How Might the Northwest Heatwave Affect Streamflow?

You are certainly on the right track thinking that there are connections between air temperature and streamflow, and here are a few examples of how those connections can play out.

Hot/dry periods can indeed enhance evaporative loss from rivers and streams, assuming constant relative humidity, wind speeds, etc. As temperatures rise, so does the vapor pressure deficit and evaporative flux. As an order of magnitude estimate, researchers in Canada have measured that streams loose on the order of 1–3 millimeters (mm) of depth per day due to evaporation during summer months in forested streams. Others have suggested evaporation rates of up to 6 mm per day for reservoirs. Using the upper end of their estimates (6 mm), that would be equivalent to reducing flow in a 10-ft wide stream by ~0.012 cubic feet per second per mile of stream (0.02 liters per second per 100 meters). To put that into perspective, summertime rates of gains and losses in the Bull Trout watershed have been measured at 100 to 1,000 times higher than that. So, it seems unlikely that evaporation from the water surface is a significant pathway to lose water from a mountain stream. However, there are other ways besides direct evaporation that air temperatures can affect streamflow in the intermountain west.

For example, we see connections between air temperatures and streamflow that are mediated through changes in soil moisture content. Hot/dry conditions drive up potential evapotranspiration in the landscape, which lowers soil moisture content and can reduce runoff ratios. That means that less rain will make it to the streams and there will be less streamflow for the same amount of precipitation. Think of it like the difference between running water on a damp (normal conditions) vs a dry (drought conditions) sponge. In this way, high air temperatures today can have impacts on streamflow generation in the future. We see this in tundra at relatively short time scales where, after a dry, warm period of just days, it has to rain more than ~15 mm before the streams start to fill. Some outstanding research led by our colleagues at the USGS Northern Rocky Mountain Science Center showed that these connection can be observed at decadal and even country time scales by analyzing tree ring data going back 1200 years. A particularly pertinent quote from a recent publication reads: 

“We find that temperature has increasingly influenced the severity of drought events by decreasing runoff efficiency in the basin since the late 20th century (1980s) onward. The occurrence of extreme heat, higher evapotranspiration, and associated low-flow conditions across the basin has increased substantially over the 20th and 21st centuries, and recent warming aligns with increasing drought severities that rival or exceed any estimated over the last 12 centuries.”

The increasing influence of air temperature on streamflow noted in that quote is further supported by research in the Upper Colorado River. Our colleagues at the USGS Utah Water Science Center recently showed that the long-term trends in the historical records are forecast to continue. Using computer models, they report in a recent paper that the majority of model scenarios resulted in decreases in streamflow in the year 2080 relative to the historical record. Pertinent to your question about mountain streams, they project the greatest decreases in streamflow will occur in headwater environments.

You also mentioned the impact on snowpack, and you are spot on. There has been a documented increase in air temperature at snow monitoring locations throughout the Western US. At the same time, there has been a decrease in snow water equivalent and a shift to earlier peak snowpack. The influence of these changes on streamflow can be seen in the recently released USGS data visualization Snow to Flow.  I particularly like the interactive graphics on that page. Additionally, watersheds in our region have been identified as “sensitive to snow supply changes” in a recent study, meaning that there could be significant social/economic impacts of alterations in our regions snowpack. The importance of snowpack and melting dynamics is further seen in a study by a USGS colleague that shows that watersheds in the western US are dominated by snowmelt inputs. Taken together, these lines of evident illustrate how air temperature can influence water resources just be changing the phase of precipitation.

Lastly, my personal favorite on how air temperatures can impact streamflow!  Jim Constantz, a USGS stalwart, published a paper in 1998 describing how water temperature can impact streamflow and groundwater exchange in alpine streams. It is a fantastic read, and the highlight for this conversation is this: In cases where increases in air temperatures result in increases in water temperatures, water viscosity is lowered, allowing for more rapid passage of water through the riverbed. This can lead to increased losses from losing reaches by effectively increasing the hydraulic conductivity of the riverbed. One result of this is that losses from streams to groundwater can be greatest in the evening when the water temperature is highest, and lower in the mornings when water temperature is lower. So, it follows that extremely high air temperatures could exacerbate losses from streams to groundwater in reaches that were already losing water.