2001 Progress Report: Riparian Reforestation in an Urbanizing Watershed: Effects of Upland Conditions on Instream Ecological BenefitsEPA Grant Number: R825798
Title: Riparian Reforestation in an Urbanizing Watershed: Effects of Upland Conditions on Instream Ecological Benefits
Investigators: Hession, W. C. , Charles, Donald F. , Hart, D. D. , Horwitz, R. J. , Johnson, T. E. , Kreeger, D. A. , Marshall, B. , Newbold, J. Denis , Pizzuto, J. E. , Velinsky, D. J.
Current Investigators: Hession, W. C. , Charles, Donald F. , Hart, D. D. , Horwitz, R. J. , Kreeger, D. A. , Newbold, J. Denis , Pizzuto, J. E. , Velinsky, D. J.
Institution: Academy of Natural Sciences , University of Delaware , University of Vermont
Current Institution: Academy of Natural Sciences , University of Delaware
EPA Project Officer: Hahn, Intaek
Project Period: June 1, 1998 through May 31, 2001
Project Period Covered by this Report: June 1, 2000 through May 31, 2001
Project Amount: $837,685
RFA: Ecosystem Restoration (1997) RFA Text | Recipients Lists
Research Category: Ecological Indicators/Assessment/Restoration , Hazardous Waste/Remediation , Land and Waste Management , Ecosystems
The objectives of this research project are to: (1) understand the effects of riparian reforestation on the structure, function, and dynamics of stream ecosystems in urbanizing watersheds; (2) define exposure-response relationships for structural and functional stream ecosystem attributes (chemical, physical, and biological) based on the level of upland or contributing watershed urbanization for streams with and without riparian forests; (3) develop models so local and state decisionmakers can prioritize riparian reforestation efforts based on expected ecological benefit; and (4) measure and evaluate the rates at which various chemical, physical, and biological variables respond to riparian reforestation, to define effectiveness criteria for a restoration activity that may take many years to be fully manifested.
As of September 2001, all project field work and laboratory counting/ID has been completed. The previous 3 years of field work included physical, chemical, and biological sampling at 12 "paired-reach" study sites composed of adjacent forested and meadow stream reaches, sampling at four forested stream reaches in highly urbanized watersheds, and sampling at two restored stream reaches. Six of the paired-reach sites, two of the forested reaches, and the two restored reaches were sampled during the 1998/1999, fall/winter field season, and are referred to as Year 1 sites. The remaining six paired-reach sites and two forested reaches were sampled during the 1999/2000 field season, and are referred to as Year 2 sites. Replicate sampling at two of the paired-reach sites sampled during Year 1, and at one of the restored sites sampled during Year 1 also was conducted during the Year 2 and Year 3 field seasons (1999/2000, and 2000/2001, respectively). Replicate sampling at the restored site was conducted to monitor ecological changes over time following restoration, and at the paired-reach sites to monitor annual variability in ecological attributes during the project period. Sampling at each site included watershed land cover, channel morphology, physical habitat, bank erosion, riparian vegetation, algae community composition, macroinvertebrate community composition, fish community composition, food web linkages, and nutrient uptake rates by benthic organisms.
During the year from October 1, 2000, to September 30, 2001, all remaining Year 3 field biological sampling and laboratory counting/ID was completed. Data were checked for quality, and entered into a computer database. Also during this period, tracer injection studies to investigate nutrient uptake by in-stream processes were completed at three of the paired-reach sites and one of the restored sites. We have now completed all field work specified in our scope of work, and are working on the final analysis, interpretation, and presentation/publication of results. The following paragraphs provide a brief description of work completed from October 1, 2000, to September 30, 2001.
Watershed Characterization. Additional work was done to characterize watershed land cover and the percent impervious surfaces. Land cover statistics were calculated using a second land cover data set obtained from the Delaware Valley Regional Planning Commission. This data, based on aerial photography, is both more recent and better represents the amount of impervious cover in watersheds as compared to the Landsat-based data previously used. Land cover statistics were calculated for each site (watershed) based on the whole watershed, land within a 100 meter buffer around streams, and land within a 300 meter buffer around streams. A "hydrologic change" index based on NRCS curve numbers also was developed and calculated for each site. This index was developed to account for local differences in the way impervious surfaces affect hydrology, and may correlate more closely with ecological metrics than direct measures of land cover. Urban development indices together with field sampling data will be used to evaluate the influence of urban development on stream ecology.
