PROCEEDINGS OF THE OKLAHOMA ACADEMY OF SCIENCE Volume 60—1980

{Page 10}

CHANGES IN THE FISH COMMUNITY OF LAKE CARL BLACKWELL, OKLAHOMA (1967-77) AND A TEST OF THE REPRODUCTIVE GUILD CONCEPT

Donald J. Orth

Oklahoma Cooperative Fishery Research Unit*, Oklahoma State University, Stillwater, Oklahoma 74078

Numerical density, biomass density, and diversity of fishes in coves of Lake Carl Blackwell were estimated in 7 years during the period 1967-77. Among fishes within the same reproductive guild, some exhibited similar trends in abundance but others did not. Some species in different reproductive guilds showed similar trends in abundance. Densities of fishes within the following groups were positively correlated with each other: (a) green sunfish (Lepomis cyanellus), longear sunfish (Lepomis megalotis), flathead catfish (Pylodictis olivaris), and white crappie (Pomoxis annularis); (b) gizzard shad (Dorosoma cepedianum), orangespotted sunfish (Lepomis humilis), black bullhead (Ictalurus melas), and largemouth bass (Micropterus salmoides); and (c) freshwater drum (Aplodinotus grunniens), river carpsucker (Carpiodes carpio), and white bass (Morone chrysops).

INTRODUCTION

The effect of reservoir water management on the abundance of fishes has been the topic of numerous investigations. Reservoir water regimens affect the fish community directly by influencing reproductive success (1, 2) and indirectly by altering plankton abundance (3, 4), benthos abundance (5, 6, 7, 8), and the establishment of aquatic macrophytes (9, 10).

Typically investigations into the factors governing density and biomass of fishes have focused on individual species. With the increasing interest in applying ecosystem modeling techniques to reservoir management, the need for a more meaningful ecological classification of fishes has become apparent. Preliminary attempts to develop reservoir ecosystem models have involved a functional classification based on food habits (11, 12, 13). The weakness here was the inability to accurately model recruitment. For instance, by maintaining a constant water level during the simulation of a reservoir cove ecosystem, Patten (13) concluded that the fish compartments appeared to be insensitive to water level variations. However, the marked effect of water level variations on recruitment of reservoir fishes has been documented in many reservoirs (14, 15, 16, 17).

Balon (18) has proposed an ecological classification scheme that is based on the reproductive strategies of fishes. Fishes are grouped into guilds based on breeding behavior and the character of their predominant spawning grounds, as well as on the morphology of embryonic and larval respiratory organs and pigments, and behavior of embryos and larvae. Balon hypothesized that, since the density of fishes is typically determined during these early stages, classification by reproductive guild may prove to be more meaningful than classification by feeding strategies. If the reproductive guild concept is valid, fishes within the same guild should exhibit similar trends in abundance. The objective of this study was to examine the changes in the structure of the fish community of Lake Carl Blackwell to see if fishes in the same reproductive guild do indeed exhibit similar trends in abundance.

STUDY AREA

Lake Carl Blackwell is a 1400-hectare impoundment located in Payne County in north central Oklahoma that has been subject to large fluctuations in water level since its impoundment in 1938 (Fig. 1). During years of declining water levels, terrestrial vegetation developed along the exposed shoreline and in years of rising water levels this vegetation was inundated. Declining water levels, unstable bottom sediments, and wave action probably in-

{Page 11}

hibit the development of aquatic macrophytes in the littoral zone.

Estimates of fish standing crops in coves were made by personnel of the Oklahoma Cooperative Fishery Research Unit in July and October 1967 and July and September 1971 (19) and during the second week of August for 1973-77. In 1967 and 1971 two coves were sampled — one in the northeast arm of the lake, and the other on the north shore of the lake; in 1973 two of the three coves sampled were in the northeast arm of the lake and the other cove was in a southern arm; and each year in 1974-77, the same two adjacent coves in the northeast arm of the lake were sampled.

Coves were mapped to determine area and volume. The mouth of the cove was blocked with a net (6.4 mm square mesh) and a 5% solution of rotenone was applied at a concentration of 2-3 mg/l. Fish were collected for a 3-day period, identified, and weighed. The numbers of dead fish of each species recovered were divided by the area of the cove to obtain numerical densities. Biomass densities were obtained by multiplying numerical densities by the mean weight of individuals of each species. Estimates were not expanded on the basis of percent recovery since percent recovery was not determined each year. In years when fish were marked and released before rotenone was applied, the recovery of marked fish ranged from 44 to 70%.

