Impacts of climate change on fish, fisheries and aquaculture

Authors


S. R. Dye, MCCIP Secretariat, Cefas, Lowestoft, NR33 0HT, UK. Email: stephen.dye@cefas.co.uk

INTRODUCTION

At the 2010 British–Irish Council meeting of Environmental Ministers, the Marine Climate Change Impacts Partnership (MCCIP, www.mccip.org.uk) launched its fourth Report Card, summarizing the current ‘state of the science’ on marine climate change impacts for decision makers (MCCIP, 2010). In discussion, the Environment Ministers (representing the UK government, the devolved administrations of Wales, Scotland and Northern Ireland, the Republic of Ireland, Isle of Man and the States of Jersey and Guernsey) agreed that a future MCCIP report card focusing on fish, fisheries and aquaculture would be extremely beneficial to both the policy and wider stakeholder community. These are issues that have generated some controversy in the media and elsewhere and have significant implications for the future management of UK seas.

The information in previous MCCIP Report Cards has been the product of peer-reviewed papers written by recognized experts in their various fields which were made available on the MCCIP web site. The publication of these three Review articles in Aquatic Conservation represents the first time information provided to MCCIP has been made available directly via the scientific literature rather than solely through the MCCIP website. The increased scientific rigour provided by the more formal review of these papers is particularly important given the contentious nature of the topics covered, not just among scientists, but also amongst policy-makers, industry and conservationists.

SETTING THE SCENE

This collection consists of three individual review papers. The first covers the ‘impacts of climate change on fish’, looking at what changes in fish and shellfish species and communities have been observed around the UK and Ireland and what could happen in the future. The second paper considers the implications of climate-driven impacts on ‘wild’ fisheries and shellfisheries for the economies of the UK and Ireland, and the livelihoods of people dependent on these industries. The final paper focuses on what climate change could mean for the future of the aquaculture industry and the ‘farmed’ species of fish and shellfish.

While acknowledging that the scientific evidence base is, in parts, fragmentary and that uncertainty exists about future climate change, the authors have been explicitly asked to also address the ‘so-what’ question that decision-makers are ultimately challenged with, i.e. what are the implications of the evidence being provided? This not only includes the economic and social implications for local fishing communities but also wider considerations such as the health of marine ecosystems, spread of non-native species and prevalence of disease (and consequent threats to human health). In order to provide informed responses to this ‘so-what’ question, decision-makers need the best possible advice from the scientific community, based on our current understanding of climate change impacts.

Many of the issues raised cut across the three review papers, which are intended to be complimentary and cross-reference each other where appropriate although the highly interconnected nature of these topics means there will be some overlap. It is also important to note that there are a variety of approaches when analysing and interpreting data and there is not always agreement on the most suitable methods. For example, many scientists investigating the influence of climate change on distributions of species and populations use ‘bioclimate envelope models’ while others feel the limitations of this method mean that other tools and explanations are required. These differences in approach are apparent in the reviews as they reflect the efforts of a large number of scientists.

Finally, because underlying ‘physical’ changes in the ocean are broadly relevant to all three topics, we set out one climate narrative below to act as the foundation for all three reviews (although some more detailed information on climate drivers is provided within the reviews where necessary). This narrative largely draws upon evidence presented in previous MCCIP scientific reviews and provides a consistent framework in which the three review papers should be considered.

CLIMATE NARRATIVE – OBSERVED AND PROJECTED CHANGES IN OCEAN CLIMATE

Sea temperature

Sea temperature is one of the key factors affecting the physiology and ecology of marine fish and shellfish (Pörtner and Farrell, 2008; Pörtner and Peck, 2010). The underlying ocean warming trend associated with anthropogenic climate change is superimposed upon natural variability on interannual to multi-decadal timescales and between regions. The result is that, even with a long-term ‘anthropogenic’ warming trend, in different regions there will be some decades in the future that will show particularly strong warming while others will exhibit little change or even cooling (Hawkins, 2011).

Relative to the underlying warming trend during the 20th century the surface waters averaged over the north Atlantic were cool in the period between 1900 and 1930, warm from 1930 to 1960, cool between the late 1960s and 1990 and then warm from 1990 to present (Holliday et al., 2011). The warming observed in the last three decades has been particularly strong in parts of the north-east Atlantic, with the sea surface around the UK and Ireland warming at rates up to six times greater than the global average (Nolan et al., 2009; Hughes et al., 2010b). Around the UK and Ireland, the warming trend is strongest in the North Sea and English Channel, with a rate between +0.6 and +0.8 °C per decade (Hughes et al., 2010b) since the mid-1980s. Over the same period the sea surface temperature trend for seas to the west and north of Ireland and Scotland has also been upward but at a slower rate of +0.2–0.4 °C per decade (Hughes et al., 2010b). It remains difficult to fully distinguish the natural variations in temperature from those due to anthropogenic emissions of carbon dioxide (CO2). In one study, Cannaby and Hüsrevoğlu (2009) analysed a nearshore time-series at Malin Head, in the north of Ireland, and estimated that about half of the warming trend over 1958–2006 at this location was contributed by anthropogenic global warming.

