of the original survey sites. The 52 years separating the two sampling intervals span a period of climatic warming. During the latter half of the 20th century, maximum air temperatures near the eastern and western ends of the Strait of Juan de Fuca near Washington state increased by ~0.2 and ~0.13 °C per decade, respectively, and mean annual water temperatures along the southern and western Vancouver Island coast warmed by ~0.08–0.11 °C per decade. Average daily maximum air temperatures during the summer, which are particularly relevant to thermal stress experienced in the intertidal zone, have warmed even more rapidly. At Victoria, on the southern end of Vancouver Island, summer average daily maxima have risen approximately linearly at a rate of 0.654 °C per decade since 1950, corresponding to an increase of 3.48 °C over 52 years, which has caused a 51% decrease in the vertical distribution of M. californianus, providing strong evidence that this change is linked to global warming. Associated with this decrease are local extinctions of M. californianus at 13% of the sites resurveyed between 2009 and 2010. Historical data for M. trossulus were only available for one site (former vertical range = 49 cm). By 2010, M. trossulus had been completely eliminated from that site, with the exception of three small juveniles found under a single rock.
From 1995 to 1999, the poleward movement of M. galloprovincialis showed a reversal concomitant with a cooling phase of the PDO (Hilbish et al. 2010). M. galloprovincialis has declined in abundance over the northern third of its geographic range (~540 km) and has become rare or absent across the northern 200 km of the range it previously colonised during its initial invasion. The distribution of the native species M. trossulus has, however, remained unchanged over the same time period. The difference in SST between warm and cold phases of the PDO is small (~1 °C), but Hilbish et al. (2010) deduced that even this minor decrease in temperature during the cold phase of the PDO may be enough to retard larval development in M. galloprovincialis, such that recruitment is handicapped in northern waters.
Geographic ranges of species are determined by climatic conditions, and this forms the basis of a number of models that use associations between aspects of climate and species’ occurrences to estimate the conditions that are suitable for maintaining viable populations (Araújo & Peterson 2012). The most frequently used climate change model is species distribution modelling (SDM), also known as environmental (or ecological) niche modelling, habitat modelling, predictive habitat distribution modelling and range mapping. SDM uses computer algorithms to predict the distribution of a species across geographic space and time using environmental data (Elith & Leathwick 2009). It can explain how environmental conditions influence the occurrence or abundance of a species, and can provide models (ecological forecasting) of a species’ future distribution under climate change, a species’ past distribution (in order to assess evolutionary relationships) or the potential future distribution of an invasive species. There are two main types of SDM: correlative SDMs, also known as climate envelope models or bioclimatic models, which model the observed distribution of a species as a function of environmental conditions; and mechanistic SDMs, also known as process‐based models or biophysical models, which use independently derived information about a species’ physiology to develop a model of the environmental conditions under which it can exist (Kearney & Porter 2009). In empirical modelling, a number of climatic variables, such as maximum and minimum temperatures and precipitation, are measured for many different locations and statistically compared to the occurrence of the focal species at these locations (Jeschke & Strayer 2008). Most bioclimatic models do not explicitly consider biotic interactions (e.g. predation and competition between species) or limitations to dispersal, and assume that species lack sufficient plasticity to adapt to environments beyond those currently occupied (Jeschke & Strayer 2008; Fuller et al. 2010). The ecological effects of climate change can only be predicted if there is an understanding of the physiology of the species in its natural environment. Evidence based on temperature indicates that species’ distributions are driven both by short‐term exposure to lethal conditions and by repeated or longer‐term exposures leading to energetic failures. The relationship between physiological performance (e.g. scope for growth, SFG) and body temperature can be depicted as a thermal performance curve (Figure 3.17). Woodin et al. (2013) define the difference between lethal and sublethal exposure limits as the transient event margin (TEM), the range of environmental conditions below the short‐term lethal limit that a species may endure on a transient basis but which would lead to mortality over longer time spans. The magnitude of the disparity (TEM) between performance and tolerance temperature thresholds relative to environmental variance determines the likelihood of failure of bioclimatic model predictions (Woodin et al. 2013). To define TEM, both CTmax SFG and LTmax data are necessary, but in the majority of data sets only LTmax estimates are available. Jones et al. (2010) used a mechanistic biogeographic model for M. edulis on the Atlantic coast of N. America based on LTmax. Physiological limits were compared against environmental temperatures and the biogeographic distribution of the species was accurately predicted. The hindcast using the model also successfully predicted the historical changes in distribution over half a decade described earlier. However, when the same model applied to Europe, it failed to predict the distribution of M. edulis; in reality, its distribution in Europe was 50% less than that predicted. When an energetics model was applied to Europe, the predicted distribution was close to the actual distribution of the species on western European coasts. This highlights the need for caution in applying the same model across large geographic distances (see details in Woodin et al. 2013).
Figure 3.17 (a) Relation between performance as scope for growth (SFG) and temperature. Lethal limits are LTmin and LTmax, performance limits are critical minimum temperature (CTmin) and critical maximum temperature (CTmax), the body temperature at any point in space and time is Tbody and the optimal temperature is Topt. The upper transient event margin (TEM) = LTmax − CTmax. Thermal response curves vary from species to species but all typically display some optimum temperature (Topt) at which performance is maximised, as well as critical minima (CTmin) and critical maxima (CTmax) beyond which mortality and/or reproductive failure occur (Helmuth et al. 2010). (b) Temperature versus frequency for environments a, b and c. TEM from (a) is expressed on yearly environmental variance curves for three environments, all with the same mean environmental temperature but differing in variance.
From Woodin et al. (2013). Source: From Woodin et al. (2013). Reprinted with permission from John Wiley & Sons Ltd. CC BY 3.0.
Closely related species with different physiological tolerances and distributions make ideal systems for documenting range shifts in response to a changing climate. The closely related species M. edulis, M. trossulus and M. galloprovincialis have distinct biogeographical ranges that are correlated with SST. Fly & Hilbish (2013) determined the SFG of these three species over a range of temperatures (5–30 °C) to determine whether energetics could predict their distributions. SFG represents energy available for growth and/or reproduction above that necessary for maintenance requirements. The results showed, as expected, reasonable correlation with the temperature regimes of habitats presently occupied, with M. trossulus, a boreal species, exhibiting positive SFG up to 17 °C, M. edulis, a cold temperate species, up to 23 °C, and M. galloprovincialis, a warm‐water species, up to 30 °C. In the spring, temperatures of peak performance (Topt) for both M. trossulus and M. edulis occurred at ~15 °C, and both species exhibited negative SFG at 25 °C, which was the Topt for M. galloprovincialis. Overall, SFG was lower in summer than in autumn and winter. M. trossulus only showed positive SFG at 10 °C, while M. edulis and M. galloprovincialis maintained positive SFG at all temperatures above 5 °C, with Topts at 20 and 25 °C, respectively (see Figure 6.15). The warm end of each species’ range correlated positively with the seawater temperature at which that species’ SFG became negative (i.e. when metabolic expenditure exceeds energy acquisition) in summer and autumn. Energetics at cold temperatures did
