The Hydrological Urban Heat Island: Determinants of Acute and Chronic Heat Stress in Urban Streams

Abstract

During and after rainfall events, the interaction of precipitation with hot urban pavements leads to hot runoff, and its merger with urban streams can result in an abrupt change in water temperature that can harm aquatic ecosystems. To understand this phenomenon and its relation to land cover and hydrometeorological parameters, we analyzed data spanning two years from 100 sites in the eastern United States. To identify surges, we first isolated temperature jumps of at least 0.5°C over 15 min occurring simultaneously with water flow increase. Surge magnitude was defined as the difference between peak stream temperature and baseflow temperature right before the jump. At least 10 surges were observed in 53 of the studied streams, with some surges exceeding 10°C. Our results demonstrate that the watershed developed area and vegetation fraction are the best descriptors of surge frequency (Spearman correlation of 0.76 and 0.77, respectively). On the other hand, for surge magnitude and peak temperature, the primary drivers are stream discharge and stream temperature immediately before the surge. In general, the more urbanized streams were found to be already warmer than their more “vegetated” counterparts during baseflow conditions, and were also the most affected by temperature surges. Together, these findings suggest the existence of a hydrological urban heat island, here defined as the increase in stream temperature (chronic and/or acute), caused by increased urbanization.

The heat is on: Predicting urban stream temperature responses to summer storms

“…Events with temperature increases were more likely to occur in the afternoon/evening, whereas stream temperature decreases typically occurred earlier in the year (before the middle of July, day of year 200). The finding that time of day is linked to the occurrence of temperature increases is in agreement with work by Zahn et al (2021), who quantified heat pulses across 100 watersheds in the midwestern and eastern US, and Hester and Bauman (2013), who observed water temperatures near urban storm sewer outfalls in Virginia, USA. Both studies observed a greater frequency and magnitude of temperature increases in the afternoon when pre‐event discharge and pre‐event temperatures were higher.…”

“…Similarly, the shape and form of these stream temperature responses may vary, and (gradual) temperature increases do not appear to be the only possible stream temperature response during a storm event. Zahn et al (2021) observed that temperature increases may be followed by a subsequent rapid decrease in water temperature, a pattern we will refer to as a ‘heat pulse’ in our study. However, further exploration of this phenomena has been limited.…”

“…It thus remains to be seen whether such patterns would hold at broad scales, which would require a comparison of a substantially larger number of sites than the six investigated here. While other studies have investigated how site characteristics, such as shading, stream width, and drainage area, are related to temperature changes (Nelson & Palmer, 2007; Somers et al, 2013; Wissler et al, 2022; Zahn et al, 2021), our study uniquely links these behaviours to event characteristics and identified hydrograph metrics (e.g., , ) as important predictors for the temperature patterns (Figures 3 and 4).…”

“…More recently, Beganskas and Toran (2021) found that the frequency and magnitude of stream temperature increases correlated well with the percentage of impervious land cover at the watershed scale. At regional scales, Zahn et al (2021) documented the highest frequency of temperature increases in highly developed watersheds with a low percentage of vegetated area.…”

Physicochemical Factors and Urban Land-Use Characteristics Associated with Resistance to Precipitation in Estuaries Vary Across Scales

“…on DO dynamics in streams and estuaries (Apple et al, 2006;Caffrey, 2003;Caffrey et al, 2014). Highlighted results link climate change and urbanization to thermal pollution of urban streams following rain events (Zahn et al, 2021), and show connections of elevated global temperatures with decreased primary production (Song et al, 2018) and increased planktonic N demand (Toseland et al, 2013). Therefore, the combined effects of watershed urbanization and global temperature increase on phytoplankton could exacerbate DO dynamics in estuaries and significantly shift the range of DO baseline.…”