When to worry about the weather: role of tidal cycle in determining patterns of risk in intertidal ecosystems


K. A. S. Mislan, tel. +803 777 3931, fax +803 777 4002, e-mail: kas.mislan@gmail.com


Species range boundaries are determined by a variety of factors of which climate is one of the most influential. As a result, climate change is expected to have a profound effect on organisms and ecosystems. However, the impacts of weather and climate are frequently modified by multiple nonclimatic factors. Therefore, the role of these nonclimatic factors needs to be examined in order to understand and predict future change. Marine intertidal ecosystems are exposed to heat extremes during warm, sunny, midday low tides. Thus, the timing of low tide, a nonclimatic factor, determines the potential contact intertidal invertebrates and algae have with heat extremes. We developed a method that quantifies the daily risk of high temperature extremes in the marine intertidal using solar elevations and spatially continuous tidal predictions. The frequency of ‘risky days’ is variable over time and space along the Pacific Coast of North America. Results show that at some sites the percentage of risky days in June can vary by 30% across years. In order to do a detailed analysis, we selected San Francisco as a study site. In San Francisco, May is the month with the greatest frequency of risky days, even though September is the month with the greatest frequency of high air temperature, ≥30 °C. These results indicate that marine intertidal organisms can be protected from high temperature extremes due to the timing of tides and local weather patterns. In addition, annual fluctuations in tides influence the frequency of intertidal zone exposures to high temperature extremes. Peaks in risk for heat extremes in the intertidal zone occur every 18 years, the length of the tidal epoch. These results suggest that nonclimatic variables can complicate predictions of shifts in species ranges due to climate change, but that mechanistic approaches can be used to produce predictions that include these factors.