The assessment of community response to changes in temperature shows geographical variations across the UK, with some areas less resilient to climate change. Community Temperature Index values for plants are much lower than those for animals and therefore are more vulnerable to increases in surface temperature in general. This will exacerbate the effect of additional anthropogenic pressures on intertidal rocky habitats.

Background

UK target on Intertidal Community Temperature Index

This indicator has been developed to assess the ‘maintain native intertidal biodiversity’ target which is set in the UK Marine Strategy Part One (HM Government, 2012). However, a full assessment against this target is not possible at this stage.

Key pressures and impacts

Climate change is one of the key pressures driving biodiversity loss and ecosystem service changes. The impacts on marine organisms can be either direct by impacting their physiology, or indirect through changes in the distribution of prey/predators. Intertidal and shallow habitats, in particular, are likely to be significantly impacted through sea level rise and temperature change.

Measures taken to address the impacts

Measures to protect benthic habitats are set out in the UK Marine Strategy Part Three (HM Government, 2015). These include those taken under the Habitats Directive (European Council, 1992), Water Framework Directive River Basin Management Plans, Urban Wastewater Treatment Directive (European Council, 1991a), Nitrates Directive (European Council 1991b), OSPAR measures on species and habitats, Marine Spatial Planning, land management schemes, catchment sensitive farming, and European Marine Site management schemes.

Monitoring, assessment and regional co-operation

Areas that have been assessed

This indicator is only applicable to intertidal rocky areas. A full assessment of the Marine Strategy Framework Directive sub-regions was not possible at this stage due to early stages of the indicator development.

Monitoring and assessment methods

The Community Temperature Index is a measure of the status of a community regarding its species composition of cold- and warm-water species. It is quantitative, easily applied, and gives a direct measurement of the response to climate and climate change across all the species in a community. The index builds on a solid foundation of understanding of changes in species abundance and presence in relation to patterns of temperature across geographical scales and over time. This information is essential to understand how anthropogenic pressure exacerbates effects from climate change. Community Temperature Index values have been calculated for 1,805 rocky shore surveys around the UK since 2002.

Assessment thresholds

The Community Temperature Index is assessed in relation to the 1.5oC limit set by the 2016 Paris Agreement of the United Nations Framework Convention on Climate Change looking at past and future Community Temperature Index changes in relation to temperature.

Regional co-operation

The indicator has not been used for the OSPAR Intermediate Assessments (OSPAR Commission, 2017) but may be used for future regional analysis.

Assessment method

Indicator metric

Community Temperature Index is a measure of the status of a community in terms of its species composition of cold- and warm-water species (Devictor and others, 2008), and gives a direct measurement of the response to climate and climate change across all species in a community. The index builds on a solid foundation of understanding of changes in species abundance and presence in relation to patterns of temperature spatially and temporally. Species distributions for the most frequently recorded UK rocky intertidal species have been compiled from literature and online sources.

The first step in producing the Community Temperature Index is the collation of information on the distribution of each species in the community to obtain the Species Temperature Index. This is the thermal midpoint of the species range: the median or 50th percentile of within-range temperatures. Species Temperature Index is usually calculated by identifying reliable sources of information on distributions of the component species, for example, the distribution of Poli's stellate barnacle (Chthamalus stellatus) in relation to average sea surface temperature is shown in Figure 1. Without a comprehensive pre-existing set of global distribution maps, the approach adopted for UK intertidal species was to examine the literature on the biogeography of each species one by one and create Geographic Information System polygon shapefiles that encompass areas where the species is known to occur. The Species Temperature Index effectively separates cold-water from warm-water species, and values range from 5.0⁰C for the Boreal Arctic barnacle (Balanus crenatus), a cold-water species, to 19.0⁰C for the purple sea urchin (Paracentrotus lividus), a warm-water species.

Figure 1. The global distribution of Poli's stellate barnacle (Chthamalus stellatus) shown as an organge polygon overlay on sea surface temperature contours of annual averages between 1982 and 2011 are from NOAA OISST V2.

Community Temperature Index calculations (Figure 2) use a weighting factor (STI). These weights describe optionally the presence (0/1) of a species, its abundance (in units comparable across many species), or occasionally the incidence of a species across many samples from a geographical area over a specified period.

Figure 2. Equation for calculating the Community Temperature Index (CTI). Where STIi is the Species Temperature Index for species i, and wi  is the weighting factor for species i.

The Species Temperature Index as a description of thermal affinities is powerful but does not capture information about the thermal range of the species, with species having closely similar thermal midpoints often having widely different upper and lower limits of within-range temperatures. A useful further index, therefore, is the Species Temperature Range. This is defined here by the difference between the 10th and 90th percentiles of within-range temperature and gives a measure of the range of temperatures occupied that is more reliable but does not represent the entire range of temperatures for the geographical range of the species.

