Ecological and man-induced factors of Tilapia (Sarotherodon) galilaea stock size in Lake Kinneret

 

Sections and subsections

            Introduction

            Statistical analysis

            Factors of biomass change

Exploitation

Hula drainage

National Water Carrier

Settling ponds

Low water levels

            Relationship of growth, fall increment and condition factor to stock size                                                     

            Discussion

                        The critical period and Peridinium blooms

                   Stocking and interspecific competition

                        Eutrophication

 

INTRODUCTION

 

While the cyprinid Mirogrex terraesanctae (meaning ‘great flock in the holy land’) is the dominant fish in today’s Sea of Galilee, the fish most closely associated with the lake, Tilapia (Sarotherodon) galilaea, known as St. Peter’s fish, amnun hagalil or musht abyad, forms much smaller populations.  Yet in past eras, under certain conditions, this herbivorous cichlid may have been dominant.  Scientist T.B. Tristram (1884) wrote about T. galilaea: “I have seen them in shoals of over an acre in extent, so closely packed that it seemed impossible for them to move..”

 

The fisheries, both purse-seine and trammel net, are described by Reich (1978).

 

STATISTICAL ANALYSIS OF CATCH AND STOCKING DATA

 

As T.galilaea is a commercially valuable fish, and thought to enhance water quality, much effort has been devoted to increasing stock size by introduction of pond-bred fingerlings.  Beginning in the 1950’s with < 1 million fingerlings per year, the stocking rate reached 6 million/year in the 1990’s.  In order to assess the effectiveness of this project, trends in annual catch, fishing intensity and ecological factors must be considered.

Figure 1. T. galilaea catches, 1968 – 2001, and combined stocking 2 & 3 years previously.  Data in Table 2.

 

The stock was not fully exploited until 1968, when gear improvement and the security situation allowed expansion of fishing effort (Landau, 1979).  Annual catch dropped abruptly between 1968 and 1974, from 230 to 73 tonnes (Figure 1, Table 2).  

 

From 1974 to 1990 there were upward trends in both yield and stocking rate, suggesting a causal relationship.  Landau (1996) tested this hypothesis by multiple regression analysis, on the assumption that annual catch from 1968 onward is proportional to stock size.  The contribution from pond-bred fish was not found to be statistically significant, and stock increase was attributed mainly to time-linked ecological factors.  However, this does not mean that pond-bred fish fail to reach the commercial fishery.

 

In the Landau (1996) study, annual catch 1968 – 1991 was regressed on stocking rate two years earlier because of the relatively high correlation (r = 0.517) between the parameters (Table 1). Most T. galilaea reach 18 cm (recruitment to commercial fishery) in their third year (Landau, 1979).

 

However, the 1996 study did not take into account the statistically significant correlation (r = 0.440; Table 1) between catch and stocking three years previously. Therefore, I recalculated the multiple regression of T. galilaea catches on stocking, combining data for stocking two and three years previously and obtained a higher correlation co-efficient (r = 0.604). 

 

In addition, the parameter for time trend was modified. In the 1996 study, the years 1968 to 1991 were assigned numbers 1 to 24, while in the present study, numbers decrease between 1968 and 1974, and increase in the following years (Table 2).  This resulted in a rise in the correlation co-efficient from 0.709 to 0.798.

 

Both variables, stocking and time trend, yielded statistically significant regressions, but time-linked factors were clearly more important than stocking (Table 2).  Due to high variance, the co-efficient of stocking, 2.55 T/million, S.D. 14.55, fails to provide a measure of the contribution of pond-bred fish to the fishery.

 

Nevertheless, this analysis does demonstrate a positive relationship between stocking and catch in the period 1968 – 1991.  In contrast, in the period 1992 – 2001 the correlation is negative (r = - 0.516) but not statistically significant.

 

Table 1. Correlation coefficients, r, for T. galilaea annual catch in relation to stocking and time trends.

