Ref. Ares(2017)5306028 - 30/10/2017

Fishfriendly Innovative Technologies for Hydropower

Funded by the Horizon 2020 Framework Programme of the European Union

D1.1 Metadata overview on response to disturbance

Project Acronym FIThydro Project ID 727830 Work package 1 Deliverable Coordinator Christian Wolter Author(s) Ruben van Treeck (IGB), Jeroen Van Wich- elen (INBO), Johan Coeck (INBO), Lore Vandamme (INBO), Christian Wolter (IGB) Deliverable Lead beneficiary INBO, IGB Dissemination Level Public Delivery Date 31 October 2017 Actual Delivery Date 30 October 2017

Acknowledgement This project has received funding from the European Union’s Horizon 2020 research and inno- vation program under grant agreement No 727830.

Executive Summary

Aim

Environmental assessment of hydropower facilities commonly includes means of fish assem- blage impact metrics, as e.g. injuries or mortality. However, this hardly allows for conclusion at the population or community level. To overcome this significant knowledge gap and to enable more efficient assessments, this task aimed in developing a fish classification system according to their species-specific sensitivity against mortality. As one result, most sensitive fish species were identified as suitable candidates for in depth population effects and impact studies. Another objective was providing the biological and autecological baseline for developing a fish population hazard index for the European fish fauna.

Methods

The literature has been extensively reviewed and analysed for life history traits of fish providing resilience against and recovery from natural disturbances. The concept behind is that species used to cope with high natural mortality have evolved buffer mechanisms against, which might also foster recovery from human induced disturbances. In contrast, species with very low natural mortality lack such buffering traits and thus, are more sensitive against human-induced mor- tality. The following life history traits have been identified reportedly indicating the species’ biological sensitivity against natural mortality: natural adult fish mortality, maximum length, maximum age, age at first maturity, fecundity (egg number), annual number of offspring per female, and migration behaviour.

Existing species trait databases (e.g. FishBase) as well as the primary literature has been ex- tensively reviewed to gather as much as possible information on each trait for all fish species occurring in European waters. A respective trait database has been compiled. When possible, missing data have been completed using analogues from closely related species or calculated values. Depending on the metric, multiple entries per species and trait were either averaged or the maximum value used for scoring. Scores were assigned at a scale of 1 (lowest sensitivity) to 7 (highest sensitivity). The classes were set trait specific based on the traits respective data distribution among all species. For each species, the scores of all metrics were averaged and rounded to the next integer. Because the two extreme classes 1 and 7 remained empty the final classification of species was adjusted to five classes. The resulting scores were used to align species according to their sensitivity against mortality and also to arrange them in a two-dimen- sional matrix of sensitivity classes and the IUCN Red List classification status to account for their conservation value.

Results

The compiled metadata overview comprises 192 fish and lamprey species. Of the 192 investi- gated species, 3 were assigned to class 5 (highest sensitivity), 37 were assigned to class 4 (high sensitivity), 82 taxa were assigned to class 3 (average sensitivity), 67 were assigned to class 2 (moderate sensitivity) and 3 taxa were assigned to class 1 (low sensitivity). The result allows for an objective and comprehensible comparison of the inherent biological sensitivity of fish and lamprey species and identified in total 40 taxa of high susceptibility to disturbance covering all biogeographic regions of Europe, which might serve as candidate species for developing novel assessment tools and for further analysing population effects.

727830 FIThydro - Deliverable 1.1 - Page 2 of 26

Table of Contents Executive Summary ...... 2 List of Figures ...... 3 List of Tables ...... 3 1. Background ...... 4 2. Methods ...... 4 2.1 Data gathering ...... 4 2.2 Identification and processing of species traits...... 4 2.2.1 Natural adult fish mortality M ...... 5 2.2.2 Fecundity ...... 7 2.2.3 Annual recruits per female and parental care ...... 8 2.2.4 Age at first maturity ...... 10 2.2.5 Maximum age...... 11 2.2.6 Maximum length and migration type ...... 12 2.3 Final scoring ...... 14 2.4 The IUCN conservation status ...... 14 3. Results ...... 16 3.1 Sensitivity analysis ...... 16 3.2 Sensitivity-IUCN matrix ...... 19 4. Conclusions ...... 21 5. References ...... 22 Appendix ...... 23

List of Figures Distribution of fecundity expressed as eggs per female among all species...... 7 Distribution of minimum and maximum number of offspring per female and year...... 9 Distribution of age at first maturity among all species...... 10 Distribution of maximum age among all species ...... 12 Distribution of maximum length among all species ...... 13 Structure and hierarchy of the IUCN threat categories...... 15 Distribution of 192 species in the final 5 sensitivity classes...... 16 Distribution of 192 fish and lamprey species among the 5 sensitivity classes...... 20

List of Tables Table 1: Classification of natural adult fish mortality...... 6 Table 2: Classification of fecundity...... 8 Table 3: Classification of minimum and maximum annual recruits/female...... 9 Table 4: Classification of age at first maturity...... 11 Table 5: Classification of maximum age...... 12 Table 6: Classification of maximum length...... 14 Table 7: Calculated and rounded results of the sensitivity analysis...... 17

727830 FIThydro - Deliverable 1.1 - Page 3 of 26

1. Background Assessing the impact of hydropower plants on different fish species and communities is a key- component of the decision support in commissioning and operating hydropower facilities. One prerequisite is identifying species most at risk and further developing methods to evaluate gen- eral susceptibility of fish species to disturbance. Therefore, as a first step, this task compiles and analyses life history traits, biological characters and environmental tolerances of European lampreys and fish species that are relevant to dispersal, meta-population dynamics, population resilience, and resistance to disturbances. Its main objective was to identify critical life history traits and reaction norms of lampreys and fish in response to human-induced mortality, e.g. here from hydropower. Meta-analyses of species traits and population developments within Europe were used to develop species-specific indices of population-biological sensitivity and conservation value that both feed into a rank classification of European lampreys and most threatened in hydropower environments.

