This is the sixth installment of my serialization of a new book chapter on  “Climate Change and Marine Communities” written with Chris Harley and Mike Burrows. It is for a new book “Marine Community Ecology and Conservation” that I’m co-editing with Mark Bertness, Brian Silliman, and Jay Stachowicz.  The book is more or less a followup to the best-selling 2001 edition (which is out of print and worth $100 used and $500 new at Amazon!).  We asked our authors to tell us what has happened over the last 10 years in their assigned subfield.  The chapters are amazing.  And I am truly blown away by how much we’ve discovered since the publication of the first edition!  Many fields have been revolutionized and many-a-paradigm has been overturned.  Cool stuff.  

For more on the basics of Ocean Acidification (OA) go here or download the awesome OA booklet (Mackie et al 2012) here.

Individual- and population-level effects of ocean acidification

Over the last ten years there has been a rapidly growing appreciation and understanding of the threat of OA to marine communities.  The OA literature now includes hundreds of experiments measuring effects of OA on a wide range of marine taxa and variables.  A meta-analysis (Fig. 8) concluded that OA had “a significant negative effect on survival, calcification, growth, development and abundance. Overall, survival and calcification are the responses most affected by acidification, with 27% reductions in both responses, while growth and development are reduced by approximately 11 to 19%, respectively, for conditions roughly representing year 2100 scenarios” (Kroeker et al. 2013).  Additionally, although OA negatively affects many taxa, some such as certain sea stars (Gooding et al. 2009) and some coccolithophores (Iglesias-Rodriguez et al. 2008) benefit from it.  These finding indicate that the impact of OA on organisms, populations, and communities could be large, although not uniform and catastrophic as once assumed.

Figure 8. Effect sizes of experimental reduction of pH by 0.5 units from a meta-analysis (Kroeker et al. 2013) for different taxa and responses.  Plotted values are means are from a weighted, random-effects model with bootstrapped bias-corrected 95% confidence intervals. The number of experiments used to calculate the means is given in parentheses. * denotes a significant difference from zero.

The well-developed mechanistic understanding of how OA affects calcification and the ecological implications of losing calcifying foundation species like oysters and corals has focused attention on this process of the many that OA could influence.   The reduction in carbonate availability caused by OA has the to potential to directly affect organisms that use calcium carbonate to build their shell or skeleton.  Indeed, in numerous laboratory experiments, carbonate ion concentration has been shown to affect rates of calcification and/or skeletal growth in taxa including single-celled crustose coralline algae, crustaceans, corals, molluscs, and echinoderms (Kroeker et al. 2013).

One striking feature of this literature is a high degree of variability in responses to reduced pH among species, higher order taxonomic groupings, life stages, and even among individuals from the same population.  Ries et al. (2009) quantified a variety of functional responses to experimental OA, e.g., positive, negative, linear, hump-shaped, etc., based presumably on how well different organisms are able to protect their shells and regulate local pH at the calcification surface (Ries et al. 2009).  Scleractinian corals, for example, appear to have substantial pH buffering capacity that enables them to maintain calcification despite reduced external reductions (i.e., in ambient seawater) of carbonate ion concentration (McCulloch et al. 2012).  This ability to up-regulate pH at the calcification site by as much as 0.6 pH units, varies among species and does not appear to be present in all taxa (McCulloch et al. 2012).

It has also been shown that corals can use bicarbonate ions (the concentration of which increases with OA) for calcification when the concentration of carbonate ions is reduced (Comeau et al. 2012).  Differential ability to employ these and other adaptive mechanisms for coping with changes in pH likely explain both the smaller than expected effects of OA on coral calcification as well as observed variance in responses to experimental OA among coral species (Edmunds et al. 2012).  In fact, experimental OA has no measurable effect on some corals such as Pacific massive Porities species (Edmunds et al. 2012) yet substantially reduces calcification of others such as Porities rus (Edmunds et al. 2012).  Such differential responses, even within genera, to OA, temperature, and other stressors emphasizes that climate change impacts are more about the changing fortunes of winners and losers (Loya et al. 2001) and altered species composition and community function, than outright community disappearance (see more on altered composition below). It is important to note that although such variability is surprising, it probably shouldn’t have been and it is certainly not unique: organismal and population responses to many other climate change-related factors vary, probably as much as responses to OA.

