183

32 .

Correlations between Indices o f Relative Abundance of Young Fish, Recruitment Size and Maturity Rate as a Basis for Annual Prediction

By

T. F. D e m e n t j e v a

VNIRO, Moscow

T he assessment of the relative abundance of fish is m ainly used for the evaluation of the annual fluctu­ations in the fish stock. A relative index is useful not for assessing the absolute size of the stock, bu t only the changes which have occurred in the stock com­position and size from one year to the next.

I f the indices of the relative abundance are extended to the total catch in compliance with time, place and fishing gear as well as the fishing intensity, a “weighted” stock can be obtained proceeding from the calculated recruitm ent size and the fishing mortality.

T he reliability of indices of the relative fish abun ­dance depends on the extent of knowledge of the ecological conditions of the habita t and the reproduc­tion of the species as well as on the reliability of the m aterial collected. The true criterion is given by the correlation between the indices of relative strength of a year-class assessed at the stage of one summer-olds and the abundance of the year-class after recruitm ent as calculated from the fishing mortality. Characteri­stics of growth are used for the assessment of changes in the biomass size and m aturity rate of fish. T he natural m ortality is determined, if possible and neces­sary, by indirect indices.

Thus the m ethod of assessment of the relative abun ­dance of fish as well as other methods includes analysis of recruitm ent, growth, fishing and natura l mortalities,i. e. four basic processes of the dynamics of fish popula­tions. M ost publications on the fishing theory and methods of assessment of the fished stock are devoted to consideration, quantitative determ ination and ana ­lysis of the said processes. T he present paper considers some propositions on the assessment of the recruitm ent size and in this way we would try to make the methods of assessment of the relative abundance of fish more accurate.

A t present m uch attention is devoted to the various methods of assessment of the total stock due to the application of m athem atical models. H e m p e l and S a h r h a g e (1961) stated tha t the m ain aims of the

numerous investigations were to find out such m athe­matical models which would include the relations between recruitm ent, growth, natural and fishing mortalities quite visually, bu t w ithout needless simpli­fication. T he dynamics of the populations can be studied experimentally on a m athem atical model, by altering one factor a t a time, while the others rem ain constant. I t should be noted, however, tha t considera­tion of a change in some factor in relation to other constant factors does not seem to secure desirable re ­sults, since the very basis of changes in the factors lie in the interaction w ith other processes and the different factors are most unlikely to be independent. Besides, hypotheses regarding which condition the extent of changes in the factors concerned, are often based on a num ber of so far unsolved biological problems, parti­cularly on the relation of the influence of the recruit­m ent size on the composition of the fish stock and on the conditions governing the strength of the year- classes.

W hen working-out the fishing theory workers often proceed from the point th a t the recruitm ent size, though being variable, depends on the stock density.

I f the annual recruitm ent size fluctuates, the mean recruitm ent size is used instead of the actual one to predict mean catches for a given period or to substant­iate the establishment of a regulated fishery. However, the fishing industry needs forecasts not only for a cer­tain average period, bu t also for every year coming. I t is quite necessary to have such a prognosis available since the catch should be scientifically planned.

T he annual recruitm ent size is known to range widely. T he determ ination of this value is the most im portant stage in the assessment of the stock. W ith this in m ind M o n a s t y r s k y (1952) worked out a m ethod of a complex evaluation of the stock with the recruitm ent composition in which he used all the ex­perience of workers accum ulated for a long period of years by studying the dynamics of fish populations, including B a r a n o v (1918) and H j o r t (1914), and he

184

tried to have all the rational dependencies confirmed by fishing practice. This m ethod is based on the pro­position tha t the stock fluctuations depend not only on the effect of the fishery bu t also on the changes in the recruitm ent size of the populations, as was stated and confirmed by a vast actual m aterial on m any species from different w ater bodies. I t was furtherm ore stated tha t the fluctuations in the recruitm ent size are mainly dependent on reproduction and biological peculiarities of the populations. Thus, the relation between the recruitm ent size and environm ent formed the basis for the application of the m ethod since all other m e­thods suggest tha t the recruitm ent be considered a constant value or dependent on the population density, i.e. on the num ber of spawners. But the latter, in its turn, depends on the fishery, thus a relation between changes in the stock and changes in the fishery itself substantiates the m ethod used. In view of this fact the models m ay include changes in the fishery.

T he difference in the backgrounds of the methods stipulates the principal divergencies in understanding the causes governing the changes in the fish stocks. Since the recruitm ent is mostly dependent on natural processes, which are never constant, the models used are most unlikely to suggest always the ready con­clusions on the recruitm ent size and causes, which bring about the changes in it. Thus, recently the concept necessitating the inclusion of the environ­m ental da ta into the models of fish populations be­came more widely spread ( G u l l a n d , 1962; et al.).