The final classification of European fish species according to their susceptibility towards hydro- power provides a set of highly sensitive candidate species from different biogeographic regions of Europe for environmental assessments and further analyses of impacts at the level of pop- ulations or communities to finally predict constellation-specific threats and effects of hydro- power operation on fish.

2. Methods 2.1 Data gathering To obtain data on the selected traits, we primarily searched the peer-reviewed literature using the search engines “Web of Knowledge” and “Google Scholar” with scientific names of species and genera as keywords. We further referred for trait information to the most comprehensive database on fish FishBase (www..org; FishBase consortium, Ver. 06/2017) and refer- ences therein. Additional grey literature, like reports and theses, were searched using “Google” and the scientific and common names of species as keyword. Finally, local experts were di- rectly contacted for information and literature. Nevertheless, for several species we could not find any data on one or more traits. In such cases, analogous conclusions were drawn from close relatives of the same . Species where such information was concordantly lacking and globally extinct species were excluded from further analyses.

2.2 Identification and processing of species traits The literature has been extensively reviewed and analysed for life history traits of fish providing resilience against and recovery from natural disturbances. The concept behind is that species used to cope with high natural mortality have evolved buffer mechanisms against, which might also foster recovery from human induced disturbances. In contrast, species with very low nat- ural mortality lack such buffering traits and thus, are more sensitive against human-induced mortality.

We identified, compiled and analysed six life-history traits that were considered as suitable metrics for the autecological or biological sensitivity of a fish/lamprey species against natural mortality. The trait values were converted into indices with 7 classes for all traits, as it promised the best compromise between the degree of resolution of the final score and comprehensibility

727830 FIThydro - Deliverable 1.1 - Page 4 of 26 of the information. For each trait, we explored the distribution of the data and identified data- dense clusters among all species which were then used to delimit the respective classes. We assigned class values from “1” to “7” in increasing order of sensitivity. For each species, all six traits (depending on data availability) were separately scored and then averaged to the overall species score.

2.2.1 Natural adult fish mortality M

Natural mortality is an important variable to evaluate demographic parameters of fish popula- tion dynamics. Average natural annual mortality of adult individuals has been occasionally re- ported from empirical observations given as percentages. However, for most species no em- pirical estimations of adult mortality have been reported. Furthermore, unbiased estimates of natural mortality in exploited fish stocks are hard to obtain, since this trait does not consider exploitation by humans. Natural adult fish mortality rates were either calculated by means of the Life-History-Tool on www.fishbase.org that uses Pauly’s (1980) equations to determine the natural mortality “M” based on parameters of the von-Bertalanffy growth functions maximum (total) length and annual average water temperature. The logarithmic scaled M value was then transferred into percentages as follows: 푀[%] = 100 ∗ (1 − 푒−푀). This approach allows for an objective comparison of mortality within and across taxa, as opposed to mortality rates obtained from studies conducted in a geographically restricted area. For several diadromous species, however, mortality rates assessed using mean annual water temperature is not particularly meaningful. The European eel and the Atlantic salmon, for example, migrate several thousand kilometers in their life, spending some parts of their life in freshwater and some in the Atlantic Ocean, thereby crossing numerous climate regions with very different water temperatures. For those species, and if data availability was satisfying, mortality rates were extracted from em- pirical observations and population models.

Anadromous spawners, like salmon, of which most specimens die after spawning were con- sidered with their average sea winter mortality. Of the catadromous eel, which migrates to the Sargasso Sea for spawning, the mortality of the freshwater feeding stage (“yellow eel”) was used.

727830 FIThydro - Deliverable 1.1 - Page 5 of 26

Figure 1: Distribution of data obtained for natural adult fish mortality among all species; light grey= relative abundance distribution of mortality rates, thin lines 10–90% percentile, dark grey bar= inter-quartile distance (25–75% percentile), dot= median.

The recorded mortalities ranged between 3% in Beluga Huso huso and 96.7% in the goby Benthophiloides brauneri. The data distribution of natural mortality rates among all species is shown in Figure 1 and revealed the following classes:

Table 1: Classification of natural adult fish mortality.

Class Mortality M (%) Remarks 1 >90 Lowest sensitivity 2 >60-90 3 >50-60 4 >40-50 5 >20-40 6 10-20 7 <10 Highest sensitivity

Species with very low natural mortality rates were considered most sensitive against human- induced mortality. In contrast, species with high natural mortality rates have evolved buffer

727830 FIThydro - Deliverable 1.1 - Page 6 of 26 mechanisms to recover and compensate such losses. Therefore, they are expectedly less sen- sitive against human-induced mortality too.

2.2.2 Fecundity

The fecundity of fishes is one of the inherent biological bottlenecks that determine recruitment, population growth and recovery after declines. Fecundity is therefore of prime interest in con- text of demographic evaluations and population dynamics. For this analysis, fecundity was de- termined as the total number of eggs produced by one mature female per year. Depending on the species, but also on the female’s age, size and condition, the number of eggs can range from a few dozen to several million. For the species studied the reported average egg numbers ranged between 26 in the small goby Benthophiloides brauneri and 5,000,000 in the flathead grey mullet Mugil cephalus.

