Interactions between climate change and ... - Semantic Scholar

Available online at www.sciencedirect.com

Marine Pollution Bulletin 54 (2007) 1845–1856 www.elsevier.com/locate/marpolbul

Review

Interactions between climate change and contaminants Doris Schiedek

a,*

, Brita Sundelin b, James W. Readman c, Robie W. Macdonald

d

a Baltic Sea Research Institute Warnemu¨nde, Seestrasse 15, 18119 Rostock, Germany Department of Applied Environmental Science (ITM), Stockholm University, 106 91 Stockholm, Sweden c Plymouth Marine Laboratory, Prospect Place, The Hoe, Plymouth PL1 3DH, UK Department of Fisheries and Oceans, Institute of Ocean Sciences, P.O. Box 6000, Sidney BC, Canada V8L 4B2 b

d

Abstract There is now general consensus that climate change is a global threat and a challenge for the 21st century. More and more information is available demonstrating how increased temperature may affect aquatic ecosystems and living resources or how increased water levels may impact coastal zones and their management. Many ecosystems are also affected by human releases of contaminants, for example from land based sources or the atmosphere, which also may cause severe effects. So far these two important stresses on ecosystems have mainly been discussed independently. The present paper is intended to increase awareness among scientists, coastal zone managers and decision makers that climate change will affect contaminant exposure and toxic effects and that both forms of stress will impact aquatic ecosystems and biota. Based on examples from different ecosystems, we discuss risks anticipated from contaminants in a rapidly changing environment and the research required to understand and predict how on-going and future climate change may alter risks from chemical pollution. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Climate change; Chemical pollution; Environmental gradients; Environmental stressors; Contaminant risks

Recently, there has been a growing concern that climate change may rapidly and extensively alter global ecosystems with, as yet, unknown consequences for humans. With the release of the fourth IPCC report (IPCC, 2007), it has become clear that the observed and projected changes are partly driven by human activities and that we may have to face more serious changes in the near future. More and more scientific evidence has been accumulated showing that temperature has risen exceptionally during the past 15–20 years, both in air (Tett et al., 1999) and water (Barnett et al., 2005), likely with consequences on the hydrological system (Zhang et al., 2007) and on the cryosphere (ACIA, 2005). The analysis of four of the world’s longest calibrated daily time series has shown that increasing trends in sea surface temperatures in the North and Baltic Seas now exceed those at any time since instru-

*

Corresponding author. Tel.: +49 381 5197205; fax: +49 381 5197211. E-mail address: [email protected] (D. Schiedek).

0025-326X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.marpolbul.2007.09.020

mented measurements began in 1861 and 1880. Temperatures in summer since 1985 have increased at nearly triple the global warming rate that is expected to occur during the 21st century and summer temperatures have risen 2–5 times faster than those in other seasons (MacKenzie and Schiedek, 2007a,b). The ability to adapt to new temperature regimes and variation among aquatic species in their present thermal tolerance limits (Somero, 2005) will be strong determining factors in the success of populations to meet the stress of rising temperature (Po¨rtner and Knust, 2007). In this regard, recent warming already exceeds the ability of some local species to adapt, which consequently may lead to major changes in the structure, function and services of ecosystems. While temperature rise – that is, thermal forcing – is clearly the component of change that has caught much of the public and scientific attention, it is by no means the only sort of change we need to consider. Marine ecosystems are being assaulted by eutrophication (Goolsby, 2000), selective overfishing (Marine Board, 2007; Myers

