The preponderance of evidence is that the earth’s climate is changing at an unprecedented rate, and many predictions are dire about the consequent changes on ecosystems. In regard to the ocean, the major physical variables of change are likely to be temperature, wave forces, sea level, and pH from increasing atmospheric CO₂ (Brierley and Kingsford 2009, IPCC 2013). Climate change must be viewed, however, within the context of numerous other stressors from increasing human populations, urbanization, intensified land use, coastal runoff of sediments, nutrients, and contaminants, species extractions, and the spread of nonindigenous species (e.g., Schiel 2009, Smale et al. 2013; see Chapter 10). These in combination may have numerous nonlinear and perhaps unpredictable effects on species distributions, community structure, disease vectoring, food and interaction webs, and ecosystem functions and services (Mooney et al. 2009, Harley et al. 2012, Smale et al. 2013). Here we use “global change” as a more inclusive term to indicate climate change within the broader context of other stressors and impacts.

It seems clear there are no “free lunches” with respect to the environment, and that almost all activities have flow-on effects to other parts of the broader ecosystem. Given this connectivity and interaction among ecosystem components, isolating the effects of (p.286) one of even a few variables acting on kelp forest dynamics is challenging. We know that impacts have already occurred and that we are not proceeding from a baseline of pristine conditions (cf. Dayton et al. 1998). Nevertheless, the impacts of extreme physical conditions, especially at the limits of kelp distributions, give clues about changes that might occur in giant kelp forests over the coming decades. As long-term studies continue in more places, we will get a sharper view of changes, their environmental correlates, and causes. Furthermore, a wide range of laboratory-based studies is beginning to show the sensitivities of many key species, for example to changes in water temperature and pH levels, although it is currently far from clear how these might translate into natural communities. The scenarios for species responses are usually designated as tolerate, move, adapt, or die (e.g., Harley et al. 2012). If this book and similar analyses stand the test of time, future readers will be able to see how accurately present-day scientists predicted effects and changes, or if we were alarmist or wide of the mark in how extreme the changes turned out to be. Certainly, there has been an exponential increase in the literature on climate change over the past few years, and this seems likely to continue as science further engages in trying to understand the mechanisms, processes, and consequences of global change. Here we present an overview of climate changes and discuss potential effects on giant kelp and associated organisms.

Climate Change Variables and Predicted Trends

The Intergovernmental Panel on Climate Change (IPCC) has gone through periodic updates of climate model refinements and predictions. The bad news, with a high degree of certainty, is that a wide range of climate-related variables are undergoing large and rapid change. The IPCC reports that the averaged globally combined land and ocean surface temperature has increased by 0.78°C during the period 2003–2012 compared to 1850–1900 (figure 13.1). There has been substantial interannual and inter-decadal variability due, for example, to strong ENSO and Pacific Decadal Oscillation periods. The IPCC (2013) reports with “high confidence” that the ocean accounts for >90% of the energy accumulated between 1971 and 2010, with the upper 75 m having warmed by 0.11°C per decade, and with 60% of this energy stored in the upper ocean (0–700 m depth). In conjunction with soot fallout and loss of ice albedo (Hansen and Nazarenko 2004), this has led to melting sea ice, thermal expansion, and increases in mean sea level by around 0.19 m over the past century (figure 13.2). Carbon dioxide concentrations have risen by around 40% since pre-industrial times, and the ocean has absorbed around 30% of this, which has caused increased acidification (IPCC 2013; figure 13.3). The rise in CO₂ concentration in the atmosphere is the greatest in several hundred-thousand years (Hoegh-Guldberg et al. 2007) and the adsorption of CO₂ has led to a decrease of around 0.2 pH units per decade over the past 30 years (Hoegh-Guldberg and Bruno 2010). There is also accumulating evidence that changes to the earth’s wind fields are leading to increased wave action in many parts of the world’s oceans (Young et al. 2011), (p.287)

which potentially will have direct impacts on kelp forests. The most recent climate models show all of these trends continuing through this century. The IPCC climate models for future scenarios are based on Representative Concentration Pathways (RCPs) carried out under the World Climate Research Programme. Based on these models, the most likely projections are for continued increases in surface temperature, loss of sea ice, increases in sea level, and increasing acidification in the ocean surface. Furthermore, IPCC (2013) concludes that most aspects of climate change are likely to last for many centuries.

