Stream temperature responses to experimental riparian canopy gaps along forested headwaters in western Oregon

Streamside (riparian) areas in the western United States and across much of North America are dominated by young, regenerating forests with closed canopies, which shade the understory and reduce light to streams. The addition of canopy gaps has been suggested as a management tool to accelerate development of riparian forest complexity, create stream light conditions that mimic those in late successional forests, and enhance in-stream productivity. Although gaps form naturally in late-successional forests, explicitly adding gaps is a concern because the increases in light that accompany a canopy gap may have the potential to increases stream temperature, which is an important ecological driver and regulatory metric in streams. The goal of this study was to determine whether and to what degree riparian forest canopy gaps that reflect localized disturbance events (mortality of one to a few dominant canopy trees) affect stream temperatures. We created experimental gaps in young regenerating riparian forests along six replicate headwater streams in western Oregon. Gaps increased light along 90 m study reaches by 3.91 (+/- 1.63) moles of photons m(-2) day(-1), similar to that of a naturally occurring gap in a late-successional forest. Using a Before-After-Control-Impact study design, we assessed stream temperature by tracking multiple responses in each reach including: maximum seven day moving average of daily maximums (T-7DayMax), maximum seven day moving average of daily means (T-7DayMean), daily maximum, and mean summer temperatures, as well as within-reach (every 30 m) responses, and downstream recovery. Over a 40-day period in summer (July 22nd August 30th), the mean response in T-7DayMax across the six replicate streams was 0.21 degrees C +/- 0.12, and the mean response in T-7DayMean was 0.15 degrees C +/- 0.14. Although the mean response in T-7DayMax (a key regulatory metric) was small, changes varied across individual study streams, and the magnitude of the relative increase in stream T-7DayMax as a result of the canopy gap was strongly negatively correlated with stream size. T-7DayMax was not correlated with the size of the canopy opening or change in reach-scale light availability (within the range of gap sizes from 514 m(2) to 1,374 m(2) (0.05 to 0.14 ha)). Overall, riparian forest canopy gaps have the potential to increase stream temperatures, but in the western Cascade Mountain headwaters studied here, gap effects were small (all < 0.5 degrees C), and temperature responses declined as stream size increased.