H52D-01
Urban Effects on Stream Hydrology and Geomorphology: Variations, Magnitudes and Implications for Stream Protection and Restoration
The hydrologic and geomorphic response to urbanization has become an increasing concern for regulatory agencies and practitioners of stream protection and restoration. However, there has been comparatively less attention to this topic in the research community. In his classic diagram of the effect of urbanization, Leopold (1968) illustrated how increases in impervious area and degree of ‘improved' storm drainage tend to shorten lag times and increase peak flows. This response, sometimes termed hydrograph modification or hydromodification, can have dramatic effects on stream morphology, inducing both accelerated erosion and/or deposition. We present field data for case studies in California that contrast hydrologic and geomorphic responses in sandy versus clay-rich soils with different underlying geology. The first case study compares the urban effect in paired watersheds with relatively sandy terrain in Southern California. Annual peak flows, measured by indirect methods, increased by up to an order of magnitude in catchments with urbanization compared to adjacent catchments without urbanization. Morphologic responses have included significant incision plus natural-levee deposition from overbank flows. The second case study, in areas with clay-rich soils in the San Francisco Bay Area, examines hydrographs in watersheds with and without urbanization. Watersheds with urbanized areas showed significant increases in stream flow in early season storms and moderate mid-winter storms, but less of an urban effect in large mid-winter storms. Near bankfull flow events capable of sediment transport also became more frequent following urbanization, especially early in the rainfall season and following mid-winter dry spells. Channel response to hydromodification included increases in bank recession and collapse associated with the early-season storms. A third case study, in the East San Francisco Bay Hills shows how even minor (approximately 1 percent) increase in impervious area, an index of urbanization, can cause up to an order of magnitude increases in early season storm flow peaks. In summary, these case studies illustrate variation in both the magnitude and type of responses due to site conditions, which can be used to guide methods of stream restoration and protection to the local setting.
H52D-02
Impacts of Urban Development on Channel Slope, Erosion, and Depositional Processes
Watershed land uses, including urbanization, influence erosion and depositional processes in a tributary to the Navarro basin in north coastal California. The Robinson Creek subwatershed is deeply incised into easily erodible Quaternary alluvial river and terrace valley-fill deposits in the vicinity of the urbanizing town of Boonville. Remnant riparian forest vegetation exists at the top of the terrace. Detailed field surveys of a 1.3 km channel reach document spatial variation in slope and suggest that a series of knickzones accommodate elevation differences between areas upstream and downstream of the urbanizing area. Average reach slope is ~0.012 whereas average terrace slope is ~0.009. This difference accounts for lower bank heights (~4.6 m) at the upstream end of the reach than at the downstream end (~8.1 m). The apparent recent increase in channel slope, an indicator of channel adjustment, leads to an increase in shear stress available to erode and transport bed and bank sediment in the relatively narrow channel. Temporal variation in bed elevation is related to episodic floods in this actively adjusting channel; up to 0.2 m of incision occurred during floods in water year 2006. Bank erosion is considerable and erosion-control structures are a typical response to land loss, threats to residential structures and domestic water wells, and undercutting of bridge piers and individual streamside trees. Existing bank erosion control structures limit geomorphic processes and aquatic and riparian restoration options over the long term. Moreover, progressive urban development including building structures near the top-of-bank further limit alternatives for restoration of geomorphic processes and associated aquatic and riparian habitat that require increases in channel width as a response to continuing channel incision.
H52D-03 INVITED
Regional Variability of Stream Responses to Urbanization: Implications for Risk-Based Assessments
Predictive scientific assessments of the geomorphic consequences of urbanization must be calibrated to the regional hydroclimatological, geologic, and historical context in which streams occur. We present examples of context-specific stream responses to hydromodification, and a general framework for risk-based modeling and scientific assessment of hydrologic-geomorphic-ecologic linkages in urbanizing watersheds. The framework involves: 1) a priori stratification of a region's streams based on flow regime, geomorphic context and susceptibility to changes in water, sediment, and wood regimes, 2) field surveys across a gradient of urban influence, 3) coupling long term hydrologic simulation with geomorphic analysis to quantify key hydrogeomorphic metrics, and 4) using probabilistic modeling to identify regional linkages between hydrogeomorphic descriptors and decision endpoints of primary interest to stakeholders and decision-makers.
H52D-04
Direct measurement of storm hydrographs from wet retention ponds: How effective are they at protecting streams from urbanization?
