H43D-1607
Experimental and Numerical Studies of the Effects of Water Sprinkling on Urban Pavement on Heat Island Mitigation
One of the main causes of gheat island phenomenonh is thought to be the artificial covers of the ground surface with asphalt or concrete which reduce greatly inherent cooling effect of water evaporation from soil surface. In this study, as a candidate method of mitigating the heat island the effects of the gwater sprinklingh on the pavements are discussed from field experiments and numerical studies. Three field experiments of water sprinkling on the asphalt/concrete pavements were performed in hot summer days in 2004-2006. For detecting the change in temperatures, the authors developed and used a 3-D measurements system which consists of two vertical planes with 6m high and 16m wide, and has network arrays of 102 thermistors distributed spatially in the planes. The temperatures measured in and around the water sprinkled area indicated that the ground surface temperature decreased 5 to 15 degrees uniformly in the water sprinkled area compared with those in the un-sprinkled area, while the relative decrease of atmospheric temperature was approximately up to 1 degree. The subsurface temperature at a depth of 14cm under the pavement decreased significantly and kept lower than that at the same depth in un-sprinkled area over the next morning. A numerical model was developed and applied to interpret the experimental results. It deals with the heat balance of radiation, sensible/latent heat transfer at the ground surface and heat conduction through the artificial and natural soil layer under ground. temperature and vapor conditions changes at and near ground surface were modeled by using the bulk formula.Good agreements between the calculated time-temperature profiles and the experimental ones were obtained by assuming adequate physical parameters and meteorological conditions. The model could be improved in order to evaluate the changes of temperature and vapor contents in atmosphere near the ground surface caused by aerodynamic turbulent diffusion.
H43D-1608
Simulation of cooling effect of newly-innovated urban pavements on water and heat budgets
People often suffer from the intense summer heat in Japan. This trend is increasing in urban areas because of the heat island effect and global warming. The present research evaluated the effect of pavements made of traditional and new materials on water and heat budgets. We coupled the NIES Integrated Catchment-based Eco- hydrology (NICE) model (Nakayama, 2007; Nakayama and Watanabe, 2004, 2006, 2007; Nakayama et al., 2006, 2007) to an urban canopy model in order to simulate the water and heat budgets in various types of natural and artificial pavements (covered by lawn, concrete, steel plate, synthetic rubber sheet, infiltration and water-holding blocks), and to evaluate the role of a new surface material in promoting evaporation and cooling to counter the heat island phenomenon (NICE-URBAN), by comparing with the simplified empirical model. Furthermore, using the heat conductivities of natural soil, we arranged these values for artificial pavement in the model equations by including the effect of the amount of water on the heat characteristics in the material. The simulated belowground water agreed with the observed value because this model includes the processes of both interception store and soil moisture store. The NICE-URBAN simulated more correctly the cooling of water-holding pavement during the intense heat of summer in an urban area than the empirical model. Because the model estimates that the air temperature at the water-holding pavement is 2 (deg.) lower than that at the lawn and 5 (deg.) lower than that at the building rooftop, it is very powerful to use this material for positive cooling effect in combination with the lawn for passive cooling effect. The simulation of NICE-URBAN showed that the decrease of surface temperature in water-holding pavement is closely related to the promotion of evaporation from the surface, the water volume of the pavement and the surface reflectance. References; 1)Nakayama,T.,Watanabe,M.,WRR,40,doi:10.1029/2004WR003174,2004. 2)Nakayama,T.,Watanabe,M.,HESSD,3,2101-2144,2006. 3)Nakayama,T.,et al.,HP,20(16),3441-3466,doi:10.1002/hyp.6142,2006. 4)Nakayama,T.,Ecol.Model.,2007(revised). 5)Nakayama,T.,Watanabe,M.,HP,doi:10.1002/hyp.6684,2007. 6)Nakayama,T.,et al.,STOTEN,373(1),270-288,doi:10.1016/j.scitotenv.2006.11.033,2007.
