Recovery in the Population Abundance of Sensitive Fish Species

Indicator Metric and Data Collection

Population abundance metrics were determined using data collected by 12 groundfish surveys carried out across two separate regions: the Greater North Sea and the Celtic Seas. Six survey data sets were available for analysis for each region (Table a).

Standard data collected on these surveys comprises numbers of each species of fish sampled in each trawl sample, measured to defined length categories (i.e. a fish with a recorded length of 14 cm would be between 14.00 cm and 14.99 cm in length). By dividing the species catch numbers-at-length by the area swept by the trawl on each sampling occasion, the catch data are converted to estimates of fish density-at-length, by species, at each sampling location in each year. Summing these trawl-sample species density-at-length estimates across all trawl samples collected within each sampling stratum in each year (e.g. ICES statistical rectangles), and dividing by the number of trawl samples within each stratum per year, gives an estimate of the density (of each species and length category) within each sampling stratum in each year. Summing these sample stratum density estimates across all sampling strata sampled in each year, and dividing by the number of strata sampled, provides estimates of the average density (N), of each species (s) and length category (l), in each year, across the whole area covered by the survey. Summing these density estimates (N_s,l / km²) across all length classes provides the required estimate of species population abundance density (N_s / km²) in each year for each survey.

Two different sensitivity metrics were used to identify species considered to be sensitive to fishing mortality. The Average Life-history Trait (ALHT) metric is the same as the metric developed by Greenstreet et al. (2012b). The ‘Proportion Failing to Spawn’ (PFS) is a new sensitivity metric developed to address flaws in the earlier metric identified by the ICES Working Group on the Ecosystem Effects of Fishing Activities (ICES, 2016). Development of the PFS metric is fully documented in a supporting paper (Greenstreet et al., 2017a). Analyses for both metrics are presented to demonstrate that assessment outcomes were not unduly affected by choice of sensitivity metric. Where choice of metric affects the assessment outcome, the principal assessment outcome should be based on the PFS metric. Both metrics rely on the availability of several life-history trait parameters for each species. For example, maximum recorded length (L_max), von Bertalanffy ultimate body length or length infinity (L_inf), von Bertalanffy growth term (K), length at maturity (L_mat), and age at maturity (A_mat). Compilation of these parameters for all 485 species encountered across the 12 groundfish surveys in the North-East Atlantic, and where missing, the methods used to estimate them, are fully described in Greenstreet et al., 2017b.

For each survey, the number of sensitive species encountered was first established. However, almost by definition, these species are among the rarest in each surveyed community, so for many species the data available were too sparse to support meaningful assessment. Only species encountered on 50% or more of occasions that each survey was carried out were considered to have been sufficiently well sampled as to support formal assessment. In the overwhelming majority of cases, box plots comparing the sensitivity metric scores of species meeting and failing to meet this 50% criterion confirm that species deemed sufficiently well sampled as to support formal assessment were representative, in terms of their sensitivity metric scores (and hence life-history trait composition), of the full suite of sensitive species encountered by each survey (Figure a and Figure b). Thus, if recovery was observed in a significant fraction of the species capable of supporting formal assessment, the same might be expected among the species sampled too infrequently to support formal assessment.

Spatial Scope: Assessment Units

For each of the 12 groundfish surveys, population abundance density time-series were determined for all sensitive species, identified by either sensitivity metric, that were deemed capable of supporting formal assessment. Individual survey-based assessments were then performed. The individual survey-based assessments across the whole of the Celtic Seas and the Greater North Sea regions were then considered to determine overall assessment outcomes.

Baselines

None of the surveys extend sufficiently far back in time as to provide an adequate reference period to establish species abundance levels commensurate with acceptable status. A trends based assessment approach, relying on the use of trends based assessment values, was therefore adopted.

Assessment Values

By virtue of their sensitivity to additional human-related mortality, the population abundance of each sensitive species sampled by each survey is assumed to have declined as a result of past human activities. There is good evidence that fishing mortality has indeed caused declines in the populations of sensitive species.