Figure 2: Distribution of fecundity expressed as eggs per female among all species; light grey= relative abundance distribution of fecundity, thin lines 10–90% percentile, dark grey bar= inter- quartile distance (25–75% percentile), dot= median.

The density and distribution among all evaluated species is shown in Table 2 and resulted in the following classification:

727830 FIThydro - Deliverable 1.1 - Page 7 of 26

Table 2: Classification of fecundity.

Class Fecundity (eggs/female) Remarks 1 >1,000,000 Lowest sensitivity 2 >100,000-1,000,000 3 >60,000-100,000 4 >10,000-60,000 5 >5000-10,000 6 1000-5000 7 <1000 Highest sensitivity

The potential to recover from population lows increases with egg number, i.e. fecundity of a species. Accordingly, fish species with naturally high fecundity per female were considered less sensitive. However, it must be noted that egg number is only one factor that determines recruit- ment and realized reproduction rate. Beside the egg numbers, recruitment is determined also by the fertilization rate as well as survival rates of eggs, larvae and juveniles. Therefore, egg numbers indicate in particular the reproductive potential of a species.

2.2.3 Annual recruits per female and parental care

As mentioned above, the reproductive potential of species according to their egg numbers be- comes rather immediately reduced by the number of unfertilized eggs and further mortality. Until the hatchlings reach the state of free feeding and swimming larvae, there are typically further three discrete periods of elevated mortality observed: shortly after egg activation, during hatching, and at final yolk resorption (Kamler 2005). In addition, the free-swimming juveniles face an initially high mortality mainly due to , temperature and oxygen stress, which exponentially declines with larvae growth (Schiemer et al. 2002). These various influences commonly result in a cumulative, overall juvenile fish mortality rate of more than 99% after the first year. Typical survival rates after the first year range between 0.01% and 2% depending on the species, offspring density, parental care as well as environmental factors, especially habitat availability and temperature (Schiemer et al. 2002).

Numerous studies have shown that parental care in form of nest building, guarding, mouth breeding and bearing (reviewed by Balon 1975) will increase the survival of eggs and juveniles. However, this advantage has hardly being quantified in offspring numbers or survival rates. Therefore, in case of parental care the survival rates have been across the board increased by a factor ten for the minimum estimate and doubled for the maximum estimate. For example, an estimated total 0+ mortality after one year of 99%-99.9% (corresponding to 0.1%-1.0% survival rate) has been lowered to 98.0%-98.9% (1.1%-2.0% survival) if a species performs some kind of parental care.

The annual recruits per female were calculated by multiplying the juvenile (0+) survival rate (based on the degree of parental care) with the average fecundity (egg number) of a species and dividing the product by the average age at first maturity. The latter step accounts for addi- tional mortality from the juvenile to the mature individual and first spawning, which increases with generation time.

727830 FIThydro - Deliverable 1.1 - Page 8 of 26

Figure 3: Distribution of minimum and maximum number of offspring per female and year; light grey= relative abundance distribution of number of offspring, thin lines 10–90% percentile, dark grey bar= inter-quartile distance (25–75% percentile), dot= median.

To calculate the final score, the respective minimum and maximum values (Fig. 4) were first classified according to Table 3 and then the median of these two scores calculated to determine the final classification. For species without any information of their breeding behavior, the two scores of minimum and maximum recruits/ per female with and without parental care were averaged.

Table 3: Classification of minimum and maximum annual recruits/female.

Class Recruits/female (min) Recruits/female Remarks (max) 1 >1000 >10000 Lowest sensitivity 2 >100-1000 >1000-10000 3 >50-100 >100-1000 4 >10-50 >50-100 5 >5-10 >10-50 6 1-5 5-10 7 <1 <5 Highest sensitivity

727830 FIThydro - Deliverable 1.1 - Page 9 of 26

Minimum and maximum estimates of recruits per female and year ranged from 0.03 to 1391.4 and 0.5 to 14,000, respectively, both in the goby Benthophiloides brauneri and the golden grey mullet Chelon aurata.

The distribution of both data sets is shown in Figure 3. The resulting classification is shown in Table 3. Species with an increasing number of annual offspring were considered less sensitive to disturbance than species with fewer annual offspring per female.

2.2.4 Age at first maturity

The age at first maturity strongly influences a species’ reproduction rate and ability of a popu- lation to recover from depletion. It often differs between males and females, with males of a species reaching maturity mostly one year earlier than females. We obtained multiple values for this trait from literature and averaged them in our analysis. Average age at first maturity reported ranged from 0.1 years in the eastern mosquitofish Gambusia holbrooki – a species not native to Europe – to 18.7 years in the Siberian sturgeon Acipenser baeri. The distribution of age at first maturity data is displayed in Figure 4. The classification scheme for this trait is shown in Table 4.

Figure 4: Distribution of age at first maturity among all species; light grey= relative abundance distribution of age at first maturity, thin lines 10–90% percentile, dark grey bar= inter-quartile distance (25–75% percentile), dot= median.

727830 FIThydro - Deliverable 1.1 - Page 10 of 26

Table 4: Classification of age at first maturity.

Class Age at first maturity Remarks (years) 1 <1 Lowest sensitivity 2 1-2 3 >2-4 4 >4-6 5 >6-8 6 >8-11 7 >11 Highest sensitivity

A fish that dies before reaching maturity does not contribute to recruitment, to the population and also to its evolution. Beside lack of spawning and overly losses of juveniles, the lack of spawners is another main reason for population declines. The longer it takes for a fish to reach maturity and to for the first time, the longer the fish will be exposed to multiple mortality risks and the higher becomes the chance to die before contributing to a new cohort at least once. Similarly, population recovery from spawner loss will take much longer if a species reaches maturity late. In contrast, reaching maturity early increases the chance to reproduce at least once before getting somehow removed from the spawners pool.