H43D-1609
Modeling the Effect of Summertime Heating on Urban Runoff Temperature
Urban impervious surfaces absorb and store thermal energy, particularly during warm summer months. During a rainfall/runoff event, thermal energy is transferred from the impervious surface to the runoff, causing it to become warmer. As this higher temperature runoff enters receiving waters, it can be harmful to coldwater habitat. A simple model has been developed for the net energy flux at the impervious surfaces of urban areas to account for the heat transferred to runoff. Runoff temperature is determined as a function of the physical characteristics of the impervious areas, the weather, and the heat transfer between the moving film of runoff and the heated impervious surfaces that commonly exist in urban areas. Runoff from pervious surfaces was predicted using the Green- Ampt Mein-Larson infiltration excess method. Theoretical results were compared to experimental results obtained from a plot-scale field study conducted at the University of Wisconsin's West Madison Agricultural Research Station. Surface temperatures and runoff temperatures from asphalt and sod plots were measured throughout 15 rainfall simulations under various climatic conditions during the summers of 2004 and 2005. Average asphalt runoff temperatures ranged from 23.2°C to 37.1°C. Predicted asphalt runoff temperatures were in close agreement with measured values for most of the simulations (average RMSE = 4.0°C). Average pervious runoff temperatures ranged from 19.7° to 29.9°C and were closely approximated by the rainfall temperature (RMSE = 2.8°C). Predicted combined asphalt and sod runoff temperatures using a flow-weighted average were in close agreement with observed values (average RMSE = 3.5°C).
H43D-1610
Assessing the Impact of Spatial Scaling on Empirical Runoff Ratio Models within a Heterogeneous Suburbanizing Watershed in Central Indiana
Suburbanized watersheds are characterized by a spatially complex mosaic of fragmented impervious and vegetated surfaces. The heterogeneous nature of land cover in and the variety of storm routing structures that accompany suburban development substantially modify surface hydrological dynamics within suburban watersheds. The hydrological consequences of land use and land cover change create a pressing issue for the management of water resources within developing watersheds. This research examines the hydrological impacts of recent population growth and accompanying suburban development within the Jack's Defeat Creek watershed in Ellettsville, IN, which is a 40 km2 basin that has experienced an approximate doubling in population in the last 25 years. Event-based, whole-basin runoff responses are determined from streamflow and precipitation data collected during 2005 and 2006 under both wet and dry antecedent conditions. Observed runoff responses are compared to multi-scale predictions of runoff ratios derived from the Soil Conservation Service Curve Number, which is an empirical model commonly used by municipal planning agencies to generate runoff estimates based largely on characterization of land cover and soil type. The comparison of observed whole-basin runoff response to predictions of the SCS Curve Number derived from a range of spatial scales addresses both (1) the accuracy of the Curve Number method as a predictor of runoff response in heterogeneously impervious landscapes as well as (2) the spatial scale at which the runoff estimates from empirical approaches best match the observed data under varying antecedent moisture conditions. These results will provide guidance regarding the best practices for employing empirical rainfall-runoff relationships to predict storm runoff responses in rapidly urbanizing watersheds.
H43D-1611
Flood Response Along a Drainage Network
Flooding in urban areas is complex. As water overtops stream banks, it comes into contact with structural obstacles on the land surface, such as bridge constrictions, that dominate flow pathways. Furthermore, at small, or local, spatial scales, other hydraulic controls such as pipe surcharge and stormwater management ponds play a significant role in flood response. A major obstacle towards a better understanding of how these controls impact flood response is the scarcity of data available to characterize them. One watershed where both hydraulic and hydrologic data is available is the Dead Run watershed in Metropolitan Baltimore, Maryland. Dead Run is a research watershed of the Baltimore Ecosystem Study (BES), part of the Long Term Ecological Research network established by the National Science Foundation. Dead Run geospatial data is available through the BES and Baltimore County; hydrologic data was collected by the authors during field campaigns in the 2003-2005 field seasons; and hydraulic information including storm drain pipes, stormwater management ponds, and bridge constrictions was digitized by the author. The availability of this data in Dead Run allows us to detail not only the impact of impervious surfaces and hydrologic forcing on flood response, but also structural aspects of the urban drainage network. In this study, we integrate the three types of observations geospatial, hydrologic, and hydraulic to characterize drainage network structure along Dead Run's tributaries. We use these characterizations and the Environmental Protection Agency's Stormwater Management Model (EPA SWMM) to estimate the 10- and 100-year floods over the drainage network. Analyses focus on two extreme floods in Dead Run: the 7 July 2004 and 28 June 2005 events. Results highlight the importance of incorporating drainage network structure into the models we use to predict flooding in urban environments.
