Effects of Urbanization on the Water Cycle and Aquatic Ecosystems II
Presiding: C Welty, University of Maryland Baltimore County; C M Swan, University of Maryland Baltimore County
NB22F-01 10:30h
Assessment of Aquatic Biological Communities Along a Gradient of Urbanization in the Willamette Valley Ecoregion, Oregon and Washington.
From late 2003 through summer 2004, the U.S. Geological Survey's National Water Quality Assessment Program sampled 28 streams within the Willamette Basin to investigate effects of urbanization on aquatic biology (fish, macroinvertebrates and algae), habitat, and water chemistry. The 28 watersheds fall along an urban land use gradient index (0 to 100, lowest to highest) based on land use and census data developed for this region. Watershed areas range from 13 to 96 square kilometers and contain greater than 20 percent of the Willamette Valley ecoregion. Ten streams were sampled for water chemistry six times during study period. The other 18 streams were sampled twice for water chemistry-once during high sustained flow, and once during summer low flow. The data will be analyzed to determine relationships to the urban gradient index and for possible detection of threshold responses. Preliminary results indicate that 57 percent of the most urbanized streams contained nonnative fish species, but only 43 percent contained salmonids. Conversely, nonnative fish species were present in 14 percent of the least urbanized streams, whereas salmonids were present in 79 percent of these streams. Population density and water chemistry variables were highly correlated with fish assemblage patterns among sites.
NB22F-02 10:45h
An Inter-City Comparison of Macroinvertebrate Assemblage Response to Urban Land use in Southeastern Australia
The relationship between macroinvertebrate assemblage composition and two measures of urban land use was assessed in 2 Australian cities: subtropical Brisbane and temperate eastern Melbourne. Average annual rainfall is similar in the 2 regions (900-1200 mm across eastern Melbourne, 1100-1200 mm across Brisbane). However, in Melbourne, rainfall is more frequent, more evenly distributed, and generally less intense. In both cities effective imperviousness (EI, including only those impervious surfaces with direct sealed connections to a stream) was a stronger predictor of assemblage composition than was total imperviousness. In both cities, compositional similarity (as portrayed by a 1-dimensional multidimensional scaling) declined linearly from sites with zero EI to a threshold of maximum degradation at 6-10% EI. The similarity of pattern in 2 cities of differing climate suggests strong impacts of stormwater runoff at very low levels of urbanization, independent of rainfall pattern. The influence of climate on instream urban impacts will be explored further with similar models being developed for two drier regions: western Melbourne and Canberra, each with approximately 600mm annual rainfall.
NB22F-03 11:00h
The Role of Native Tree Species on Leaf Breakdown Dynamics of the Invasive Tree of Heaven ( Ailanthus altissima) in an Urban Stream
Anthropogenic disturbance of ecosystem processes is increasingly being explored in urban settings. One profound impact is the striking increase in the distribution of invasive plant species. For example, Tree of Heaven (Ailanthus altissima, TOH), introduced into the U.S. from Asia in 1784, is a successful colonist of recently deforested habitats. As a result, remnant patches in urban ecosystems have become overrun with this tree species, excluding native species via fast growth and allelopathy. While suffering from human-induced degradation, urban streams still support food webs that function to process riparian-derived organic matter (e.g., leaves, wood). The purpose of this study was to (1) estimate leaf litter breakdown of native tree leaves and those of TOH in an urban stream, (2) study the detritivore feeding rate of the same leaf species, and (3) determine if increasing native species richness of leaf litter can alter breakdown of TOH leaves. Field manipulations of leaf pack composition were done in a highly urbanized stream (>30% upstream urban land use) in Baltimore County, Maryland, USA. This was complimented by a series of laboratory feeding experiments employing similar leaf treatments and local shredding invertebrate taxa. Breakdown of TOH alone was extremely rapid, significantly exceeding that of all native tree species employed. Furthermore, mixing TOH with native tree species, red maple and white oak, substantially reduced TOH decay compared to decay of TOH alone. However, supporting laboratory studies showed that TOH was a preferred resource by shredding invertebrates over all native species. Subsequent analysis of the structural integrity of all leaf species revealed that TOH was the least resistant to force, possibly explaining the counterintuitive decrease of TOH decay in mixtures. We interpret this as meaning the stream invertebrates, while preferring to consume TOH, appeared not to influence TOH decay in mixtures with native species. Instead, the relatively tougher nature of native species appeared to slow TOH breakdown by armoring the invasive from the highly-variable flow regime characteristic of urban streams. Therefore, the presence of native tree species in urban riparian zones may be critical to how invasive trees, like TOH, could alter carbon flux in urban streams.