 

Catch 1968 – 1991 vs combined stocking  2 & 3 years earlier

0.604**

Catch 1968 -1991vs stocking 3 years earlier

0.440*

Catch 1968 -1991vs stocking 2 years earlier

0.517**

Catch 1968 – 1991 vs stocking 1 year earlier

0.456*

Catch 1968 – 1991 vs stocking same year

0.414*

Catch 1968 – 1991 vs years

0.709**

Catch 1968 – 1991 vs time trends (see text)

0.798**

Catch 1992 – 2001 vs combined stocking 2 & 3 years earlier

-0.516

 

** p< 0.01;  * p< 0.05

Table 2.  Multiple regression analysis of T. galilaea catch and stocking data

 

Based on ‘Israel Fisheries in Figures’, Israel Dept of Fisheries reports.

Calculated yield, Y^ = 212.0 + 18.78 X1 + 2.55 X2

Y = T. galilaea annual catch (T), 1968 – 1991

X1 = time trend

X2  = combined stocking 2 and 3 years previous to catch, millions.

 

Regression due to X1       F =  36.7**                                     

Regression due to X2     F =  21.0**

Coefficient of X1, bY.12 =  18.78 +/- 9.80;  st. deviation =   4.73;  t = 4.0**

Coefficient of X2, bY21  =    2.55                 st. deviation = 14.55;  t = 0.175 (not sig.)

 

 

Year

T. galilaea catch,  Y

Time trend, X1

Stocking  X2

Calculated yield, Y^

Y – Y^

X3

X4

1968

230

0

2

217.1

   12.9

116

1.04

1969

198

-1

2.32

199.1

-    1.1

116

0.93

1970

176

-2

1.79

183.6

-    7.6

113

1.17

1971

153

-3

  .83

159.5

-    6.5

159

1.93

1972

115

-4

1.13

139.8

-  24.8

  97

2.45

1973

101

-5

2.85

125.4

-  24.5

  79

1.09

1974

 73

-6

4.28

110.2

-  37.2

  51

1.04

1975

121

-5

3.68

127.5

-    6.5

  84

0.25

1976

280

-4

3.42

145.6

  134.4

146

3.19

1977

181

-3

3.14

163.7

    17.3

  98

3.85

1978

163

-2

3

182.1

-   19.1

116

3.00

1979

267

-1

3.4

201.9

    65.1

104

3.22

1980

309

0

3.9

221.9

    87.1

  83

0

1981

223

1

4.25

241.6

-   18.6

  64

0

1982

304

2

3.8

259.3

     44.7

 

 

1983

233

3

3.6

277.5

-    44.5

 

 

1984

113

4

3.5

296.0

-  183.0#

 

 

1985

258

5

4.06

316.3

-    58.3

 

 

1986

316

6

4.32

335.7

-    19.7

 

 

1987

291

7

7.46

362.5

-    71.5

 

 

1988

249

8

7

380.1

-  131.1

 

 

1989

445

9

5

393.8

     51.2

 

 

1990

581

10

6.5

416.4

   164.6

 

 

1991

503

11

5.8

433.4

     69.6

 

 

# Greatest deviation Y – Y^. Previous winter, 1983, was especially cold.
X 3 and X4 are S. aureum catch (T) and stocking 2 years previously (millions), resp.

 

Pond-bred fingerlings of the cichlid Sarotherodon (Oreochromis) aureum have also been introduced to L. Kinneret since the 1950’s.  Regression of annual catch 1960 – 1981 on stocking 2 years previously indicates a yield of 21 T/ million fingerlings (r = 0.59). 

 

It has been claimed that stocking of S. aureum is detrimental to T. galilaea, the more valuable stock (Gophen et al 1983; Gophen,1984).  However, T. galilaea catches are not co-related to S. aureum stocking 2 years previously.  Catches of the two species both reached minima in 1974 (Table 2), suggesting that they are subject to the same population factors, but their correlation co-efficient, r = 0.303 is not statistically significant. 

 

FACTORS OF BIOMASS CHANGE

 

Rising exploitation rates

In Israel’s early years, tilapia stocks were not highly exploited.  T.  galilaea age composition data for 1950 and 1951 (Ben-Tuvia, 1959) compared to later data (Landau, 1979) indicate a rise in the rate of exploitation from ~10 to ~40%, and a decrease in stock size between the late 1940s and the early 1970’s from >1000 T to ~250 T/a.