(p.288)

(p.289) (p.290) Taken together, these changes and their interactions provide a dramatically altered environmental envelope in which ocean ecosystems are likely to operate. The potential changes to kelp forest ecosystems are far less certain and largely surmised, due not only to the, as yet, unpredicted and complex ways that climate variables may act on species and their interactions, but also in how they may act in a more direct way on giant kelp itself with its complex requirements and physiological–ecological responses across its life stages (see Chapters 2–4). Helmuth et al. (2006) raised several questions that must be addressed in forecasting the impact of climate change on the distribution and abundances of organisms. Among these are, how close are individuals living to the limits of their physiological tolerances within their range? Essentially, are they at the limit of their “fundamental niche”? Other questions involve knowing where distributions are determined by biotic interactions that may be affected by climate indirectly, whereby boundaries and limits determine the “realized niche” space, understanding whether species can adapt and evolve, and how patterns of fertilization success and propagule dispersal are affected. Because of the complexities involved in answering these questions, it is often the case that community changes do not necessarily correlate well with any single factor. With reference to predictions of future responses, Harley et al. (2012) termed these “gaps in the ecophysiological framework.”

These complexities are exemplified by a reconstruction of giant kelp distribution and biomass in southern California since the last glacial maximum, around 20,000 years ago. Graham et al. (2010) made the assumption, which they recognized as untestable, that the responses of Macrocystis to environmental forcing have stayed relatively constant over this long time period. Their model showed that the area cover of rocky reef and the biomass of giant kelp peaked around 13,500 years ago, but then declined by 40–70% to present levels. Peaks in kelp abundance were the result of increases in habitable forest area and a transition to more productive ocean conditions. Production was surmised from the known relationship between temperature and nutrients in that region. A transition to a sand-dominated system in some areas around 4000–6000 years ago was associated with the demise in kelp distribution and abundance. This reconstruction is based on known, present-day responses of kelp, but, of course, these constitute only one of a suite of potential responses of giant kelp to environmental forcing. How instructive this might be for the rapid environmental changes that are occurring now cannot be determined at this stage.

The complexity of processes relating to global change encompasses most of the topics of this book. These span broad spatial changes in ocean conditions to local-scale effects within coastlines and kelp forests, temporal changes over decades to short-term effects within seasons, and potentially cumulative effects that compound impacts (Pandolfi et al. 2011, Harley et al. 2012, Maxwell et al. 2013). Here, however, we discuss potential changes to giant kelp forests under the rubric of climate-related stressors, and refer back to relevant chapters for more thorough treatment of the effects of these factors on giant kelp.

(p.291) Temperature

Brierley and Kingsford (2009) argued that temperature has the most pervasive climate-related influence on biological function. It seems clear that temperature is a major factor in die-off events of major habitat formers such as corals (e.g., McClanahan et al. 2004, Donner et al. 2005) and kelp (Wernberg et al. 2010), although there are potentially numerous contributing factors such as increases in solar radiation and decreases in nutrients (Hoegh-Guldberg et al. 2007). An increase in extreme thermal events (e.g., Denny et al. 2009) contributes to the frequency of these die-offs. For kelps, diebacks have tended to occur at the limits of biogeographic distributions where physical tolerances may be exceeded (e.g., Wernberg et al. 2010). For example, Helmuth et al. (2006) reported that the poleward contraction of a cold-water kelp, Alaria esculenta, from its southern limit in Britain has reflected a poleward shift in the summer SST isotherm of 16°C, which can be fatal to mature plants of this species (Widdowson 1970). The occasional change in the southern, warm-water limit of Macrocystis has been well described along the west coast of North America. It recedes from its southern limit off Baja California, Mexico, during El Niño years, most likely because of a combination of high temperature, low nutrients, and extreme storm events (see Chapter 5, Edwards and Estes 2006). Similarly, canopy dieback during summer in southern California is correlated with both higher seawater temperatures and nutrient depletion (Jackson 1977, 1987) but may also be related to natural frond senescence (Rodriguez et al. 2013). Because numerous factors act in concert in natural situations, there can be some lability of kelp responses even during extreme events, whereby some local populations of giant kelp survive (e.g., Foster and Schiel 1993, Ladah and Zertuche-González 2004). These refugia may act as source populations for recovery.