http://userpages.umbc.edu/~cmswan
NB22F-04 11:15h
Effects of Urbanization on Allochthonous Inputs to Small Puget Sound Lowland Streams
Urbanization produces fundamental changes to abiotic and biotic functions in streams, with shifts affecting everything from flow regimes to benthic communities. However, development patterns are manifested in changes to riparian vegetation communities as well, and these changes alter the sources and fate of organic matter in streams. This ongoing study in the Puget Sound Lowlands evaluates the influences of urbanization on allochthonous inputs to small streams. The study area includes 16 litterfall sites with riparian vegetation composition spanning the disturbance spectrum. Vegetation types include minimally altered conifer and mixed forests (western hemlock, big leaf maple, red alder), deciduous (red alder), landscape trees associated with residential development, and minimal forest cover with prevalent invasive species. While conifer/mixed and red alder forests produce similar peak litterfall rates, as much as 10 g m-2d-1, red alder litter contains more nitrogen, resulting in higher nitrogen loads to streams. Short-term laboratory leaching studies indicate that red alder leaves lose 25% of their mass within 24 hours, as compared with 1% for conifer materials. Highly disturbed areas with little or no riparian vegetation produce low allochthonous inputs (peak 0.1 g m-2d-1); grass clippings, for example, do not compensate for the loss of tree cover.
NB22F-05 11:30h
Effect of Riparian Impervious Cover on Leaf Litter Decomposition Rates and Lotic Benthic Macroinvertebrate Diversity
Riparian impervious cover has been implicated in reducing benthic biodiversity. We examined the impact of impervious cover on the decomposition and colonization of leafpacks in 6 sites located on three streams. The six sites ranged in impervious cover, in the drainage basin from 98% upstream at Meeting of the Waters Creek to 25% upstream on Bolin Creek. 5g leafpacks of dogwood leaves (Cornus florida), were placed at each site on October 15, 2004 and 5 replicates per site were collected on Oct. 25, Nov. 8 and Nov. 22. Decomposition rates were highest (AFDM remaining lowest) where forested cover was greatest, and lowest where impervious cover was greatest. The diversity of the initial macroinvertebrate colonization was highest with least impervious cover, and was dominated by chironomids where impervious cover was greatest. The negative impact of increasing impervious cover on stream community diversity is supported by this study. The ecosystem level process of leaf litter decomposition is also negatively affected.
NB22F-06 11:45h
Benthic macroinvertebrate populations of urban freshwater tidal wetlands in the Anacostia River, Washington D.C.
This study characterizes the benthic communities establishing themselves on recently reconstructed urban freshwater tidal wetlands along the Anacostia River in Washington, D.C. in comparison to a similar relic wetland as well as to a reference wetland in the adjacent Patuxent River watershed. The study's focus is the two main areas of Kingman Marsh, which were reconstructed from Anacostia dredge material by the U.S. Army Corps of Engineers in 2000. Populations from this 'new' marsh are compared to those of similarly reconstructed Kenilworth Marsh (1993), as well as to the relic Dueling Creek Marsh on the Anacostia and the outside reference Patuxent Marsh in an adjacent watershed. Benthic organisms were collected using selected techniques including the Ekman bottom grab sampler, sediment corer, D-net and Hester-Dendy sampler. Samples were collected seasonally from tidal channels, tidal mudflats, three vegetation zones (low, middle and high marsh), and pools. Data collected from this study can provide valuable information on the extent that benthic macroinvertebrate communities can serve as an indicator of the relative success of freshwater tidal marsh reconstruction.