CHAPTER 15

STATUS OF THE CAPTURE FISHERIES

IN FOUR ASIAN RESERVOIRS AND
A VOLCANIC LAKE
1
Boonsong Sricharoendham, 2* Upali S. Amarasinghe,
2
R.P. Prabath K. Jayasinghe & 3 Simeona M. Aypa

1
Inland Fisheries Resources Research and Development Institute, Inland Fisheries
Research and Development Bureau, Department of Fisheries, Ministry of Agriculture
and Cooperatives, Bangkok, Thailand
2
Department of Zoology, University of Kelaniya, Kelaniya 11600, Sri Lanka
3
Formerly of Aquaculture Division, Philippine Bureau of Fisheries and Aquatic
Resources (BFAR), 860 Quezon Avenue, Metro Manila 3008, Philippines
(* Author for Correspondence: Tel.: +94-11-2903397; Fax: +94-11-2911916;
E-mail: zoousa@kln.ac.lk)

Key Words: tropical reservoirs, cyprinidae, cichlidae, clupeidae, artisanal fisheries,

Thai river sprat

Abstract
A comparative assessment of the fisheries of five Asian lacustrine water bodies was
undertaken. In Lake Taal, Philippines, where a wide range of gear types is used, monthly
fish catches are influenced mainly by the proportion of the clupeid, Sardinella tawilis,
with high catches of this species during April through October, perhaps reflecting its
peak recruitment to the fishery. In Ubolratana reservoir, Thailand, almost all species
are exploited. Fish yield in this reservoir was negatively influenced by water level and
positively by water inflow. Due to the attraction of part-time fishers to the fishery during
the non-cultivation agricultural season, total fishing effort increases from January to May,
while a ban on the use of certain fishing gears during the spawning season of fish species
from mid-May to mid-September reduces fish catches in the reservoir. In three Sri Lankan
reservoirs, namely Minneriya, Udawalawe and Victoria, catch rates of individual boats
vary according to weather patterns, reservoir water levels and lunar phases possibly due
to the changes in behavioural patterns of dominant fish species. In Udawalawe reservoir
where cyprinids dominate landings, lunar phase and rainfall had significant influences
266 B. Sricharoendham, U.S. Amarasinghe, R.P.P.K. Jayasinghe & S.M. Aypa

on catch per unit effort (CPUE). Water level was a major factor influencing CPUE in
Minneriya and Victoria reservoirs, where fisheries are dominated by exotic cichlids. High
variations in catch rates of individual boats are evident in the three Sri Lankan reservoirs
studied, which bring about high variabilities of income derived from fishing. As such,
in addition to seasonal and spatial variations in catch rates, their variations in individual
boats should be considered when defining management strategies for these fisheries.

Introduction
In Asia, reservoir construction for energy and agriculture has altered the landscape of
the region (De Silva & Amarasinghe, 1996). The paucity of natural lakes compared to
most other continents (Fernando & Holčik, 1991), extensive availability of freshwater
river systems (Shiklomanov 1993; Nguyen & De Silva, 2005) and historically high
dependence for paddy cultivation on irrigation schemes in most countries such as Sri
Lanka (Siriweera, 1994) have been contributory factors for reservoir construction.
Tropical lakes and reservoirs are important rural development resources because most of
the immediate beneficiaries of their fisheries are rural poor. In spite of the importance of
reservoir fisheries in the context of rural development, this sector is given a lower priority
in national development plans of most Asian countries, mainly because reservoir fisheries
are almost always a secondary activity (De Silva, 1996).
Fisheries output in tropical lakes and reservoirs varies according to factors such as
limnological differences (Moreau & De Silva, 1991), the nature of fish communities
(Fernando & Holčik, 1982, 1991) and differences in fishing pressure (Bayley, 1988;
Welcomme, 2001). These differences in turn affect the livelihoods of fisheries-dependent
rural communities. As such, it is imperative to investigate the relevance of such unique
characteristics of individual reservoir fisheries for defining their management strategies.
In the present chapter, a comparative assessment of the status of the fisheries of four
reservoirs and a volcanic lake is undertaken, with a view to identifying important aspects
to be considered in defining management strategies.