Therefore, sensitivity of a species has been ranked higher with increasing age of first maturity.

2.2.5 Maximum age

Old growing species are considered to have population structures evolved that depend on these old specimens and their long lifetime fecundity for their resilience and recovery. Maximum age is inversely correlated with the spawning frequency. For example, European sturgeons spawn only every second to forth year, the older ages less frequent. Further, old females have been frequently shown to contribute most to the offspring cohort, due to their significantly higher egg quality and other maternal effects.

Thus, maximum age was considered a suitable metric to indicate species’ resilience to disturb- ance. With increasing maximum age taxa were ranked more sensitive. Within the European fish fauna, the maximum ages reported ranged from 1 year in the goby Benthophiloides brauneri to 118 years in beluga Huso huso. The reported maximum age distribution is illustrated in Figure 5, its classification shown in Table 5.

727830 FIThydro - Deliverable 1.1 - Page 11 of 26

Figure 5: Distribution of maximum age among all species; light grey= relative abundance dis- tribution of maximum age, thin lines 10–90% percentile, dark grey bar= inter-quartile distance (25–75% percentile), dot= median.

Table 5: Classification of maximum age.

Class Maximum age (years) Remarks 1 <5 Lowest sensitivity 2 5-6 3 >6-8 4 >8-10 5 >10-15 6 >15-25 7 >25 Highest sensitivity

2.2.6 Maximum length and migration type

Maximum length is strongly correlated to maximum age. Although large fish are less prone to predation and other natural mortality factors, they are especially threatened by human-induced mortality, as e.g. fisheries attractiveness and harvest increases with size as well as mortality risk at hydropower or pumping stations. Further, movement distances significantly increase with fish size (Radinger & Wolter 2014). Therefore, bigger fish have a wider activity range,

727830 FIThydro - Deliverable 1.1 - Page 12 of 26 which increases the encounter probability with potential mortality factors, as e.g. a pumping station. Accordingly, bigger fish were considered more prone to mortality, i.e. more sensitive.

Fish length is also an important growth factor to calculate adult fish mortality according to Pauli (1980). Here we agreed on the maximum reported length, because an observed “common length” for a species has a higher variability throughout Europe. Thus, the maximum length went also into the calculations of the natural mortality factor M.

Figure 6: Distribution of maximum length among all species; light grey= relative abundance distribution of maximum length, thin lines 10–90% percentile, dark grey bar= inter-quartile dis- tance (25–75% percentile), dot= median.

In addition to the length related swimming ability and movement distance of a fish, many spe- cies show obligatory migrations as part of their life cycle. Most significant are spawning migra- tions. Diadromous species for example, obligatorily have to migrate between marine and fresh- water habitats to complete their life cycle. These species are most threatened by interruptions of their migration routes. Here we distinguish anadromous and catadromous species, which hatch in freshwaters migrate to the sea for growth and maturity and return back to freshwaters for spawning, or vice versa. Most prominent examples are Atlantic salmon and European eel for anadromous and catadromous species, respectively. Potamodromous species show also obligatory spawning migrations as part of their life cycle, but migrate only within freshwater habitats, within . All obligatory migrating species essentially depend on the ecological continuity of streams and face a higher risk of encountering a migration barrier or a mortality

727830 FIThydro - Deliverable 1.1 - Page 13 of 26 factor. As mentioned above, the capability of dispersal depends on body length and thus, large- bodied migratory species were considered most sensitive and small resident species least sen- sitive.

Species were scored more sensitive according to their maximum length and then weighted, obligatory migrants by 1 times the respective score value and resident species by 0.2 times the respective score value. If there was no information on the migration type available, the score of maximum length was multiplied with the median of 0.2 and 1. Maximum length of European lampreys and fishes ranges between 3.2 cm in the Delta dwarf goby Knipowitschia cameliae and 800 cm in Beluga Huso huso. The distribution of the data is shown in Figure 6, the resulting classification is displayed in Table 6.

Table 6: Classification of maximum length.

Class Maximum length (cm) Remarks 1 <5 Lowest sensitivity 2 5-15 3 >15-25 4 >25-40 5 >40-50 6 >50-60 7 >60 Highest sensitivity

2.3 Final scoring For each species, the resulting 6 trait scores were averaged and rounded to the next integer. The theoretically possible overall sensitivity score for a taxon is therefore also a number be- tween 1 and 7.

After averaging the single scores, no taxon scored higher than 5.58 or lower than 2.32 on our initial 7-class scale, so we adjusted the classification system to five classes for the final sensi- tivity score. From 1-5, they are named “low sensitivity”, “moderate sensitivity”, “average sensi- tivity”, “high sensitivity” and “highest sensitivity”. To illustrate the process of scoring and classi- fying the taxa, three examples are described in detail: The highest biological sensitivity score, 5.58 (adjusted to the final 5th class), was assigned to two sturgeon taxa (Acipenser baeri and Acipenser gueldenstaedtii) and the common barbel, barbus (see Table 7). They feature a comparably low natural mortality of 9.5, 6.3 and 9.1%. All three mortality rates were scored 7. Their maximum lifespan is 63, 48 and 20 years, scored 7, 7 and 6. All three species under- take migrations, so their maximum lengths of 200, 300 and 120 cm (scored 7, 7 and 6) were averaged with full weight. They mature at around 19, 13 and 4 years (scored 7, 7 and 3) and produce roughly 740,000, 220,000 and 8000 eggs per female and year (scored 2, 2 and 5), resulting in about 30 to 400, 16 to 160 and 2 to 20 offspring per female and year (scored 3.5, 3.5 and 5.5). They are classified EN, CR and LC in the IUCN Red List.

