Post-release stranding rates of stocked allis shad (Alosa alosa) larvae exposed to surface wave action

Post-release stranding rates of stocked allis shad (Alosa alosa) larvae exposed to surface wave action

By S. Stollⁱ,2 and P. Beeck³

I Limnologicallnstitute. University of Konstanz. Konstanz. Germany; 2 Biodiversitdt und Klima Forschungszentrum (BiK F). . Frankfurt. Germany; 3Stijtung Wasserlau;: Geschiljisstelle EU Life Projekl Maifisch. Aquazoo Uibbecke Museum. Diisseldolj Germany

Summary

The objectives of the present study were to test the effects of sudden wave action caused by ships on newly released larval shad thriving along shallow river beaches because heavy shipping traffic has developed in many shad river systems.

Experiments were conducted in a wave tank simulating a cross section of the river shore. Up to 17% of the one week old allis shad stranded within the first 7 h after their release, but with increasing age of the released larvae, stranding rates decreased.

Stranding rates were always highest directly after the release of fish and decreased throughout the experimental period of 7 h.

The stranding rates were further influenced by weather, with higher stranding rates when it was sunny and when water temperatures were low. These results can be used to improve release strategies and stocking procedures of shad larvae and inay probably be applied to other larval fish species as well.

Introduction

Allis shad (A/osa alosa) were distributed across all Western Europe 100 years ago. However, their distribution range has diminished dramatically due to overfishing, water pollution and river development (Bagliniereet ai., 2003). Currently, efforts are made to re introduce allis shad into different rivers of their former distribution range, e.g. the Rhine system (Beeck et aI., 2009), where in the years from 2008 to 2010, approximately 4.8 million allis shad larvae were released. The re introduction project is based on the experience with the restoration of the closely related American shad (Alosa sapidissima) populations in freshwater systems on the North American East Coast. There, millions of I 4 weeks old shad larvae are released into the rivers every year and as a consequence shad populations recovered, e.g. from a few hundred to more than two hundred thousand individuals in the Susquehanna system (Hendricks, 2003).

However, the abiotic environment of the river Rhine and other rivers has changed since allis shad disappeared at the beginning of the 20th century. Many of them became intensely used shipping routes. At the middle and lower Rhine, on average every fifth minute a ship passes, creating surface waves with potential effects for the released allis shad larvae. Thus.

knowledge 011 the eHects of waves on the released shad larvae is valuable to optimize stocking routines. Insights gained from allis shad larvae may also be transferred to stocking of other fish species as well.

In this study, we tested whether wave induced stranding of allis shad larvae does occur and which environmental variables mediate stranding rates.

Materials and methods Allis shad

Allis shad larvae and juveniles were provided by MIGAOO (Association pour la restauration et la gestion des poissons migrateurs du basin de la Garonne et de la Oordogne). Adult fish were caught in the river Garonne during spawning migration. Fish were transported to the hatchery and eggs were obtained by artificial reproduction. Eggs were incubated at 20°e. Yolk sac larvae were transported to the Limnological Institute at the University of Konstanz where the experiments took place. Fish were kept in round 20 I holding tanks with gentle circular current. Water temperature was always 20°e.

Fish were fed with artemia and commercial powder food for fish larvae. Three age classes of allis shad were tested in the experiments, I week (8 12 days, 11.6 ± 1.3 mm (mean ± SO», 2 weeks (15 20 days, 13.5 ± 1.8 mm), and 3 weeks old individuals (21 27 days, 15.5 ± 1.9 mm).

Experimental setup and procedures

A wave mesocosm of the Limnological Institute, University of Konstanz was used for the study on habitat use patterns of allis shad. The mesocosm had a base dimension of lO x I m and a water depth of 0.84 m (Fig. I). A slope was installed at one end, simulating a littoral zone. The slope was constructed using a metal grid, covered by a thick canvas and topped with a 10 15 cm deep layer of gravel and stones. The grain sizes used, I 2 cm and 6 20 cm, are representative of the natural substrata that dominate in many eulittoral areas of the river Rhine.

A wave machine was situated on the opposite side of the mesocosm. Waves were generated in pulses of I min, followed

20 em

1100 em 210 em

s i l l ^{230 em 310 em 130 em} 1

=

1000 em

Fig. I. Design of the wave mesocosm. A Wave generator, B Water inlet, C Water outlet, D Net bag for collection of stranded fish, El E4 Position of the temperature loggers First publ. in: Journal of Applied Ichthyology ; 27 (2011), S3. - S. 41-44

Konstanzer Online-Publikations-System (KOPS) URL: http://nbn-resolving.de/urn:nbn:de:bsz:352-204440

42

by 4 min wave pause. This frequency of 12 wave pulses per hour imitated the average number of ship passages on the Rhine (11.3 ships per hour, averaged number of ship passages from Upper, Middle and Lower Rhine, personal observation).