Riparian Vegetation. Quantitative measures of riparian vegetation at each site were reviewed and additional indices of riparian "condition" were developed. In particular, efforts were made to incorporate information about the nutritional quality of various leaf types into an index. Other calculated indices include basal area of trees, canopy density, and the abundance of native versus exotic species.
Geomorphology. Progress was made in data analysis and interpretation. Preliminary results suggest that riparian vegetation is more important than urban development in terms of influencing channel morphology, and that bank erosion rates are greater in meadow reaches than forested reaches.
Chemistry/Nutrients. Nutrient cycling experiments using injected bromide, nitrogen, and phosphorus tracers were conducted at three of the paired-reach sites and one restored site. The one restored site was studied during all 3 years of this project to assess site variability. Samples from these experiments currently are being analyzed for the concentrations of dissolved bromide (tracer), ammonium, and soluble reactive phosphorus. These experiments will provide information about nutrient uptake in forested and nonforested stream reaches along an urban gradient. Once all samples have been analyzed, computer modeling will be used to assess spiral lengths and uptake of N and P.
Preliminary statistical analysis of routine water quality data (i.e., nutrient data) from the study sites do not indicate a strong relationship with impervious surfaces. Concentrations of dissolved ammonium and soluble reactive phosphorus were lower in open reaches than forested reaches at some sites, suggesting greater uptake by benthic algae or bacteria. However, this was no consistent trend throughout all sites. Further analysis will be performed to assess relationships between land use, biological characteristics, and water quality.
Algae. Year 3 sampling at selected sites was conducted during the early fall of 2000. Algae and diatom samples from Year 3 were identified and counted in the laboratory. Progress also was made with data analysis and interpretation. Preliminary results suggest that forested reaches had higher Shannon-Weiner diversity values and lower percent dominance values compared with open reaches. Algal biomass was significantly higher in open reaches, and was associated with a proliferation of green alga Cladophora sp. and Spirogyra sp. Results of ordination and Monte Carlo analysis show that the environmental variables basin area, percent forest, percent urbanization, N:P and pH contributed significantly to the variability in diatom assemblages between sites.
Macroinvertebrates. Year 3 sampling at selected sites was conducted during the winter of 2001. Sampling was conducted in February to coincide with the period of maximum larval abundance in low order streams of southeastern Pennsylvania. All macroinvertebrate samples from Year 3 were counted and identified, and biomass was estimated. Progress also was made with data analysis and interpretation. Preliminary results suggest that forested reaches have slightly greater species diversity than open reaches. The EPT index was lower, and Chiromonid Dominance was significantly higher in urban watersheds than in nonurban watersheds.
Fish. Year 3 sampling at selected sites was conducted during the fall of 2000. Fish sampling was conducted in the fall to survey a full season's production and to avoid sampling during low-flow periods. Preliminary results suggest a greater abundance of fish in most nonintermittent meadow reaches. There are differences between urban and nonurban fish communities reflected in species richness, abundance, and/or biomass of tolerant species and intolerant species, but these are complicated by differences in stream size, habitat, and other factors. Small headwater streams have low richness of native, tolerant species, but larger species show more variability, related to size, habitat, urbanization, and other factors.
Food Web Linkages. Food web samples from additional sites (leaf litter, algae, macroinvertebrates from various feeding niches, fishes from various feeding niches) were submitted for analysis of carbon and nitrogen isotopic composition to determine food web linkages. Gross biochemical characterization (i.e., total proteins, carbohydrates, lipids) also was determined to help elucidate food quality for various consumer groups. The carbon and nitrogen isotopic composition of the various producers and consumers suggest that the periphyton is a dominant source of carbon to macroinvertebrates and most fish during the summer. Leaf litter (both fresh and microbially altered) was not a major food source during this time period but may be more in the winter. Further analysis of winter samples at selected reaches is ongoing. In addition, the nitrogen isotopic composition of various fish species increased as the amount of urban area increased. This may be due to changing sources of N in more urban areas or increased microbial processes resulting in higher nitrogen isotopic compositions of primary N.
During the period from the fall of 2001, to June 2002, we will focus on data analysis, interpretation, and publication. Significant effort will go towards the synthesis of the data, and the understanding of urbanization and riparian effects at the ecosystem level.