1846

D. Schiedek et al. / Marine Pollution Bulletin 54 (2007) 1845–1856

and Worm, 2003), destructive practices like bottom trawling and blast fishing (Meysman et al., 2006; Morton and Blackmore, 2001), introduction of invasive species (Lewis et al., 2003; Boudouresque and Verlaque, 2001; Leppa¨koski et al., 2002) and, globally, by acidification due to anthropogenic CO2 loading (Steffen, 2006). Increased intensity and frequency of storms is also projected to be a consequence of temperature rise forced by CO2 emissions, as is sea level rise. Although the IPCC predicts only a fraction of meter in sea-level rise for the coming century, Hansen et al. (2007) warn that a much more rapid rise may be in store for us. More and more information is accumulating on how increased temperature may affect aquatic ecosystems and living resources or how increased water levels may impact coastal zones and their management. There is now consensus in the science community that climate change is a global threat and a challenge for the 21st century. In this Review article our intention is to highlight an additional anthropogenic impact, i.e. contaminants, acting in concert with many of the changes listed above, and to express our concern that we are likely to encounter nasty surprises unless we address contaminants in the context of global variability and change. Human activities release contaminants from land based sources via rivers and the atmosphere to the ocean where they may accumulate and recycle in sediments and organisms. Many of these contaminants are ecologically harmful and thus are also referred to as pollutants or hazardous substances. It is well known that persistent biomagnifying chemicals can accumulate in the marine food web to levels that are toxic to organisms, and where they also present health risks to humans especially those who depend strongly on the sea as a source of food (Van Oostdam et al., 1999). Many other chemicals are less persistent but, nevertheless, cause concern as they can affect, for example, hormone and immune systems (Colborn et al., 1997). Here, we caution that whether or not a contaminant exceeds an effects threshold somewhere in the ocean, and thus becomes a pollutant, depends on the exposure to the contaminant(s) and the susceptibility of the individual or population to that exposure: both exposure and susceptibility may be strongly affected by climate variability and change. In the report to the European Water Directors on Marine and Coastal Dimensions of Climate Change in Europe (2006), it is stated that ‘‘eutrophication is, and will remain, an issue for the coastal zone because both agriculture and in particular urbanisation are the main drivers’’. Pollution (e.g. from toxic chemicals) is considered of lesser importance although the report considers the legacy of past contaminations an important issue. As noted there, ‘‘toxic chemicals are stored in waste dumps, behind dams, in soils and are present in deposited sediments in catchments’’. These natural and man-made repositories are, in principle, subject to erosion and further transport in the direction of the coastal zone. Changes in the hydrological regime (e.g. through climate change) can mobilize these stored contam-

inants. The recent example of the massive flooding of New Orleans during Hurricane Katrina clearly illustrates the risks lurking in poorly-contained contaminant repositories (Rykhus, 2005). In our opinion, contaminants are not merely a ‘‘legacy of the past’’. ‘‘New’’ persistent organic contaminants, such as flame retardants (Stapleton et al., 2005; de Wit, 2002) and perfluorinated compounds (Yamashita et al., 2005), together with what seems to be a continuous array of ‘‘emerging’’ contaminants (such as those used in personal care products, see Muir and Howard, 2007) are being produced and released and we are only just beginning to understand their biological effects. Concerning many of the ‘‘old’’ persistent organic contaminants, as stated in the report mentioned in the above paragraph, these continue to cycle in marine ecosystems (e.g. PCBs or DDT), often with much of the inventory held by the organisms themselves. Indeed, in the case of long-lived top predators, like killer whales (Orcinus orca) that were born before the rise of organochlorine chemicals, present body burdens and exposure to biomagnifying chemicals are a product of lifetime contaminant accumulation (Hickie et al., 2007). A concern is that these archived contaminants may ‘kick in’, for example, when stress on the salmon food supply occurs due to overfishing or temperature rise, and the whales mobilize large quantities of stored fat reserves to get them through the period of stress. Understanding the relationship between orcas and anadromous fish may be even more complex; bioaccumulating contaminants are collected by these fish in the ocean and then focussed into nursery lakes with, as yet, unknown effects on the lake ecosystems in which they hatch and rear (Blais et al., 2007). 1. Contaminants and changes in environmental gradients due to climate change Temperature has long been known to modify the chemistry of a number of chemical pollutants resulting in significant alterations in their toxicities, e.g. for fish. It is also generally accepted that a higher temperature increases the rate of uptake of pollutants via changes in ventilation rate in response to an increased metabolic rate and decrease in oxygen solubility (Kennedy and Walsh, 1997). For a variety of freshwater fish species it has been shown that the upper temperature tolerance limits are decreased in the presence of certain organic chemicals (Cossins and Bowler, 1987; Patra et al., 2007). A study on the common shore crab Carcinus maenas in Stavanger Fjord (Norway) revealed that physiological functions such as heart rate appear to be more vulnerable to contaminant exposure (i.e. copper) at seasonal temperature extremes (Camus et al., 2004). The observed erratic heart rate was interpreted as a response to enhanced copper toxicity at higher temperature (25 °C). In the deposit-feeding Baltic amphipod Monoporeia affinis temperature was found to act synergistically with the fungicide fenarimol resulting in increased numbers of females with dead eggs. A more than four-fold