The experimental setup delivered near harmonic waves with a maximum wave height, H of 0.13 m; wave period, T of 1.2 s and wave number, k of 2.8 m-^I at the bottom of the slope and 3.1 m -I near the surf zone. Maximum near bottom orbital velocities associated with surface waves, IIllla, were estimated using linear wave theory (Kundu and Cohen, 2002).

The resulting maximum current velocity was 0 0.05 m S-I at the bottom part of the mesocosm, around 0.10 m S-I in the midwater areas, and 0.15 up to 0.3 0.4 m S-I at the water surface, with increasing velocities in the shallow water (Beeck et aI., 2009).

With the limited dimensions of the mesocosm, the waves that were generated were shorter than typical waves produced by ships (Stoll and Fischer, 20 II). Furthermore, with the current velocities of the waves were at the lower end of the velocity spectrum induced by ships, which can reach 0.8 m S-I

(Arlinghaus et aI., 2002).

The area close to the wave machine was separated from the rest of the mesocosm by a frame strung with a 4 mm knot to knot mesh in order to prevent the fish from getting crushed in the paddle of the wave machine. With this mesh size, the mesh was not a real physical barrier to the small and slender fish.

Therefore the mesh barrier was continuously monitored throughout the experiments through the glass wall on one side of the mesocosm to check for larvae penetrating the mesh.

However, the vast majority of the larvae accepted the mesh as a wall and only very few 1 week old larvae

«

Fish that passed the net wall were immediately fished out of the wave machine compartment with a hand held dip net and transferred back to the other side of the barrier.

At the upper end of the slope, a net bag was installed. The opening of the net bag was well above the water level in calm conditions without waves. Approximately 2 I of water washed over the sill and into the net bag with every wave (Fig. I).

Water temperature was measured throughout the experi ment by Onset™ temperature loggers at four locations spread out in the mesocosm with a resolution of 12 h-^I.

The mesocosm was fed with water from Lake Constance in a flow through system, with a complete water exchange every 24 h. Water was introduced into the mesocosm on the slope side, and drained via an outlet situated at the opposite end of the mesocosm near the wave machine.

Per day, one experiment using 100 allis shad larvae was run.

To transfer the larvae from the holding tanks to the wave mesocosm, the larvae were scooped out of the holding tank with a I I beaker, and gathered in a 10 I bucket, where the water temperature was matched to the water temperature in the experimental tank over a period of I h. At 9:30, the larvae' were introduced into the mesocosm. After an acclimatization period of 30 min, the fish larvae were exposed to the wave scenario from 10:00 to 17:00. After periods of I h, the number of stranded fish in the net bag at the upper end of the slope was counted. Additionally, for each observation, the weather condition was noted, differentiating between sunny (direct sunshine), cloudy (less than two thirds of the sky covered with clouds, sun is hidden), overcast (more than two thirds of the sky is covered with clouds, sun is hidden) and rainy weather.

At the end of the experiment, the larvae were caught again with a dip net and transferred back to the laboratory. Larvae were only used once for the experiments.

Data analysis and statistics

Ln(x + I) transformed numbers of fish stranded at each observation were analyzed with a General Linea~ Model with fish age, time since release, weather, and water temperature as variables, using JMP software (SAS Institute Inc., Cary, NC, USA).

Results Abiotic variables

During the experiments, the water temperature ranged between 15.5 and 26.2°C. The waves prevented the establish ment of a strong temperature gradient in the water column. On average, the temperature difrerence between the warmest and the coldest part of the mesocosm was only 0.2°C. The four different weather conditions that were differentiated (sunny, cloudy, overcast, and rain) occurred with similar frequencies in experiments with one and 3 weeks old fish (expected propor tion: 25%, realized proportions: 18 35%) for each weather condition in both age classes. In experiments with 2 weeks old fish, however, sunny weather prevailed during 60% of the measurements, and no rainy weather occurred.

Stranding events

During the transfer of the larvae to the experimental tank, throughout the acclimatization period and the 7 h experimen tal period, no dead larvae or larvae with conspicuous swimming patterns that may be suggestive of injury were observed. Further, no stranding occurred in the acclimatiza tion period without waves.

Up to 17% of the fish used in the experiments stranded during the 7 h in the mesocosm (Fig. 2a). Stranding rates decreased significantly with fish age (Table I; Fig. 2a). The stranding rate also decreased with the time after the release of the fish into the mesocosm; 25% of the fish that stranded did so already within the first hour after the release (Table I; Fig. 2b). Projecting the linear decrease of stranding rate with the time since release, no further noteworthy stranding is assumed later than 8 h after release. Further, 65% more fish stranded at sunny weather compared to the average of the other weather types (Table I; Fig. 2c) and stranding rate was inversely related to water temperature (Table I; Fig. 2d).