Data used and quality assurance

Survey sites used in the Marine Biodiversity and Climate Change project (knonw as ‘MarClim’) are located in areas of extensive, exposed intertidal rocky reef or artificial hard coastal structures or defences, away from areas of coastline heavily developed or utilised for social or economic purposes, and avoiding riverine and estuarine outputs (see Figure 3). Rocky shore species (47 species of ectothermic invertebrates and 30 species of macroalgae) assessed during MarClim surveys, since 2002, are recorded using a 7-point categorical abundance scale, (known as the SACFOR scale) that gives scores ranging from ‘Not Seen’, through ‘Superabundant’, ‘Abundant’, ‘Common’, ‘Frequent’, ‘Occasional’, ‘Rare’, with 'Superabundant' and ‘Extremely abundant’ at the top of the scale. For the Community Temperature Index calculation, these categories are replaced with the integer numbers 0 to 8, allowing the combined effect of changes in abundance of different kinds of species to be assessed.

Figure 3. Sites surveyed by the Marine Biodiversity and Climate Change programme since 2002. Coloured circles indicate the survey leaders: Marine Biological Association in Plymouth: pink and light green; Scottish Association for Marine Science in Oban: blue, pale yellow and red; National University of Ireland in Cork and Galway: bright green. Grey diamonds show sites surveyed mostly annually from 2002 to 2016.

Knowledge of the distribution of Community Temperature Index anomalies over time from long-term monitoring sites allows determination of the minimum number of survey sites required to detect a given change in the Community Temperature Index value. This represents a given shift in community thermal trait composition with a specified level of detectability and statistical significance (in this case P=0.05). Detectability is otherwise known as the power of the test (1 - β, where β is Type II error: the probability of erroneously accepting an otherwise false null hypothesis of no change). This minimum number of sites for a given change in temperature and level of significance critically depends on the standard deviation of the anomalies. For the sites in south-west Britain from 2002 to 2014, the standard deviation of anomalies was 0.409⁰C, giving a range of numbers of paired surveys needed to detect a change in the Community

Temperature Index of a given magnitude:  

  • 22 sites are required to detect a change of 0.3⁰C with a power of 0.9
  • for a power of 0.8.

Uncertainty in Community Temperature Index estimates is generated by the randomness of species encountered at each survey site. This can be distinguished from uncertainty produced by the surveys themselves. For Marine Biodiversity and Climate Change surveys, additional uncertainty is generated by the variable application of survey methods, differences in effort across shore levels and among taxa, mistakes in identification and varying levels of taxonomic skill across surveyors. Where cross-calibration among surveyors has been done (Burrows and others, 2017), levels of similarity for abundance estimates and species richness are reassuringly high: 83% of SACFOR abundance estimates are within one category and 92% within two categories.

For single surveys, a useful estimate of uncertainty is derived from the standard deviation of the Species Temperature Index values divided by the square root of the number of species in the community: the standard error of Community Temperature Index. 95% confidence intervals for Community Temperature Index can be calculated by multiplying the standard error by the appropriate Student’s t value (for p=0.05 and n-1 where n is the number of species in the community).

A full description of the method can be found in Burrows and others (in press).

Results

Findings from the 2012 UK Initial Assessment

This indicator was not considered as part of the Initial Assessment (HM Government, 2012).

Latest findings

Status assessment

The approach for the development of the Community Temperature Index accounts for prevailing physiographic, geographic, and climatic conditions as required under the Marine Strategy Framework Directive Descriptor on Biodiversity (European Commission, 2008), and aims to assess the condition of the typical species and communities for intertidal rocky habitats in UK waters within the context of climate change. Thus, the indicator target should reflect that species are present as expected for the habitat type and the prevailing physical conditions.