A meta-analysis of the response of scleractinian corals to experimental acidification (Chan and Connelly 2013) concluded that “under business as usual conditions, declines in coral calcification by end-of-century will be ~22%.”  This outcome is similar to the meta-analysis by Kroeker et al. 2013, but contrasts with Edmunds et al (2012), who found little evidence of a general relationship between seawater pH and coral calcification.  Such variable outcomes and interpretations unfortunately characterize even the synthetic literature on OA experiments (e.g., see Hendriks et al. 2010 versus Kroeker 2010), obscuring simple take home lessons relevant to policy makers and non-specialists.

The experimental OA literature is also equivocal about whether OA could act additively or synergistically with other stressors, particularly rising temperature.  Kroeker et al. (2013) found interactions between these two stressors for some taxa, but no general synergistic effect.  There is also substantial ambiguity over the relative sensitivity of different life stages; calcification and survival of some early larval stages appear more sensitive to OA in some taxa such as corals and molluscs (Doropoulos et al. 2012, Talmage 2012) but not in others such as echinoderms or crustaceans and not in general (Kroeker et al. 2013). Interestingly, larval settlement and recruitment appears to be influenced by OA through multiple independent ecological mechanisms.  For example, Doropoulos et al. (2012) found that OA can influence coral recruitment by reducing the cover of crustose coralline algal species that facilitate settlement and by altering the settlement behavior and substrate choices of coral larvae.  Recent work is also examing “carry-over” effects of larval exposure to OA.  For instance, negative effects of OA on larvae of the Pacific Olympia oyster Ostrea lurida appear to carry over and affect settled juveniles long after pH has returned to normal levels (Hettinger et al. 2012). This work highlights the potential long-lasting effects of early and short-term exposure to lowered pH.

Regarding the overall state of the field, a crucial caveat is that very few acidification experiments measure effects on reproduction, overall fitness, susceptibility to other stressors (i.e., in multi-factorial designs) or otherwise attempt to put the results into an ecologically meaningful context – thus we could very well be over- or underestimating the potential impacts of ocean acidification.  In fact, we know next to nothing about how or whether the documented effects of acidification scale up to population dynamics.  In other words, at this point, we have no idea what the population-level significance of a 25% reduction in calcification would be (Hendriks and Duarte 2010).

We also do not understand the potential costs of mechanisms that enable calcification in low pH environments, although theoretical work suggests such energetic costs are minimal (McCulloch et al. 2012).  And it is important to note that even species that show no response to OA for one response variable, often respond substantially for another.  For instance, although OA does not affect the calcification of massive Porities species it does reduce respiration and photochemical efficiency (Edmunds 2012).  Finally, the near complete absence of field data on both natural patterns of seawater pH variability and changes due to carbon emissions has greatly hampered the field of OA.  New instrumentation (Hofmann et al. 2011) has recently made fine scale and long term field measurements possible and has revealed a huge range of surprisingly variability (Fig. 9).  Such variability of pH across space and time could be used to parameterize lab experiments and as treatments in natural field experiments to better understand responses including acclimatization to OA.  There is clearly a need to fill in numerous gaps in this rapidly progressing field and developing a better understanding of seawater pH in nature is certainly one of them.

The literature Cited for the entire chapter is here as a PDF

This is the fifth installment of my serialization of a new book chapter on  “Climate Change and Marine Communities” written with Chris Harley and Mike Burrows. It is for a new book “Marine Community Ecology and Conservation” that I’m co-editing with Mark Bertness, Brian Silliman, and Jay Stachowicz.  The book is more or less a followup to the best-selling 2001 edition (which is out of print and worth $100 used and $500 new at Amazon!).  We asked our authors to tell us what has happened over the last 10 years in their assigned subfield.  The chapters are amazing.  And I am truly blown away by how much we’ve discovered since the publication of the first edition!  Many fields have been revolutionized and many-a-paradigm has been overturned.  Cool stuff.  

Population-level effects of ocean warming

Temperature also has less-intuitive yet generally predictable effects on several population-level processes (Kordas et al. 2011).  For example, population growth rate often follows the same unimodal response to temperature as enzyme activity and individual growth (Fig. 6).  By altering individual fitness, reproductive output, and generation time, temperature can potentially play a role in how quickly a population may recover from a disturbance and the rate at which a population can adapt to a changing environment.