T he most representative knowledge of the recruit­m ent size can be obtained from the above-mentioned m ethod of stock calculation w ith recruitm ent size and composition. This m ethod aims a t determ ining of this value, as a variable and independent on the stock density and seems to be advantageous as most com­posite and dynamic. Along with the advantages it re­quires thorough versatile observations on changes in the biological characteristics of the population, its composition, the fishing methods and fishing intensity, as well as in the regime and productivity of the waters.

The m ain elements of the m ethod are as follows : -

1. Usual determ ination of age and size composition of the stock by sampling expressed either as the total catch (when the fishing intensity remains the same) or as the catch per fishing effort (when different fishing power is applied).

2. D eterm ination of the correlation between the growth rate and the m aturity rate of the recruit­m ent (by spawning marks) and evaluation of the causes affecting the fluctuations in growth and m aturity rates.

3. Determ ination of the predicted strength of the year-classes by counting one summer-olds and by finding a correlation of these indices with the

strength of the year-classes a t different stages of life.

4. Determ ination of the density of the population in the spawning or w intering concentrations.

5. Calculation of brood strengths and fishing m or­tality rate.

6. Determ ination of the ratio between the actual catches, the density of population and the strength of the year-classes.

7. D eterm ination of the lowest level of the stock and the value of possible or permissible catch.

The process of calculation of the said indices and of collection of the m aterial needed is well known. I t is w orth m entioning tha t samples are most representative for the determ ination of the fishing stock in the period of gathering for spawning migrations, on the way to spawning places and a t the spawning.

Growth and m aturity

In the analyses of the peculiarities of growth and m aturity it is desirable to have the following items in mind : -

(a) M aturation is bound with the length of the fish, b u t not with the age.

(b) A t a lower rate of growth the fish, as an excep­tion, reach m aturity when shorter-sized due to some acceleration in the developm ent of the fish.

(c) As a rule, the specimens with a higher rate of growth, especially in the early stages, reach m aturity sooner. For example, in the N orth Caspian bream , the m aturity is reached by 30-40 °/0 of 3 year-olds when the growth ra te is high, while a t a lower rate of growth only 10—20°/0 of specimens of this age reach m aturity . T he Baltic cod usually reach m aturity a t the age of 3, bu t a t the spawning places the catches include 13-17 °/0 of the same year-class with a lower rate of growth, while in the years of intensive growth 32 °/0 of 3 year-olds occurred in the catches ( T o k a r e v a , 1963).

T he growth of fish depends both on the quantity of food in the w ater and on the num ber of consumers. As shown from the studies carried out by Z e m s k a y a (1958), changes in the abundance of consumers affect the growth of fish sooner than the fluctuations in the food resources. T he fluctuations in the fish growth are also the result of different rates of metabolism. Any rise in tem perature to a certain lim it increases the “ appetite” , i.e. promotes m ore active movements of fish in search for food and more intensive assimilation of food.

T he interrelation of the tem perature regime in the w ater body, m etabolic intensity and growth rate in

185

T able 1.

D eviations o f annual increm ents in the length o f 4 year-old North Caspian bream from the long-range average (in cm )

(From T . F. D e m e n t j e v a , 1952; K . Z e m s k a j a , 1958)

Year- Year of growthclass 1940 1941 1942 1943 1944 1945 1946 1947

193 9 ......... +1-4 +0-6 -0 -1 -194 0 ......... +0-1 +0-9 +0-3 -0 -1 -194 1 ......... - - 0 -2 - 0-3 0 -0 -2194 2 ......... - - -0 -1 -0 -2 -0 -9 -0 -1194 3 ......... - - 0-4 -0 -2 - 0 - 4 -0 -5194 4 ......... - - - - 0 - 4 -0 -2 +0-6 +0-7194 5 ......... - - - 0 - 4 0 +1-6194 6 ......... - _ _ _ _ _ +0-4 +0-1194 7 ......... - _ _ _ _ _ _ +0-1

27 (4-1-3) +1-3 -0 -2

fish is often m entioned by m any authors (A l m , 1959; N i k o l a e v , 1961 ; P e g e l , 1950; H a n s e n , 1954; N ü - MANN, 1959; K a r z i n k i n , 1952; etal.). T he difference in the average length of fish at the same age in relation to metabolic intensity m ay be very high, for example, in the Baltic herring it amounts to more than 3 cm- (D e m e n t je v a , 1952).

The data on growth m ay be back-calculated from otoliths and scales. T he increments obtained are se­parately com pared w ithin each most numerous age- group, as a result of which a higher degree of accuracy can be obtained, the age peculiarities of growth being of no importance. I f the increments are expressed by deviations from the long-range average it is possible to get a sum of deviations for each year of growth for all the year-classes of the stock in a given year. This is illustrated in T able 1.

