Assessment Protocol

This pre-core indicator and its threshold values are yet to be commonly agreed in HELCOM.
The indictor is included as a test indicator for the purposes of the 'State of the Baltic Sea' report, and the results are to be considered as intermediate.

Due to the lack of data, at this stage the assessment is rather descriptive. The basic principles to be used once the underlying data can be further completed is described. Good status is achieved if the incidental by-catch numbers of all assessed species within a given assessment unit are below the removal target used as the threshold value taking also other human-induced mortality into account. A population-specific evaluation is applied to all HELCOM level 2 sub-basins in which i) the population occurs and ii) fishing methods causing incidental by-catch are spatially overlapping with the distribution of that population.

The evaluation for a single sub-basin is done using the 'one-out, all-out' principle, which for instance is applied in the EU Water Framework Directive (European Commission 2000). In this indicator, this means that good status is not reached if incidental by-catch for a single population contributes to exceeding the pre-defined threshold value of human induced mortality for that population (for waterbirds; for harbour porpoises the removal target is related explicitly to incidental by-catch).

It must be taken into account that not all species are distributed throughout all sub-basins. Consequently, for areas outside the distributional range, no conservation or removal target for the species are needed in the particular sub-basin and the number of species assessed varies among the sub-basins. For the two populations of harbour porpoise and the three waterbird species initially assessed in this indicator the occurrence in the sub-basins is shown in Assessment protocol table 1. Assessment protocol table 2 indicates published data on incidental by-catches per species and sub-basin.

Besides assessing incidental by-catch on a population scale (see below: assessment units), it may be desirable for management purposes to downscale information in order to implement measures on a smaller scale (e.g. sub-basin, HELCOM scale 2). Difficulties exist both in measuring incidental by-catch and population size to a sufficiently high degree of accuracy on such a small scale. If this information becomes available, the assessment units may be downscaled for management purposes. In Assessment protocol table 1, some examples of mammal and waterbird distributions are downscaled (not quantitatively) from populations to HELCOM assessment unit scale 2.

Assessment protocol table 1. Distribution of some marine mammals and waterbirds on the level of HELCOM level 2 sub-basins (after Durinck et al. 1994, Skov et al. 2011, Sveegaard et al. 2011, ASCOBANS 2016b). 

Assessment Table 1.png

Assessment protocol table 2. Indication of published data on incidental by-catch related to species and sub-basin. 

Assessment Table 2.png

x = species occurring, number = reference to incidental by-catch:

1: Oldén et al. 1988: gill nets for cod and herring; 90-95% common guillemot, 3-7% great cormorant, <1% long-tailed duck

2: Lynneryd et al. 2004: Common guillemot in post-breeding season mainly around Gotland, in the Skerries and in the Åland Sea – areas with gillnet fishing. Also common guillemots caught in gillnets with pingers in Hanö Bight.

3: Olsson et al. 2002 and Österblom et al. 2002 (summarzied in Erdmann et al. 2005): map with ring recoveries of common guillemots found in fishing gear (mostly gillnets). Because data are used from 1912 onwards, only sub-basins with many recoveries (i.e. probably including recent ones) are considered.

4: Schirmeister 2003 and Erdmann et al. 2005: Long-tailed ducks very often caught in gillnets off Usedom, to a smaleer amount also greater scaups and common guillemots.

5: Kirchhoff 1982: long-tailed ducks and greater scaups by-caught by gillnet fishery in Kiel Bay.

6: Mentjes & Gabriel 1999: weak to strong bycatch of "ducks" (no species mentioned) west and south of Fehmarn; can be refered to a general bycatch risk for long-tailed ducks in those areas.

7: Kowalski & Manikowski 1982: in gillnets between Dzivnow and Pobierowo (Pomeranian Bay) bycaught birds comprised 53% long-tailed ducks and 0.3% common guillemots.

8: Stempniewicz 1994: in gillnets between Hel, Gdynia and Vistula mouth (Gdansk Basin) bycaught birds comprised 48.3% long-tailed ducks, 7.7% greater scaups and 0.8% common guillemots.

9: Kies & Tomek 1990: in gillnets in Zatoka Pucka (Gdansk Basin) bycaught birds comprised 41% long-tailed ducks, 0.5% greater scaups and 20.5% common guillemots.

10: Dagys & Žydelis 2002: off the Lithuanian coast, the majority of bycaught birds were long-tailed ducks (61%), Alcidae 1%.

11: Urtans & Priednieks 2000: proportions of bycaught birds in the Gulf of Riga: long-tailed duck 35.4%, "diving ducks" 22.9%, auks 5.2%; Baltic Sea (belonging to Eastern Gotland Basin): long-tailed duck 42.5%, "diving ducks" 39.6%. Auks probably include common guillemots and diving ducks probably include greater scaups, although both species are not mentioned in the text.

12: Dagys et al. 2009: Many long-tailed ducks bycaught in Estonian part of Gulf of Finland. There were bycaught long-tailed ducks in other areas of Estonia, and because at least test fishing in the Estonian part of Northern Baltic Proper revealed bird bycatch and the test fishing areas overlap with long-tailed duck distribution (Skov et al. 2011) it can be assumed that this species was actually bycaught there. – Latvia: bycatch occurrence of common guillemot and long-tailed duck in both Gulf of Riga and Eastern Gotland Basin (Latvian parts), greater scaup only in Gulf of Riga. – Lithuania: long-tailed duck 57%, alcids 8% (probably including common guillemot).

