Defense Date


Document Type


Degree Name

Master of Science



First Advisor

Dr. Leonard A. Smock


Twenty five first through third order streams in the Coastal Plain of Virginia were sampled for benthic macroinvertebrates and fishes to determine whether a predictable relationship between areas of Unfragmented Natural Land Cover (UNLC) and biotic integrity could be established. I hypothesized that as the area of UNLC increased in a watershed at either the whole catchment or riparian scale, biotic indices measuring stream water and habitat quality would increase. Biotic integrity was measured through the scores from the Coastal Plain Macroinvertebrate Index (CPMI) for benthic macroinvertebrates and the VCU Index of Biotic Integrity (IBI) for fishes. Using GIS, the percentage of UNLC at the catchment and riparian scale was calculated for each stream's watershed. Physicochemical parameters, habitat metrics and other environmental data were also analyzed to determine if relationships existed between those parameters and biotic integrity. Unfragmented Natural Land Cover ranged from 7% to 82% at the catchment scale and 10%to 96% in the riparian area. There were no significant correlations between the biological assessment scores for either the benthic macroinvertebrate or the fish communities and UNLC at either scale. Analyses of physicochemical parameters and habitat metrics did show some significant correlations between those variables and biotic metrics. Dissolved oxygen (DO) and pH were positively correlated with the CPMI and DO was positively correlated with the IBI scores. Several habitat metrics were significantly correlated with the CPMI, including pool variability, which was positively correlated with the CPMI, and bank stability, sediment deposition, and channel flow status, which were negatively correlated with the CPMI. The results of this study indicated that streams with unconstrained channels score significantly lower on the CPMI and have significantly lower DO concentrations than streams with constrained channels despite some streams with unconstrained channels having high percentages of UNLC in the watershed. Although there were other biotic and abiotic factors that may have introduced variability into the study, such as severe weather, beaver activity, and changing land use, it is likely that the CPMI was not an appropriate bioassessment tool for swampy Coastal Plain streams. It is therefore imperative from assessment and management perspectives for state agencies and researchers to develop appropriate bioassessment indices for Coastal Plain streams that have limiting water quality influenced by natural processes.


© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission

June 2008

Included in

Biology Commons