The Influence of Un-ionized Ammonia on the Long-term Survival of Sockeye Salmon Eggs. Fisheries and Marine Service Technical Report No.

Transcription

1 The Influence of Un-ionized Ammonia on the Long-term Survival of Sockeye Salmon Eggs D. Paul Rankin Department of Fisheries and Oceans Fisheries and Marine Service Resource Services Branch Pacific Biological Station Nanaimo, British Columbia V9R 5K6 December 1979 Fisheries and Marine Service Technical Report No. 912

2 Fisheries and Marine Service Technical Reports These reports contain scientific and technical information that represents an important contribution to existing knowledge but which for some reason may not be appropriate for primary scientific (i.e. Journal) publication. Technical Reports are directed primarily towards a world wide audience and have an international distribution. No restriction is placed on subject matter and the series reflects the broad interests and policies of the Fisheries and Marine Service, namely, fisheries management, technology and development, ocean sciences, and aquatic environments relevant to Canada. Technical Reports may be cited as full publications. The correct citation appears above the abstract of each report. Each report will be abstracted in Aquatic SCiences and Fisheries Abstracts and will be indexed annually in the Service's index to scientific and technical publications. Numbers in this series were issued as Technical Reports of the Fisheries Research Board of Canada. Numbers were issued as Department of the Environment, Fisheries and Marine Service, Research and Development Directorate Technical Reports. The series name was changed with report number 715. Details on the availability of Technical Reports in hard copy may be obtained from the issuing estaqlishment indicated on the front cover. Service des peches et de la mer Rapports techniques Ces rapports contiennent des renseignements scientifiques et techniques qui constituent une contribution importante aux connaissances actuelles mais qui, pour une raison ou pour une autre, ne semblent pas appropries pour la publication dans un journal scientifique. II n'y a aucune restriction quant au sujet, de fait, la serie reflete la vaste gamme des interets et des politiques du Service des peches et de la mer, notamment gestion des peches, techniques et developpement, sciences oceaniques et environnements aquatiques, au Canada. Les Rapports techniques peuvent etre consideres comme des publications completes. Le titre exact paraitra au haut du resume de chaque rapport, qui sera publie dans la revue Aquatic Sciences and Fisheries Abstracts et qui figurera dans I'index annuel des publications scientifiques et techniques du Service. Les numeros de cette serie ont ete publies it titre de Rapports techniques de I'Office des recherches sur les pecheries du Canada. Les numeros , it titre de Rapports techniques de la Direction generale de la recherche et du developpement, Service des peches et de la mer, minis,tere de I'Environnement. Le nom de la serie a ete modifie it partir du numero 701. La page couverture porte Ie nom de I'etablissement auteur ou I'on peut se procurer les rapports sous couverture carton nee.

3 ERRATA - TECHNICAL REPORT NO. 910 Page 2 - AtZorchestes angusta Page 3 - " of pollution (Read et a ). Page 3 - A. angusta ERRATA Page 5-1. Line 2: "0.36 and 1.S2 mg N~-N.l "should be: "0.403 and 1.93 mg NH3 -N 1-1" Page S Table 1. Columns 1 and 2 should read: ;rreatment Theoretical Water control Buffer c"ontrol 1 2 o.os mg/-t

4 Fisheries and Narine Service Technical Report 912 December 1979 Tlili INFLUENCE OF UN-IONIZEIJ AMMONIA ON THE LONG-TERH SURVIVAL OF SOCKEYE SALMON E~GS by D. Paul Rankin Department of Fisheries and Oceans Fisheries and Marine Service Resource Services Branch Pacific Biological Station Nanaimo, British Columbia V9R SK6

5 - ii - (c) Minister of Supply and Services Canada 1979 Cat. no. F s 97-6/912 1SSN