Temperature also plays an important role in dispersal dynamics. As with processes at other life history stages, the underlying temperature-dependence of enzyme activity strongly influences the developmental rate of marine invertebrate larvae (O’Connor et al. 2007).  This in turn largely determines Pelagic Larval Duration (known as “PLD”)(Palumbi and Pinsky chapter), which influences larval dispersal distance (all else being equal), survival, and even population connectivity (Shanks 2009).  Thus, in warmer water, marine larvae tend to develop more quickly, experience a reduced PLD, and have much shorter dispersal than congeners in colder water (O’Connor et al. 2007).  This can exert important controls on the limits to range distribution (see Sanford, this volume)

There are essentially three ways to measure (or estimate) the effects of temperature on population-level processes: (1) experiments (by far the easiest and most common approach), (2) models parameterized with experimental results, field data or a mechanism-based theory such as MTE, and (3) field data, relating population fluctuations (e.g., declines) with variability in temperature in space and / or time.  The latter approach takes advantage of variable temperature in the field to perform natural experiments that are imperfect due to confounding factors, yet are commonly used since manipulating temperature at relevant spatial scales in situ in the ocean is nearly impossible.  Experiments certainly lead to cleaner results but for many species cannot be used to directly measure population responses.  Testing for effects of anthropogenic ocean warming on populations is even more challenging and generally requires time series data – often several decades of population data, e.g., size structure or density, and water temperature.

An excellent example of the multi-scale influence of temperature from cellular metabolism to individual performance and on up to population and community scale dynamics is the large body of work by Peter J. Edmunds on juvenile corals.  This is also one of the few studies to have combined field data, experiments and models to understand temperature effects on populations.  Pete and his collaborators have shown in laboratory experiments that temperature affects the respiration (Edmunds et al. 2011), growth (Edmunds 2005), and survival (Cumbo et al. 2013) of coral larvae, generally with a parabolic response that peaks around 28º C (Fig 5).  Long-term field studies (Edmunds 2004, 2007) have found ocean warming has exceeded this 28°C threshold and is related to reduced growth and survival of juvenile corals (Fig. 7).

Literature Cited for the entire chapter is here as a PDF

This is the fourth installment of my serialization of a new book chapter on  “Climate Change and Marine Communities” written with Chris Harley and Mike Burrows. It is for a new book “Marine Community Ecology and Conservation” that I’m co-editing with Mark Bertness, Brian Silliman, and Jay Stachowicz.  The book is more or less a followup to the best-selling 2001 edition (which is out of print and worth $100 used and $500 new at Amazon!).  We asked our authors to tell us what has happened over the last 10 years in their assigned subfield.  The chapters are amazing.  And I am truly blown away by how much we’ve discovered since the publication of the first edition!  Many fields have been revolutionized and many-a-paradigm has been overturned.  Cool stuff.  

Individual-level effects of ocean warming: ecophysiology

Temperature directly affects organismal physiology and performance through its control of the rate of biochemical reactions.  Temperature determines diffusion rates and enzyme kinematics, which in turn control the reaction rates for enzyme-driven biological activity.  Typically, warming increases the rates at which enzymes encounter and bind with substrate molecules, thereby increasing the rate of enzyme-catalyzed reactions, however, when temperatures become too high, the enzymes themselves become damaged, and reaction rates drop.  The resulting thermal response curve for enzyme activity is therefore unimodal, with an accelerating rise from low temperatures up to some thermal optimum, and then an increasingly steep drop-off once the optimum temperature has been exceeded.  Due to these fundamental biochemical constraints, metabolic rate has a similar relationship with temperature (Fig 5).  For marine ecotherms like invertebrates and most fishes, warming up to a point means higher metabolism, increased energetic demands, and a variety of other life history changes (Kordas et al. 2011).

Figure 5. Temperature effects on respiration of larvae of the coral Pocillopora damicornis.  From Edmunds et al 2011.   