The analysis of the increments shown in T able 1 indicates tha t in the same year the trends of growth m ay be similar in a num ber of year-classes, i.e. in different age-groups in compliance with the peculiari­ties of their growth. Thus, we m ay assume that some general conditions in a given year caused changes in the growth of both young and adu lt specimens, though their feeding grounds were different. This paper does not propose to consider the causes of the changes in the growth shown in the example. I t is w orth m en­tioning, however, tha t as shown in Table 1 the growth of the bream was greatest in 1940, 1946 and 1947, while in 1943-45 the growth was lowest. T he growth indices shown horizontally enable us to judge the character of growth of each year-class within their span of life.

M aturity and span o f life in fish

Proceeding from the relation between the m aturity rate and the recruitm ent size we m ay state tha t the

-0 - 7 -1 - 7 -1 -1 +0-5 +2-5

earlier a fish reaches m aturity , the shorter is its span of life, i.e., the sooner the fish of a given year-class complete their life cycle and go out of the fishery. I t m ay be assumed tha t if the acceleration in growth of the population and the earlier m aturity are observed w ithin a period of several years, it m ay lead to re­juvenation of the fishing stock. T he rejuvenation is expressed by an increase in the strength o f younger ages and a corresponding decrease in the abundance of the older specimens of the fished stock. This regular­ity should be borne in m ind when analysing the causes of rejuvenation of the fishing stock in spite of the fact tha t it is the intensive fishing tha t plays the leading role in it.

Examples of differential mortalities when fastly- growing specimens reach m aturity and die sooner than the slowly-growing individuals can be found in any species ( H i l e , 1936; S m ith , 1956; etal.). The values calculated for the Baltic cod and the N orth Caspian bream are shown in Table 2.

T he two generations of bream had a different growth rate in the early life. In the long run the year-class of 1933 proved to occur in the catches for a longer period than the brood of 1934, in spite of the fact th a t it was not strong. Z e m s k a ja ’s data (1958) indicate the same ratio in the year-classes of 1939 and 1942 and others.

All the m aterial collected suggest the necessity of studying the biological peculiarities of the fish popula­tions, changes which affect the recruitm ent rate in the period of forming the fishing stock and fishing m or­tality.

P redicting the strength o f a generation

W hen the percentage of recruits in relation to the m aturity rates is calculated, it is necessary to find an index of relative abundance of the year-class, from which the calculation of the recruitm ent size and the

186

T able 2.

I. E xploited generations o f the B altic cod at the follow ing ages:

(After T o k a r e v a , 1963)

3 and 4 year-olds 5 and more year-olds N um ber of1 year-classes

0 / 0 // 0 / 0

W ith intensive growth as 2 year-olds 70-0-77-2 22-4-30-0 5W ith poor growth, as 2 year-olds. . . 56-4-65-0 35-0-43-6 3

II. Percentage age com position o f catches o f bream

Year- GrowthAge Size of year-classes

in millions of specimensclasses 1-3 years

2 3 4 5 6 7 8 0110

19331934

poormedium

0-6 5-244-8

71-139-4

14-911-3

5-23-1

2-91-4

0-10-05

100100

34-0590-65

rem ainder of the fished stock could be based. Counts of young fish m ay serve for this purpose. Indices of counting are the m ean catches of one summer-olds per un it of time. They indicate a density of their distri­bution over a given area. T he period of counting one summer-olds is advantageous to other m ethods pro­viding counts of other age-groups for the following reasons : -

(a) T he heaviest mortalities take place a t the earliest stages of development.

(b) A t the stage of one summer-olds and older the num ber of the individuals dying is in proportion to the generation sizes, if there is no great in ­crease in strength of predators or some other na tura l circumstances.

(c) T he one summer-olds are far less active and try to avoid trawls to a lesser extent, thus their catchability is greater than th a t of one or two year-olds. I f the necessity of counting the latter arises, then the design of a trawl and towing rate should be changed.

Before commencing the counts of young fish it is expedient to study the distribution of young fish by seasons in relation to depths, currents, salinity, distri­bu tion of food organisms and predators, and other factors. In a num ber of species a correlation between indices of quantita tive counts of young fish and the sizes of the year-classes which were born in the same years (calculated from the catches) was found. Such correlations are illustrated by da ta on the N orth Caspian bream and the Baltic cod (Figures 1 and 2). M aterial is available to allow to develop similar curves for the Baltic herring, Barents Sea cod and for other species.