2.4 The IUCN conservation status In addition to the life history data, we obtained the conservation status for each species ac- cording to the European Red List of Freshwater Fishes (Freyhof and Brooks, 2011) by the

727830 FIThydro - Deliverable 1.1 - Page 14 of 26

International Union for Conservation of Nature and Natural Resources (IUCN). The IUCN threat categories are primarily based on population trends of a species, but also takes the area of occupancy, total population size (here “population” refers to all adult specimen of a species) and a quantitative threat analysis (which is partly based on a taxon’s life-history traits) into account. The result quantifies the species’ relative extinction risk. Within this method, species are either extinct, extinct in the wild, regionally extinct, critically endangered (“CR”), endangered (“EN”), vulnerable (“VU”), near-threatened (“NT”), least concern (“LC”), data deficient (“DD”), not applicable or not evaluated (“N/A”), see Figure 7.

The conservation status provides a proxy for the conservation value of a species. As such, it provides an additional assessment criterion for human-induced mortality besides the species’ autecology.

Figure 7: Structure and hierarchy of the IUCN threat categories. Obtained from www.iu- cnredlist.org

727830 FIThydro - Deliverable 1.1 - Page 15 of 26

3. Results 3.1 Sensitivity analysis The sensitivity scores could be assigned to a total of 192 fish and lamprey species occurring in Europe. Of all analysed taxa, 3 were classified of “highest sensitivity” (class 5), 37 of “high sensitivity” (class 4), 82 of “average sensitivity” (class 3), 67 of “moderate sensitivity” (class 2), and 3 of “low sensitivity” (class 1), see Figure 8.

The complete trait database and classification for each species including the references and literature database (Endnote) is available at IGB and becomes publicly available at the project website after publication.

Particularly sensitive and therefore assigned to the highest sensitivity class are the two stur- geon species Acipenser baeri and Acipenser gueldenstaedtii as well as the common barbel, Barbus barbus, with a score of 5.58 each.

Figure 8: Distribution of 192 species in the final 5 sensitivity classes. Numbers represent spe- cies counts.

The high sensitivity class includes all other analysed sturgeon species, several salmonids like the Adriatic trout Salmo obtusirostris (5.33), brook trout (Salvelinus fontinalis, 5.25), brown trout (Salmo trutta, 4.83), Atlantic salmon (Salmo salar, 4.92), grayling (Thymallus thymallus, 4.75), and the whitefish (Coregonus maraena, 4.92). Furthermore, this class contains several of the typically riverine cyprinids, like ide (Leuciscus idus, 5.08), chub ( cephalus, 4.75), com- mon nase ( nasus, 4.92), and the Mediterranean barbel species. As expected, further migratory species belong to this class too: sea lamprey (Petromyzon marinus, 4.75), lamprey (Lampetra fluviatilis, 4.67) and the European eel (Anguilla anguilla, 5.00).

The average sensitivity class contains many rather common species, but also several species of high conservation value listed in the Habitats Directive (92/43/EEC), as e.g. asp (Leuciscus

727830 FIThydro - Deliverable 1.1 - Page 16 of 26 aspius, 4.17), ( alosa, 4.08), twait shad (Alosa fallax, 4.08), and zingel (, 3.92).

The complete species list including the individual scores is shown in Table 7.

Table 7: Calculated and rounded results of the sensitivity analysis for 192 fish and lamprey species. The species are arranged from highest (class 5) to lowest (class 1) sensitivity.

Class Score Species Class Score Species

5.58 Acipenser baeri 4.35 Phoxinellus alepidotus 5.58 Acipenser gueldenstaedtii 4.33 Cyprinus carpio

sens. highest highest 5.58 Barbus barbus 4.33 Ctenopharyngodon idella 5.42 Acipenser nudiventris 4.33 Lethenteron zanandreai 5.42 bocagei 4.30 Petroleuciscus borysthenicus 5.33 Salmo obtusirostris 4.25 lucioperca 5.25 Acipenser stellatus 4.25 Rutilus rutilus 5.25 Polyodon spathula 4.25 Pelecus cultratus 5.25 Salvelinus fontinalis 4.25 Eudontomyzon danfordi 5.17 Huso huso 4.23 Phoxinellus dalmaticus 5.17 Acipenser sturio 4.22 Leuciscus leuciscus 5.17 Acipenser oxyrinchus 4.22 Barbus meridionalis 5.17 Platichthys flesus 4.22 Squalius pyrenaicus 5.08 Acipenser ruthenus 4.22 Scardinius dergle

5.08 Leuciscus idus 4.18 eurycephalus 5.08 Luciobarbus sclateri 4.17 Leuciscus aspius 5.08 Luciobarbus comizo 4.13 Mesogobius batrachocephalus 5.00 Anguilla anguilla 4.13 Squalius janae 5.00 Acipenser naccarii 4.13 Barbus petenyi

average sensitivity average

5.00 Hucho hucho 4.10 Cottus gobio

high sensitivity high 4.92 Chondrostoma nasus 4.08 Ballerus ballerus 4.92 Coregonus maraena 4.08 Alosa fallax 4.92 Salmo salar 4.08 Osmerus eperlanus 4.83 Salmo trutta 4.08 Carassius carassius 4.75 Squalius cephalus 4.08 Alosa alosa 4.75 Hypophthalmichthys nobilis 4.08 Rutilus heckelii 4.75 Thymallus thymallus 4.08 Luciobarbus steindachneri 4.75 Barbus plebejus 4.07 Micropterus salmoides 4.75 Petromyzon marinus 4.05 Gymnocephalus schraetser 4.75 Pseudochondrostoma polylepis 4.05 Barbus carpathicus 4.75 Squalius tenellus 4.00 Lota lota 4.67 lucius 3.98 Abramis brama 4.67 Lampetra fluviatilis 3.98 Ameiurus melas 4.58 Mylopharyngodon piceus 3.98 Cottus poecilopus