The Community Temperature Index has been developed with sampling data across UK rocky shore communities (Figure 3) as a practical measure of community response to changes in temperature along geographical gradients and over time. Communities of animals on rocky shores are comprised of species with distributions centred in warmer areas. Plant communities are generally comprised of cold-water species, therefore making a negative response to climate change more likely for plants. The Community Temperature Index for animals is lower on wave-sheltered shores than on wave-exposed shores, likely due to the reduction of temperatures underneath macroalgal canopies. The patterns of Community Temperature Index values generally follow isotherms around the UK but with local variations and potential for biogeographic discontinuities (Figure 4). For example, in Scotland, enclosed areas such as the Clyde Sea and the Firth of Forth have lower Community Temperature Index’s for animals than nearby open coasts. Community Temperature Index for plants (Figure 4b), in contrast, follows more of a North-South gradient with a greater dominance of warm-water species in the English Channel than in southern Ireland. The negative thermal bias of macroalgae across most of the UK (except for eastern Scotland and north-east England) may indicate communities are more vulnerable to the effects of warming as most species are in the cold half of their range and therefore likely to decline with further warming. This suggests that climate warming may have increasingly negative impacts on intertidal macroalgae in the south-west. Communities of intertidal animals have positive thermal bias across the region, with only local patches of modestly negative thermal bias. This suggests that, while individual cold-water species may suffer, overall, intertidal animals may benefit from the effects of warming since they have distributions centred in warmer areas. Areas of locally negative thermal bias suggest areas where species of animals from warmer climates may be likely to invade.

Figure 4. Regional patterns in Community Temperature Index (CTI) around the UK (a) for animals, and (b) for plants. Community Temperature Index values are averaged across all surveys in 0.5-degree grid cells.

Repeated surveys of sites show changes in Community Temperature Index that match 15-year trends in temperature since 2000. A power analysis has shown the minimum number of sites needed to detect a change in Community Temperature Index given the variability in Community Temperature Index values over time: comparisons across 61 sites can detect a 0.15°C change with an 80% detection rate.

Trend assessment

The trend is unknown: The indicator was not considered as part of the Initial Assessment (HM Government, 2012).

Further information

Patterns of Community Temperature Index for animals and plants follow gradients of sea surface temperature, with that for animals changing more rapidly with temperature than that for plants. Where thermal bias is negative, the community is dominated by cold-water species and where positive, dominated by warm-water species. The dominance of cold-water versus warm-water species relative to local sea surface temperatures is expressed by the thermal bias of the local communities, given by the local Community Temperature Index value less the local Sea Surface Temperature (Figure 5). Various interpretations are possible. Negative thermal bias may indicate communities vulnerable to the effects of warming (Stuart-Smith and others, 2015), such that most species are in the cold half of their range and are likely to decline with further warming. This appears to be the case for macroalgae across most of the UK, except for eastern Scotland and north-east England (Figure 5b), with the discrepancy between Community Temperature Index and Sea Surface Temperature growing towards the south-west of Britain and Ireland

Figure 5. Regional patterns in Thermal Bias (TB), the difference between Community Temperature Index and local Sea Surface Temperature, around the UK for animals (a), and for plants (b). Community Temperature Index values are averaged across all surveys in 0.5-degree grid cells. Negative thermal bias indicates communities comprised of species with distributions centred in colder waters, while positive thermal bias indicates communities dominated by warm-water species.

Local habitat variation

Where surveys across a region span a range of conditions of wave exposure, the Community Temperature Index for animals is lower on wave-sheltered, macroalgal-dominated shores (Figure 6), with plant species showing no such patterns with wave exposure. A plausible explanation is that seaweed canopies protect those organisms living on the rock surface from the direct heating effects of sunlight and from desiccation during periods of low tide, allowing cold-water species to thrive at higher average temperatures than otherwise would be possible. In-situ data logger records show that temperatures under algal canopies is generally lower than that on the open rock, typically by 5 °C during a deployment in South West England (Moore and others, 2017), and by a similar amount during May to September in West Scotland (M.T. Burrows, personal observations). The Community Temperature Index for macroalgae appears unaffected by these differences, providing but not necessarily benefitting from the effects of shade. Although only seen as a correlation at this time and untested through experiments, this direct influence on community structure is further evidence of the direct responsiveness of intertidal species to climate, both directly and potentially by influencing the outcome of biological interactions (Wethey, 1984). The effect also highlights the potential for biological mitigation of the negative effects of climate change through the encouragement of structuring species providing shade for cold-affinity species. Decline and loss of macroalgae through warming may produce a more rapid community shift toward warm-water species as the protective canopy effect is lost.

Assessment of the response of rocky shores communities to climate can be improved by accounting for the wave exposure of the site. This is most easily achieved using the latitude and longitude of the site to extract values from a map-based dataset of wave fetch (the summed distance to the nearest land around a coastal grid cell), such as that generated by Burrows (2012), available at Scotland’s National Marine Plan Interactive website.

Figure 6. Community Temperature Index (CTI) for animals (a) and plants (b) versus wave fetch. The blue line in (a) is the regression line for animal species Community Temperature Index at West Scotland sites (slope = 0.8460, n=220, p < 0.001), showing 1.6°C increase in Community Temperature Index across two orders of magnitude of wave fetch. Wave fetch values from Burrows (2012), available at Scotland’s National Marine Plan Interactive website.