Diversity of the fish communities was measured by the Shannon-Weaver index:

where p_i is the proportion of fish (as either individuals or biomass) in the ith species and s is the total number of species (20, 21).

Correlation coefficients were computed between the numerical densities of all possible combinations of species, and between diversity (d)¯ calculated from numerical units and the numerical densities of each species.

The nest-spawning lithophil guild of Balon (18) contains the greatest number of fish species (six) from Lake Carl Blackwell (Table 1); other guilds are represented by only one to three species.

Twenty species were collected from the coves, of which 13 were collected in all seven years (Tables 2 and 3). Gizzard shad made up the greatest proportion of the fish community in both numbers and biomass in almost every year. Abundance of all fish species varied from year to year as did the total fish standing crop (Tables 2 and 3; Fig. 2). Total fish standing crop increased markedly after 1973, from an average of 156 kg/hectare for 1967 and 1971, to an average of 387 kg/hectare for 1974 through 1977. Several species (notably gizzard shad and orangespotted sunfish) produced a large year class in 1973 as evidenced by the increase in numerical density, and decrease in mean weight per individual (Fig. 2).

{Page 12}

Diversity as calculated from numerical units declined from 1967 to 1973, and then increased through 1977. Diversity as calculated from biomass units was higher than diversity calculated from numerical units in all years except 1975, when the gizzard shad reached its highest biomass (Fig. 3). Diversity for all years averaged 2.01 and 2.57, respectively, based on numerical and biomass units. Diversity calculated from numerical units was positively correlated with numerical densities of white crappie (r = 0.77; P = 0.04), green sunfish (r = 0.75; P = 0.05), and channel catfish (r = 0.77; P = 0.04).

The matrix of correlation coefficients (Table 4) indicated which fish species exhibited similar trends in abundance. Three lithophils (longear sunfish, green sunfish, and flathead catfish) had similar trends in abundance.

The hypothesis examined here is that fishes which use similar reproductive strategies to exploit the same resources should exhibit similar trends in abundance. These

{Page 13}

data on the changes in the abundance of fishes of Lake Carl Blackwell show that, among fish species within the same reproductive guild, some exhibited similar trends in abundance but others did not. Also, some fishes in different reproductive guilds showed similar trends in abundance.

Among the nest-spawning lithophils, the longear sunfish, green sunfish, and flathead catfish showed similar trends in abundance. Two other nest-spawning lithophils, orangespotted sunfish and black bullhead, also showed similar trends in abundance, but the abundance of neither was correlated

{Page 14}

with the abundance of other lithophils. There were, however, no significant correlations between the abundance of bluegills and other lithophils. It is possible that not all members of the nest-spawning lithophil guild are similarly affected by the same factors. Furthermore, some factors may be common to members of different guilds.

If fishes with similar trends in abundance were combined, they would form three groups: (a) green sunfish, longear sunfish, flathead catfish, and white crappie; (b) gizzard shad, orangespotted sunfish,

{Page 15}

{Page 15 Consists entirely of Table 4.}

{Page 16}

black bullhead, and largemouth bass; and (c) freshwater drum, river carpsucker, and white bass. Although some of these groups contain members of the same reproductive guilds, the correspondence with the guild is not close enough to support the application of the reproductive guild concept for modeling communities of fishes in this reservoir. However, the close correlations within these groups may indicate that either similar factors influence fishes of the same group, or the different factors which influence fishes of the same group are themselves correlated.

A possible explanation for the lack of correspondence of the three groups distinguished in this study with reproductive guilds could be that short-term population trends are determined by random environmental factors. Fishes that showed similar short-term trends in abundance may be similarly affected by these random environmental factors. This explanation is supported by the predominance of significant positive correlations over significant negative correlations in Table 4 (12 vs. 2). The dramatic changes in density (Fig. 2) may be related to water level changes (Fig. 1) and the concomitant changes in habitat and biota. Diversity also increased as total fish density increased and increases in diversity were correlated with increased densities of white crappie, green sunfish, and channel catfish. Average diversities and the variation in diversity were comparable to those reported by Wilhm (21) for rotenone sampling of Oklahoma reservoirs.