6 - iii - ABSTRACT Rankin, U. P The inf hlt~nce of un-ionized anuuonia on the long-term survi val of sockt!ye salmon eggs. t.'ish. Mar. ::>erv. Tech. Rep. 912: 17 p. Sockeye salmon (Oncorhynchus nerka Walbaum) eggs were exposed, from fertilization to hatching, to constant levels of un-ionized anmonia (NH3) from to 4.05 rug NH3-N l-l at 100e and a ph of 8.4. Eggs were also raised in a water-only control (ph - 7.4) and a buffer-only control (ph 8.4) to determine base mortalities. Total (100%) mortality occurred at concentrations above mg NH3-N l-l. The median tolerance limit for the entire period from fertilization to hatching (62 days) was estimated at 0.11 rug NH3-N l-l. There was a linear increase in the time to 50% mortality frow 23 days at 4.05 rug NH3-N l-l to 34 days at 0.40 rug NH3-N l-l. Un-ionized ammonia and TRI::> buffer influenced hatching rates; 9.4% of tht! alevins exposed to 0.U99 mg NH3-N l-l hatched before any of those in the water-only control. In addition, ammonia and high ph affected size at hatching: alevins in the treatment and buffer-only control were significantly smaller than those in the water-only control. Key words: Un-ionized ammonia, survival, sockeye development, stress, 62-day TLm.

7 - iv -,, KESUI1E l{ankin, D. P The influence of un-ionized anmlonia on the long-term survival of sockeye sallllon eggs. Fish. Har. Servo Tech. l{ep. 912: 17 p. Des oeufs de saumon sockeye (Onchorhynchus nerka Walbaum) ont ete exposes, depuis la fertilisation jusqu'a l'eclosion, a des teneurs constantes d'ammoniac non ionise (NH3), situees entre 0,099 et 4,05 rug de NH3-N.l-l, a looe et a ph 8,4. Des oeufs ont aussi ete places soit dans l'eau (ph = 7,4), soit dans une solution tampon (ph = 8,4) afin de determiner leur taux de mortalite naturelle. Aux teneurs sup~rieures a 0,099 mg de NHrN.l-l, Ie taux de mortalite a ete de 100%.,,, La limite de tolerance moyenne au cours des 62 jours ecoules entre la fertilisation et l'eclosion a et~ evalu~e a 0,11 mg de NH3-N.l-l. Le taux de mortalite a augmente de fa~on lin~aire pour atteindre 50% apres 23 jours a 4.05 mg de NH3-N.l-l, et apr~s 34 ~ 0,40 mg de NH3-N.l-l. Dans les solutions d'ammoniac et tampon TRIS, l'eclosion s'est produite plus rapidement: 9,4% des alevins exposes a 0,099 mg de NH3-N.l-l etaient deja apparus avant qu'on en ait d~cele en milieu temoin (eau). La presence d'ammoniac et un ph eleve ont eu des effets sur la taille des alevins, ceux des solutions d'ammoniac ou tampon ayant ete' beaucoup plus petits que ceux qui n'avaient sejourne que dans l'eau. Mots cles: Ammoniac non ionise, survie, developpement du sockeye, stress, TLm 62 jours.

8 INTRODUCTION Ammonia is an end product of amino al!id metabolism. In its un-ionized form it is toxic to terrestrial and aquatic vertebrates (Warren and Schenker 1962; Rice and Stokes 1975). Terrestrial and marine vertebrates have evolved detoxification processes that incorporate ammonia in urea or uric acid. Freshwater fishes, which are constantly excreting water, do not detoxify ammonia (Campbell 1973). The excreted ammonia, which is highly soluble in the surrounding water, is diluted and rapidly carried away. In water, ammonia is present in two forms: ionized (NH4+) and un-ionized or free (NH3) ammonia. Together these two forms are referred to as total ammonia. The relative concentrations of the two forms are determined by temperature and ph (Emerson et al. 1975). Under natural conditions (neutral ph and moderate (10-12 ) temperatures), the non-toxic NH4+ usually prevails. However, increased alkalinity and/or an increase in temperature will increase the proportion of the toxic un-ionized form. Some work has been done on the short-term toxicity of ammonia to juvenile salmonids, particularly rainbow trout (Salmo gairdneri Richardson). Rice and Stokes (1975) estimated the 24-hr median tolerance limits (24-hr TLm) for the eggs and alevins of rainbow trout. They found that eggs and alevins up to 10 days after hatching (at 10 C) were not affected by un-ionized ammonia levels below 2.95 mg.l-1 The 24-h TLm was 1.05 mg l days after hatching. Thereafter, the 24-hr TLm decreased exponentially to mg l-1 40 days after hatching. Present salmonid incubation techniques concentrate eggs in limited flows of perfusing water. Under such conditions total ammonia concentrations in the water will increase as the eggs develop (Rice and Stokes 1974). Although most of the ammonia will be present in the ionized form (because of the ph and the water temperature), some ammonia will be present in the toxic un-ionized form. The purpose of this investigation was to estimate threshold levels of un-ionized ammonia causing egg mortality. In addition, the effect of ammonia and ph on daily hatching rates and the size of the surviving alevins was examined and results presented. To provide the maximum stress possible, the eggs were exposed from fertilization to hatching; the results approximate a 60-day TLm.