H43D-1612
Hydrologic Response of Watersheds to Urbanization Across the Upper Great Lakes Region, USA
The Great Lakes region has experienced significant urbanization in the last century, resulting in a sharp increase in Impervious Surface Area (ISA). ISA is significantly different from the natural land cover in hydrological response, and has great potential influence on watershed hydrology. The Variable Infiltration Capacity (VIC) model is employed to investigate the hydrological responses in urbanized watersheds in the Great Lakes region. Previously, VIC did not consider urban as a distinct land cover, either treating the developed areas as bare soil or removing them by rescaling vegetation area. In this study, a simple urban ”°slab" algorithm has been added to the VIC model to make it suitable for the simulation of urbanized watersheds. Approximately 15 small watersheds with drainage areas between 10 to 150 square miles in the White River watershed (Indiana) that have experienced significant urbanization between 1983 and 2001 have been selected for analysis. The change in annual streamflow statistics (e.g. frequency change of high flow peaks and low flow events, and daily variation of streamflow) are examined from 1980 to 2003. Using 1983 and 2001 land cover maps of the White River watershed, the VIC model with urban ”°slab" algorithm is evaluated by comparing the simulated change in discharge statistics due to land use with the observed change in the small watersheds. In addition, surface temperatures from the VIC model simulation are compared to MODIS surface temperature products to evaluate the ability of the ”°slab" algorithm to represent urban heat islands. Subsequently, the feasibility of the predicted impact of increasing ISA to regional hydrological processes is investigated, including the change pattern of the Bowen Ratio for the White River watershed. The hydrologic impact of future land cover change is then quantified for other Great Lakes regional urban watersheds (e.g. the Detroit, Milwaukee and Illinois Rivers), utilizing a 2020 land cover map created by the Land Transformation Model. The results of this work will increase our understanding of the impacts of urban expansion to hydrological processes at regional scales.
H43D-1613
A Case Study of Differing Effects of Urbanization on Streamflow From Two Proximate Watersheds
The effects of urbanization on streamflow from two proximate watersheds (Little Lehigh Creek (LLC) and Monocacy Creek (MC)) are investigated. Despite close similarities in rainfall, population growth, land use, imperviousness, and geology of the watersheds, streamflows at the LLC gage have changed markedly over the past 50 years, while those at the MC gage have not. In LLC, there are significant increasing trends in annual stormflow volume, annual maximum flow, and flashiness, but there are no significant trends in these measures in MC. Neither stream shows significant trends in annual baseflow volume or low flow. It appears that the distinct difference in response to urbanization of these two streams can be ascribed to differences in 1) watershed geomorphology, 2) spatial distribution, composition, and infiltration characteristics of carbonate bedrock, and 3) the spatial pattern of land development in each watershed with respect to the gage location. In regards to geomorphology, there is a steeper main channel and narrower floodplains in LLC than in MC. Carbonate soil and bedrock (primarily dolostone) are distributed throughout much of LLC watershed but only in the lower half of MC watershed; however the lower MC watershed (primarily limestone) has much more abundant sinkholes and karst features than in the LLC watershed. Finally, residential and commercial development is concentrated in the upper two thirds of the LLC watershed, where travel times are such that these areas contribute to the peak flows measured at the gage. Development is concentrated in the lower third of the MC watershed, where it has had less effect on peak flows at the gage. Overall, the study indicates that relatively subtle differences between watershed characteristics and development patterns can result in significant differences in runoff and in how streamflow regimes may change in response to urbanization.
H43D-1614
Predictive Modeling of Urbanization Impacts on Flow Regimes in a Semi-arid Watershed in Southern California
In semi-arid southern California, large areas of vegetative land cover are disappearing due to rapid and extensive urbanization. The anthropogenic process paves natural land surfaces, altering the partitioning of precipitation, and subsequently affecting regional hydrologic processes and water supply. This research aims to improve the understanding of the impacts of urbanization in semi-arid watersheds. We focus our efforts on the Upper Santa Clara River basin in northern Los Angeles County which is undergoing rapid transition from a natural state to an urbanized state. The Hydrologic Simulation Program-Fortran (HSPF) model has been parameterized with the land use, soil, and channel characteristics of the study watershed. Model parameters related to hydrologic processes are calibrated at the daily time-step using both lumped and distributed precipitation inputs, respectively, over a six-year period. Calibration results from lumped inputs indicate that model performance is reasonable during wet years and wet seasons, but is less satisfactory in dry years and dry seasons. Model calibration with distributed inputs is ongoing. Potential urbanization scenarios are generated by assigning various percentages of developed area in the watershed. Hydrologic processes under each scenario are simulated via the calibrated model. Simulation results and discussion on the hydrologic impacts of urbanization in this rapidly expanding watershed will be presented.
