11th International Conference on Urban Drainage, Edinburgh, Scotland, UK, 2008
Effect of Urban Stormwater BMPs on Runoff Temperature in
Trout Sensitive Regions
M.P. Jones1*
1
Biological and Agricultural Engineering Department, North Carolina State University, NCSU
Box 7625, Raleigh, NC, USA
*Corresponding author, e-mail Matthew_Jones@ncsu.edu
ABSTRACT
In recent years, a variety of stormwater BMPs have been developed to reduce the impact of sediments,
nutrients, and metals on receiving waters; however, one area that has received little attention is the
effect that these systems have on the temperature of stormwater runoff. Warm stormwater runoff is a
concern because it can raise the temperature of coldwater stream environments, negatively impacting
the aquatic organisms that live there, such as trout. A monitoring study was conducted in North
Carolina at 4 bioretention areas, a stormwater wetland, and wet pond to determine the effect of these
systems on runoff temperature and identify any design criteria that affect effluent temperatures. It was
found that stormwater wetlands and wet ponds increased runoff temperatures. Temperatures at the
bottom depths of the stormwater wetland and wet pond were cooler than existing effluent
temperatures, suggesting the implementation of modified outlet structures that draw water from these
depths. Bioretention areas were able to reduce maximum runoff temperatures; however, storm median
temperatures sometimes increased. Results of this study show that the thermal impact of stormwater
treatment can be minimized and sometimes runoff temperatures can be reduced when thermal
pollution is considered in stormwater BMP design.
KEYWORDS
Aquatic Environment; Infiltration; Stormwater Management; Temperature; Wetland
INTRODUCTION
As advances in aquatic ecology have coincided with increased urbanization, it has become
evident that warm stormwater runoff can negatively impact coldwater stream environments.
Due to the thermal properties of asphalt, pavement surfaces can reach temperatures in excess
of 60 °C, with much of the heat concentrated near the surface (Asaeda et al., 1996). During a
storm event, runoff collects this heat and distributes it throughout the watershed. Although
water temperature in streams naturally fluctuate in response to changes in solar radiation and
the ambient air temperature, temperature changes are typically limited to a 0.6-0.8°C
temperature increase for every 1°C increase in air temperature (Morrill et al., 2005). The
discharge of stormwater runoff into coldwater creeks can raise stream temperatures above
normal levels, with runoff temperatures elevated even during the night and early morning
(Kieser et al., 2004; Lieb and Carline, 2000).
Heated stormwater runoff is a concern because of the negative impact it can have on the
aquatic ecosystem. Because many aquatic organisms lack the ability to regulate their body
temperature, temperature increases in the aquatic environment can impair major biological
functions. Increases in stream temperature can negatively impact the behavior, metabolism,
reproduction, growth, and vulnerability to disease of many aquatic organisms. Moreover,
because various organisms respond differently to elevated temperatures, thermal pollution can
Jones
1
11th International Conference on Urban Drainage, Edinburgh, Scotland, UK, 2008
have substantial impacts on community structure. Trout and salmon, which serve as
important game fish in many parts of the United States, are among the fish most sensitive to
elevated water temperatures. Due to a variety of complex factors, it is difficult to predict the
specific effects of elevated temperatures on fish behavior and survival (Hocutt et al., 1980).
In spite of these complexities, it has been determined that trout generally avoid water
temperatures in excess of 21°C (Coutant, 1977).
Despite evidence that the thermal impacts or urban stormwater runoff has a negative effect on
the aquatic ecosystem, there are few mechanisms available to cool stormwater runoff.
Stormwater best management practices (BMPs) have been installed throughout the United
States in order to reduce the impacts of sediments, nutrients, metals, and other pollutants;
however, the effect of these systems on runoff temperature is not well understood. A study of
a wet pond found that temperatures increased between the inlet and outlet (Kieser et al.,
2004). Because wet ponds are not typically shaded, solar radiation can heat the water within
the pond above the temperature of the ambient air. A stormwater wetland has been shown to
reduce runoff temperatures when well shaded, with the net heat reduction attributed to
evapotranspiration and infiltration (Kieser et al., 2004).