 

Because of high demand for T. galilaea it can be assumed that exploitation remained close to the maximum, ~42%, from 1968 onward.  In years of lower abundance, fishing effort was reduced, which alleviated overfishing (Landau, 1979). 

 

Introduction of nylon nets in 1957, in both the trammel net and the purse-seine fisheries, doubled T. galilaea catches (Fig.2). Technological advances since 1968, such as larger nets and sonar equipment, increased efficiency but had little impact on total catch.

 

A    Hula drainage, 1954 – 1957, organic effluents from the Kinneret watershed. The lavnun (Mirogrex) biomass expanded, increasing competition for zooplankton and zoobenthos, the food supply of young T. galilaea.

 

B    Six-day-war, 1967.  Without fear of Syrian guns, fishing effort expanded.  Nevertheless, Tilapia biomass dropped steeply in the 1970’s.

 

C    National Water Carrier and salty water conduits, 1965 - 1971. Decreased chlorinity of Kinneret water probably accounts for the steep drop in Tilapia biomass.  A high rate of exploitation removed fast-growers from the stock, reducing the average growth rate.

 

D     Settling ponds in Kinneret watershed, 1982.  Lavnun population declined, reducing competition for food of young T. galilaea.

 

E     Cold winters in 1983 and 1992 caused fish mortality.

 

F      High Peridinium biomass, 1983 – 1989.  Peak catches of  S. galilaeus were taken in 1989, 1990 and 1991. In most years, scarcity of Peridinium in late summer and fall limits T. galilaea growth and survival.

 

Increased stocking of Tilapia fingerlings from 1985 probably contributed to yields in this period. 

 

G      Failure of Peridinium blooms, 1996 – 2001.  An abrupt decline in S. galilaeus yields was followed by decreased growth potential. High stocking rates were ineffective in this period.

 

Narrow littoral zone due to low water levels since 1992. Tilapia fingerlings depend on zoobenthos; therefore the Kinneret population is limited by bottom area within the aerobic zone in summer and fall.

 

 

 

Hula drainage, eutrophication, Mirogrex biomass changes

Drainage of Lake Hula and the Hula swamps, 1954 – 1957, brought nutrients into L. Kinneret (Avnimelech, 1980).  Effluents of fish ponds increased eutrophication until the late 1970’s, while organic pollutants from settlements in the Kinneret watershed continued until 1982.

 

T. galilaea catches declined from 300 T in 1957 to ~120 T/a in the mid-1960’s.  From 1976 – 1988, a period of full exploitation, the mean catch was 245 T/a, indicating a total stock of ~600 T, i.e. less than 60% of the 1940’s stock.

 

Among the consequences of eutrophication was an increase in the Mirogrex (lavnun, or Kinneret sardine) population from ~2000 T in the 1950’s to ~3000 T in the mid-1960’s, and a peak of 15 – 20 thousand in the 1980’s (Landau, 1991; Mirogrex biomass). 

 

Using data for 1975 – 1991, Landau (1996) tested the hypothesis of Gophen (1984) that the sardine negatively affects the tilapia stock. The sardine abundance index (catch per unit effort corrected for diversion of effort to other species) was not found to be a statistically significant factor.  However, the consumption index, which takes into account the mean size of sampled specimens, did yield a significant negative co-efficient in multiple regression analysis.  This gives credence to Gophen’s hypothesis, which is based on observed feeding of the young of both species on zoobenthos and zooplankton on the littoral.  T. galilaea requires an animal source of food until a size of at least 5 g (Gophen, 1980).

 

The re-analysis described below strengthens the argument that the sardine can affect the T. galilaea stock. However, sardine competition was not a primary factor in the decline of T. galilaea catches 1957 - 1962 as Mirogrex biomass was relatively small in those years.  Other effects of Hula drainage and eutrophication are implicated.  A second decline, 1968 –1974, is associated with construction of the National Water Carrier, as described in the next subsection.

 

Data for the period 1976 – 1991 yield significant negative correlation coefficients (Table 3).  CI values are correlated to T. galilaea catches of the same year, challenging the assumption that the cichlid stock is affected by Mirogrex competition with fingerlings a year or two before they enter the fishery.  However, regression of 1976 – 1991 data (Fig. 3) is dependent on 3 points only, the 1989 – 1991 values.  Mirogrex biomass was in steady decline between 1987 and 1991 due to recruitment failure in earlier years (Landau, 1991, 1996; Mirogrex biomass).

