Abstract

During and subsequent to a rainfall event, stormwater runoff can transport heat from an urban area (which can be significantly warmer than its surrounding region) to a receiving water body, such as a stream, river, lake, or a wetland. Excess heat can affect an aquatic ecosystem in the following ways (Thomann and Mueller, 1987):

• direct lethal effect on sensitive plants or animals;
• indirect long-term effects on the aquatic ecosystem by affecting growth and/or reproduction; and
• indirect effects through changes in species distribution in the ecosystem.

Observations have been made that support the idea that stormwater runoff can contain enough heat energy to warm the water of a receiving body above temperatures that can harm biota living within it (Galli, 1990; Herb et al., 2009; James and Xie, 1999; Kieser et al., 2004; Li and James, 2004; Nelson and Palmer, 2007; Van Buren, 1999; Verspagen, 1995). Consequently many jurisdictions have implemented regulatory guidelines on temperature limits of urban stormwater runoff. For example, in the Stormwater System Design Manual that is produced by the Ontario Ministry of the Environment it is stated (Ontario Ministry of the Environment, 2003, Section 3.3.4.2):

Where temperature is a significant concern it is recommended that the designer consult with the local conservation authority, the federal Department of Fisheries and Oceans (Fisheries and Habitat Management) and the Ontario Ministry of Natural Resources, during the design process.

A challenge for municipalities is to predict under what conditions storm-water effluent could violate regulatory limits. If such conditions could be predicted with sufficient accuracy then management practices could be developed to mitigate the adverse impacts on the receiving water bodies. To support this endeavour, several computer models have been developed to evaluate the transport of heat energy in urban stormwater systems (or components within them). Results from these models should be compared to measured values from representative systems to demonstrate that the important processes have been adequately simulated and to establish the validity of these models for this type of problem.

The focus of this chapter is on using a low cost sensor array to measure temperature at different depths throughout a water column in wet detention ponds, i.e. ponds that retain a permanent pool of water between rainfall/runoff events. Wet detention ponds have been used in the management of urban stormwater runoff for several decades and are considered essential components in many stormwater systems. For many of these ponds, the ratio of the water surface area to the pool volume is large enough that the water temperature responds quickly to changes in atmospheric conditions. In particular, these ponds can heat up during summer days and become warmer than nearby natural water bodies.

Van Buren et al. (2000) observed that an on-stream detention pond, in Kingston, Ontario, with an average depth of 1 m added thermal energy to the creek into which it discharged. They measured the temperatures at the inlet, outlet and at one point in the pond (1 m below the waterline, just above the pond bottom) at 20 min intervals from May to September, 1997. A series of discrete water temperature surveys was also conducted in the pond where temperature was measured at depths of 0.10 m, 0.45 m and 0.90 m below the water surface, and 0.10 m above the pond bottom, at nineteen different locations. They found that temperature varied at the different locations and depths. During rainfall/runoff events, the pond had two distinct flow zones in the horizontal plane (Shaw et al., 1997) and temperature differences were observed between these zones. In the advection zone, flow was from the inlet to the outlet (with higher velocities than average) in a direct route and the water temperature was up to 2.0 °C cooler than water in the recirculation zone, where flow velocities were considerably less than average. Between rainfall/runoff events and under conditions of low flow and calm winds, it was observed that the average temperature at the water surface was 3.6 °C warmer than at the bottom of the pond (where temperature was monitored continuously). These observations suggest that the assumption of well mixed conditions, i.e. a single representative temperature for water temperature in the pond, needs careful consideration as variations in temperature can exist.

In a study conducted by Herb et al. (2009), a wet detention pond in Woodbury, Minnesota, was instrumented with a string of thermistors, which collected temperature measurements at seven elevations through a 2.5 m water column in the pond, from June 3 to August 25, 2005. The observations showed that the vertical temperature gradient in the pond varied during the course of the monitoring season and there could be several degrees centigrade difference in water temperature between the top and bottom of the pond. Temperature measurements were also taken near the inlet and outlet of the pond and it was observed that the overall average outlet temperature was 1.2 °C higher than the inlet temperature but with significant variations among the rainfall events. Although higher outlet temperatures were observed in some cases, the primary function of the pond is to attenuate peak flow which means that heat energy added to a downstream water body would be over a greater duration than if the pond was not in the system. Therefore, the use of a wet pond to mitigate the thermal impacts from urban runoff on a water body results in a tradeoff between reducing the magnitude of the temperature increase on the water body and lengthening the duration of this increase.

To build upon the work done by Van Buren et al. (2000) and Herb et al. (2009), more monitoring of water temperatures in wet detention ponds is being planned by the authors of this chapter. The main driver for this is to examine the effects of pond bathymetry and the locations of inlets and outlets on the water temperature within and exiting the pond. A better understanding of these relationships could help in the design of wet detention ponds to better mitigate the thermal effects of urban stormwater runoff on water bodies.

It is realized that to build a sufficient database of measurements for this endeavour will require several thermal arrays to measure vertical temperature profiles in the ponds. The thermal sensor array proposed in Section 6.2 was designed to minimize cost while maintaining the ability to collect good quality temperature measurements. This array was tested in a wet detention pond in Regina, Saskatchewan, from September 9 to October 16, 2009, and the results of this evaluation are presented in Section 6.3. Conclusions on the evaluation as well as recommendations for further deployments of the thermal sensor array are provided in Section 6.4.

This paper is only available in PDF Format:

  View full text PDF