human population on the cover of emergent vegetation in Cootes Paradise Marsh, a degraded coastal wetland in Lake Ontario, Canada. Data for 20 observations between 1934 and 1993 were used in the analysis. First, we calculated the inundated area based on the DEM, a derived measurement that reflected the bathymetry of the marsh and the mean water level for a particular year. Then Mantel correlations and regression analyses were used to analyze the relationships between emergent cover and corresponding water level, inundated area, and human population, respectively. Results of the simple and partial correlations indicated that areal change of emergent plants was significantly correlated with inundated area after controlling for the effect of water level fluctuation; however, there was no significant correlation between emergent cover and water level after controlling for inundated area. This is an important consideration when multiple sites from the same Great Lake are compared since the same water level may correspond to vastly different inundated areas for different marshes. Changes in emergent cover were also significantly correlated with human population after controlling for water level effects. Altogether, inundated area explained 83.1% of the variation, human population explained 4.2%, and the interaction between population and inundated area explained an additional 4.3% of the remaining variation in areal emergent cover. This indicates that the synergistic effect of high water level (expressed as inundated area) and increased human population induced greater detrimental impact on the emergent plants than did either stressor alone. © Springer 2005.

publication date

  • August 2005