727830 FIThydro - Deliverable 1.1 - Page 17 of 26

Class Score Species Class Score Species 4.58 Rutilus frisii 3.92 Zingel zingel 4.58 Pseudochondrostoma duriense 3.92 Vimba vimba 4.58 Alburnus chalcoides 3.92 Alosa immaculata 4.55 Squalius zrmanjae 3.92 Eudontomyzon vladykovi 4.52 Lampetra planeri 3.85 Telestes souffia 4.50 Dicentrarchus labrax 3.83 Mugil cephalus 4.43 Telestes turskyi 3.83 Hypophthalmichthys molitrix 4.42 Salmo labrax 3.83 Liza haematocheilus 4.42 Pseudochondrostoma willkommii 3.83 Sabanejewia romanica 4.42 Parachondrostoma toxostoma 3.80 Lepomis gibbosus 4.42 Rutilus virgo 3.80 Squalius svallize 4.42 Squalius illyricus 3.77 elongata

average sensitivity average 4.42 Luciobarbus microcephalus 3.75 Blicca bjoerkna 4.38 Squalius microlepis 3.75 Alburnus alburnus 3.75 Eudontomyzon mariae 3.35 Squalius torgalensis 3.73 Romanogobio vladykovi 3.33 Ballerus sapa 3.73 Rhodeus amarus 3.32 Gobio lozanoi 3.73 Cobitis calderoni 3.32 Cobitis paludica 3.72 Scardinius plotizza 3.32 Pungitius pungitius 3.72 Chondrostoma knerii 3.32 Knipowitschia panizzae 3.68 Gobio hettitorum 3.27 Perccottus glenii 3.68 Romanogobio benacensis 3.25 Gymnocephalus cernua 3.67 Carassius gibelio 3.25 Chelon aurata 3.67 Chelon ramada 3.23 Achondrostoma salmantinum

3.65 Scardinius erythrophthalmus 3.23 valsanicola 3.65 Phoxinus phoxinus 3.22 Sander volgensis 3.65 Gobio obtusirostris 3.18 Rutilus aula 3.65 Chelon saliens 3.18 Syngnathus abaster 3.62 Perca fluviatilis 3.18 Ponticola kessleri

average sensitivity average 3.62 Ameiurus nebulosus 3.17 Carassius auratus

3.60 Ponticola syrman sensitivity moderate 3.15 Romanogobio albipinnatus 3.58 Alburnoides bipunctatus 3.15 Romanogobio uranoscopus 3.57 Oncorhynchus mykiss 3.15 Romanogobio belingi 3.57 Cobitis illyrica 3.15 Knipowitschia croatica 3.57 Cobitis elongatoides 3.10 Alburnus neretvae 3.52 Rutilus basak 3.08 Gymnocephalus baloni 3.52 Barbatula barbatula 3.08 Sabanejewia bulgarica 3.52 Achondrostoma oligolepis 3.08 Iberochondrostoma lemmingii 3.50 Rutilus pigus 3.08 Pomatoschistus microps 3.50 Aulopyge huegelii 3.02 Iberochondrostoma lusitanicum

e

ty

se

nsi 3.48rat Silurus glanis 2.98 Squalius aradensis

tivi

de-

mo

727830 FIThydro - Deliverable 1.1 - Page 18 of 26

Class Score Species Class Score Species 3.48 Padogobius bonelli 2.98 Romanogobio kessleri 3.45 Alosa maeotica 2.98 Sabanejewia baltica 3.43 Neogobius melanostomus 2.98 Iberochondrostoma almacai 3.43 Babka gymnotrachelus 2.98 Anaecypris hispanica 3.43 Atherina boyeri 2.98 Knipowitschia longecaudata 3.42 Alosa tanaica 2.98 Benthophilus stellatus 3.42 Clupeonella cultriventris 2.93 Iberocypris alburnoides 3.42 Alburnus arborella 2.93 Neogobius fluviatilis 3.42 Cobitis taenia 2.92 Gambusia holbrooki 3.40 Tinca tinca 2.90 Pseudorasbora parva 3.40 Gasterosteus aculeatus 2.90 Gasterosteus gymnurus 3.40 Cobitis jadovaensis 2.90 Benthophiloides brauneri 3.40 Cobitis narentana 2.85 Umbra krameri 3.40 Achondrostoma occidentale 2.82 Leucaspius delineatus 3.40 Sabanejewia balcanica 2.80 Misgurnus fossilis 3.40 Knipowitschia mrakovcici 2.73 Proterorhinus semilunaris 3.40 Pungitius platygaster 2.73 Knipowitschia caucasica 3.38 Squalius carolitertii 2.70 Knipowitschia cameliae

3.35 Gobio gobio 2.40 Knipowitschia radovici 3.35 Zingel 2.35 Zosterisessor ophiocephalus

3.35 Achondrostoma arcasii sens. low 2.32 Percarina demidoffi

3.2 Sensitivity-IUCN matrix The risk classification of the investigated 192 species according to the IUCN Red List of Euro- pean freshwater fish (Freyhof & Brooks 2011) revealed that most species are of least concern (LC), i.e. they are currently not threatened (Fig. 9). The correspondence between the IUCN risk classification and the sensitivity scoring based on the species’ life history traits was rather weak. For example, of the 17 and 13 species classified critically endangered and endangered, re- spectively, only 2 and 9 are of highest and high sensitivity, respectively against human induced mortality (Figure 9).