Thus trends based assessment values related to population recovery constituted the primary basis for assessment; sensitive species should be increasing in abundance. However, should this primary assessment give an unacceptable outcome, or the results prove uncertain, a secondary assessment can be performed. The secondary assessment addresses an alternative question of whether further decline in the population abundance of sensitive species has at least been halted (Greenstreet et al., 2012b).

OSPAR Intermediate Assessment (IA) 2017 Indicator Assessment values are not to be considered as equivalent to proposed European Union (EU) Marine Strategy Framework Directive (MSFD) criteria threshold values, however they can be used for the purposes of their MSFD obligations by those Contracting Parties that wish to do so.

Primary Assessment: Recovery

Abundance trends for most species are not monotonic, so a simple parametric trends based approach (e.g. linear regression) is not appropriate; instead a non-parametric approach has been used. Essentially, the entirety of each survey time series acts as the reference period. The assessment value is set as: abundance in the assessment year must lie in the upper 25% of all abundance values observed throughout the time series.

For the purposes of the Intermediate Assessment (IA) 2017, the assessment yearwas the last year in each survey time series for which data were available. However, to explore recent trends, preceding years back to 2010 were also defined as the assessment year. The approximate period when data would have been available for the initial assessments required in 2012 under the European Union Marine Strategy Framework Directive (MSFD) was 2010. In assigning each consecutive preceding year as the assessment year, the length of the time series available to determine the upper 25% of all target abundance values diminishes. This places a limit on how far back in time the assessment can be pushed, particularly for surveys that only commenced more recently.

The surveys therefore provide abundance density estimates for each sensitive species in each year: a species-specific abundance metric for each sensitive species. These data can be ranked, and the ranking representing the lower boundary of the upper 25^th percentile of all data for each species establishes the species-specific abundance metric-level assessment valuefor each species-specific abundance metric. In the stipulated assessment year, each species-specific abundance metric ranking should be greater than the ranking representing the species-specific abundance metric-level assessment value.

The sensitive species abundance indicator, for any given survey, is defined as the number of species in any assessment year whose species-specific abundance metric meets or exceeds its species-specific abundance metric-level assessment value. The metric meets or exceeds the assessment value when it is within the upper 25^th percentile of all the abundance data over the whole survey time series, up to and including the year defined as the assessment year.

The sensitive species abundance indicator-level assessment value is defined based on probability. Random walk simulations suggest that the probability of the last abundance estimate datum in a time series falling into the upper 25^th percentile of all data is 0.332. Knowing the number of assessed sensitive species in each survey, and the probability of any one species-specific abundance metric meeting its upper 25^th percentile species-specific abundance metric-level assessment value, allows the sensitive species abundance indicator-level assessment value to be defined as the value that represents a significant (p<0.05) departure from the binomial distribution.

Secondary Assessment: Halt Further Decline

The logic underlying the alternative secondary assessment is identical to the logic described in relation to the primary assessment of population recovery objectives. The same simple non-parametric trends based approach can be used, except that in this instance, for each sensitive species in each survey, abundance in the assessment year must lie outside the lower 25% of all abundance values observed throughout the time series for the metric to meet its assessment value.

Following the same logic described for the primary assessment, the annual species-specific abundance metric data are ranked and the ranking representing the lower boundary of the upper 75^th percentile of all data established as the new species-specific abundance metric-level assessment value; thus, in the stipulated assessment year, each species-specific abundance metric rank should be greater than the ranking representing the species-specific abundance metric-level assessment value.