Wet stormwater retention ponds (SRP) have become the method of choice for controlling storm runoff, and are a significant component of suburban hydrology. Despite widespread use, little data has been collected to assess their success in controlling stormwater quantity. We have collected data on flows from four SRPs and a forested watershed in James City County (JCC), Virginia. Our goal is to assess whether these SRPs function as designed and release flows comparable to natural conditions. All monitored SRPs are in suburban developments of seven to 34 acres, and are designed to meet 1) JCC criteria requiring 24 hr retention of runoff produced by the one year, 24 hr storm and 2) Virginia regulations requiring the reduction of the two year, post- development peak discharge to pre-development conditions. We have analyzed data for from July 2004 through September 2007. Storms with recurrence intervals of one to two years consistently generate peak inflows into the SRPs greater than predicted by engineering design, resulting in outflows to streams greater than the maximum allowed under Virginia stormwater regulation. In addition, during storms similar to the 1 yr, 24 hr design storm, none of these SRPs retained runoff for the required 24 hr period, the primary requirement for stream protection in JCC. Comparison of outflow hydrographs from SRPs to a forested watershed during high-frequency storm events indicates peak flow reduction to predevelopment conditions is sometimes achieved, but total flow volume is much greater. During a near 100-year storm, SRPs passed runoff with only minor attenuation, resulting in peak and total flows into receiving streams substantially greater than natural conditions. The results suggest SRPs receive and release higher peak discharges than intended, often release stormwater over periods shorter than required, and are far from mimicking natural flows. Underprediction of developed runoff, inadequate pond design, and lack of maintenance contribute to this poor performance. As these ponds were designed using standard engineering methods, our results suggest that these shortcomings in SRP performance are widespread.
H52D-05
A simulation tool for predicting thermal impacts of stormwater runoff on trout streams
Our current work is focused on investigating the impact of urban development on the temperature of trout streams in urbanizing watersheds. The goal of the project is to characterize this impact, and to produce a robust simulation tool to quantify the impact. A preliminary version of the tool, MINUHET (MINnesota Urban Heat Export Tool), has been developed. MINUHET tracks heat transfer processes associated with the routing of storm water through a development-scale watershed, producing a time series of flow rate and temperature at the development outlet that could potentially serve as input to a stream temperature model. Currently, the tool is capable of modeling runoff volume and temperature from a variety of land uses and the effect of various management practices, including storm water ponds, storm sewers, and rain gardens. A case study will be presented in which MINUHET has been applied to a housing development in Plymouth, MN. The case study includes a comparison of simulated and measured runoff flow rate and temperature, and an analysis that characterizes the uncertainty of simulated runoff based on the uncertainty of watershed parameters.
H52D-06
Cation Chemistry in Baltimore Streams: Another Syndrome Symptom?
Urban streams are degraded due to a variety of physical and chemical factors including impervious surfaces, sewer cross-connections, and multiple chemical contaminants. There is particular interest in the loading of anionic nutrients and heavy metals into urban streams, especially those draining to nutrient sensitive receiving waters. However, there has been less characterization of cation chemistry in urban streams. We examined metal concentrations in archived water samples as part of the stream monitoring program of the Baltimore Ecosystem Study (BES), an urban component of the U.S. National Science Foundation Long Term Ecological Research network. Preliminary analysis suggests sewer cross-connections (storm and sanitary) influence metal concentrations (e.g. Si, Ca, Mg). For example, while silica content is elevated in water draining BES agricultural watersheds, it is also elevated in watersheds with extensive impervious surface coverage, areas with the least water-rock interaction. These results raise questions about the effects of urbanization on chemical weathering, the stoichiometry of urban storm and sewer waters, and the potential for human inputs to function as hydrologic tracers in urban watersheds.
H52D-07
Mercury Mobilization in Urban Stormwater Runoff
Higher mercury (Hg) concentrations in urban rivers have been found compared to rural rivers with the majority of the increase associated with particles (HgP) during periods of stormflow (Lawson et al. 2001; Hurley et al., 1998). It is unclear whether the increase in concentrations is a result of internal mobilization (e.g. bank erosion) or from external loading from stormwater runoff. The objectives of this research are: to use high frequency sampling to characterize the water and Hg export dynamics from an urban micro-catchment during individual rain events under a range of conditions; determine the relationship between suspended sediments and Hg transport; assemble event-scale mass balances of atmospheric Hg inputs, surface Hg associated with street dust, and Hg export in runoff for the micro-catchment to evaluate the relative importance of rainfall-derived Hg and surface- derived Hg in runoff; and finally, to compare the yield of Hg from the urban micro-catchment to that of a much larger mixed land-use urban catchment to evaluate the feasibility of scaling the results. Overall, the results from this research show that large Hg loads are delivered to an adjoining waterway during storm events and that the majority of the load was HgP. A significant relationship was also observed between the TSS and Hg concentrations. Comparison of the inputs of Hg in rain and the outputs in runoff indicated that the urban surfaces could act as both sinks and sources depending on the rain and runoff characteristics.
H52D-08
Spatial and Temporal Impacts of Urbanization on Local Hydrological Processes - A Case Study in the Central Taiwan
A simulated urban area with size 15x15 km2 representing Chiayi area of central Taiwan can produce up to 3 times more sensible heat flux around noon time than that of a non-urban area. This effect has significant impact on locations of local summer thunderstorms and precipitation over Taiwan western plain by sensitivity tests. We used precipitation and stream gauges in west, east, and northeast of Chiayi area to examine the impact on local hydrological processes including suspended sediment transport. Both monthly mean precipitation and precipitation days have been found greater in the east and northeast of the Chiayi area than the west during the summer periods from 1984 to 2006. In addition, we also found that the decadal mean precipitation has been decreasing in the further east toward the mountain rage since 1980s. This spatial distribution matches the model prediction based on different simulated urban sizes indicating more summer thunderstorms happening before they reaching the mountain range by the effect of heat island. This spatially and temporally increasing trend in precipitation days and total amount of precipitation in the study area has posted a new challenge on the estimation of suspended sediment transport.