9 - 2 - MATERIALS AND METHODS Spawning sockeye salmon were beach-seined from Kennedy Lake, British Columbia, on November 9, The eggs and milt stripped from the adults were placed in separate plastic containers in styrofoam coolers. The coolers, containing the gametes and commercial freezer packs, were shipped to the Pacific Biological Station, Nanaimo, B.C. The temperature of the gametes upon arrival was Y.3 C. The eggs were fertilized approximately 11 hours after collection in the laboratory (November 9, 1970; 2131 hr P.S.T) and placed immediately in Plexiglass incubators. The incubators provide a flat velocity front and microturbulent flow, perfusing one layer of eggs normal to the direction of flow. Water flowed past the eggs at a bulk velocity of 310 cm h-1 (500 ml.min-1). Three incubators were connected in series and immersed in the contents of a 40-1 tank. Water was pumped from the tank, through the incubators and back into the tank for temperature and oxygen equilibration (10 C, saturation) using peristaltic pumps. A pulsing-flask between the pump and the first incubator steadied the flow of water. The order of the replicate incubators on line was changed daily to remove any bias associated with position in the flowing system. Seven tanks were set up as described to provide two controls and five trials: a water-only control, a 0.05 M "TRIS" buffer-only control (tri[hydroxymethyl] amino-methane) and five concentrations of un-ionized ammonia (0.08, 0.41, 0.82, 2.06, and 4.12 mg.l- 1 ). The ammonia solutions were made up using NH4Cl. The ph of the test solutions, with the exception of the water control (mean ph = 7.35) was maintained at approximately 8.4. The experiment at 0.82 mg NH3-N l-l was terminated prematurely because of a malfunction in the temperature control device. The contents of each of the seven 40-1 tanks were exchanged by the addition of fresh solution at the rate of 20 ml min-1 ( h-1). The components of the exchange solutions were contained in seven overhead reservoirs. One of the reservoirs, a overflowing tank, contained dechlorinated water. A second reservoir (100 1) contained stock buffer solution, which when mixed with water, produced a 0.05 M solution with ph 8.4 (Mahoney 1966). The other five reservoirs (2-1 glass bottles) contained stock NH4Cl solutions at concentrations appropriate to produce the required ammonia levels in the five test tanks. Water from the water reservoir was delivered to the water-only control at 20 ml min- 1 Water and buffer were delivered to the remaining six tanks at 8 and 10 ml min- 1, respectively. Five of these tanks received stock ammonia solution, from the appropriate ammonia reservoirs at the rate of 2 ml min-1 to give a total flow of 20 ml.min-1 Flow rates were controlled using a Manostat cassette pump, pulsing-flasks and flow meters. Ammonia and buffer stock solutions were prepared daily. The ph of the solutions in the seven tanks was measured every second day using a Fischer Accumet Expanded Scale Meter. The ph was adjusted, if necessary, by varying the concentration of HCl in the buffer stock solution.