The sensitive species abundance indicator, for any given survey, is defined as the number of species in any assessment year whose species-specific abundance metric meets or exceeds its species-specific abundance metric-level assessment value of being within the upper 75^th percentileof all their own abundance data over the whole survey time series, up to and including the year defined as the assessment year. The random walk simulation suggest that the probability of the last abundance datum in a time series falling into the upper 25^th percentile of all data is 0.332 for the primary assessment. Similarly, for the secondary assessment the probability of the last abundance datum in a time series falling into the lower 25^th percentile of all data is also 0.332 and thus the probability of the last abundance datum in a time series falling into the upper 75^th percentile of all data is 1 - 0.332 = 0.668. Following an identical logic to the recovery primary assessment, the number of assessed sensitive species and the probability of any one species-specific abundance metric meeting its upper 75^th percentile metric-level target is known for surveys in the secondary assessment, allowing to define the sensitive species abundance indicator-level assessment value as the sensitive species abundance indicator value that represents a significant (at p<0.05) departure from the binomial distribution.

Assessment Integration Approaches

Two integration approaches, ‘probabilistic’ and ‘averaging’ were applied to determine integrated assessment outcomes at the regional scale.

Probabilistic Integration

Since the p<0.05 significance level, used to identify significant departures from a binomial distribution, is applied s the basis for setting sensitive species abundance indicator-level assessment values for each individual survey assessment, it follows that the probability of observing any one survey sensitive species abundance indicator actually meeting its sensitive species abundance indicator-level assessment value by chance is p≤0.05.  Therefore the binomial distribution can again be used to determine the number of surveys where the indicator meets the indicator-level assessment value for a given number of individual survey assessments for this to represent a significant deviation from the binomial distribution. The probability of observing the sensitive species abundance indicator-level assessment value being met by chance is low. This results in the reliability of the probabilistic integration approach becoming questionable when the number of individual survey assessment outcomes that require integration becomes smaller.

Six survey assessment outcomes were available for both the Celtic Seas and Greater North Seas regions. This was considered to be just sufficient to allow for the use of the probabilistic integration method. Observing sensitive species abundance indicator-level assessment values met in two or more of the six surveys would represent a significant departure from the binomial distribution in both regions.

Averaging Integration

To apply an averaging integration approach, all survey sensitive species abundance indicator values were first converted to a Common-scale Indicator Score (CIS) by expressing sensitive species abundance indicator values as a fraction of their sensitive species indicator-level assessment values. Where individual indicator assessment values were met or exceeded the CIS was ≥1, where assessment values were not met the CIS was <1. These CIS values could then be averaged across all surveys carried out within each region to derive averaged integrated assessment outcomes. Where the averaged CIS was ≥1, the averaged integrated assessment outcome conferred acceptable status.

The lack of historical data precludes the use of absolute assessment values. The absence of survey data at times early enough to be considered suitable as reference periods precludes the use of an empirical approach to set assessment values for the absolute abundance of sensitive fish species, most of which are known to have become depleted well before the onset of any of the currently extant surveys. For the current assessment a trends based approach has been adopted, but this can only ever be a temporary stop-gap. Once such trends based assessment values are achieved, then the need for absolute abundance targets will become paramount because continuous increase in any species is biologically impossible. At some point each species’ population carrying capacity will be reached and resource limitation will cap any further population growth. In the absence of empirical data to support absolute abundance assessment value setting, populations dynamics models parameterised to represent specific fish communities could provide the only viable alternative. Part of the motivation behind the development of the PFS sensitivity metric is that the fraction of any cohort that actually survives to spawn at least once is a key variable in any population dynamics model.

A further knowledge gap relates to the effects of climate change on the potential for depleted sensitive species populations to recover. In many organisms, individuals in colder environments grow more slowly, but are larger as adults. This widespread pattern is embodied by two well-established rules: Bergmann’s rule, which describes the association between temperature and body size in natural environments, and which gives rise to gradients of increasing body size with latitude, and the temperature-size rule, which describes reaction norms relating temperature to body size in laboratory experiments (Anguilletta and Dunham, 2003). Large ultimate body size is a key life-history trait determining the ‘sensitivity’ of fish species to fishing mortality, and these temperature-size rules imply that warming seas might well inhibit population growth and reduce the potential for population recovery among large-bodied sensitive species.