 

Table 3.  Correlation coefficient, R, of T. galilaea annual catches, 1976 – 1991 to Mirogrex consumption index, CI

 

CI same year

R = -0.61*

CI previous year

R = -0.54*

CI 2 years previously

R = -0.53*

CI 3 years previously

R = -0.02

Combined CI, one and two years previously       R = -0.61*

Regression of T. galilaea catches, 1976 – 1991 (Y, in tons) on combined CI one and two years previously (X): 

Y = 660 – 965 X.  St. error of intercept = 130;  st. error of  X variable = 335.

 

 

Figure 3.   T. galilaea annual catch 1974 – 1991 plotted against Mirogrex consumption index (CI) one and two years previously.

 

The regression of T. galilaea catches on combined CI (Table 3) crosses the X-axis at CI = 0.68.  This implies a great impact of lavnun, and suggests that the T. galilaea stock would have been demolished in the mid-1980’s if mean sardine size was 20 g rather than ~30 g, as observed  (CI values were obtained by dividing the abundance index, A’, by mean weight to the power of 0.8).  However, this regression exaggerates the influence of Mirogrex competition and its removal. In the late 1980’s the T. galilaea stock was enhanced by increased stocking and improved food supply (see subsection on settling ponds).  Nevertheless, the low level of the T. galilaea stock (~600 T) between 1976 and 1988 must be largely attributed to competition for food in the fingerling stage.

 

In the low-lake-level years of the 1990’s, when Mirogrex biomass rose to a peak of ~30,000 T, its combined CI exceeded 2.00 (the dominant size class was ~10 g). Nevertheless, the T. galilaea stock was not demolished by competition, and continued to yield >200 T annually until 2001.  T. galilaea’s continuation, albeit diminished, supports the hypothesis that lavnun consume ‘lake snow’ and small organisms which proliferate in eutrophic conditions (see Mirogrex biomass).  This does not preclude the possibility that sardine competition was a factor in the decline of T. galilaea in low-lake-level years (see subsection 5).

 

The National Water Carrier

When salty-water conduits were constructed for the National Water Carrier (1965 –1971) chlorinity of the L. Kinneret epilimnium dropped from 350 to 210-235 mg/l (Annual reports, Kinneret Limnological Laboratory).  This change was accompanied by an abrupt decline in T. galilaea catches to a nadir of 73 T in 1974. 

 

With the decline in stock size, there was a decrease in average growth rate between the 1950’s and 1970’s (Landau, 1979, comparing data of Ben-Tuvia, 1959).  Increased fishing intensity would lower the mean growth in the population, as faster-growing specimens are more heavily exploited. But declining stock size indicates stress, which would also affect growth pattern.

 

Pumping into the National Water Carrier led to a lowering of lake level, which, though not as severe as in the 1990’s, did affect water quality and algal biomass (Serruya & Pollingher, 1977).  These changes would tend to reduce the T. galilaea population, as described below, in the section entitled ‘Low lake levels’.

 

After 1974, there was a rise in T. galilaea catches to ~245 T/a between 1976 and 1987.  Adaptation of the fish to lower salinity can be postulated as an important factor in this period.

 

Settling ponds

The settling ponds that began operation in the Kinneret watershed in 1982 reduced eutrophication caused by Hula drainage and other factors.  The lavnun (Kinneret sardine) population declined steeply due to recruitment failure, which began in 1979 and became more severe in 1982 (Mirogrex biomass changes; Landau, 1991).  The dinoflagellate Peridinium cintum increased in biomass (Berman et al, 1992).

 

 In 1989 – 1991 T. galilaea catches were exceptionally good, ~ 500 T/a.  Winter mortality brought the catch down to 250 T in 1992.  At least three factors were influencing the T. galilaea population in the late 1980’s.  Firstly, stocking rate rose from < 2 million to >3 million/year.  Secondly, sardine biomass began to drop in 1986/87 reducing competition for food of young tilapia (subsection 2). 