Although bioretention areas have been shown to reduce runoff flows, their effect on runoff
temperature has not been thoroughly examined (Hunt et al., 2006). Soil temperature is largely
governed by radiation and convective heat transfer near the soil surface. Because thermal
conduction through a soil column is relatively slow, deeper layers of soil are able to maintain
relatively stable temperatures over long periods of time (Mohseni et al., 2002). During
irrigation with warm water, it has been shown that water temperature approaches the
temperature of the surrounding soil as it infiltrates, resulting in minimal temperature
variations at soil depths greater than 100 cm (Wierenga et al., 1970). At shallow soil depths,
heat exchanges between the soil and infiltrating water may have little impact on water
temperatures. For example, seasonal mean water temperatures between rooftop runoff and
underdrain outflow were not significantly different at a relatively shallow (60 cm)
bioretention area in the northeastern United States (Dietz and Clausen, 2005).
In order to examine the effect of urban stormwater BMPs on runoff temperature, a monitoring
study was conducted at 6 locations in western North Carolina, along the south-eastern extent
of United States trout populations. Effectively managing thermal pollution is important in
this region, since many coldwater stream environments approach the temperature limits of
trout under natural conditions. With an understanding of the effect stormwater BMPs have on
runoff temperature, it should be possible to ascertain which BMPs are best suited when
thermal pollution is a concern and identify any design specifications that may influence water
temperatures.
METHODS
All monitoring sites were paired based on proximity, with 2 bioretention areas located in
Brevard, a bioretention area and stormwater wetland located in Asheville, and a bioretention
area and wet pond located in Lenoir, North Carolina. Monitoring was conducted during the
summer months, generally from May through mid-October. Data were collected at the
Asheville wetland, Asheville bioretention area, and Lenoir north bioretention area during
2005, and all monitoring sites during 2006 and 2007.
2
Effect of Urban Stormwater BMPs on Runoff Temperature in Trout Sensitive Regions
11th International Conference on Urban Drainage, Edinburgh, Scotland, UK, 2008
To measure outflow at all bioretention areas and the stormwater wetland, pulley-float stage
recorders were used in conjunction with v-notch weirs. Pulley-float stage recorders with vnotch weirs or measured rainfall were used to measure inflow at all monitoring sites. Due to
complications with the flow monitoring equipment, flow data were primarily used to identify
periods of flow over the temperature sensors. Rainfall data were collected for each
monitoring pair using tipping bucket rain gauges with a resolution of 0.25 mm. HOBO™
Water Temp Pro (H20-001) and HOBO™ 4 channel loggers (H08-008-04 & U12-008) with
temperature sensors attached (TMCX-HD), which were all manufactured by Onset Computer
Corporation were used to collect temperature measurements. All temperature and flow
measurements were logged at a 5 minute interval.
Statistical analysis was conducted using SAS® software, Version 9 (SAS Institute Inc., Cary,
NC). The potential thermal impact to trout populations was ascertained by comparing water
temperatures to 21° C, the temperature at which trout begin to experience thermal stress,
using a signed rank test. Comparisons of influent, effluent, and soil temperatures were
conducted using the Wilcoxon Rank Sum test (Wilcoxon, 1945). Storm medians and
maximums were used in the analysis of BMP influent and effluent temperatures.
Site Description
The Asheville bioretention area (35°36'46"N, 82°33'54"W) covered 45m² and received runoff
from 280 m² of asphalt parking lot. During the summer of 2005, when the bioretention area
and contributing parking lot were constructed, a light colored chip seal was applied to the
parking surface in an attempt to reduce pavement temperatures. The bioretention area was
located approximately 6 m from the parking lot and runoff was routed into the BMP within an
asphalt channel. Perforated PVC pipes, 10 cm in diameter, were used to drain the
bioretention area and 38 cm smooth walled corrugated plastic tubing (CPT) discharged the
bioretention effluent into nearby Reed Creek. The bioretention area did not contain mulch
and naturally progressed from no vegetative cover in June of 2005 to an estimated 55%
vegetative cover in late August, 2006. Temperature probes were installed at the inlet channel,
inside the outlet weir box, and at 5 evenly spaced depths within the soil column from the
surface to the underdrain depth of 135 cm.
The Lenoir bioretention area (35°55'20"N, 81°31'24"W) covered a 30 m² area. Runoff
originated from a 675 m² asphalt parking lot and two sections of concrete sidewalk, which
covered a combined area of 95 m². The bioretention surface was covered by hardwood mulch
and vegetation within the BMP shaded 79% of the area, while the bioretention area and
adjacent parking surface were shaded by a mature tree canopy. A network of 12 cm
perforated CPT underdrains located 95 cm below the ground surface collected infiltrating
runoff and discharged the water onto a street curb located approximately 10 m away.
Temperature probes were installed in the inlet pipe, outlet weir box, and at 4 evenly spaced
depths within the soil column from the surface to the underdrain depth. Soil temperatures
were only monitored at this site during the summer of 2007.