The temperature dependence of fundamental biological processes like metabolism, photosynthesis, and life span are formally unified under the Metabolic Theory of Ecology (Brown et al. 2004).  MTE is based on biophysical principles and is useful for predicting how biological activity scales with body mass and temperature.  MTE can also be used to predict and understand the primary and secondary effects of natural spatial and temporal variation in temperature as well as those of anthropogenic global warming (O’Connor et al. 2011b).  For example, increasing water temperature from 14°C to 28°C roughly doubles oxygen demand (Fig. 6A) by the green sea urchin Lytechinus semituberculatus, a common herbivore on shallow, subtidal reefs in the Galapagos Islands where temperature swings of that magnitude can occur among seasons and years.  And as predicted by theory (O’Connor et al. 2011a), warming-induced increases in metabolism cause a four fold increase in grazing rate (Fig. 6B), which could affect standing algal biomass and thus a number of ecosystem attributes.  Thus temperature modifies not only the metabolism of individuals but also their ecosystem function (Sanford 1999).

Figure 6. Temperature effects on (A) metabolism of the urchin Lytechinus semituberculatus measured as urchin oxygen consumption, and (B) the rate of consumption of the green alga Ulva sp. by the urchins.  Values are means ± SE.  Based on laboratory mesocosm measurements from Carr and Bruno in review. 

The Literature Cited for the entire chapter is here as a PDF

This is the third installment of my serialization of a new book chapter on  “Climate Change and Marine Communities” written with Chris Harley and Mike Burrows. It is for a new book “Marine Community Ecology and Conservation” that I’m co-editing with Mark Bertness, Brian Silliman, and Jay Stachowicz.  The book is more or less a followup to the best-selling 2001 edition (which is out of print and worth $100 used and $500 new at Amazon!).  We asked our authors to tell us what has happened over the last 10 years in their assigned subfield.  The chapters are amazing.  And I am truly blown away by how much we’ve discovered since the publication of the first edition!  Many fields have been revolutionized and many-a-paradigm has been overturned.  Cool stuff.

Ocean warming

Sunlight naturally warms the upper layers of the ocean.  When the earth’s energy budget is in equilibrium, this heat is eventually returned to the atmosphere though thermal convection (because the atmosphere is generally cooler than the ocean surface).  But an energy imbalance leads to the oceans either gaining or losing heat.  Because it is in contact with the atmosphere, the ocean is also warmed by the supercharged greenhouse effect. In fact, approximately 84% of the excess heat being retained via climate change is going into the oceans (Levitus 2005).

Ocean warming is occurring at all depths (Fig. 2A) and has been since at least since the 1980s (Purkey and Johnson 2010, Levitus et al. 2012).  Heat gained in surface layers is transferred to the deep ocean by vertical circulation.  The global average warming rate (since 1960) for the upper 700 m of the oceans is estimated to be 0.1º C / decade (Casey and Cornillon 2001, Burrows et al. 2011, IPCC 2007), while the deep ocean (700-2000 m) is warming more slowly (Purkey and Johnson 2010).  However, such global averages obscure enormous differences among regions and years.  For example, the warming in the arctic has been much greater (Fig. 2 B) while some small regions such as the upwelling region off the west coast of North America has cooled somewhat (Fig. 2B).  The same patchiness in temperature change has been observed in the deep sea.  For instance, the deep southern ocean appears to be warming relatively quickly, at about 0.03 / decade (Purkey and Johnson 2010).

Figure 2. (A) Changes in the heat content of the land and sea as a result of greenhouse gas emissions. (From Nuccitelli et al. 2012)  (B) Trends in ocean surface temperatures for 1980–2011 (Hadley Centre data set Had1SST 1.1). (C) Cumulative number of weeks with sea surface temperature anomalies >1º C (1985–2005) in the Indo-Pacific. (From Selig et al. 2010) (D) Estimated change in sea surface carbonate ion concentration between the pre-industrial period (1700s) and the 1990s. Global Ocean Data Analysis Project.  (From Hoegh-Guldberg and Bruno 2010)

Variability in the long-term trend is a fundamental characteristic of how greenhouse gas emissions are altering the physical and chemical properties of the ocean.  This patchiness is apparent even at relatively small (by oceanographic standards) spatial scales (Figs. 2 B & C).  For example, the average size of high temperature anomalies that cause coral bleaching are only ~ 50 km² (Selig et al. 2010).  This can lead to striking differences among neighboring reefs in terms of historical temperature patterns, the frequency and severity of anomalies, and the biological responses to them (Berkelmans 2002, Berkelmans et al. 2004).  Furthermore, these fine-grained hot spots do not appear to be stationary as once assumed.  If true, this would pose a huge challenge for coral reef management, much of which is premised on idea that warming extent is spatially constant and predictable.