T he above-mentioned correlation enables us to con­clude as follows : -

1. W hen the fish becomes older than one sum m er the na tu ra l m ortality becomes m uch less pro ­nounced, and

2. By the m ethod of analogy it is possible to predict the size of a generation by counting the one summer-olds. Nevertheless, the correlations thus obtained should be perm anently checked in view of: —

(a) changes in the environm ent and distribution of one summer-olds;

(b) lack of adaptation in one summer-olds to severe winters;

(c) excessive catch of young fish which does not seem to be subject to statistical accounts and some other causes.

T he existence of correlations between the abun ­dance of a generation as calculated from the fishing m ortality and the da ta on the density of concen­trations of young fish enable us to suggest a relatively insignificant na tu ra l m ortality in the m ain fishable age-groups. Thus, in its turn , it enables us to assume th a t in some cases we m ay reckon the fishing m ortality as a total mortality.

T he assessment of the ra te of fishing m ortality is carried out by using the above-m entioned m ethod in relation to the fished stock obtained from the catches, and thus, it reflects only a lower level of the stock. This level is known to fluctuate annually as well as in the total stock in relation to the recruitm ent size and the fishing power. Knowledge of a range of the stock fluctuations in conjunction w ith the evaluation

187

1 2 0

5 3 *

5 7 «

70 1 3 0 1 5 09 0 110 1 7 0 1 9 0CO

C a t c h o f 3 - 4 y e a r - o l d s p e r h o u r o f t r a w l i n g

Figure 1. Baltic cod.

100

8 0

6 0

400 1 0 0 2 0 0 3 0 0N u m b e r o f y o u n g f i s h p e r

h o u r o f t r a w l i n g

B

5 3 »

5 7 »

of biological peculiarities of the population and their changes will contribute to gaining more accurate predictions.

References

A lm , G., 1959. “ Connection between m aturity, size and age in fishes” . Rep. Inst. Freshw. Res. Drottning., No. 40: 5-145.

B e v e r t o n , R. J . H . , & H o l t , S. J ., 1957. “ The dynamics of ex­ploited fish populations” . Fish. Invest., Lond., Ser. 2, 19.

D e m e n t j e v a , T . F., 1952. “ A method of compiling prognosis of catches of bream in the northern Caspian” . T rud . V N IR O , 21: 163—84. (In Russian).

D e m e n t j e v a , T . F., 1952. “ The growth of f ish in relation t o the problem of dynamics of populations” . Zool. Zh., 31: 634-37.

G u l l a n d , J . A., 1962. “ Population studies” . Proc. roy. Soc., A 265: 351-54.

H a n s e n . P. M ., 1954. “ The stock of cod in Greenland waters during the years 1924-52” . R app. Cons. Explor. M er, 136: 65-71.

H e m p e l , G., & S a h r h a g e , D., 1962. “ Neuere Modellvorstellun­gen über die Dynamik der Grundfischbestände” . Ber. dtsch. Kom m . Meeresforsch., 16: 51-89.

H ile , R ., 1936. “ Age and growth of the casco, Leucickthys artedi (Lesueur) in the lakes of the northeastern highlands. Wisconsin” . Fish. Bull.. U.S.. 48: 211-317.

K a r z i n k i n , G. S., 1952. “ Bases of biological productivity of a water body” . Pishchepromizdat.

M o n a s t y r s k y . G. M . , 1952. “ Dynamics of populations of com­mercial fish” . T rud. V N IR O , 21: 3-162. (In Russian).

N i k o l a e v , J . J . , 1961. “ Meteorology and fishery m anagem ent” . Vop. Ikhtiol., No. 2: 46-56. (In Russian).

N ü m a n n , W ., 1 9 5 9 . “ Das Wachstum der Blaufelchen und seine Berechnung bei Erfassung aller Individuen der einzelnen J a h r ­gänge in Abhängigkeit von der Stärke des Jahrganges, den jährlich variierenden Aussenfaktoren, der Eutrophierung und den regionalen Unterschieden des Sees” . Arch. Fisch Wiss., 10: 5 - 2 0 .

P e g e l , V. A., 1950. “ Physiology of digestion in fish” . Tomsk Univ. R i c k e r , W. E., 1958. “ H andbook of computations for biological

statistics of fish populations” . Bull. Fish. Res. Bd Can., No. 119.

S m it h , S. H., 1956. “ Life history of lake herring of Green Bay, Lake M ichigan” . Fish. Bull., Û .S . , 57: 87-138.

T o k a r e v a , G. I., 1963. “ Some data on the peculiarities of growth in the Baltic cod” . T rud. A tlantN IR O .

Z e m s k a j a , K., 1958. “ Growth and sexual m aturity of the North Caspian bream in relation to changes in its abundance” . T rud. V N IR O , 34: 63-86. (In Russian).

>0 1 0 0 15 0 2 0 0Sizes of ye a r - c la sse s in mil l ions

Figure 2. Caspian bream.