Adjacent bioretention areas were monitored in Brevard, NC (35°14'20"N, 82°43'52"W).
Water from these bioretention areas, as well as others in the parking lot, was routed by an
underground drainage network to Kings Creek, which was adjacent to the property. Both
systems were mulched with pine straw. The Brevard east bioretention area covered a 37 m²
area and received runoff from approximately 525 m² of asphalt parking lot. Water left the
bioretention area through a 12 cm perforated CPT underdrain network located approximately
48 cm below the soil surface. An estimated 43% of this bioretention area was shaded by low-
Jones
3
11th International Conference on Urban Drainage, Edinburgh, Scotland, UK, 2008
lying plants. Temperature probes were located in the inlet weir box and outlet pipe. Soil
temperature was measured at a depth of 48 cm during the summer of 2007.
The Brevard west bioretention area covered a 36 m² area and received runoff from
approximately 325 m² of asphalt parking lot. Water left the bioretention area through a 12 cm
perforated CPT underdrain network located approximately 43 cm below the soil surface. An
estimated 43% of this bioretention area was shaded, primarily from 2 maple trees. Periods of
inflow and outflow were identified with weir boxes mounted at the inlet and outlet, where
temperature probes were also located.
The stormwater wetland was located in Asheville, NC (35°36'52"N, 82°33'48"W) and
primarily received runoff from an estimated 7,350 m² asphalt parking lot. The stormwater
wetland covered a 724 m² area, of which an estimated 70% was covered by vegetation during
mid-summer of 2006. Water was conveyed from the parking surface to the stormwater
wetland through a 168 m long, 71 cm diameter buried corrugated metal pipe and a separate
vegetated channel. Water was discharged from the wetland through a flashboard riser
structure and corrugated metal pipe into Reed Creek, approximately 10 m away. Temperature
probes were located inside a drop inlet at the parking lot, the corrugated metal pipe inlet to the
wetland, the vegetated channel inlet, and the outlet structure. Additionally, 4 probes were
spaced at 30 cm depths from the normal pool elevation to the base of a deep pool near the
outlet structure.
The wet pond was located in Lenoir, NC (35°54'1"N, 81°31'18"W), and captured runoff from
an estimated 56,500 m² of rooftop and asphalt parking lot. The wet pond’s surface area was
2,545 m², and it received water from a 122 cm (48 in) diameter corrugated metal pipe and a
122 cm (48 in) diameter reinforced concrete pipe. The corrugated metal pipe had a perennial
discharge into the wet pond, which likely originated from groundwater interception. The
outlet structure was designed for water to exit the wet pond through two 60 cm holes in the
concrete structure; however, poor seals lowered the normal pool elevation approximately 120
cm below these outlet holes. There was a consistent flow through 25 cm diameter corrugated
plastic tubing at the base of the outlet structure, likely due to a leak in the emergency
drawdown pipe. This wet pond discharged water directly into a heavily urbanized creek.
There was a substantial amount of algae covering the pond for much of the year, in part
because the wet pond received essentially no shading from vegetation. Temperature probes
were located inside a drop inlet at the parking lot, inside the corrugated metal pipe inlet to the
wet pond, inside the reinforced concrete pipe inlet, inside the outlet pipe, as well as 4 depths
near the outlet structure. Stormwater runoff entering at the monitored drop inlet travelled
through 320 m of buried pipe and was combined with flow from a number of additional drop
inlets before entering the wet pond through the reinforced concrete pipe inlet.
RESULTS AND DISCUSSION
Runoff temperatures directly leaving the parking surfaces were significantly* (p<0.05)
warmer than 21°C during the months of June through August at all monitoring sites. Runoff
temperatures were typically warmest at the beginning of a storm event and cooled as the
storm progressed. The coolest runoff temperatures were observed leaving the parking surface
at the Asheville bioretention area, likely due to the presence of a light colored chip seal.
At the stormwater wetland and wet pond, water temperature decreased as runoff was
transported through buried pipes (Figure 1). At the stormwater wetland, both median and
4
Effect of Urban Stormwater BMPs on Runoff Temperature in Trout Sensitive Regions
11th International Conference on Urban Drainage, Edinburgh, Scotland, UK, 2008
maximum water temperatures were significantly** (p<0.01) cooler after travelling through a
buried metal corrugated pipe to the wetland inlet for the period from May through August.