 

Thirdly, an increased amount of food was available in late summer and fall, a critical period for T. galilaea, especially juveniles.  Until recently, Peridinium cintum was the dominant alga of L. Kinneret and the main food item in T. galilaea guts (Spataru, 1976).  The fish completed most of their growth during the Peridinium bloom season (Landau, 1979), usually February to June; in latter months, guts were less full (Spataru, 1976) while fat content and condition factor (weight/length3) declined (Landau, 1979). 

 

In former years, Peridinium blooms were so massive that fish predation had no impact upon them (Pollingher & Serruya, 1976).  Nevertheless, fluctuations in Peridinium biomass were relevant to T. galilaea survival.  Peridinium was selectively consumed by adult T. galilaea during the summer and fall months when it was scarce (Spataru, 1976) but still constituted a considerable portion of phytoplankton biomass (Berman et al, 1992).  Thus the high Peridinium biomass 1983 – 1989 (ibid) was a factor in producing peak T. galilaea catches, 1989 – 1991.

 

 Low water levels

T. galilaea mean annual catch was 335 T between 1992 and 2000, but descended to 110 T in 2001 and remained at this level until 2005.  Winter mortality decreased catches in 1992 and 1993 but the overall descent of the stock must be attributed to the lowering of L. Kinneret water levels in the 1990’s, resulting in eutrophication and a receding shoreline (homepage),

 

Along with the decline in yearly yield, the fisheries suffered from the disappearance of larger specimens. In the following section, entitled "Relationship of growth capacity to stock size" (joint work of Ruth Landau and Tzvi Snovsky, Israel Dept of Fisheries) it is shown that a decrease in growth capacity, or dwarfing, followed a decline in stock size in the period 1997 – 2004 as well as in the 1970's.. 

 

In the 1970’s, T. galilaea ranged from 16 to 36 cm total length and modal length was between 20 and 25 cm (Landau, 1979). Between 2002 and 2004 more than 70% of the fish were under 20 cm total length.  Cichlids of this size are barely marketable, so the fishery had virtually collapsed.

 

Low-water-level conditions appear to be the primary cause of the collapse of the fishery.  This hypothesis is supported by the failure of stocking to have any effect on the catch even after an increase from 3 to 6 million fingerlings/year.. 

 

Eutrophication brought the Mirogrex (sardine) population to a peak of ~30,000 T in 1998.   The impact of competition for zooplankton on cichlid survival was offset by an increased availability of alternative food sources for the sardine (subsection 2 above and Mirogrex biomass).  On the other hand, narrowing of the littoral meant reduced benthos and increased competition for this resource.

 

Even in years when water levels were normal, that is, above -212 m at onset of the rainy season, the littoral was narrow, limiting the benthos available to young T. galilaea and other organisms.  With water levels as low as –214.5 m in recent years, the shoreline is very close to the depths and the anoxic hypolimnium. Young-of-the-year T. galilaea are found on the littoral from the time of spawning, spring/summer, through winter.  They enter a critical stage between late summer and fall.  Guts of fingerlings taken on the littoral October and November contained nothing but detritus, sand particles and algal material;  their preferred foods are crustaceans (Landau, 1979; Gophen, 1980).  

 

As pond-bred fingerlings are usually introduced to the lake between July and November, they are affected by this ‘bottleneck’, especially in low-water-level years.  This accounts for the lack of correlation between stocking and catch after 1991.

 

The effects of eutrophication on adult T. galilaea are not easily defined.  There were record high blooms of Peridinium in 1994 and 1995, but in the following years, up to 2001 at least, no dinoflagellate bloom appeared (Berman et al, 1998; internal reports of Kinneret Limnological Laboratory). How did T. galilaea survive?  In laboratory experiments T. galilaea had maximum feeding rates on zooplankton, Peridinium cinctum as well as some intermediate-sized nanoplankton (Vinyard et al, 1988).  This indicates adaptability to changes in food supply.  After 1994 primary production rates were generally higher than in previous years and there were short-term blooms of various algae (Berman et al, 1998).  Nevertheless, disappearance of a preferred food source, dinoflagellate blooms, is bound to have a negative impact on cichlid stocks.