One of the longest studies involving giant kelp in which temperature was manipulated involved a small cove (18 m maximum depth, 2 km shoreline, 15 ha area) in central California that was heated by an average of 3.5°C above ambient seawater temperatures for 10 years (Schiel et al. 2004, Steinbeck et al. 2005). The heating was the result of a once-through cooling unit (see Chapter 10) of the Diablo Canyon nuclear power plant, which draws seawater from outside the cove, uses it in a cooling system to recondense freshwater steam, and discharges it into the cove at a maximum rate of 9.5 × 10⁹ L day⁻¹, creating a thermal plume that spreads throughout the cove with the warmest water near the sea surface. As this cove was near the biogeographic boundary of Point Conception between central and southern California, it was hypothesized that cold-water species might decline and warm-water species might dominate, as predicted by other research (Sagarin et al. 1999). Although 150 of the 172 subtidal species recorded changed significantly in abundance with elevated temperatures relative to controls, there was no consistent pattern of change in abundance related to geographic affinities. Most species were cosmopolitan, with broad geographic distributions, and most of these underwent significant changes in abundance. Subtidal understory kelps, especially Pterygophora (p.292) californica and Laminaria setchelli, declined by 82%, from an average of 7 m⁻² to 1 m⁻² (figure 13.4). The cold-water bull kelp Nereocystis luetkeana, which had dominated the surface canopy in the cove, was almost entirely replaced by Macrocystis. This study showed that a rise of around 3.5°C in temperature can trigger wide-ranging changes in communities, but also that changes involve complex biotic interactions. Although control areas underwent large changes over 20 years, they were remarkably similar to initial conditions by the end of the study. The changes to thermally affected areas, however, were great. Not only was there replacement of the large kelps, but the understory kelps also had no major recruitment over the 10 years of temperature increase; sea urchins increased for 3 years and then decreased as foliose understory (mostly red) algae came into dominance. These fluctuating changes in understory algae and invertebrates in the thermally elevated cove have continued since Schiel et al. (2004) completed their study (J. Steinbeck, pers. comm.). It is interesting to note that although the once-dominant kelp Nereocystis recruited annually into the cove, the plants did not grow to reach the canopy, presumably because of thermal stress in the water column. Although it is unwarranted to ascribe a single cause to these persistent changes, it nevertheless is instructive that the major factor that initiated change was the elevation in temperature within the cove. Furthermore, because this was set up as a before–after control-impact study, there was thorough sampling for 10 years in both the cove and outside control areas before the heating treatment affected the cove, thereby providing robust statistical analyses of changes, rather than just correlations between disparate time periods.

One other study is particularly instructive for discussion of temperature effects on giant kelp communities. The studies of Craig Johnson and colleagues in Tasmania, discussed in a previous chapter (see Chapter 8), presented clear mechanistic ways in which environmental effects on a major herbivore contributed to the significant demise of kelp forests. In this case, the onshore movement of a warm water current brought larvae of the sea urchin Centrostephanus rodgersii to the coast of Tasmania. Because urchin larvae had suitable temperatures and development time in this current, they were able to survive the transport from the Australia mainland and establish large populations along coastal Tasmania (Johnson et al. 2011). Their arrival and then population expansion coincided with both a long-term decline in giant kelp and a large increase in fishing of their major predator, the lobster Jasus edwardsii, especially of the large individuals that are capable of eating mature urchins (Ling et al. 2009). It is unlikely anyone could have predicted these types of extensive changes to giant kelp communities because they were the result of complex interactions between kelp dynamics, the behavior of a current, larval life history, and transport of a key grazer that had not previously been important in Tasmanian waters, buildup of urchin populations over time resulting in intense grazing effects, and size-selective overfishing of a key predator. As in the example of the warming of a cove above, temperature played a key role in triggering ecological changes, but the resultant cascades through the kelp communities were dictated by a wide range of other factors.

(p.293)

(p.294) It is clear that even with the potentially quick recovery of kelp after disturbances, short-term changes can have long-term consequences. As has been argued for intertidal assemblages (Schiel 2011), these may be due to sporadic recruitment, changes to benthic substrata such as occupation by corallines and foliose algae, alteration of canopy structure and therefore of the light environment to benthic areas but also, as seen in subtidal kelp assemblages, the continued presence of stressors such as sediments (Connell 2005). Collectively, these impacts may alter the resilience of kelp communities and their competitive and facilitative interactions, as well as make them more vulnerable to a changing climate (Wernberg et al. 2010).

There is also evidence that some kelp-dominated areas are becoming “tropicalized.” Verges et al. (2014) showed that regions with continuous tropical-to-temperate coastlines influenced by Western Boundary Currents, such as Japan, the eastern United States, eastern Australia, and southeastern South Africa, are vulnerable to kelp declines and replacement by barren communities or even corals. They concluded that these phase shifts are largely due to warm-water intrusion and range-shifting tropical herbivorous fishes that severely graze temperate macrophytes, including kelps. These types of effects have not yet been seen in giant kelp forests, but this study offers instructive scenarios on how temperature-mediated complex interactions can drastically alter warm-temperate communities.