Changes over time

Changes in south-west Britain since 2002 from annually surveyed sites (Figure 7) have been shown by calculating annual anomalies from means of all Community Temperature Index values at each site (Figure 8). Declines in Community Temperature Index were seen at each site with the downward trend matching the local trend in sea surface temperature over the same period (Figures 9). This suggests that rocky shore communities have been responding to even the modest decrease over the last 15 years. The period of slight cooling in the south-west followed a period of a more positive trend in temperatures from 1980 to 2000.

Figure 7. Sites in south-west England and Wales surveyed since 2002. CTI: Community Temperature Index.

Figure 8. Community Temperature Index annual anomalies: the difference between single survey Community Temperature Indexes and the mean Community Temperature Index for each site surveyed (Figure 7), plotted against the year of the survey.

Figure 9. Changes in sea surface temperature (SST) as annual means for 49 to 53 °N, 1 to 6 °W from the Hadley Centre HadISST v1.1 dataset, showing trends from 1980 to 2000 (red) and 2000 to 2015 (blue).

Community Temperature Index in relation to pre-industrial temperature and future trends

An assessment of Community Temperature Index present values in relation to the 1.5oC limit set by the 2016 Paris Agreement was undertaken to evaluate the extent of Community Temperature Index changes in relation to temperature increases (relative Community Temperature Index). Since the Index does not change in direct proportion to temperature, the comparison was made with expected future values for pre-industrial increased by 1.5°C (see Figures 10 and 11). Increases in Community Temperature Index with climate change warming will increase the proportion of the coast that exceeds the threshold of changes under a “pre-industrial plus 1.5°C” scenario.

Figure 10. Community Temperature Index values for (a) animals and (b) macroalgae, less expected future Community Temperature Index values for pre-industrial sea surface temperatures plus 1.5°C, with a 0°C threshold

Figure 11. Distribution of Community Temperature Index values for (a) animals and (b) macroalgae relative to thresholds of 0°C for the “pre-industrial plus 1.5°C” Community Temperature Index estimates.

Conclusions

The Community Temperature Index can be compared with prevailing temperature conditions to produce useful further measures, such as thermal bias, with implications for the sensitivity of such communities to further climatic warming, providing a unique way to assess indirect ecological effects which improve resilience to warming-related biodiversity change.

The relative Community Temperature Index is an assessment of the Community Temperature Index values in relation to the 1.5⁰C limit set by the 2016 Paris Agreement of the United Nations Framework Convention on Climate Change looking at past and future Community Temperature Index changes in relation to temperature. The use of Community Temperature Index-derived measures as indicators of Good Environmental Status are explored by reference to (1) differences between observed present-day Community Temperature Index and expected values of Community Temperature Index for pre-industrial sea surface temperatures, between 0.4 and 1.2⁰C below present-day temperatures, and (2) differences between present-day and expected Community Temperature Index values for “pre-industrial plus 1.5⁰C” temperatures, the limit set by the 2016 Paris Agreement on climate change. Figures 10 and 11 indicate that communities are very close to pre-industrial plus 1.5⁰C” and further warming will increase the vulnerability of communities and their ability to withstand the effects of other anthropogenic pressures.

Further information

The Community Temperature Index has many advantages over current ways of assessing climate change responses. Being driven by patterns in data removes undue reliance on expert judgement on what constitutes a climate-sensitive species and widens the applicability of the method to conservation practitioners. The metric averages responses over a large number of species, removing the need to weave a fabric made from individual species responses. Range shifts are notoriously difficult to establish, given how hard it can be to judge where a range ends and how changes in sampling effort over time can render assessment of change extremely hard (Stuart-Smith and others, 2015). The approach is versatile too, being able to accommodate different kinds of data from presence-only records to detailed abundance estimates, and is well suited to the categorical abundance methods used on rocky shore for many decades.

Knowledge gaps

The key knowledge gaps are:

  • Data is only available for a limited number of species during different time periods
  • Exploration of the implications of biogeography of rocky shore communities around the UK has yet to be done in scientific literature.
Further information

A difficult problem is when data is only available for a limited number of species during different time periods. Because the focus was often on species distributions and to maximise the effectiveness of early surveys around the UK, for example, Southward and Crisp (1954) often sampled only the climate-sensitive species, such as the topshells of the genus Osilinus and Gibbula, the barnacles Semibalanus and Chthamalus and the limpets. Variable numbers of species will have a strong influence on Community Temperature Index estimates and further methods of dealing with this issue are needed before the application to earlier datasets from the 1950s onwards. Another drawback is that the specific mechanisms driving changes are not immediately obvious, mostly since the approach ‘averages out’ responses across such a wide range of species. This area is beginning to yield some more insight, for example, responses to climate by species are emerging as critically dependent on the location within the species range, with species in the cold half of their range responding positively to warming, while those in their warm half responding negatively.