10 - 3 - AMMONIA MEASUKEMENT Total ammonia levels were measured every second day using an ammonia-sensitive gas electrode (Urion 95-10) in conjunction with the ph meter. The electrode was placed in a stirred 100-ml sample. One milliliter of 10M NaUH was added to the sample to convert all ionized ammonia to NH3. The electrode was calibrated before each series of measurements by recording the electrode potentials of standard ammonia solutions (0.S2, 41.17, S2.4 mg total ammonia.l-l ) and construction of a calibration curve. The electrode potential of samples from each tank was determined following the same procedure, and total ammonia was calculated using the calibration curve. After total ammonia had been determined, the amount of un-ionized ammonia present at the sample ph and temperature was calculated from the following formulas (Emerson et ale 1975): where T is temperature in ok and pka = S T-l where f is the fraction of total ammonia present as NH3-N. The accuracy of the standards was checked after dilution using the Nesslerization Method (APHA 1976). Since the electrode functions through the diffusion of ammonia gas across a permeable membrane, the reaction time is dependent on the ammonia concentration in the sample. Below 0.07 mg total ammonia l-1, reaction time is very slow. l The number of opaque (dead) eggs in each incubator was recorded daily. The incubators were opened approximately every 4 days and the dead eggs removed and preserved. These eggs were examined to determine if fertilization had occurred, and the embryonic stage at death was recorded using the stages described by Vernier (1969). During hatching, numbers of alevins hatched and post-hatch mortality were recorded daily. At the time of 50% hatch 20 alevins were removed from the three incubators which contained surviving embryos. The alevins were anaesthetized with MS-222, the fork length was recorded, and the fish were returned to the incubator after recovery. When all embryos in each incubator had either hatched or died, mean egg mortality, mean number of infertile eggs, and mean number hatched were calculated for each treatment and control. ITotal ammonia concentration in the water-only control was 0.19 mg l-1 ( msnh3-n.l-l).

11 - 4 - RESULTS Un-ionized ammonia levels ranged from mg 1-1 (water-only control) to 4.05 mg NH3-N 1-1 (Table 1). The mean ph in the buffer-only control and the buffered trials ranged from 8.42 to 8.45; the ph in the water-only control was Total mortality ranged from 21.3% in the water control to 100% in the trials where un-ionized ammonia concentrations exceeded mg NH3-N.l-l- (Table 2). DISTRIBUTION OF MORTALITY The probit mortality-log time response curves (Fig. 1) indicated that the mortality rate did vary with time. In both the and 1.93 mg NH3-N l-1 treatments mortality decreased between days 20 and 40. In the mg NH3-N.l-1 trial there was a rapid increase in egg mortality after day 48: 30% of the total mortality occurred in the next 4 days. A similar increase occurred at that time in the 1.93 mg NH3-N.l-1 trial, however 96% of the eggs were already dead. Estimates of the time to 50% mortality derived from the probit distributions resemble a rectangular hyperbole as described by Alderdice and Brett (1957) (Fig. 2). This curve was derived by a least-squares regression of the reciprocal of the time to 50% mortality on the reciprocal of the un-ionized ammonia concentration to which the eggs were exposed. On this basis the un-ionized ammonia concentration expected to produce 50% mortality between fertilization and hatching, at 10 C, was estimated to be 0.08 mg NH3-N l-1. HATCHING RATES AND ALEVIN SIZE In addition to causing mortality, un-ionized ammonia and high ph may have affected the daily hatching rates (Fig. 3a-c). Although there were no differences in the time to 50% hatch, alevins began hatching earlier at the higher ammonia and ph levels. In the buffer-only control (0.023 mg NH3-N l-1) approximately 11% of the hatch emerged before the alevins in the water-only control started hatching. Approximately 9.4% of the hatch at mg NH3-N l-1 occurred before hatching commenced in the water-only control. The size of alevins at hatching was also influenced by ammonia and ph levels. The eggs exposed to ammonia produced alevins which were significantly shorter (18.65 mm at mg NH3-N.l-1 and mm at mg NH3-N l-1) than those in the water-only control (20.78 mm; t , p<o.ool). However, a four-fold increase in the ammonia concentration (from to mg NH3-N l-1) produced no significant difference in size (t = 1.29, p)0.05).