incidence was recorded compared to exposure to only one of the stressors. Exposure to both stressors also resulted in a negative intrinsic rate of increase (Jacobson et al., in press). For the intertidal clam Mya arenaria from the Saguenay Fjord, Gagne´ et al. (2007) have shown that electron transport in mitochondria is more sensitive to incremental temperature increases when clams are subjected to pollution stress. The examples listed above underscore the importance of considering chemical pollution as but one factor of stress adding to other factors imposed by global change. A proper view of these other factors and how they may interact with contaminants is required before we can correctly assess the risks from contaminants under present and future projected climate changes. In brackish water systems such as the Baltic Sea, salinity is one of the key factors affecting species distribution. The salinity gradient is particularly pronounced in the transition zone in the western Baltic (25–30) towards the brackish Baltic Proper (down to 5). Due to freshwater inflow, salinity is further reduced in the most northern part to more or less freshwater conditions. As a consequence, biodiversity in the Baltic Sea is clearly lower compared to other seas (Bonsdorff, 2006). On the other hand, organisms are subjected to multiple anthropogenic impacts including chemical pollution. Organisms living there, thus, not only have to cope with salinity gradients, but also with the presence of hazardous substances. The concentrations of ‘‘classical’’ contaminants (DDT, PCBs, or heavy metals) have been monitored regularly in the Baltic Sea since the 1970s, and over the past 20–30 years the loads of several hazardous substances have been reduced considerably. Nevertheless, chemical contamination is still of concern (HELCOM, 2002, 2003, 2004): perch show an increasing trend of EROD induction over the last decade (Olsson et al., 2004), sea eagles do not recover in the Gulf of Bothnia (Helander, 2006), and frequency of embryo malformations in amphipods increases (Sundelin et al., 2006). Some novel contaminants as e.g. hexabromocyclododecane and the extremely persistent PFAS (perfluorinated compounds) show disturbing increases during the last decade (Holmstro¨m et al., 2005). As in other regions, numerous potentially toxic substances present in the Baltic Sea ecosystem remain unidentified or poorly characterized. Novel contaminants (such as pharmaceuticals, agrochemicals and PCB replacements) are being released and can be assumed to have a potential effect on biota. Residence time of persistent chemical pollutants is high because of the specific topographical (basins with narrow connections) and hydrographic conditions (thermohaline stratification, salinity and oxygen gradients). Deposition in the sediments and re-mobilisation are common processes in the comparatively shallow Baltic Sea. The impact of contaminants present in the Baltic on a variety of species has been documented in a number of studies (e.g., see the overview by Lehtonen and Schiedek, 2006) and a range of biological responses has been assessed