Discussion

It was demonstrated that even low intensity surface waves bear a potential for substantial stranding of fish larvae, especially in the first hours after their release. Adams et al. (1999) point out that the likelihood of stranding is related to the behavioural response of fishes to receding water levels. Species that typically occur in littoral and backwater areas swam with the currents or passively drifted, and stranded only in low numbers, whereas the young of main channel fishes often exhibited a positive rheotaxis, swimming against receding waves in shallow waters and thus were more likely to become stranded. However, in the post release period, stranding rates decreased fast, which suggests that after a short orientation phase, the fish find their preferred open water habitat and accidental stranding ceases.

We showed that up to 17% of the total number of released allis shad larvae strand within the first 7 h after the release in a littoral area with moderate waves. Since with up to 80 cm S-I,

43

(a) (b)40

16

if

-; 12 R' = 0.91; P < 0.001 30 R' = 0.99; P < 0.001 OJ c

'5 _c

~ 8

20

en

4 10

0 5 10 15 20 25 30 5 6 7

Fish age (days) Time since release (h)

(c) 2 (d)10

A

~

0.21; P < 0.001 B 8

.'@ w ₆

OJ c ¹

'5 ₄

c ~

en

0 Fig. 2. Dependence of stranding rates

on fish age (a), the time since the release of the larvae in the mesocosm (b), weather (c), and water tempera ture (d). Error bars depict standard errors

Sun Clouds Overcast Rain 19 23 27

Table I

Results of General linear model analysis on stranding events.

R~ 0.44

Factor df SS F P

Fish age I 6.48 26.2 <0.001

Time since release I 3.40 13.7 <0.001

Weather 3 2.78 3.7 0.012

Temperature 1 2.36 9.5 0.002

the maximum current velocities that may be reached near shipping routes (Arlinghaus et aI., 2002) may even exceed the current velocities induced in this experiment (max. 30 40 cm S-I), the stranding rates in the field may even be underestimated by this experiment. Thus stranding through wave wash can kill a substantial proportion of the released fish larvae in the post release period.

In line with Leach and Houde (1999), our results suggest that careful choice of the release site is crucial in order to reduce losses. The results indicate that fish should be released in areas without waves, e.g. in oxbow lakes and side canals without navigation. In such habitats, the fish are protected during the short post release period in which they were susceptible to waves. Where such protected habitats are not available, it may be a good option to stock with older fish which have been shown to be less sensitive to waves. Also ship based stocking off the shoreline may produce better results than release of larvae from the shore in shallow water.

Howcvcr, also off shore stocking close to shipping routes may produce mortality, as Killgore et al. (200 I) have shown that shear stress in the vicinity of boat propellers can induce severe fish larvae mortality. The 'killing volume' (50% of the organisms killed during I min exposure) for a typical barge tow was estimated to be 22 ffi" per meter passage (Kennedy et aI., 1982; as cited in Holland, 1986), which, depending on the cross section of the river can lead to severe mortality.

Further, Crecco and Savoy (1984, 1985) showed that hydrodynamic stress, as it is induced by waves but also by

Weather Temperature (0C)

river flow does not only induce incidental mortality in the post release period, but may also cause long term effects, which were not captured by this study. They demonstrated that hydrodynamic stress hampers growth and survival of Amer ican shad.

Such long term effects have also been shown in different juvenile fish species, since hydrodynamic stress alters food accessibility (Stoll et aI., 2010), leading to changes in their energy budget and growth rates (Gabel et aI., 2011; Stoll and Fischer, 2011) and ultimately, to changes in their habitat choice (Stoll et aI., 2008).

Furthermore, water temperature and weather eondition affected the post release stranding rates of allis shad, with increased stranding rates at sunny weather and at lower water temperatures. Thus, the weather conditions should be consid ered if the stocking time table is flexible enough. Also Leach and Houde (1999) point out that timing of the stocking can be critical for return rate of American shad. They advise to avoid too early stocking, when water temperatures are still low, as growth and survival of American shad increases with temper ature. However, they also discourage late stocking in summer when predation rates on shad larvae are highest, decreasing their survival rates. Johnson and Ringler (1998) showed that predation mortality of released American shad larvae are highest when larvae are released at daytime, as most of their predators are optically oriented, and shad larvae are most vulnerable in the post release orientation phase. Thus also stocking at night is a good option.

Acknowledgements

The study was conducted within the framework of the EU Life Project LIFE 06 NAT / D / 00005 and financed by the LIFE financial instrument of the European Community and the research funding program 'LOEWE Landes Offensive wr Entwicklung Wissenschaftlich 6konomischer Exzellenz' of Hesse's Ministry of Higher Education, research, and the Arts.

Martin Wolf provided technical support with the wave generator. Wolfgang Nikolaus Probst, Daniela Harrer,

44

Michael Donner helped with the rearing of the fish larvae.