12 - 5 - DISCUSSION Mortality rates varied with time in the water-only eontrol, the 0.36 and 1.82 mg NH3-N I-1 treatments. Mortality rates did not vary in the buffer-only control, the or the 4.05 mg NH3-N l-1 treatments. It is possible that the observed changes in daily mortality rates were the result of eggs, which had died sometime earlier, turning opaque when disturbed during removal of "dead" eggs. The sockeye eggs did not survive in un-ionized ammonia concentrations above and including mg NH3-N l-1. Burkhalter and Kaya (1977) found that rainbow trout (Salmo gairdneri) egg mortalities were not affected by ammonia concentrations up to 0.37 mg NH3-N l-1 during 25-and 33-day exposures (one-half the exposure time used in the present experiment). Mortality occurring in the ammonia treatments consists of three components: natural mortality, mortality due to the increased ph, and mortality caused by the presence of ammonia. Although little data is available, I want to estimate each of the component mortalities mentioned above. Natural mortality (mean total mortality in the water-only control) was 21%. The mortality caused by the buffer alone must be separated from the mortality due to ammonia (0.023 mg NH3-N l-1) in the buffer. The mortality caused by the buffer is estimated by determining what the mortality would be in a mg NH3-N.l-1 treatment (the ammonia level in the water-only control). Extrapolation of the line derived from the total mortality in the buffer-only control and the mg NH3-N l-1 treatment suggests that the mortality at mg NH3-N l-l would be 12%. This is an estimate of the buffer-only control mortality at the free ammonia level observed in the water-only control. Therefore, mortality due solely to the buffer is estimated as: Water-only control mortality minus buffer-only control mortality (at mg NH3-N L-l) = 9% To obtain the mortalities due solely to ammonia, the control mortalities (water-only = 21%; buffer-only = 9%) are summed (30%) and used to apply Abbott's correction (Finney 1952). represented by the formula P' P-C = 1-C Abbott's correction is where P' = that proportion dying because of the treatment; P = total mortality in the treatment; C = total mortality in the control (30%). The corrected mortalities are given in Table 2. Corrected total mortalities plotted against the square root of the un-ionized ammonia concentration (to linearize the relationship) produced a TLm estimate of 0.11 mg NH3-N l-1 for sockeye eggs at 10 C and 02 saturation (Fig. 4).

13 - b - The m('challisltls by which morlality results are poorly understood. Un-ionized ammonia can increahe the blood pi! and interfere with the ion transfer aerosh cell membranes, tlmh affecting oxidative metabolism (Campbell 1973). Ammonia may also inhibit the ability of haemoglobin to ~ombine with oxygen (Brockway 1950; Sousa and Meade 1977). In addition to causing mortality, the combination of TRIS buffer and ammonia also appeared to influence the initiation of hatching and alevin size. Hatching started much earlier in the buffer-only control and the ammonia treatment. This may be a response to adverse physical conditions. However, since only 9.4% of the alevins had emerged before those in the water control, the effect may not be significant. There were no significant differences in the mean size among alevins hatched in the ammonia treatments and the buffer-only control. However, alevins in these tanks were significantly (p(o.001) smaller than those in the water-only control. The size discrepancy may have resulted from increased energy having been shunted into excreting ammonia or in maintaining ionic gradients across cell membranes. Other workers have noted that low levels of ammonia retard salmonid growth (Rice and Stokes 1974, Burkhalter and Kaya 1977). In conclusion, our results show that developing sockeye eggs cannot survive continued exposure to un-ionized ammonia greater than or equal to mg NH3-N I-l. Since the presence of ammonia and buffer affects egg mortality and alevin size at hatching, further experiments without buffer are needed to more closely simulate natural conditions. The 62 day TLm (0.11 mg NH)-N I-l) for sockeye eggs at lo C and a ph of 8.4 is considerably higher (three orders of magnitude) than the un-ionized ammonia levels observed in pink (Q. gorbuscha) and coho (. kisutch) salmon hatcheries (P. Rankin, unpublished data; J. Jensen, personal communication). If all other environmental parameters (e.g. 02 levels, ph and temperature) are optimal, it is unlikely that un-ionized ammonia represents a serious problem to developing salmonid eggs in hatcheries along the B. C. coast. ACKNOWLEDGMENTS I would like to thank Drs. D. F. Alderdice and J. G. Stockner, and Messrs. R. Bams and J. Jensen who critically reviewed the manuscript. Messrs. F. Velsen and J. Myhill-Jones provided valuable assistance during the experiments.