Nutrients and Wave Forces

A fundamental relationship in the coastal zone around giant kelp forests is the negative correlation between seawater temperature and nutrients (see Chapter 3). This is particularly well described along southern California where seasonal upwelling brings a rapid drop in seawater temperature and a corresponding increase in upwelled nitrogen (Jackson 1977, McPhee-Shaw et al. 2007, Lucas et al. 2011). The combination of these two factors can have a great effect on giant kelp, and it is the frequency and magnitude of high SST and low nutrients with climate change that are likely to produce the greatest effects, particularly if SSTs get beyond around 26°C (Hernández-Carmona et al. 2001).

As has been shown in strong El Niño periods, Macrocystis is vulnerable to extended periods of low nutrients because of its limited nutrient storage capacity of around 2 weeks (Gerard 1982b). As described in Chapter 3, water column stratification can lead to warm, nutrient-depleted surface water (Lucas et al. 2011) and deterioration of plants (Gerard 1982a). The reversal of these conditions during upwelling episodes with low SST and great pulses of nitrogen sweeping through kelp forests is a principal driver of giant kelp forest growth and the persistence of giant kelp at the equatorward limits of its range in Mexico (Hernández-Carmona et al. 2001) and Chile (review in Vásquez et al. 2006). Any alteration of the periodicity, intensity, or chemical characteristics of upwelling may therefore interact with larger-scale increases in temperature and (p.295) decreases in nutrients to further truncate the biogeographic limits of Macrocystis under future climate change scenarios, especially as they impinge on the physiological tolerances of giant kelp.

Extensions of the biogeographic distribution of Macrocystis in the opposite direction toward the poles could conceivably occur in the northern hemisphere. This would depend not only on the extent of warmer SSTs but also to a considerable extent on the ability of giant kelp to compete with a diverse range of resident northern kelps in the lower light and truncated seasons of the north (Chapter 3). The southern hemisphere poleward distribution of Macrocystis is unlikely to change in future scenarios. The current southern distribution is set by the absence of shallow rocky reefs south of the subantarctic islands (see Chapter 5), and considerable warming of the Antarctic continent as well as considerable dispersal would need to occur for giant kelp to occupy it.

Exposure to waves has large effects on the local-scale distribution and dynamics of Macrocystis (see Chapter 3). In very exposed situations, such as the subantarctic islands and many outer coast sites, Macrocystis is limited to the lee side or protected inlets. Reed et al. (2011) showed that winter storms remove most of the canopy of giant kelp annually in central California, essentially overwhelming many other processes affecting primary production in forests. El Niño storms may provide the best mimics of a changing wave climate, whereby increasing temperatures, decreasing nutrients, and increased storm waves combine to remove large tracts of giant kelp forests, with recovery taking up to several years (Edwards 2004). The wind fields and wave climate are changing across the globe. Young et al. (2011) showed that there has been a significant increase in wave height, particularly of the largest waves (95th and 99th percentiles), since 1985. If this trend persists, then exposed outer coasts will continue to be challenging environments for developing kelp forests, regardless of what other changes occur in temperature increases and nutrient provision.

Ocean Acidification (OA)

Effects of Climate Change on Non-Algal Species

A review of the climate change literature highlights both the potential for massive changes in algal-dominated systems through direct effects on organisms and the importance of interactive effects across the wide spectrum of physical, biological, chemical, and ecological changes. There are numerous examples of changes in the distributions of marine fishes in response to increases in seawater temperature (e.g., Perry et al. 2005, Hsieh et al. 2009, Verges et al. 2014). Temperature effects were examined, for example, in southern California by Holbrook et al. (1997), who assessed changes in fish assemblages at two sites between a cold-water period from 1960 to 1975 and a warm-water period from 1975 to 1995 when average SST increased by 1.0–1.5°C. They found that fish species richness declined by 15–25% and abundances by >90% (p.298) between these two periods. They attributed most of this decline to poor recruitment, which coincided with a decrease in macro-zooplankton biomass in the California Current. Other studies have shown that recruitment success of species at higher trophic levels can depend greatly on synchronization of larvae and pulses of planktonic production. The so-called match–mismatch of marine pelagic communities can be affected by climate warming where responses differ over the seasonal cycle, with a consequent mismatch between trophic levels and different functional groups (Edwards and Richardson 2004).