Reiner Eckmann kindly gave the permission to conduct these experiments at the facilities of the Limnological Institute of the University of Konstanz. Hilmar Hofmann provided an ADV to measure current velocities in the experimental tank and was a much appreciated dialogue partner on wave physics.

Conflict of interests

The authors have declared no potential conflict of interests.

References

Adams, S. R.; Keevin, T. M.; Killgore, K. J.; Hoover, J. J., 1999:

Stranding potential of young fishes subjected to simulated vessel induced drawdown. Trans. Am. Fish. Soc. 128, 1230 1234.

Arlinghaus, R.; Engelhardt, c.; Sukhodolov, A.; Wolter, c., 2002:

Fish recruitment in a canal with intensive navigation: implications for ecosystem management. J. Fish BioI. 61, 1386 1402.

Bagliniere, J. L.; Sabatie, M. R.; RochaI'd, E.; Alexandrino, A.;

Aprahamian, M. W., 2003: The allis shad A/osa a/osa: biology, ecology, range and status of populations. In: Biodiversity, status and conservation of the world's shads. K. E. Limburg, J. R.

Waldmann (Eds). American Fisheries Society, Symposium 35, Bethesda, Maryland, pp. 85 102.

Beeck, P.; Lages, E.; Stoll, S.; Hofmann, H.; Eckmann, R.; Jatteau, P., 2009: Der Einfluss von Oberflachenwellen und Bodensubstrat auf die Ei und Larvenentwicklung von Maifischen (A/osa a/osa).

Jahrestagung del' Deutsche Gesellschaft fur Limnologie (DGL) 2008. Deutsche Gesellschaft fur Limnologie, 329 333.

Crecco, Y.; Savoy, T., 1984: Effects of fluctuation in hydrographic conditions on the year class strength of American shad (A/osa sapidissima) in the connecticut river. Can. J. Fish. Aquat. Sci. 41, 1216 1223.

Crecco. V.; Savoy, T., 1985: Effects of biotic and abiotic factors on growth and relative survival of young American shad, A/osa sapidisima. Can. J. Fish. Aquat. Sci. 42, 1640 1648.

Gabel, F.; Stoll, S.; Fischer, P.; Pusch, M. T.; Garcia, X. F., 2011:

Waves affect predator prey interactions between Ilsh and benthic invertebrates. Oecologia 165, 101 109.

Hendricks, M. L., 2003: Culture and transplant of alosines in North America. In: Biodiversity, status and conservation of the world's shads. K. E. Limburg, J. R. Waldman (Eds). American Fisheries Society, Symposium 35, Bethesda, Maryland, pp. 303 312.

Holland, L. E., 1986: Eflccts of barge traffic on distribution and survival of ichthyoplankton and small fishes in the upper mississippi river. Trans. Am. Fish. Soc. 115, 162 165.

Johnson, J. H.; Ringler, N. H., 1998: Predator response to releases of American shad larvae in the Susquehanna river basin. Ecol.

Freshw. Fish. 7, 192 199.

Kennedy, D.; Harber, J.; Littlejohn, J., 1982: Effects of navigation and operation / maintainance of the upper mississippi river system nine foot channel on larval and juvenile fishes. Upper Mississippi River Basin Commission, Minneapolis, MN.

Killgore, K. J.; Maynord, S. T.; Chan, M. D.; Morgan, R. P. 11,2001: Evaluation of propeller induced mortality on early life stages of selected fish species. N. Am. J. Fish Mgmt. 21, 947 955.

Kundu, P. K.; Cohen, I. M., 2002: Fluid mechanics, 2nd edn.

Academic Press, London, UK, pp. 730.

Leach, S. D.; Houde, E. D .. 1999: Effects of environmental factors on survival, growth, and production of American shad larvae. J. Fish BioI. 54, 767 786.

Stoll, S.; Fischer, P., 20 II: Three diflerent patterns of how low intensity waves can affect the energy budget of littoral fish: a mesocosm study. Oecologia 165,567 576.

Stoll, S.; Fischer, P.; Klahold, P.; Scheif11acken, N.; Hofmann, H.;

Rothhaupt, K. 0., 2008: Effects of water depth and hydrody namics on the growth and distribution of juvenile cyprinids in the littoral zone of a large pre alpine lake. J. Fish BioI. 72, 100 I 1022.

Stoll, S.: Hofmann,. H.: Fischer, P., 2010: Effect of wave exposure dynamics on gut content mass and growth of young of the year fishes in the littoral zone of lakes. J. Fish BioI. 76, 1714 1728.

Author's address: Stefan Stoll, Biodiversitat und Klima Forschungs zentrum (BiK F), Senckenberganlage 25, 60325 Frankfurt, Germany.

Email: Stefan.Stoll@Senckenberg.de