The full species-specific correspondence matrix between IUCN threat categories and biological scores of sensitivities against mortality is provided as Appendix.

727830 FIThydro - Deliverable 1.1 - Page 19 of 26

100% 90% 80% 70% 60% 50% 40%

30% Proportion per class per Proportion 20% 10% 0% highest high average moderate low sensitivity sensitivity sensitivity sensitivity sensitivity CR 1 6 4 6 0 EN 1 3 5 4 0 VU 0 5 9 8 1 NT 0 2 5 0 1 LC 1 18 53 47 1 DD 0 2 0 1 0 N/A 0 1 6 1 0

Figure 9: Distribution of 192 fish and lamprey species among the 5 sensitivity classes ranging from highest sensitivity (5th) to low sensitivity (1st) and the IUCN Red List categories. CR: Critically endangered, EN: Endangered, VU: Vulnerable, NT: Near threatened, LC: Least con- cern, DD: Data deficient, N/A: Not evaluated. While the sensitivity index derived here solely relies on biological and life history traits of spe- cies, which provide resistance and resilience against natural mortality, the IUCN Red List clas- sification is based on the observed extinction risk of a species, taking criteria into account like long-term and short-term population trends, area of occupancy and total population size. How- ever, considering the IUCN threat status here allows also accounting for conservation value, which is in accordance with national or international species conservation laws, as e.g. the EU Habitat Directive (92/43/EEC).

Based on both classifications, the following 16 species are of particular conservation and pro- tection interests, because they are of at least high sensitivity against mortality and already threatened at the European scale: the sturgeon species Huso huso, Acipenser gueldenstaedtii, A. baeri, A. naccarii, A. nudiventris, A. stellatus, A. sturio, and A. ruthenus, the paddlefish Polyodon spathula, the migratory whitefish Coregonus maraena, the European eel Anguilla anguilla, the Danubian salmon Hucho hucho, the Adriatic trout Salmo obtusirostris, the Luciobarbus comizo, the Iberian chub Squalius tenellus, and the Northern straight- mouth nase Pseudochondrostoma duriense.

727830 FIThydro - Deliverable 1.1 - Page 20 of 26

Accordingly, we found for all biogeographic regions of Europe at least one species of high sensitivity and conservation value, which might serve as target species or indicator in environ- mental assessments and planning.

The full sensitivity-conservation matrix is provided in the Appendix.

4. Conclusions Quantifying and comparing the species’ inherent biological sensitivity allowed to identify those species, which are less resilient against disturbances and thus, most sensitive to mortality. Accordingly, species found highly sensitive against mortality should provide especially diag- nostic indicators for environmental assessments of human alterations and target species for environmental planning and rehabilitation.

Six life history traits related to species’ resilience and recovery from disturbances have been compiled for as much as possible fish and lamprey species occurring in European freshwaters. Although the data availability was rather poor for a lot of rare and endemic species and several trait values had to be calculated from proxies or concluded from closely related species, we finally compiled life-history data for 192 European fish and lamprey species. Each trait was classified based on its character distribution among all species and scored at a scale of seven. For each species the scores of all traits were averaged resulting in a species-specific resilience, respectively sensitivity to mortality. All 192 species were classified according to their scores at an adjusted scale of five, from 1= low sensitivity to 5= highest sensitivity.

In total 40 species were assigned to the high and highest sensitivity classes representing all biogeographic regions and all major river catchments of Europe. This allows using similarly derived, comparable and calibrated sensitive indicators throughout Europe for environmental assessments and planning, e.g. in relation to hydropower projects.

It further provides a bunch of sensitive species as potential candidate species for further pop- ulation modelling and assessments of population impacts at regional scales. Population effect models for candidate species identified will be developed in Task 1.2 of WP1. The candidate species will also serve to develop the Decision Support tool in Task 1.4 of WP1 as well as habitat suitability models in WP3 and impact assessment tools in WP4.

The complete trait database forms the biological basis for the Population Hazard Index, which will be developed in Task 1.3 of WP1.

727830 FIThydro - Deliverable 1.1 - Page 21 of 26

5. References Balon, E. K. (1975). Reproductive Guilds of Fishes: A Proposal and Definition. Journal of the Fisheries Research Board of Canada, 32(6), 821-864.

Freyhof, J. and Brooks, E. 2011. European Red List of Freshwater Fishes. Luxembourg: Pub- lications Office of the European Union.

Kamler, E. (2005). Parent–egg–progeny Relationships in Teleost Fishes: An Energetics Per- spective. Reviews in Fish Biology and Fisheries, 15(4), 399-421.

Pauly D. 1980. On the interrelationships between natural mortality, growth parameters, and mean environmental temperature in 175 fish stocks. ICES Journal of Marine Science 39: 175- 192.

Radinger, J. & Wolter, C. (2014) Patterns and predictors of fish dispersal in rivers. Fish and Fisheries 15: 456-473.

Schiemer, F., Keckeis, H., & Kamler, E. (2002). The early life history stages of riverine fish: ecophysiological and environmental bottlenecks. Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology, 133(3), 439-449.

727830 FIThydro - Deliverable 1.1 - Page 22 of 26

Appendix Table A1: Species sensitivity – IUCN conservation status matrix for 192 fish and lamprey species occurring in Europe. DD: Data deficient. The species are arranged from highest (class 5) to lowest (class 1) sensitivity.