At the time of writing (March 2017), the patterns presented remain unpublished, so as yet, little exploration of the implications for the biogeography of rocky shore communities around the UK has been done in the scientific literature. The index captures the broad patterns effectively: a progressive switch in dominance from cold-water species in the north to warm-water species in the south. Separating macroalgae and animal species shows important differences in patterns for these two different taxa. Both follow gradients of sea surface temperature, but the difference in spatial pattern is striking, with warmer-water animal communities found all around south-west Britain and the West Coast of Ireland, but warmer-water communities of macroalgae are mostly restricted to Channel coasts.

References

Burrows MT (2012) ‘Influences of wave fetch tidal flow and ocean colour on subtidal rocky communities’ Marine Ecology Progress Series, 445: 193-207 (viewed on 27 November 2017)

Burrows MT, Twigg G, Mieszkowska N, Harvey R (2017) ‘Marine Biodiversity and Climate Change (MarClim): Scotland 2014/15’ Scottish Natural Heritage Commissioned Report Number 939 (viewed on 27 November 2018)

Burrows MT, Mieszkowska N (in review) Development of an MSFD intertidal rocky shore indicator for climate change response and an interim assessment of UK shores Scottish National Heritage Commissioned Report.

Devictor V, Julliard R, Couvet, D, Jiguet F (2008) ‘Birds are tracking climate warming, but not fast enough’ Proceedings of the Royal Society of London B: Biological Sciences, 275: 2743-2748 (viewed on 27 November 2018)

European Council (1991a) ‘Council Directive 91/271/EEC of 21 May 1991 concerning urban waste-water treatment Official Journal of the European Union L 135, 30.5.1991, pages 40–52 (viewed on 8 October 2018).

European Council (1991b)’ Council Directive 91/676/EEC of 12 December 1991 concerning the protection of waters against pollution caused by nitrates from agricultural sources Official Journal of the European Union L 375, 31.12.1991, pages 1–8 (viewed on 8 October 2018).

European Commission (2008) ‘Directive 2008/56/EC of the European Parliament and of the Council of 17 June 2008 establishing a framework for community action in the field of marine environmental policy (Marine Strategy Framework Directive)’ Official Journal of the European Union L 164, 25.6.2008, pages 19-40 (viewed 21 September 2018)

HM Government (2012) ‘Marine Strategy Part One: UK Initial Assessment and Good Environmental Status’ (viewed on 5 July 2018)

HM Government (2015) ‘Marine Strategy Part Three: UK Programme of Measures’ December 2015. (viewed on 5 July 2018)

Moore P, Hawkins SJ, Thompson RC (2007) ‘Role of biological habitat amelioration in altering the relative responses of congeneric species to climate change’ Marine Ecology Progress Series, 334: 11-19 (viewed on 27 November 2018)

OSPAR Commission (2017) ‘Intermediate Assessment 2017’ (viewed on 21 September 2018)

Southward AJ, Crisp DJ (1954) ‘Recent changes in the distribution of the intertidal Baracles Chthamalus stellatus Poli and Balanus balanoides L. in the British Iles’ Journal of Animal Ecology 23(1): 163-117 (viewed on 28 November 2018)

Stuart-Smith R. D, Edgar GJ, Barrett NS, Kininmonth SJ, Bates AE (2015) ‘Thermal biases and vulnerability to warming in the world’s marine fauna’ Nature, 528: 88-92 (viewed on 27 November 2018)

Wethey DS (1984) ‘Sun and shade mediate competition in the barnacles Chthamalus and Semibalanus: a field experiment’ Biology Bulletin, 167: 176-185 (viewed on 27 November 2018)

Acknowledgements

Assessment Metadata

Please contact marinestrategy@defra.gov.uk for metadata information

Recommended reference for this indicator assessment

Michael T. Burrows1, Nova Mieszkowska2, John Baxter3, Cristina Vina-Herbon4, Gemma Singleton4, Karen Robison5 and Mike Young6 2018. Intertidal community index (MarClim). UK Marine Online Assessment Tool, available at: https://moat.cefas.co.uk/biodiversity-food-webs-and-marine-protected-areas/benthic-habitats/intertidal-community-index/

1Scottish Association for Marine Science

2Marine Biological Association

3Scottish Natural Heritage

4Joint Nature Conservation Committee

5Natural Resources Wales

6Natural England