During some storms, runoff temperatures decreased by more than 7°C and cooled water
below the 21°C temperature threshold for trout before water entered the BMP. Similar
temperature reductions were observed when runoff was conveyed through a buried concrete
pipe from the parking surface to the wet pond inlet in Lenoir.
29
Temperature (°C)
27
25
23
21
19
17
15
07/04/06
07/06/06
07/08/06
07/10/06
Parking Drop Inlet
07/12/06
07/14/06
07/16/06
Wetland Pipe Inlet
Figure 1. Temperature of runoff before and after conveyance in the buried metal pipe at the
Asheville stormwater wetland
At the wet pond, the effluent temperature remained above 21°C for the entire period from
June through August. There was no evidence of a significant* difference between the wet
pond effluent temperature and the temperature of direct runoff from the parking surface for
the entire monitoring period. Because water was cooled as it was transported from the
parking surface to the wet pond inlet, median effluent temperatures were significantly**
warmer than the water entering the wet pond. At times, temperature differences exceeded
10°C, implying that the wet pond served as a source of thermal pollution (Figure 2). For the
entire monitoring period, temperatures at all pond depths (0 cm, 40 cm, 80 cm, and 120 cm)
were significantly** different from each other, with the warmest temperatures near the
surface. Diurnal fluctuations in water temperature were also greatest near the surface. These
large temperature fluctuations near the surface were indicative of the unpredictable nature of
effluent temperatures early in a storm as surface water is flushed through the outlet structure.
Jones
5
11th International Conference on Urban Drainage, Edinburgh, Scotland, UK, 2008
35
8
33
7
31
Temperature (°C)
5
27
25
4
23
3
21
Cumulative Rainfall (cm.)
6
29
2
19
1
17
15
07/01/06
0
07/06/06
Runoff
07/11/06
07/16/06
Concrete Inlet
07/21/06
Metal Inlet
07/26/06
Outlet
07/31/06
Rain
Figure 2. Influent and effluent temperatures at the Lenoir wet pond during July, 2006
Water leaving the current wet pond outlet structure was significantly** warmer than the water
at a depth of 120 cm for June through October. These results suggest that a modified outlet
structure that draws from lower depths in the water column would result in cooler effluent
temperatures. Despite these cooler temperatures, the water at a depth of 120 cm was still
significantly** warmer than 21°C for the months of June through September, which indicates
that effluent could still impact trout populations when the coolest water in the wet pond is
discharged. Not only was the water at a depth of 120 cm warmer than the trout temperature
threshold, it was also significantly* warmer than the metal pipe influent for the entire
monitoring period.
At the stormwater wetland, median effluent temperatures were significantly** warmer than
piped inflow temperatures and the 21°C trout threshold for the period of June through
September. Maximum effluent temperatures were significantly** warmer than runoff directly
leaving the parking surface during the same period. These temperature results implicate the
stormwater wetland as a source of thermal pollution. Despite increases in temperature,
reduction in runoff volume due to infiltration and evapotranspiration should effectively reduce
the thermal load to Reed Creek. Although complications with the flow monitoring equipment
made it difficult to quantify exact volume reductions, a number of storms during the
monitoring period were captured entirely by the wetland without generating outflow.
Water temperatures at the bottom of the stormwater wetland were the coolest during the
monitoring period and also exhibited the smallest diurnal fluctuations. For the months of
May, June, September, and October, water at the bottom of the stormwater wetland (90 cm)
was significantly** cooler than 21°C. There was no significant* difference in temperature
between the piped influent and water at a depth of 90 cm for every month in the monitoring
6
Effect of Urban Stormwater BMPs on Runoff Temperature in Trout Sensitive Regions
11th International Conference on Urban Drainage, Edinburgh, Scotland, UK, 2008
period. These results suggest that a modified outlet structure may discharge temperatures
suitable for trout; however, because inflow temperatures were already cool, it is unlikely that
substantial temperature reductions will result from wetland treatment.
At the Asheville bioretention area, there was no significant* difference between median
effluent temperatures and the 21°C threshold when examining the entire monitoring period;
however, median effluent temperatures were significantly* cooler than 21°C when the month
of August was excluded. Both influent and effluent temperatures exhibited substantial
seasonal variations with the warmest influent and effluent temperatures during the month of
August. There was no significant* difference between median influent and effluent
temperatures at the Asheville bioretention area, indicating that the bioretention area was not
able to consistently reduce runoff temperatures over the course of an entire storm. Although
maximum effluent temperatures were not significantly* different from the 21°C threshold,
maximum effluent temperatures were significantly* cooler than the maximum influent. A
reduction in maximum water temperatures suggests that the bioretention area was able to
reduce the initial runoff temperature spike, but unable to convey the benefits of cooler runoff
as a storm progresses.