Biological sensitivity highest sensitivity high sensitivity average sensitivity moderate sensitvitiy low sensitivity

Acipenser gueldenstaedtii Acipenser naccarii Telestes turskyi Cobitis jadovaensis Acipenser nudiventris Cobitis illyrica Iberochondrostoma almacai Acipenser stellatus Gobio hettitorum Iberochondrostoma lusitanicum Acipenser sturio Phoxinellus dalmaticus Knipowitschia cameliae Anguilla anguilla Knipowitschia mrakovcici

Critically endangered Huso huso Romanichthys valsanicola Acipenser baeri Hucho hucho Aulopyge huegelii Achondrostoma occidentale

Salmo obtusirostris Cobitis calderoni Achondrostoma salmantinum Squalius tenellus Phoxinellus alepidotus Anaecypris hispanica Romanogobio benacensis Squalius torgalensis Endangered Squalius microlepis Acipenser ruthenus Pseudochondrostoma willkommii Cobitis narentana Knipowitschia radovici Coregonus maraena Alosa immaculata Achondrostoma arcasii Luciobarbus comizo Chondrostoma knerii Cobitis paludica

IUCN RedIUCN List classification Polyodon spathula Cyprinus carpio Iberochondrostoma lemmingii Pseudochondrostoma duriense Luciobarbus microcephalus Iberocypris alburnoides Luciobarbus steindachneri Knipowitschia croatica Vulnerable Parachondrostoma toxostoma Squalius aradensis Squalius janae Umbra krameri Squalius svallize

d

ar ar

Ne

thr

ene Acipenser oxyrinchus Barbus meridionalis Percarina demidoffi eat-

727830 FIThydro - Deliverable 1.1 - Page 23 of 26

Biological sensitivity highest sensitivity high sensitivity average sensitivity moderate sensitvitiy low sensitivity Squalius zrmanjae Hypophthalmichthys molitrix Sabanejewia romanica Scardinius dergle Squalius illyricus Barbus barbus Alburnus chalcoides Abramis brama Alburnus arborella Zosterisessor ophiocephalus Barbus plebejus Achondrostoma oligolepis Alburnus neretvae Chondrostoma nasus Alburnus alburnus Alosa maeotica Dicentrarchus labrax Alosa alosa Alosa tanaica Esox lucius Alosa fallax Atherina boyeri Lampetra fluviatilis Ameiurus melas Babka gymnotrachelus Lampetra planeri Ameiurus nebulosus Ballerus sapa Leuciscus idus Ballerus ballerus Benthophilus stellatus Barbatula barbatula Carassius auratus

Luciobarbus sclateri Barbus carpathicus Chelon aurata Petromyzon marinus Barbus petenyi Clupeonella cultriventris Platichthys flesus Blicca bjoerkna Cobitis taenia

Least concern Pseudochondrostoma polylepis Carassius carassius Gambusia holbrooki Rutilus frisii Chelon ramada Gasterosteus aculeatus Salmo trutta Chelon saliens Gasterosteus gymnurus Squalius cephalus Cobitis elongata Gobio gobio Thymallus thymallus Cobitis elongatoides Gobio lozanoi Cottus gobio Gymnocephalus baloni LR/LC Cottus poecilopus Gymnocephalus cernua Salmo salar Eudontomyzon danfordi Knipowitschia caucasica Eudontomyzon mariae Knipowitschia longecaudata Eudontomyzon vladykovi Knipowitschia panizzae

727830 FIThydro - Deliverable 1.1 - Page 24 of 26

Biological sensitivity highest sensitivity high sensitivity average sensitivity moderate sensitvitiy low sensitivity Gobio obtusirostris Leucaspius delineatus Gymnocephalus schraetser Misgurnus fossilis Lepomis gibbosus Neogobius fluviatilis Lethenteron zanandreai Neogobius melanostomus Leuciscus aspius Padogobius bonelli Leuciscus leuciscus Pomatoschistus microps Lota lota Ponticola kessleri Mesogobius batrachocephalus Proterorhinus semilunaris Micropterus salmoides Pseudorasbora parva Mugil cephalus Pungitius platygaster Osmerus eperlanus Pungitius pungitius Pelecus cultratus Romanogobio albipinnatus Perca fluviatilis Romanogobio belingi Petroleuciscus borysthenicus Romanogobio kessleri Phoxinus phoxinus Romanogobio uranoscopus Ponticola eurycephalus Rutilus aula Ponticola syrman Sabanejewia balcanica Rhodeus amarus Sabanejewia baltica Romanogobio vladykovi Sabanejewia bulgarica Rutilus basak Sander volgensis Rutilus heckelii Silurus glanis Rutilus pigus Squalius carolitertii Rutilus rutilus Syngnathus abaster Rutilus virgo Tinca tinca Salmo labrax Zingel streber Sander lucioperca

727830 FIThydro - Deliverable 1.1 - Page 25 of 26

Biological sensitivity highest sensitivity high sensitivity average sensitivity moderate sensitvitiy low sensitivity Scardinius erythrophthalmus Scardinius plotizza Telestes souffia Vimba vimba Zingel zingel

Hypophthalmichthys nobilis Benthophiloides brauneri DD Mylopharyngodon piceus Salvelinus fontinalis Alburnoides bipunctatus Perccottus glenii

Carassius gibelio Ctenopharyngodon idella Liza haematocheilus

Not evaluated Oncorhynchus mykiss Squalius pyrenaicus

727830 FIThydro - Deliverable 1.1 - Page 26 of 26