Of more direct influence on kelp forests are the potential effects of climate change on sea urchins. Dupont et al. (2013) showed that acclimation time affected the responses of the sea urchin Strongylocentrotus droebachiensis to increased seawater CO₂. This urchin species had impaired egg production and increased larval mortality when exposed to high partial pressures of CO₂ (1200 μ‎atm) compared to 400 μ‎atm pCO₂. However, these effects disappeared when urchins had acclimated over 16 months. In a review that included corals, polychaetes, molluscs, and echinoderms, Byrne (2011) found that warming of 4–6°C did not impair fertilization success. Similarly, the great majority of the species she reviewed showed that fertilization was robust to pH levels of 7.4–7.6 (pCO₂ ≥ 1000 ppm). Although there is concern about the effects of OA on the development of calcifying larval stages because of reduced aragonite and calcite saturation in seawater, there is only limited knowledge about these effects. Some echinoderms can increase their metabolic rates and ability to calcify in response to increasing seawater acidity. However, this may come at the cost of muscle wastage (Wood et al. 2008). Abalone larvae may be quite sensitive to climate change, with embryo development being impaired in conditions of increasing temperature and decreasing pH (Byrne 2011). For many taxa, there is also evidence for phenological shifts induced by ocean warming that create mismatches between larval production and their food (Schofield et al. 2010).

Global sea level rise over the next 100 years is likely to be around 0.5 m due primarily to the melting of the Greenland and Antarctic ice sheets and thermal expansion, although this is considerably across the globe (IPCC 2013). This would undoubtedly have great impacts on the coastal environment, but, other than potentially providing more hard substrata for giant kelp, it seems unlikely it will have much of an effect on its distribution. However, as in the distant past (Graham et al. 2010), associated changes in habitats, such as sand inundation, could affect patch structure and abundances.

The overwhelming conclusion of studies on climate change is that “the impact of humans on the biotic systems of the earth is dramatic and is accelerating” (Mooney et al. 2009), that responses of species are quite variable, and that the “proximate causes of species decline relative to resilience remain largely obscure” (Moritz and Agudo 2013). (p.299) It seems likely that in many places, we will be dealing with new mixtures of species, as alien species continue to establish and spread across the globe, perhaps acclimatizing better than native species (Occhipinti-Ambrogi 2007), species shift in response to changing conditions, and ecological relationships and dynamics change as new assemblages are formed. Many processes are, of themselves, quite complex. For example, Roleda, Boyd, et al. (2012) offer a reminder that before OA is discussed there must be “calcifier chemistry lessons.” They highlight that many published OA studies have overlooked the fundamental issue that most calcifying organisms do not rely on carbonate from seawater to calcify, but can use bicarbonate (HCO₃⁻) or metabolically produced CO₂. Mollusc shell carbonate comes from three sources, and calcifying seaweeds also vary in their substrate for calcification. It seems we are still some way from understanding both fundamental processes across taxa and how these might combine to act across communities.

Given the projections in changes to SST over the 21st century, it may well be the case that the distribution of Macrocystis contracts along its present equatorward limits (figure 13.5). SST changes are projected to be in the range of 1–2°C, even with the conservative RCP4.5 models, and above the global average in much of the area occupied by giant kelp (IPCC 2013). Given that giant kelp has limited nutrient storage capacity, and tends to decline at temperatures greater than around 23°C where nutrients are usually below 1 μ‎mol L⁻¹ (Zimmerman and Robertson 1985), there may be considerable shrinking of its range away from the equator. This has been seen, for example, during El Niño years in the Mexico and southern California range of Macrocystis, which has contracted up to several hundred kilometers during these periods (e.g., Edwards and Estes 2006). Even if upwelling remains the same in the future or perhaps increases (Sydeman et al. 2014), the higher temperatures between events may prevent recovery of temperature-and nutrient-stressed populations. Other areas of potential contraction are in the Peru–northern Chile region, the northern range of giant kelp along Argentina, the warm-water area along South Africa, and along the coast of Tasmania where warm-water events are increasing. Any expansion of the range of giant kelp toward the poles is likely to be light limited, because of shorter seasons and the low angle of incidence. Jackson’s (1987) modeling (see Chapter 3) seems fairly convincing that long periods of low or no light in far northern latitudes probably limits giant kelp to where it is now, with competition from other species only making things more difficult for Macrocystis to expand its range in that direction.

More localized, but potentially global, threats to giant kelp forests are likely to be cumulative impacts across multiple changes and stressors, which can occur even where there is some form of management protection (Maxwell et al. 2013). Our best guess is that Macrocystis will continue to survive and thrive along most of its present mid-latitude distribution in the face of these threats because of its broad physiological tolerances, fast growth, and massive reproductive output. On the other hand, the ecological (p.300)

(p.301) surprises and lessons from Tasmania are that overfishing, combined with changes in currents, warming seawater, and the population explosion of a formerly rare sea urchin, can have devastating consequences. Time will tell whether lessons are learned, stressors under our control can be reduced, new and more effective types of management can be enacted, and the degree to which these will make a difference in the face of global climate change. (p.302)