Anguilletta M.J.Jr., Dunham, A.E. 2003. The temperature size rule in ectotherms: simple evolutionary explanations may not be general. American Naturalist, 162, 332-342.

Frisk, M.G., Miller, T.J., and Fogarty, M.J. 2001. Estimation and analysis of biological parameters in elasmobranch species: a comparative life-history study. Canadian Journal of Fisheries and Aquatic Sciences, 58: 969-981.

Gislason, H., Pope, J.G., Rice, J.C. and Daan, N. 2008. Coexistence in North Sea fish communities: implications for growth and natural mortality. ICES Journal of Marine Science, 65: 514-530.

Greenstreet, S.P.R., Fraser, H.M., Rogers, S.I., Trenkel, V.M., Simpson, S.D., and Pinnegar, J.K. 2012a. Redundancy in metrics describing the composition, structure, and functioning of the North Sea demersal fish community. ICES Journal of Marine Science, 69: 8–22.

Greenstreet, S.P.R. and Hall, S.J. 1996. Fishing and the ground-fish assemblage structure in the north-western North Sea: an analysis of long-term and spatial trends. Journal of Animal Ecology, 65: 577-598.

Greenstreet, S.P.R., Holland, G.J., Fraser, T.W.K., and Allen, V.J. 2009. Modelling demersal fishing effort based on landings and days absence from port, to generate indicators of “activity”. ICES Journal of Marine Science, 66: 886–901.

Greenstreet, S.P.R. and Rogers, S.I. 2000. Effects of fishing on non-target fish species. Effects of Fishing on Non-Target Species and Habitats: Biological, Conservation and Socio-economic Issues. (eds M. J. Kaiser and B. de Groot), pp. 217-234. Blackwell Science, Oxford, U.K.

Greenstreet, S.P.R. and Rogers, S.I. 2006. Indicators of the health of the fish community of the North Sea: identifying reference levels for an Ecosystem Approach to Management. ICES Journal of Marine Science, 63: 573-593.

Greenstreet, S.P.R., Rogers, S.I., Rice, J.C., Piet, G.J., Guirey, E.J., Fraser, H.M., and Fryer, R. J. 2011. Development of the EcoQO for the North Sea fish community. ICES Journal of Marine Science, 68: 1–11.

Greenstreet, S.P.R., Rossberg, A.G., Fox, C.J., Le Quesne, W.J.F., Blasdale, T., Boulcott, P., Mitchell, I., Millar, C., and Moffat, C.F. 2012b. Demersal fish biodiversity: species-level indicators and trends based targets for the Marine Strategy Framework Directive. ICES Journal of Marine Science, 69: 1789–1801.

Greenstreet, S.P.R., Spence, F.E. and McMillan, J.A. 1999a. Fishing effects in northeast Atlantic shelf seas: patterns in fishing effort, diversity and community structure. V. Changes in structure of the North Sea groundfish assemblage between 1925 and 1996. Fisheries Research, 40: 153-183.

Greenstreet, S.P.R., Spence, F.E., Shanks, A.M., and McMillan, J.A. 1999b. Fishing effects in Northeast Atlantic shelf seas: patterns in fishing effort, diversity and community structure. 2. Trends in fishing effort in the North Sea by UK registered vessels landing in Scotland. Fisheries Research, 40: 107–124.

Hobday, A.J., Smith, A.D.M., Stobutzki, I.C., Bulman, C., Daley, R., Dambacher, J.M., Deng, R.A., Dowdney, J., Fuller, M., Furlani, D., Griffiths, S.P., Johnson, D., Kenyon, R., Knuckey, I.A., Ling, S.D., Pitcher, R., Sainsbury, K.J., Sporcic, M., Smith, T., Turnbull, C., Walker, T.I., Wayte, S.E., Webb, H., Williams, A., Wise, B.S. and Zhou, S. 2011. Ecological risk assessment for the effects of fishing. Fisheries Research, 108: 372-384.

