Date of Award


Document Type

Open Access Thesis


Earth and Ocean Sciences

First Advisor

Jean Taylor Ellis


Rising sea level and increases in temperature are causing biogeographic shifts in intertidal zones and can also lead to shifts in the local vertical zonation of organisms due to changes in body temperature during aerial exposure during low tide. Swash is an important determinant of aerial body temperatures, and vulnerability of intertidal zones to changes in climate could potentially depend on how much and how often animals are cooled by waves. While wave heights (and thus swash) are generally expected to increase with climate change, anthropogenic physical structures, such as breakwaters and wave energy converters, can decrease wave height and swash. Here I explore the ability of a simple biophysical model of aerial body temperature to predict how changing wave climates might affect intertidal zones in California. A key limitation is that biophysical models frequently rely on input data from weather stations that are limited in their spatial coverage. An attractive alternative is to use larger-scale reanalyzed data with global coverage, but the accuracy of the models that use these data has not been evaluated in the context of sensitivity to swash.

This study examines the sensitivity of mussel (Mytilus californianus) aerial body temperature to changing atmospheric, oceanographic, and morphological parameters. Specifically, it compares model predictions using coarse gridded weather data from NOAA NCEP Climate Forecast System Reanalysis (CFSR), versus local weather station data as environmental inputs. A sensitivity analysis was conducted to determine whether precise details of mussel morphology and site topography are needed, or whether more generic estimates can be used. The model was used to explore the potential impacts of changing wave climates on mussel temperatures by considering current and potential future wave heights at four locations in California (Bodega Bay, Hopkins, Alegria and Coal Oil Point). In all cases model results were ground-truthed against in situ measurements of temperature made using biomimetic sensors using both standard measures of error (average absolute difference and RMSE) as well as metrics of physiological state based on mussel performance curves.

Model results demonstrated that body temperatures predicted using rough estimates of mussel absorptivity, mussel size, cloud cover, absolute shore level (ASL), effective shore level (ESL) slope and ASL plus ESL slope were very close to those generated using organism- and site-specific parameters. At Bodega Bay, the model successfully predicted maximum daily aerial body temperatures to within a few degrees using both in situ weather station (average absolute error 3.2°C) and CFSR data (error 3.3°C). Model error using CFSR data for the other locations was considerably higher (error: Hopkins Marine Station, 6.8°C; Alegria, 6.5°C; Coal Oil Point, 8.5°C). Physiologically, the model ran cold at Bodega and Hopkins but ran hot at Alegria and Coal Oil Point. Sensitivity analyses suggested that mussel temperatures were most sensitive to changes in the nearshore wave climate at Bodega Bay and Hopkins and least sensitive at Alegria and Coal Oil Point. Due to the overall inaccuracy of this model when using CFSR data, the use of this method by decision makers should be approached with caution. This study suggests that additional analyses using a tailored site-specific model are required.