H43D-1615
Channel Responses and Hydromodification in Southern California
Hydromodification (changes in watershed hydrologic characteristics, and the resulting hydraulics and channel forms due to urbanization) is ubiquitous in Southern California. In this region, the effects of hydromodification are driven and compounded by the arid/semiarid climate, high relief, erodible soils, high urbanization rates, and relatively low frequency of retention/detention. We conducted a preliminary survey of over 50 stream reaches along a gradient from least disturbed to fully urbanized. All stages of the Channel Evolution Model (CEM) of Schumm et al. (1984) were observed, from stable to degrading, widening, aggrading, and quasi-equilibrium channels. Several sites have CEM stages II through V in close proximity due to headcutting, hardpoints, and infrastructure. We also observed channels in undeveloped watersheds impacted by downstream urbanization via headcutting. A range of intervention measures was observed, with the frequent evolutionary endpoint as a concrete engineered flood control channel. We also observed multiple channel evolution sequences that deviate from the CEM for single-thread, incising channels. An alternative channel response, particularly on smaller urbanized streams is a stabilized, vegetation encroached low-flow channel with regular baseflow supplied by residential irrigation runoff. The limited cases of unimpacted streams that remain tend to be high gradient, high energy systems that are naturally proximate to the transition between braided and meandering form for a given sediment size.
H43D-1616
Observed Changes in Stream Morphology Due to Hydromodification in Southern California
We conducted field investigations of over 50 streams in Southern California as part of a project evaluating the impact of hydromodification on channel morphology. An initial step of the project involves developing quantitative tools for evaluating the relative susceptibility of channel types and risk of channel instability in response to watershed urbanization. Field investigations indicate that most Southern California streams naturally transition from single-thread to braided channels in the presence of certain boundary conditions, and that watercourses near the single- to multi-thread threshold are potentially more susceptible to hydromodification. Based on the field investigations, we present a general channel-reach morphology sequence, and a general modeling framework for assessing the risk of various channel response trajectories associated with urbanization in Southern California.
H43D-1617
A Scenario Based Assessment of Future Groundwater Resources in the Phoenix Active Management Area
The availability of future water supplies in central Arizona depends on the interaction of multiple physical and human systems: climate, hydrology, water and land-use policy, urbanization, and regulation. The problem in assessing future water supplies requires untangling these drivers and recasting the issue in a way that acknowledges the inherent uncertainties in climate and population growth predictions while offering meaningful metrics for outcomes under alternative scenarios. Further, the drivers, policy options, and outcomes are spatially heterogeneous surface water supplies, new urban developments and changes in land-use will not be shared uniformly across the region. Consequently, different geographic regions of the Phoenix metropolitan area will be more vulnerable to shortages in water availability, and these potential vulnerabilities will be more or less severe depending on which factors cause the shortage. The results of this research will make several contributions to existing literature and research products for groundwater conservation and future urban planning. It will provide location specific metrics of water vulnerability and offer a novel approach to groundwater analysis; it will demonstrate the XLRM framework with an application to central Arizona Water resources. Lastly, it will add to the WaterSim climate model by spatializing the groundwater component for the Phoenix Active Management Area.
H43D-1618
Impact of urbanization on sediment chemistry in small-scale watersheds, southeast Virginia.
The state of Virginia contains only two natural bodies of water with the rest being comprised of mill ponds, farm ponds and impoundment lakes. These man made water bodies are ubiquitous along the eastern seaboard. Southeast Virginia (Williamsburg, Jamestown and surrounding counties) was the locus of early European settlement and many of the local ponds date back to the early colonial period. As such, the sediment record in these ponds provides a unique historical record of the impact human activity can have on small watersheds. Two small man-made ponds (Lake Matoaka and Jolly Pond) were studied. Both ponds lie within the James River basin, a major feeder to the Chesapeake Bay. Both drainage basins cover ~600 ha but differ significantly in the level of development. Lake Matoaka was originally dammed ca. 1720. The Matoaka drainage is currently experiencing rapid development (~22% high population residential/commercial) and includes the College of William & Mary's campus. Jolly Pond and the two dams that created it first appear on maps dating back to 1863. The Jolly Pond basin is largely dominated by forests and agricultural land. Sediment cores were taken from both ponds using a Russian peat corer, Matoaka to 1.5m depth and Jolly Pond to 0.9m dpeth. Sediment splits were analyzed for total exchangeable lead concentration and lead isotopes (207Pb/206Pb & 208Pb/206Pb), as well total carbon and nitrogen. Carbon/Nitrogen ratios were used to indicate changes in water level and dam height increases. Lead concentrations in both cores show a peak associated with leaded gasoline use. [Pb] in Matoaka are significant (~320 ppm) and show a clear anthropogenic isotopic signature. Jolly Pond [Pb] are surprisingly low and are close to background values. These data indicate that even minimal development has significant impact on sediment chemistry. Likely lead sources for Lake Matoaka include road and building runoff and lead aerosols.