If niche breadth is a function of population density as suggested by Vandermeer (22), fish species will fit more distinctly into reproductive guilds at low densities than at high densities; at the higher densities, reproductive strategies and breeding grounds of fish species may encompass more than one reproductive guild. For specific predictions of short-term changes in abundance, the grouping of reservoir fish species into reproductive guilds may not be advantageous. The reproductive guild concept has, however, been applied with success to long-term population changes (23, 24) and further tests on long-term population trends in other reservoirs would be informative.

ACKNOWLEDGMENTS

Funds for this project were provided by Federal Aid to Fish and Wildlife Restoration, Oklahoma Dingell-Johnson Project F-36-R, Job 3. Eugene K. Balon, Michael D. Clady, Howard R. Jarrell, George W. Luker, O. Eugene Maughan, Harold E. Namminga, and Robert C. Summerfelt reviewed the manuscript and made many useful suggestions.

REFERENCES

1.   M. C. BROSS, Proc. Okla. Acad. Sci. 48: 194-199 (1969).

2.   C. H. WALBURG and W. R. NELSON, U.S. Bur. Sport Fish. Wildl., Res. Rep. 69, 1966, 30 pp.

3.   F. C. JUNE, U.S. Fish Wildl. Serv., Tech. Pap. 71, 1974, 57 pp.

4.   M. P. TAYLOR, in: G. E. HALL (ed.), Reservoir Fisheries and Limnology, Am. Fish. Soc., Washington, D.C., 1971, pp. 209-217.

5.   L. R. AGGUS, Ibid., pp. 139-152.

6.   N. G. BENSON and P. L. HUDSON, Trans. Am. Fish. Soc. 104: 526-528 (1975).

7.   B. C. COWELL and P. L. HUDSON, in: C. E. LANE, JR. (chairman), Reservoir Fishery Resources Symposium, So. Div. Am. Fish. Soc., Washington, D.C., 1968, pp. 541-555.

8.   B. G. ISOM, in: G. E. HALL (ed.), Reservoir Fisheries and Limnology, Am. Fish. Soc., Washington, D.C., 1971, pp. 179-191.

9.   C. E. BOND, Ibid., pp. 153-166.

10.   J. H. B. DEGRUCHY, Water Level Fluctuation as a Factor in the Life of the Higher Plants of a 3300 Acre Lake in the Permian Redbeds of Central Oklahoma, Ph.D. Thesis, Okla. State Univ., Stillwater, 1952.

11.   G. R. LEIDY and R. M. JENKINS, The Development of Fishery Compartments and Population Rate Coefficients for Use in Reservoir Ecosystem Modeling, Final Rep. by U. S. Fish Wildl. Serv. for U.S. Army Eng. Waterways Exp. Stn., Vicksburg, Mississippi, 1977.

12.   B. C. PATTEN, Trans. Am. Fish. Soc. 104: 594-617 (1975).

13.   B. C. PATTEN, Ann. Okla. Acad. Sci. 5: 1-10 (1976).

14.   L.R. AGGUS and G. V. ELLIOTT, in: R. H. STROUD and H. CLEPPER (eds.), Black Bass Biology and Management, Sport Fishing Inst., Washington, D.C., 1975, pp. 317-322.

15.   D. B. JESTER, in: G. E. HALL (ed.), Reservoir Fisheries and Limnology, Am. Fish. Soc., Washington, D.C., 1971, pp. 265-285.

16.   R. C. SUMMERFELT and K. E. SHIRLEY, Proc. Okla. Acad. Sci. 58: 54-63 (1978)

17.   C. E. VON GELDERN, JR., Calif. Fish. Game 57: 228-245 (1971).

18.   E. K. BALON, J. Fish. Res. Board Can. 32: 821-864 (1975).

{Page 17}

19.   J. N. JOHNSON, Effects of Water Level Fluctuations on Growth, Relative Abundances and Standing Crop of Fishes in Lake Carl Blackwell, Oklahoma. M.S. Thesis, Okla. State Univ., Stillwater, 1974.

20.   J. L. WILHM, Ecology 49: 153-156 (1968).

21.   J. WILHM, Ann. Okla. Acad. Sci. 5: 29-46 (1976).

22.   J. H. VANDERMEER, Ann. Rev. Ecol. Syst. 3: 107-132 (1972).

23.   E. K. BALON, W. T. MOMOT, and H. A. REGIER, J. Fish. Res. Board Can. 34: 1910-1921 (1977).

24.   E. K. BALON and E. M. P. CHADWICK, Arch. Hydrobiol. 85: 543-547 (1979).