Ocean acidification

Approximately 25% of the CO₂ emitted since the industrial revolution is now dissolved in seawater.  When CO₂ is absorbed by the ocean it combines with water to form carbonic acid (H₂CO₃), which then dissociates to become bicarbonate (HCO₃^-1) and hydrogen ions (H⁺).  Some of the liberated hydrogen ions further interact with carbonate (CO₃^-2) to create more bicarbonate (Fig. 3). The remainder of the hydrogen ions remain in solution, resulting in a reduction in pH (Orr et al. 2005).  This process is called “ocean acidification” (OA), although, technically, seawater is not expected to become an acid (defined as pH < 7).  The average pH of the surface ocean has fallen from 8.2 in 1750 to 8.1 in 2000 (Doney et al. 2009b); a 30% increase in acidity.  Like ocean warming, the degree to which the concentration of carbonate ions (and pH) has been reduced by acidification varies greatly (Fig. 2D).

Figure 3.  Ocean acidification.  (left) Time series data of atmospheric CO₂, ocean surface pCO₂ and seawater pH (Doney et al. 2009a).  (right) The chemical process of ocean acidification. 

(more…)

This is the second installment of my serialization of a new book chapter on  “Climate Change and Marine Communities” written with Chris Harley and Mike Burrows. It is for a new book “Marine Community Ecology and Conservation” that I’m co-editing with Mark Bertness, Brian Silliman, and Jay Stachowicz.  The book is more or less a followup to the best-selling 2001 edition (which is out of print and worth $100 used and $500 new at Amazon!).  We asked our authors to tell us what has happened over the last 10 years in their assigned subfield.  The chapters are amazing.  And I am truly blown away by how much we’ve discovered since the publication of the first edition!  Many fields have been revolutionized and many-a-paradigm has been overturned.  Cool stuff.

What is climate change?

When most people think of climate change, they think about the greenhouse effect and global warming. The greenhouse effect is caused by gases that heat the atmosphere by “trapping” infrared radiation that would otherwise escape into space.  Roughly half of the solar radiation that reaches the atmosphere is reflected back to space or absorbed by clouds, gases, and particles like soot pollution.  The other half reaches the earth’s surface and is used in photosynthesis, melts ice and evaporates water, and warms the land and lower atmosphere.  This heating emits infrared radiation, some of which is absorbed by greenhouse gas molecules and re-radiated back towards the earth’s surface.  This further warms the land and atmosphere.

Although the greenhouse effect is essential to life on earth (without it the surface temperature would be roughly -18º C), human activities have intensified this natural process by increasing the concentration of greenhouse gases in the atmosphere.   The two primary gases causing anthropogenic climate change are carbon dioxide (CO₂) and methane.  Other natural and anthropogenic greenhouse gases include water vapor, nitrous oxide, ozone, and chlorofluoracarbons (CFCs).

CO₂ is a less potent greenhouse gas than methane (on a per molecule basis), but the concentration of the former is more than 200 times greater (as of February 2013, the concentration of CO₂ was 397 ppm, compared to 1.8 ppm for methane).  Because increased CO₂ concentration accounts for nearly two thirds of anthropogenic warming, it is considered the most important greenhouse gas in terms of emissions mitigation (and catastrophe avoidance).

Atmospheric CO₂ concentration increased by 1.9 ppm per year between 2000 and 2008 (Le Quéré et al. 2012).  This rate increased in each of the last four decades, e.g., up from 1.5 ppm per year during the 1990s.  The more recent increased emissions rate is primarily due to economic growth in China and other developing nations and the global shift towards coal as an energy source.  CO₂ concentration is expected to double relative to the preindustrial baseline of 278 ppm during the latter half of the 21^st century (IPCC 2007).   The resulting global average land surface warming (called the equilibrium climate sensitivity) is “likely to be in the range of 2 to 4.5°C with a best estimate of about 3°C” (quote from IPCC 2007, also see Knutti and Hegerl 2008).  The uncertainty around climate sensitivity is due to potential feedbacks in the Earth’s climate system, some of which are not well understood.

Figure 1. Some important abiotic changes to the oceans caused by greenhouse gas emissions.  Redrawn from (Harley et al. 2006)

The Literature Cited for the entire chapter are here as a PDF

Go here or here to learn more.