1847

in bivalves (e.g. blue mussel, Mytilus spp.) and fish (eelpout, flounder, perch) in a recent EU funded study (BEEP). An overview of the main results is given in Lehtonen et al. (2006). Systems like the Baltic, exhibiting strong hydrographic gradients, appear to be particularly prone to hazardous substances under changing environmental conditions. Many of the species living there are of marine origin with a limited capacity to adapt to this brackish water environment. Most of them already live at the edge of their physiological tolerance range in terms of salinity and many of them are boreal (i.e. cold adapted) organisms. Many glacial relics such as Monoporeia affinis, Saduria entomon, and Pallasea qudrispinosa as well as the marine Pontoporeia femorata are stenotherm coldwater species. M. affinis, for instance, has an upper limit of 8-9° above which reproduction is seriously disturbed. Therefore, small temperature increases will influence these species. According to a recent comparative assessment, the BACC report (BACC Author Team, 2007), the mean sea surface temperature of the Baltic Sea is projected to increase by about 2 °C–4 °C by the end of the 21st century, depending on the model and scenario used. According to the model results of Meier (2002) the maximum warming will occur during autumn at 20-50 m depth due to higher wind speeds. Because there is a more or less permanent oxygen deficiency at depths below 50 m, this zone of expected maximum warming is also the preferred living depth of Monoporeia affinis, a key Baltic species. Furthermore, the projected temperature maximum, which occurs in the autumn, coincides with the temperature-sensitive gonad maturation period (see Eriksson-Wiklund and Sundelin, 2001; Eriksson Wiklund and Sundelin, 2004) making it highly likely that M. affinis will be affected by such warming of the Baltic Sea. Thus, M. affinis could find itself trapped in a dwindling habitat between too warm shallow water and too low oxygen-containing deep water. Temperature increase in the Baltic Sea will influence the length of the ice season, which would decrease by 1–2 months in the northern parts (e.g. Bothnian Bay, Gulf of Finland) and by 2–3 months in the central parts. In association with the warming, changes in precipitation patterns are expected both geographically and seasonally. In some of the regional scenario simulations the average salinity of the Baltic Sea is projected to decrease (Raïsa¨nen et al., 2004; Gustafsson, 2004; Meier, 2006; BACC Report, 2007). A systematic change towards warmer temperatures and reduced salinity will have an impact on the distribution and acclimation capacity of native biota. The general fitness of many species will clearly be reduced and thus also their capability to cope with contaminants. Considering the specific hydrographic conditions in the Baltic, which have the consequences of restricting marine species to the southern part and glacial relicts sensitive to higher temperatures to the northern part, the following scenario is likely: the range of marine species such as bivalves and polychaetes will be forced southwards and coldwater species will be diminished to be replaced by invading species. The ability

1848


to cope with the additional stress from pollution under a changing climate regime is also likely to differ among species. We have already seen a dramatic increase in invading polychaetes, i.e. Marenzelleria sp., less sensitive to oxygen deficiency, temperature increase and contaminant exposure. In addition, this species was found to be more efficient than the co-occurring Monoporeia affinis in bioturbating sediments resulting in increased release of sediment bound contaminants (Hedman, 2006). Climate related changes in fish communities (MacKenzie et al., 2007) might also result in a modified transfer of contaminants within the food web. We probably also have to be prepared to deal with species common in southern waters (e.g. anchovies, sardines or red mullet) and presently becoming more and more abundant in the North Sea (Beare et al., 2004; Perry et al., 2005) and the Western Baltic (Enghoff et al., 2007). Generally, distributions of both exploited and non-exploited North Sea fishes have responded markedly to recent increases in sea temperature (Perry et al., 2005). For species with northerly or southerly range margins in the North Sea, half have shown boundary shifts with warming, and all but one shifted northwards. Besides latitudinal range expansions of species or extinction of native species due to changing temperature conditions, the impact of non indigenous species (NIS) or invasive species are likely to increase. Climatic driven changes may affect both local dispersal mechanisms due to the alteration of current patterns, and competitive interactions between NIS and native species because of new thermal optima and/or different carbonate chemistry (OcchipintiAmbrogi, 2007). Presently we do not know how organisms might respond to contaminant exposure at their future climate change driven boundaries of distribution. The projected changes in abiotic factors are also likely to influence processes involved in the metabolism of toxic substances: higher temperatures result in increased turnover rates or generally higher metabolic rates. Baltic blue mussels clearly respond to changes in temperature or salinity by, for example, altering metabolic rates and enzyme activities (Pfeifer et al., 2005). Higher temperatures and/ or lower salinity will affect the species’ ability to cope with toxic substances (Cherkasov et al., 2006; Lanning et al., 2006) and the various physiological regulation processes involved in the detoxification of hazardous substances. A broad overview concerning the interactions between various classes of chemicals and different environmental factors (e.g. salinity, temperature or nutritional state) in aquatic organisms is given by Heugens et al. (2001). The analysis of a greater number of field and laboratory studies revealed that increasing temperature raised toxicity, and decreasing salinity resulted in increased metal toxicity but reduced toxicity of organophosphates. Moreover, organisms living under conditions close to their tolerance limits appeared to be more vulnerable to additional chemical stress. Hypoxia is another stress that coastal and estuarine organisms have to cope with. In the southern part of the Baltic Sea its occurrence is often associated with high