 

Other aspects of eutrophication such as oxygen deficiency and lower water transparency may have affected the T. galilaea stock as well.

 

  RELATIONSHIP OF T. GALILAEA GROWTH AND CONDITION TO STOCK SIZE

 

Most of the growth information given below was included in the joint presentations (English and Hebrew) of Ruth Landau and Tzvi (Gregory) Snovsky (Israel Dept of Fisheries) at the meeting of the Israel Zoological Society in December 2004 and the meeting of the Israeli Association of Aquatic Studies in May 2005.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


Size-at-age data for 1950 and 1951 (Ben-Tuvia, 1959) indicate a final length, L-inf, of 35 cm.  In the 1970’s L-inf was estimated as 28 cm from measurements of radii of annuli on scales (Landau, 1979). From specimens collected in 2004, L-inf = 25.5 cm was calculated.

 

Decline in growth capacity of the T. galilaea population is connected to changes in stock size and rate of exploitation.

 

In the 1950’s the T. galilaea stock was relatively large, ~ 1000 tonnes, and exploitation was low, with an annual catch of ~200 T (based on Ben-Tuvia, 1959).  In the 1970’s annual catch dropped to a nadir of ~70 T, while exploitation - 40% of total stock, approached the maximum sustainable rate (Landau, 1979)

 

 

T. galilaea: changes in size distribution and annual catch

 

 

In the period 1997 – 2002 size distribution peaked at 20 cm and extended to 40 cm.  In the following two seasons, 2002/03 and 2003/04, fish of over 23 cm were rare. The peak, at 18 – 19 cm, comprised fish that had passed two summers (age 2), as well as some age 3.

 

The shift in size distribution was abrupt: it followed the abrupt decline in total catch of T. galilaea from ~400 T/a to 110 T in 2001. This is shown by changes in median size and annual catch (lower diagram).

 

A similar trend was observed in the 1970’s.  The lowest catches, ~70 T/a, were those taken between 1973 and 1975.  The lowest median size, 19 – 20 cm, was observed in the 1975/76 and 1976/77 seasons (Landau, 1979).

 

In both 1971 and 2001 a combination of natural and man-made factors reduced the T. galilaea stock.  Fishing effort remained at the same level; therefore rate of exploitation increased. In these conditions, the ability of small specimens to escape from nets has an impact on population growth parameters.

 

 

Disappearance of fast-growers from the stock

 


 Measurements of the radii of annual rings on the scales of T. galilaea taken in 2004 indicate the disappearance of faster growing specimens by the end of the 2nd growth season, at age 2.  Among the older cohorts, L-2 over 18 cm was not observed. The older fish showed a smaller increment in the 2nd year than age 2 of the same initial length, L-1.

 

Conclusion:  the vulnerability to fishing nets of T. galilaea from the size of 18 cm leads to elimination of fast-growers, especially when

the rate of exploitation is high.

 

Research results summarized in this paper have implications for fisheries management:  High exploitation lowered mean growth potential in the T. galilaea population. The use of small mesh-size was a result rather than a cause of the high proportion of ‘undersized’ fish in the catch

 

In order to improve the stock, fishing pressure should be reduced whenever there is a downward trend in abundance.  This cannot be accomplished by controlling mesh size of the nets, but by limiting the number of fishing licenses.

 

Growth increment in fall months

 

In the 1970's, October to November were months of low food intake, growth cessation, loss of visceral fat reserves, and decline in fish condition (Spataru, 1976; Landau, 1979).  In contrast, data for the 2004/5 season (above diagram) indicate an appreciable growth during October and November.

 

Condition factor C of T. galilaea in L. Kinneret

Period         Mean C     n

1974-1977    2.20        242

1997-2002    2.24        201

2004-2006    1.98        664

S.D.             0.1 – 0.2

 

 

 

Since 2004 condition factor, C = g / mm 3 total length x 105, is much lower than in the 1970's and the period 1997 -2002 (data collected by Tsvi Snovsky). In the 2005/6 season there was a marked decrease between November and March.