UKCP09 (Lowe et al., 2009) projected that over the coming century, sea-surface temperature will continue to rise all around the UK and Ireland. Their model, using a medium emissions scenario, projects annual temperature increases of ~1.5–2.5 °C in open ocean, shelf edge, and northern North Sea regions by 2070–2098 (relative to the 1961–1990 average). Larger increases of ~2.5–4 °C are projected for the Celtic, Irish and southern North Sea over the same period. Warming is projected to be most pronounced in the autumn, with waters off the south-east coast seeing the largest sea temperature increases (Lowe et al., 2009). These projections present just one physically plausible future, derived from a single model, and therefore can not provide a full assessment of uncertainty. Since fishing fleets from the UK and Ireland also operate outside European Union waters, temperature and ocean-atmospheric changes in other regions of the world ocean are also covered in the three review papers where required.

Other physical changes in ocean climate

Other changes in ocean conditions that may affect fish, fisheries and aquaculture, include changes in the frequency and intensity of severe weather events, stratification, and salinity although there is lower confidence in linking variation in these factors to climate change.

The incidence of severe winds and wave heights in waters to the west and north of the UK and Ireland increased over the second half of the 20th century, although this could be part of long-term natural cycles operating over many decades (Alexander et al., 2005; Wolf and Woolf, 2006; Nolan et al., 2009). Confidence in future wind and storm projections is very low, with some models suggesting that Europe might experience fewer major storms and others suggesting an increase in the future (Woolf and Wolf, 2010).

Climate change could affect wave heights by changing the intensity of storms, or their tracks, but as there is very low confidence in projections of storms under climate scenarios there is a correspondingly wide range in potential changes in wave height. Model projections (Lowe et al., 2009) demonstrate this wide range in the potential changes for both seasonal means and annual extremes (changes in the winter mean wave height of between 35 cm decrease and 5 cm increase, and changes of the annual maxima of −1.5 m to +1 m).

Trends in salinity are subject to regional differences and decadal scale variability. During the first decade of the 21st century, sea surface salinity has been high in the Atlantic waters to the west and north of the UK and Ireland but this is less evident in the shelf seas (Nolan et al., 2009; UKMMAS, 2010; Hughes et al., 2010a). Atlantic waters generally also had high salinities before the mid-1970s, when they freshened strongly and remained relatively low in the 1980s and 1990s. Future changes in salinity of the waters around the UK and Ireland will be determined by the balance of saline oceanic (Atlantic circulation) and fresh water contributions (rivers and rainfall) and how these vary as the climate changes. There is currently low confidence in projections of future changes in salinity.

Other changes in large-scale ocean processes and seasonal cycles of ocean stratification in the shelf seas may be related to climate change. For example, there is some evidence that temperature stratification over some parts of the north-western European shelf is beginning slightly earlier in the year (Sharples et al., 2006).

Acidification and deoxygenation

In addition to changes in ocean temperature, the chemistry of sea water is also affected by greenhouse gases, most notably through ocean acidification and deoxygenation. Ocean acidification is recognized as an important, but not well understood, threat to marine ecosystems and associated fisheries resources. As a direct consequence of increasing levels of atmospheric CO2, sea water is becoming more acidic, with the pH of surface water having decreased by 0.1 units since 1750, representing a 30% increase in hydrogen ion (H+) concentration. It is almost certain that the ocean will continue to get more acidic, possibly by 150%, given projected future increases in atmospheric CO2 (Orr et al., 2005; Turley and Boot, 2011).

As a consequence of global warming, it is also thought that the ocean will become less oxygenated because of lower solubility of oxygen in warmer water and an increase in upper ocean stratification (Greenwood et al., 2010; UKMMAS, 2010). Oxygen is the fundamental molecule for the survival of marine organisms and plays a direct role in nutrient cycling in the ocean (Keeling et al., 2010).

The full scale and extent of the implications of ocean acidification and deoxygenation are difficult to quantify and describe at this time. It is nevertheless important to take them into account as best as possible when considering the medium- to long-term future of seas around the UK and Ireland (Greenwood et al., 2010; UKMMAS, 2010).