13: Bønløkke et al. 2006: report of a greater scaup drowned in fishing net near Rostock (according to German ring recovery database in 1970).

14: Grimm 1985: annual bycatch of 2800 greater scaups in gillnets in Wismar Bay.

15: Larsson & Tydén 2005: incident of 998 bycaught long-tailed ducks on Hoburgs Bank.

16: Žydelis et al. 2006: many beached long-tailed ducks in Lithuania show signs of net entanglement, also some alcids (probably including common guillemot).

17: Bardtrum et al. 2009: turbot gillnet-fishing at east coast of Gotland produces few bycaught common guillemots, and bottom-set gilnnets for cod produced only few bycaught long-tailed ducks on Hoburgs Bank. Two (old?) ring recoveries of greater scaup in fishing gear are reported, but unclear whether east or west of Gotland.

18: Degel et al. 2010: gill net fisheries investigated around Aerø suggest that long-tailed ducks and common guillemots are drowing in gillnets in marine areas belonging to Kiel Bay and Great Belt.

19: ASCOBANS 2016b: Puck Bay incidental catch in semi-driftnets, set gillnets and others

20: ICES 2015, 2016a: estimated bycatch number for ICES subdivisions 22,23 and 24 available, bycatch known in all other areas but without estimate


Assessment units

The indicator is applicable in the whole Baltic Sea, as it is known that incidental by-catches of birds and mammals in fisheries occur in the whole area. The indicator is assessed using HELCOM assessment scale 2 which consists of 17 Baltic Sea sub-basins. The assessment units are defined in the HELCOM Monitoring and Assessment Strategy Annex 4.

Assessments concerning the incidental by-catch of mammals and birds face the challenge that on the one hand the situation of marine areas needs be assessed on a scale that allows for identification of problem areas where actions should be taken (e.g. within the MSFD framework), but on the other hand the methods available need to be exercised on the level of populations which often additionally are impacted by anthropogenic activities outside the assessment area (especially in migratory waterbirds). Given the high mobility of marine mammals and waterbirds, and the distributional range of populations, assessments will necessarily need to incorporate a scale of the range of a population or management unit, but also needs an adjustment to HELCOM assessment units, with Scale 2 appearing to be an appropriate one.

For example, in the case of the harbour porpoise, two management units exist: 1) the Baltic Proper population and 2) the population of the Western Baltic, Belt Seas and Kattegat. A third management unit is the North Sea population which extends into the Kattegat. Due to the comparatively low fraction of its range and porpoise numbers occurring in the area covered by HELCOM, no separate assessment is made here. Instead we refer to OSPAR M6 indicator. Certain high-density areas (probably representing key habitats) have been identified in the Baltic (Sveegaard et al. 2011; Carlström & Carlén 2016). The preliminary distribution maps produced within the SAMBAH project make it possible to draw a contour around an area where the probability of detecting a porpoise within a given month is e.g. 20% or higher. Based on this, the area around the Midsjö offshore banks south-east of Öland seem to be of crucial importance during the summer months when the Baltic Proper porpoise population is spatially separated from the population of the Western Baltic, Belt Sea and Kattegat. This approach is similar to choosing certain kernel contours based on satellite transmitter data of tagged porpoises (as used in Sveegaard et al. 2011). Incidental by-catch risk assessment can be made combining this data with available information on fishing effort with gear types known for high incidental by-catch risk (e.g. gillnets with large mesh size). A BRA was initially developed for cetaceans at an ICES Workshop (ICES 2010) in order to identify areas and fisheries that are likely to pose the greatest conservation threat to by-caught cetacean species, taking into account the uncertainty of the population structure. The BRA highlights areas where the greatest problems occur and enables educated fisheries management decisions. For an area in the Skagerrak, Kindt-Larsen et al. (2016) demonstrated a clear correlation between incidental by-catch risk and the products of porpoise densities and fishing effort (in terms of soak time). A BRA has been carried out for all Swedish waters and the results will be presented in the revised Swedish action plan for harbour porpoises.

The tentative assessment for this core indicator is made using HELCOM assessment units. For the population of the Western Baltic, Belt Sea and Kattegat, the HELCOM open sea assessment units Kattegat, the Sound, Great Belt, Kiel Bay, Bay of Mecklenburg and Arkona Basin should be combined. Because the assessment for the Kattegat is to be made by OSPAR within their by-catch indicator M6 (which in contrast to this indicator does not include any birds and seals), the threshold should be harmonised between OSPAR and HELCOM.

For the Baltic Proper population, a combination of the assessment units Arkona Basin, Bornholm Basin, Western and Eastern Gotland Basin, Gdansk Basin, Northern Baltic Proper and Åland Sea is necessary (Table 1). More northern and eastern regions may be added as information becomes available if these areas are inhabited by harbour porpoises. In the overlapping area where both populations occur (i.e. Arkona Basin), incidental by-catches should be assigned to both populations as a precautionary approach.

Some waterbird populations extend into areas outside the Baltic. The tentative assessment is made in sub-basins in which the species is known to occur (Assessment protocol table 1) attempting to also consider pressures from areas outside of these areas into account.