14 - 7 - REFERENCES American Public Health Association Standard methods for the examination of water and wastewater. 14 ed. American Public Health Association, Washington, D.C.: 1193 p. Alderdice, D. F., and J. R. Brett Some effects of Kraft Mill effluent on young Pacific salmon. J. Fish. Res. Board Can. 14: Brockway, D. R Metabolic products and their effects. Prog. Fish-Cult. 12: Burkhalter, D. E., and C. M. Kaya Effects of prolonged exposure to ammonia on fertilized eggs and sac fry of rainbow trout (Salmo gairdneri). Trans. Am. Fish. Soc. 106: Campbell, J. W Nitrogen excretion, p In C. L. Prosser [ed.] Comparative animal physiology. 3rd Edit10n. W. B. Saunders, Toronto, Ont. 966 p. Emerson, K., R. C. Russo, R. E. Lund, and R. V. Thurston Aqueous ammonia equilibrium calculations: effect of ph and temperature. J. Fish. Res. Board Can. 32: Finney, D. J Probit analysis. 2nd Edition. Cambridge. 318 p. University Press, Mahoney, R Ltd., London. Laboratory techniques in zoology. 404 p. Butterworth and Co. Rice, S. D., and R. M. Stokes Metabolism of nitrogenous wastes in the eggs and alevins of rainbow trout, Salmo gairdneri Richardson, p In J. H. S. Blaxter [ed.] The early life history of fish. Springer-Verlag, Berlin, tleidelberg, New York. 765 p Acute toxicity of ammonia to several developmental stages of rainbow trout, Salmo gairdneri. Fish. Bull. 73: Sousa, R. J., and T. L. Meade The influence of ammonia on the oxygen delivery system of coho salmon hemoglobin. Compo Biochem. Physiol. Vol. 58A: Vernier, J.-M Chronological table of the embryonic development of rainbow trout, Salmo gairdneri Rich Ann. Embryol. Morphog. 2: (Transl. from French by Fish. Mar. Servo Transl. Sere No. 3913, 1976). Warren, K. S., and S. Schenker Differential effect of fixed acid and carbon dioxide on ammonia toxicity. Am. J. Physiol. 203:

15 Table 1. Mean un-ionized ammonia levels, ph, and mortality of sockeye salmon embryos in the different treatments. Estimates of the median tolerance limits at several ammonia concentrations are also given. Percent mortality (mean) TLm Un-ionized ammonia (NH -N) 3 ph Fertilization During Total (From Treatment Theoretical Actual Theoretical Actual to hatching hatching mortality fertilization) Water (±0.0009)b 7.35 (+0.21) control Buffer (±0.008) (±0.11) control mg/l (±0.035) (±0.12) (±0.074) (±0.14) (±0.654) (±0.14) (±1.37) (±0.14) Median tolerance limits in days. bactual ammonia and pr values are means of approximately 30 observations between fertilization and 100% hatch (or mortality). This treatment was terminated prematurely following equipment failure.

16 - 9 - Table 2. Total mortality and mortality due to ammonia after Abbott's correction. Mortality (%) Un-ionized Treatment ammonia Total Ammonia Water - only mg control Buffer - only control

17

18 B c 50 - ~ -"-as -~ 0 ~ 98 -t:: Q) 0 ~ 2 Q) 0 E F a '~--~~~-?~~r---'-~~~TT~~--~~~~TT~ Time from fertilization (days) Fig. 1. Cumulative percent mortality (on a probit scale) of sockeye eggs from fertilization to hatching. Lines fitted by eye. fa = water-only control, B = buffer only control, C = mg NH 3 -N.l-, D = mg NH 3 -N.l-l, E = 1.93 mg NH 3 -N.l-l, F = 4.05 mg NH3-N l-l).

19

20 Y = 100 x ( X) o o ~----~ 5.0 Fig. 2. Time from fertilization to 50% mortality at different un-ionized ammonia levels.

21

22 A t 50-;' HATCH t 70 <J) ~ > ILl 100 B.J <t IL 80 0 <J) II: 60 ILl III ~ ::;) 40 z C TIME (DAYS FROM FERTILIZATION) Fig. 3. Mean numbers of a1evins hatching per two controls. Vertical bars indicate 2 S.E. [n = 3], B = buffer-only control [n = 1], C = [n = 3]). day in one treatment and (A = water-only control mg NH3-N 1-1

23

24 Y = {1X ) , = ~ '" r... 0 ~... '" ~ 4.0 o N H3 - N ('Vmg. L -1 ) Fig. 4. Total mortality (in probits, after Abbott's correction) from fertilization to hatching at different free ammonia concentrations. Line fitted by least-squares regression.