Sea level

Sea level, adjusted for land movement, rose around the UK coast at a rate of about 1.4 mm year−1 during the 20th century (Woodworth et al., 2009; UKMMAS, 2010). Some shorter periods within the century experienced rates of sea-level rise that were faster or slower than this century-long trend. For example, the sea level around the UK rose by 3–4 mm year-1 in the 1990s (UKMMAS, 2010) and has risen around Ireland by around 3.5 mm year-1 since 1994 (Dunne et al., 2008). UKCP09 projections of sea-level rise around the UK coast between 1990 and 2095 are based on projections of sea-level rise given by the IPCC Fourth Assessment (IPCC, 2007) and on further analysis of the multi-model ensemble of projections they used. UKCP09 gives a range between 12 and 76 cm (5th percentile of model projections for lowest emission scenario to 95th percentile for highest emission scenario used, excluding vertical land movement) (Lowe et al., 2009). These projections do not include any sea-level rise from any future glaciologically accelerated flow of the ice sheets, as there is limited current understanding and modelling capability of this process. Sea-level rise could thus be higher than the above projections. UKCP09 give an upper bound for the UK, thought to be highly unlikely but physically tenable, of 1.9 m.

The rise in sea level actually experienced at any location on the coast (known as relative sea-level rise; RSLR) also depends upon changes in the level of the land itself. The level of the land in the UK and Ireland is changing through local processes such as sediment compaction or groundwater extraction, and through regional land movements, such as those resulting from millennial scale adjustment to the removal of ice since the end of the last glacial period (‘isostatic readjustment’). Previously glaciated areas are rising, while areas that were further from the centre of mass of the ice are subsiding. Central and western Scotland is rising at a maximum rate of ~1.4–1.6 mm year-1 and north-eastern Ireland is also rising. Areas that are subsiding include the south-west of Ireland, the Shetland Islands (by more than 0.5 mm year-1), and south-west England at ~0.6–1.2 mm year-1, (Shennan and Horton, 2002; Shennan et al., 2009; UKMMAS, 2010). For Scottish coasts Rennie and Hansom (2011) recently highlighted that directly measured rates of uplift over the last two decades (Bingley et al., 2007; Bradley et al., 2009) are lower than those based on the last 1000 years (Shennan et al., 2009). Their analysis of tide-gauge measurements between 1992 and 2007 suggest that the compensating effect of land uplift may be less significant in reducing the rate of RSLR on Scottish coasts than previously assumed and is near zero in some places (Rennie and Hansom, 2011).

The local impact of sea-level rise may lead to significant coastal habitat loss, which is of particular relevance when considering the suitability of aquaculture sites at or near the shoreline. Parts of the Atlantic coast of Ireland that are generally more storm affected are currently retreating at a rate of 0.5–1.0 m year-1, with an expectation that this rate will increase as sea levels rise (Cooper and Pilkey, 2004; Lozano et al., 2004; Nolan et al., 2009). The frequency of extreme sea levels that can cause flooding at the coast will change in the future due to both RSLR and any changes to storms and their consequent surge. For most of the coasts around Ireland and the UK, UKCP09 standard projections suggest that RSLR will be the more important factor (Lowe et al., 2009).

Precipitation, freshwater input and pollution

Since river discharges respond to changes in precipitation regimes, climate change may ultimately affect the flow and quality of freshwater inputs to marine and coastal habitats (Osborn et al., 2000; Roessig et al., 2004; Graham and Harrod, 2009).

Average annual precipitation over England and Wales has not significantly changed since 1766, but its annual distribution and intensity has shifted (Jenkins et al., 2008). In the winter, daily rainfall intensity has increased (Osborn et al., 2000; Maraun et al., 2008), precipitation events becoming heavier across the UK, particularly in north and west Scotland. The proportion of summer rain that fell as heavy events declined between 1961 and 1995, though this trend has now reversed (Maraun et al., 2008). Evidence presented in Jenkins et al. (2008) also supports a shift in annual distribution of rainfall, indicating that since the late 19th century, England and Wales have seen a decrease in precipitation during July and August with a concomitant increase in winter.

The UKCP09 central estimate projection (Jenkins et al., 2009) for precipitation shows a reduction in mean summer rainfall up to 10–20% across most of the northern UK, and greater reductions further south, up to 40% in the south-west, by 2070–2099 compared with 1961–1990 values under a medium CO2 emissions scenario. However, the occurrence of extreme precipitation events in winter, spring, and autumn are projected to increase by up to 30% in the UK (Fowler and Ekstrom, 2009). Across Ireland, models project drier summers but increases in autumn and winter precipitation of 15–20% with a suggestion of greater increases of 20–25% in the north (Dunne et al., 2008). Changes to stratification of coastal waters caused by freshwater inputs from the land could change in the future but current models cannot predict this with any certainty (Lowe et al., 2009; Sharples et al., 2010).

ACKNOWLEDGEMENTS

The climate narrative provided here benefited greatly from the information originally collated by review authors, particularly William Cheung, Suzanne E. Grenfell, and Anouska F. Mendzil. The text was reviewed by Sarah Hughes, Craig Wallace and Jonathan Tinker. The development of the review papers and the publication of the summary report card for policy-makers based on these reviews was supported by the Marine Climate Change Impacts Partnership (MCCIP).

Ancillary