 

 

 

DISCUSSION

 

          The critical period and Peridinium blooms

Most fishes must pass a critical period in the larval stage, when yolk has been consumed and the slow-swimming larva must find sustenance and escape predators.  As a mouth-breeding cichlid, T. galilaea can overcome dangers of the larval stage better than other fish and reproduce abundantly.  Therefore, it is unlikely that spawning and nursery area limit the population of L. Kinneret despite the receding shoreline of recent years.  On the other hand, extent of the littoral would affect survival in the fingerling stage by limiting the benthic organisms available.

 

Before the lowering of water level below -212 m, the critical stage for T.galilaea, especially young-of-the-year, was October and November. In respect to the stocking program, this ‘bottleneck’ could have been been avoided by introducing pond-bred tilapia fingerlings in late November or December, especially if they contained fat reserves (Landau, 1979; 1983).  A large proportion of tilapia fingerlings are stocked in summer due to crowding in the breeding ponds.

 

Data for growth during the fall months in 2004 indicate changed conditions.  Fall blooms of Dinophyta that did not occur before destabilization of the ecosystem (Zohari, 2004) are benefiting T. galilaea.  On the other hand, the absence of Peridinium blooms in spring have created a new critical period.  A low Peridinium biomass in both spring and fall was observed in 2000 and in 2001 (ibid) followed 2 years later by exceptionally low T. galilaea catches, indicating the strong impact of Peridinium on young tilapia.

 

Low condition factors in recent years are also related to low Peridinium biomass, especially in the 2005/06 season.

 

            Stocking programs and interspecific competition

The multiple regression analysis given above indicates a positive relationship (until the 1990’s) between numbers of stocked T. galilaea fingerlings and catches 2 and 3 years later. Because the co-efficient of the stocking parameter is not statistically significant, no conclusion can be drawn from this analysis on the efficiency of the stocking program in pre-low-water-level years.  In the conditions that prevailed after 1990, stocking was clearly ineffective.

 

Catches of Sarotherodon aureum, a closely related tilapia, is largely dependent on stocking in L. Kinneret; statistical analysis (above) indicate a yield of  ~20 T/million fingerlings.  The claim of Gophen (1984) and Gophen et al (1983) that S. aureum ‘displaced’ T. galilaea by competing for animal food was not borne out by statistical tests. While the lavnun stock is measured in 1000’s, S. aureum amounts to no more than a few hundred tonnes, and would not be an effective competitor under normal conditions.  However, along with other stocked fishes, silver carp and mugilids, S. aureum may interact with T. galilaea at critical periods, especially in low-water-years when the littoral is very limited.

 

          Eutrophication

In the last 50 years T. galilaea yearly catches have varied between 73 and 581 tonnes, 8 times.  Manipulation of the environment far outweighed natural variation as a causal agent, affecting fish stocks mainly by increasing or alleviating eutrophication.  Mirogrex expands during eutrophication due to increased availability of rotifers and other small organisms that are essential for larval survival (Landau et al, 1989 and others). 

 

In the same conditions, the T. galilaea stock declines. When sardine biomass rose above 3000 T, competition for zooplankton and benthos became an important factor for T. galilaea growth and survival.

 

Eutrophication also affected T. galilaea through the shift from Peridinium to smaller algae.   Physical parameters such as oxygen deficiency may also have been involved.

 

Extinction of cichlid species in Lake Victoria, Africa, is attributed primarily to the introduction in the 1950’s of a predator, the Nile perch.  With the disappearance of herbivorous cichlids, water quality deteriorated and the ecosystem destabilized (Ogari & Asila, 1992 and others). Eutrophication of near-shore waters by inflow pollution (http://www.lvemp.org/start.htm ) may have played a role in cichlid depletion.

 

In Lake Kinneret, no exotic predator has been introduced (as yet!).  Eutrophication was clearly an initial factor in cichlid depletion since the causal agents - Hula drainage, etc. - preceded reduction, and an alleviating factor, settling ponds, preceded a rise in cichlid biomass.

 

When Tristram (1884) observed L. Kinneret fish, the area around the lake was sparsely populated, water level and salinity were not regulated, and fishing activity was negligible. Under these conditions, T. galilaea flourished.

 

Homepage

 

Mirogrex biomass changes