H43D-1619
Quantifying Land Use and Land Cover Effects on Urban Runoff Water Quality.
The impact of non-point source pollution on urban storm runoff is of major concern in the Southwest where water resources are scarce, episodic rainfall is intense and runoff recharge is a water management strategy. The objectives of this study are to 1) determine the extent to which specific types of urban land use impact the quality of monsoonal rainfall-runoff, and 2) identify pollutant source and modification during transport within urban washes of different types. We installed autosamplers at the outlet of four watersheds in the Tucson, AZ basin, with land uses representative of growing urban centers in the southwest U.S.: 1) commercial; 2) medium and high density residential; 3) low density residential; and 4) mixed use. At each outlet, storm runoff samples were collected at 20 minute intervals during several monsoonal storms. To characterize how pollutants were modified during transport, we installed autosamplers at upstream and downstream locations of a wash. Samples were analyzed for nutrients, organic pollutants, metals, anions, cations and fecal indicator bacteria (E. coli). Preliminary data show that nitrate concentrations were highest in the commercial and low density watersheds (median = 2.53 mg/L and 2.81 mg/L NO3-N, respectively) and lowest in the medium density watershed (median = 1.68 mg/L). Ammonium concentrations were also highest in the commercial and low density watersheds (median = 1.84 mg/L and 1.75 mg/L NH4-N, respectively) and lowest in the medium density watershed (1.28 mg/L). E. coli counts were highest in the commercial (median = 4500 CFU/ml) and lowest in the medium density watershed (median = 61.26 CFU/ml). Over the season, E. coli concentrations decreased in all except the mixed density watershed where they increased as the monsoon progressed. We observed distinct pollutant concentration response patterns to storm events among watersheds. Pollutant concentrations in runoff from commercial and low density watersheds peaked within the first 40 minutes of a storm event and subsequently tapered, whereas concentrations in the middle density watershed increased throughout a storm event. Our study demonstrates that land use type directly and distinctly impacts storm runoff chemical composition, which has significant implications for basin wide pollutant fate and transport. Our data also suggests that the type of runoff drainage system may play an important role in contaminant degradation and subsequent transport.
H43D-1620
EFFECTS OF URBANIZATION ON NITROGEN DYNAMICS IN STREAM (CITY OF LYON, France)
The objective of this project was to investigate the impact of urbanization on water exchange and nitrogen dynamics in the surface and-hyporheic waters. The watershed of the Yzeron stream, close to the city of Lyon (France), was selected. It covers rural and periurban areas and has a total surface area of 150 km2. Two stream reaches were sampled: a rural reach and a periurban reach. In every reach, surface and hyporheic waters were collected at six different points up and downstrean geomorphic riffles. Six sampling campaigns were carried out between February 2004 and June 2005. Similar water delta D and delta18O values in surface and hyporheic waters indicate fast exchange between these two compartments in the rural reach. In contrast, water isotopic composition differs between surface and hyporheic waters in the periurban reach. This suggests slower exchange probably related to weaker slope and slight different stream-bed grain-size, compared to rural reach. Nitrate concentration is rather low with a relatively higher level in the rural reach. Inorganic nitrogen is essentially composed of nitrate with occasional occurrence of ammonium in the hyporheic zone of the periurban reach. High delta15N values in nitrate (up to 13.20 per mil), coupled with rather low nitrate concentrations, reflect the process of denitrification in the hyporheic waters. Large fluctuations in nitrate delta15N values (-2.2 to 13.20 per mil) in the rural reach suggest the occurrence of nitrification-denitrification patches. The results of this field study stress the importance of geomorphic riffles in surface-hyporheic water exchange and in stream-bed denitrification pattern.