Testing the effects of wave exposure, site, and behavior on intertidal mussel body temperatures: applications and limits of temperature logger design

The rocky intertidal mussel Mytilus californianus is exposed to potentially damaging thermal conditions during low tide. However, because the temperatures of ectothermic organisms are driven by multiple climatic factors, we do not fully understand what the body temperatures of intertidal invertebrates are under field conditions, or how thermal stress varies between intertidal sites. We designed a temperature logger that thermally matches (similar size, color, morphology and thermal inertia) living mussels of the species M. californianus, and tested its ability to provide realistic measurements of body temperature in the field. As part of these tests, we examined the propensity of M. californianus to "gape," a behavior in which the mussel opens its shell valves, and which may permit evaporative cooling. Because our instruments were unable to mimic this behavior, we tested the degree to which gaping contributes to animal cooling by exposing M. calfornianus to a range of climatic conditions while recording body temperatures, gaping behavior and water loss. Results indicated no significant influence of gaping on body temperature. Tests comparing temperatures of loggers to those of real mussels under laboratory and field conditions showed that thermally matched loggers recorded temperatures within a few degrees of living animals and that unmodified loggers regularly incurred errors of up to 14degreesC. We then deployed a series of thermally matched loggers at two sites in central Oregon (Boiler Bay and Strawberry Hill) previously hypothesized to display site-specific differences in aerial temperature, and at various wave-exposure regimes within each site. Significant differences were demonstrated between sites using a subset of temperature metrics in a multivariate analysis. Yearly peaks in maximum temperature, average daily maximum temperature, and degree hours were useful in discriminating between sites. In 2001, wave-exposed sites at Strawberry Hill displayed fewer degree-hours than wave-protected sites, but an equivalent or greater maximum temperature. In 2002 both of these metrics were significantly lower in wave-exposed sites. Boiler Bay and Strawberry Hill differed in thermal regime, but not in a way that would indicate one was "hotter" than the other.