Huston, M.A. 1994.  Biological diversity: the coexistence of species on changing landscapes. Cambridge University Press, Cambridge, UK 

ICES. 2016. Report of the Working Group on the Ecosystem Effects of Fishing Activities (WGECO), 6–13 April 2016, Copenhagen, Denmark. ICES CM 2016/ACOM:25. 110 pp.

Jennings, S., Greenstreet, S.P.R. and Reynolds, J. 1999a. Structural change in an exploited fish community: a consequence of differential fishing effects on species with contrasting life histories. Journal of Animal Ecology, 68: 617-627.

Jennings, S., Alvsvíg, J., Cotter, A. J., Ehrich, S., Greenstreet, S.P.R., JarreTeichmann, A., Mergardt, N., Rijnsdorp, A.D. and Smedstad, O. 1999b. Fishing effects in Northeast Atlantic shelf seas: patterns in fishing effort, diversity and community structure. 3. International fishing effort in the North Sea: an analysis of spatial and temporal trends. Fisheries Research, 40: 125–134.

Jennings, S., Reynolds, J. D. and Mills, S. C. 1998. Life history correlates of responses to fisheries exploitation. Proceedings of the Royal Society of London, 265: 1-7.

Le Quesne, W.J.F. and Jennings, S., 2012. Predicting species vulnerability with minimal data to support rapid risk assessment of fishing impacts on biodiversity. Journal of Applied Ecology, 49: 20-28.

MacArthur, Robert H.; Wilson, E. O. 1967. The Theory of Island Biogeography. Princeton, New Jersey: Princeton University Press.

May, R. 1976. TheoreticalEcology: Principles and Applications. Blackwell Scientific Publishers. ISBN 0-632-00768-0

Modica, L., Velasco, F., Preciado, I., Soto, M., and Greenstreet, S.P.R. 2014. Development of the large fish indicator and associated target for a Northeast Atlantic fish community. ICES Journal of Marine Science, 71: 2403–2415.

Philippart, C.J.M. 1998. Long-term impact of bottom fisheries on several by-catch species of demersal fish and benthic invertebrates in the south-eastern North Sea. ICES Journal of Marine Science, 55: 342–352.

Piet, G. J., and Jennings, S. 2005. Response of potential fish community indicators to fishing. ICES Journal of Marine Science, 62: 214-225.

Reznick D., Bryant M. J., Bashey F. 2002r- and K-selection revisited: the role of population regulation in life history evolution. Ecology 83, 1509–152010.1890/0012-9658(2002)083[1509:RAKSRT]2.0.CO;2 (doi:10.1890/0012-9658(2002)083[1509:RAKSRT]2.0.CO;2

Rijnsdorp, A.D., Van Leeuwen, P.I., Daan, N., and Heessen, H.J.L. 1996. Changes in abundance of demersal fish species in the North Sea between 1906–1909 and 1990–1995. ICES Journal of Marine Science, 53: 1054–1062.

Roff, D. 1993. The Evolution of Life Histories. Chapman and Hall, New York. xii + 535 pp. ISBN: 0-412-02381-4, ISBN: 0-412-02391-1 (paper).

Shephard, S., Reid, D.G. and Greenstreet, S.P.R. 2011. Interpreting the large fish indicator for the Celtic Sea. ICES Journal of Marine Science, 68: 1963–1972.

Stearns, S.C., 1977 Evolution of life-history traits - critique of theory and a review of data.  Annu. Rev. Ecol. Syst. 8: 145–171. doi:  10.1146/annurev.es.08.110177.001045

Stearns, S.C., 1992 The evolution of life histories. New York, NY: Oxford University Press

van Strien, A.J., van Duuren, L., Foppen, R.P.B. and Soldaat, L.L., 2009. A typology of indicators of biodiversity change as a tool to make better indicators. Ecological Indicators, 9: 1041-1048.

Walker, P.A. and Hislop, J.R.G. 1998. Sensitive skates or resilient rays? Spatial and temporal shifts in ray species composition in the central and north-western North Sea between 1930 and the present day. ICES Journal of Marine Science, 55: 392-402.