At the Lenoir bioretention area, median and maximum influent temperatures were
significantly** warmer than 21°C, suggesting potentially negative impacts if runoff were
directly discharged into trout waters. Median and maximum effluent temperatures were both
significantly** lower than influent temperatures at the Lenoir bioretention area, which
indicates that the bioretention area was able to reduce the thermal impact associated with the
stormwater runoff. Despite the reduction in water temperature resulting from bioretention
treatment, median effluent temperatures were not significantly* different from 21°C and
maximum effluent temperatures were significantly** warmer than 21°C. Although thermal
impacts were reduced, the effluent from this bioretention area still posed some risk to the
thermal environment of trout waters.
Median and maximum inlet temperatures were significantly** warmer than 21°C at the
Brevard east and Brevard west bioretention areas. At the Brevard east bioretention area,
median and maximum effluent temperatures were significantly** cooler than influent
temperatures; however, effluent temperatures were also significantly** warmer than 21°C
(Figure 3). Similar to the Lenoir bioretention, the Brevard east bioretention was able to
reduce, but not eliminate the thermal impact to a coldwater stream environment. At the
Brevard west bioretention area, there was no significant* difference between median influent
and effluent temperatures; however, there was a significant** difference between maximum
temperatures. Similar to the Asheville bioretention area, these temperature results suggest
that the bioretention area was able to reduce the initial spike in runoff temperatures, but could
not convey the benefit of cooler runoff temperatures later in a storm. Median and maximum
effluent temperatures were both significantly** warmer than 21°C, indicating potential
thermal impacts to a coldwater stream environment. Seasonal trends in influent and effluent
temperatures were not as well defined during 2007, possibly due to the impact of the drought
throughout the region.
Jones
7
11th International Conference on Urban Drainage, Edinburgh, Scotland, UK, 2008
Figure 3. Influent and effluent temperatures at the Brevard east bioretention area
Despite significantly** warmer median influent temperatures at the Brevard east bioretention
area, there was no significant* temperature difference between median effluent temperatures
at the Brevard east and Brevard west bioretention areas. The primary difference between
these bioretention areas was the size of the contributing watershed, since they shared
immediately adjacent locations, similar soil properties, drain depth, and bioretention size.
The area occupied by the Brevard east bioretention area was equivalent to 7% of the
contributing watershed, while the area of the Brevard west bioretention was equal to 11% of
the contributing watershed. Based on temperature results alone, there appears to be minimal
thermal pollution benefits associated with moderately oversizing a bioretention area. One
benefit of a proportionally larger bioretention area that does not directly relate to effluent
temperatures is the ability of a larger system to infiltrate runoff without generating outflow.
When water exfiltrates, or leaves the bioretention area through the underlying soil and not the
drainage pipes, the thermal impact from that runoff is effectively eliminated. Although the
adjacent bioretention areas received the same rainfall, measurable outflow occurred in
response to 76% of rainfall events at the Brevard east bioretention area, but only 27% of
rainfall events at the Brevard west bioretention area.
Analysis of temperature trends at specific soil depths provided insight into optimum
underdrain depths. With the exception of night time surface temperatures, soil temperatures
were coolest at the bottom of the bioretention soil column during the summer months. Soil
temperatures at the bottom of the bioretention areas also exhibited the smallest fluctuations in
response to storms or diurnal and seasonal temperature changes. However, the stability of
soil temperatures at greater depths poses a potential risk of increasing the temperature of
infiltrating water during later portions of a storm when cooler runoff prevails. Although there
were several exceptions during late night storms, soils at a depth of 90 cm or greater remained
cooler than soils at shallower depths for most storms. Beginning in September, soil
8
Effect of Urban Stormwater BMPs on Runoff Temperature in Trout Sensitive Regions
11th International Conference on Urban Drainage, Edinburgh, Scotland, UK, 2008
temperatures were often warmest at the bottom of the bioretention areas, which raises
potential concerns of warm effluent from deep bioretention areas during spawning seasons in
the fall when temperature preferences are cooler.