temperatures in the summer, whereas in the Baltic Proper, there is more or less permanent hypoxia on sublittoral bottoms below 50-60 m. Above 20-30 m water depth, hypoxia is more frequent during early autumn (August–October), the same period when there is a temperature increase. In many coastal areas or estuaries, hypoxia is a common feature during low tide and species living there have successfully adapted to cope with low oxygen conditions, e.g. by switching to anaerobic energy production (Zebe and Schiedek, 1996). However, with increasing temperature many species are likely to reach their thermal acclimation limits (Po¨rtner, 2001). Of the invertebrates, crustaceans are most susceptible to hypoxia and organic matter loading (Gray et al., 2002; Lenihan et al., 2003). In the Baltic Proper a decline of the amphipod Monoporeia affinis population has been observed since the 1970s and today populations are just 1/10 of the abundance 25 years ago. M. affinis is extremely sensitive to oxygen deficiency during oogenesis, which occurs from August to November. As mentioned before, this period is characterised by impoverished oxygen conditions resulting in increased frequency of dead M. affinis broods (Eriksson-Wiklund and Sundelin, 2001; Eriksson Wiklund and Sundelin, 2004). Studies on the clam Macoma balthica from its southern to northern distribution limit have shown that the species is able to cope with pollution caused, for example, by trace metals and low oxygen conditions (Hummel et al., 2000). However, there is some indication that with the increasing temperatures during recent years, this species is losing its ability to survive at the uppermost limits of its southern distribution where pollution is also high (Jansen et al., 2007). Generally, areas affected by anoxia and hypoxia in aquatic ecosystems throughout the world have expanded, and have become more frequent in recent decades (Diaz and Rosenberg, 1995, 2001; Wu, 1999; Gray et al., 2002; Karlson et al., 2002; Levin, 2003; Goolsby, 2000). Global warming will increase the problem due to increased precipitation and temperature bringing nutrient-rich, fresh and relatively warm water to sensitive areas where water stratification will increase (Harley et al., 2006). Hypoxia caused by eutrophication and organic pollution is considered to be amongst the most pressing water pollution problems in the world (GESAMP, 1990; Goldberg, 1995; Wu et al., 2002). Hypoxia is often caused by increased loads of organic matter and various organism groups respond differently to increased TOC and sulfide. For benthic groups that are favoured by high TOC and less sensitive to hypoxia and sulfide (e.g., annelids and echinoderms), the effects of contaminants will be reduced while for arthropods, which are more sensitive to TOC and hypoxia, the effects of contaminants will be increased (Lenihan et al., 2003). Furthermore, there are indications of reduced feeding rates during hypoxic events (Ripley and Foran, 2007 and references herein) and starved organisms are more sensitive to contaminants. Thus, we can expect changes in benthic community composition due to hypoxia combined with contaminant stress.


Factors such as hypoxia and anoxia may act as co-varying stressors with contaminants, causing damage to organisms that are more dramatic than the effects of any one of these stressors by itself. Contaminant concentrations in sediment are often orders of magnitudes higher than in the overlying water (Jonsson, 2000). In most mass balance calculations of contaminants in the Baltic Sea, sediments are generally considered to be a sink (e.g. Wania et al., 2001). Physical processes, such as resuspension, storm events, dredging and fish trawling, as well as biological processes, such as bioturbation by benthic organisms, may disrupt the equilibrium partitioning between water and sediment. These disruptive processes can, in turn, cause an increased net release of sequestered contaminants from sediments to overlying waters. Thus, benthic organisms are often exposed to the combined stress of sediment-bound contaminants and hypoxia due to their niche habitat in the sediments where hypoxia is more frequent occurring than in the water phase. Manganese (MnO2), one of the most abundant naturally-occurring metals, is widely concentrated in oxic marine surface sediments through redox processes (Aller, 1994; Gobeil et al., 1997). Hypoxic episodes (