CONCLUSIONS
Based on monitoring results, it appears that bioretention areas were able to best mitigate the
thermal impacts of urban stormwater runoff. The superior performance of standard
bioretention areas can be attributed to their ability to cool runoff through contact with deep,
insulated soils, and their capacity to reduce runoff flows. Although these systems effectively
reduced thermal pollution from urban stormwater runoff, thermal impacts were not
eliminated, since effluent temperatures were often warmer than the 21°C trout temperature
threshold. Although stormwater wetland and wet ponds provide a variety of water quality
benefits, they were not able to effectively reduce runoff temperatures and typically increased
the temperature of water they were treating. Utilization of a modified outlet structure for
these systems that draws from the bottom waters should result in cooler effluents than existing
structures. Although this type of outlet configuration could cause a stormwater wetland to
mitigate thermal pollution, temperature reduction in a wet pond appears to be unlikely.
Drawing water from the bottom of the water column also merits further investigation due to
water quality concerns associated with decreased dissolved oxygen levels and potentially
higher pollutant concentrations. Conveyance in underground pipes served as a major runoff
temperature reduction mechanism; however, the benefits of this cooler runoff were typically
diminished by the warmer water contained in the stormwater wetland and wet pond. It is
possible that conveyance in underground pipes could be implemented after BMP treatment to
better mitigate thermal pollution. Results of this research show that without careful
consideration, stormwater BMPs can exacerbate thermal pollution concerns from urban
stormwater runoff. When runoff temperature is considered in BMP design, it should be
possible to reduce runoff temperatures or minimize temperature increases, while maintaining
the water quality and quantity benefits of these treatment systems.
ACKNOWLEDGEMENT
This research was funded by the North Carolina Department of Environment and Natural
Resources (NCDENR), Division of Water Quality. The authors would like to thank Dr. Bill
Hunt, Jonathan Smith, Dr. Dan Willits, Dr. Garry Grabow, Dr. Aziz Amoozegar, Jon
Calabria, Allen Caldwell, Eric Caldwell, Seth Nagy, and Jason Zink, all of whom are
currently or formerly of NC State University, for their assistance in selecting research sites
and support throughout the project.
REFERENCES
Asaeda, T., Ca, V. T., and Wake, A. (1996). Heat Storage of Pavement and its Effect on the Lower Atmosphere.
Atmospheric Environment, 30(3), 413-427.
Coutant, C. C. (1977). Compilation of Temperature Preference Data. Journal of the Fisheries Research Board of
Canada, 34 740-745.
Dietz, M. E., and Clausen, J. C. (2005). A Field Evaluation of Rain Garden Flow and Pollutant Treatment.
Water, Air, and Soil Pollution, 167 123-138.
Hocutt, C. H., Stauffer, Jay R., Jr., Edinger, J. E., Hall, Lenwood W., Jr., and Morgan, Raymond P., II. (1980).
Power Plants: Effects on Fish and Shellfish Behavior, Academic Press, New York, NY.
Hunt, W. F., Jarrett, A. R., Smith, J. T., and Sharkey, L. J. (2006). Evaluating Bioretention Hydrology and
Nutrient Removal at Three Field Sites in North Carolina. Journal of Irrigation and Drainage
Engineering, 132(6), 600-608.
Jones
9
11th International Conference on Urban Drainage, Edinburgh, Scotland, UK, 2008
Kieser, M. S., Spoelstra, J. A., Feng Feng, A., James, W., and Li, Y. (2004). Stormwater Thermal Enrichment in
Urban Watersheds.
Lieb, D. A., and Carline, R. F. (2000). Effects of Urban Runoff from a Detention Pond on Water Quality,
Temperature and Caged Gammarus minus (Say) (Amphipoda) in a Headwater Stream. Hydrobiologia,
441(1), 107-116.
Mohseni, O., Erickson, T. R., and Stefan, H. G. (2002). Upper Bounds for Stream Temperatures in the
Contiguous United States. Journal of Environmental Engineering, 128(1), 4-11.
Morrill, J. C., Bales, R. C., and Conklin, M. H. (2005). Estimating Stream Temperature from Air Temperature:
Implications for Future Water Quality. Journal of Environmental Engineering, 131(1), 139-146.
Wierenga, P. J., Hagan, R. M., and Nielsen, D. R. (1970). Soil Temperature Profiles During Infiltration and
Redistribution of Cool and Warm Irrigation Water. Water Resources Research, 6(1), 230-238.
Wilcoxon, F. (1945). Individual Comparisons by Ranking Methods. Biometrics, 1(6), 80-83.
Notation
* Denotes significance at 5% level
** Denotes significance at 1% level
Dates are given as follows: July 1st, 2006 appears as 07/01/06
10
Effect of Urban Stormwater BMPs on Runoff Temperature in Trout Sensitive Regions