A brief description of the five Asian water bodies studied and their
fisheries
The five Asian water bodies studied differ from each other with respect to origin, extent,
multiple uses as well as other morphological characteristics and land-use features
of catchment areas (see Chapters 2, 3 & 4). The reservoir fisheries of Sri Lanka have
distinctive characteristics, namely that the craft used are non-mechanised out-rigger
canoes approximately 6 m in length, manned by two persons; the major gear type is
gillnets ranging in mesh size from 7.5 to 12.7 cm, with each net piece being 60 m long
and 1.5 m high, respectively, and with the average number of 18 net pieces per boat; and
the landings are dominated by exotic cichlid species, Oreochromis mossambicus and O.
niloticus (De Silva, 1988; Amarasinghe, 1998). The major characteristics of the fisheries
of the five Asian water bodies studied are given below.
Chapter 15 – Status of the capture fisheries in four Asian reservoirs and... 267

Lake Taal, Philippines

In Lake Taal, the total number of fishers was reported to rise from 1,206 in 1995 (de Jesus,
1998; Villanueva et al., 1995) to 2,761 in 1998 (Rapera, 2000) and to about 3,000 in 2000.
Fishers use motorised and non-motorised boats for fishing in open waters while many
nearshore fishers use bamboo rafts. In 1995, there were 400 motorised boats and 238
non-motorised boats in Lake Taal. On the basis of the amount of investment for fishing
in Lake Taal, the fishing community can be divided into two groups, namely subsistence
and commercial fishers. The majority of fisherfolk operate on a subsistence basis and use
gillnets, hook-and-line, spear guns, fish traps and lift nets. On the other hand, commercial
fishers operate the motorised push nets, ring nets, beach seines, several sets of gillnet of
various mesh sizes. The minimum permissible mesh size of the fishing nets imposed by
the government is 3 cm. Nevertheless, fishers use motorised push nets and beach seines
of finer mesh sizes to catch juvenile Sardinella tawilis during April-September. Major
species caught in the fishery of Lake Taal are S. tawilis, silverside and tilapias.

Minneriya reservoir, Sri Lanka

Although decaying submerged tree stumps are frequent in most of Sri Lanka’s lowland
reservoirs, Minneriya is free of such impediments to fishing. Minneriya is a shallow
irrigation reservoir with a mean depth of 5.8 m. During the months of low water level,
illegal beach seines are used sporadically (Fernando, 1967; Amarasinghe & De Silva,
1992). About 100 fishers operate in this reservoir, where over 90% of the landings from
this reservoir comprise O. mossambicus and O. niloticus.

Udawalawe reservoir, Sri Lanka

The mean depth of this reservoir is 7.9 m. Due to the presence of decaying tree stumps,
use of fishing gears other than gillnets is virtually impossible. There are about 90 fishers
using large mesh (22.5 cm) gillnets. Being situated close to a state-owned fish breeding
centre, this reservoir is regularly stocked with fingerlings of Indian major carps. Dominant
species caught are O. mossambicus, O. niloticus, Labeo dussumieri, L. rohita and Catla
catla.

Victoria reservoir, Sri Lanka

Victoria reservoir is situated in the central hill country of Sri Lanka. Its mean depth is 30.5
m and it has a very steep basin. As water is released regularly to generate hydroelectricity,
drastic water level fluctuations occur. Gillnets of mesh sizes ranging from 7.6 cm to 11.4
cm are used by about 40 fishers. The main species landed are O. mossambicus, O. niloticus
and Cyprinus carpio.

Ubolratana reservoir, Thailand

At present there are over 5,000 fishers in Ubolratana reservoir, practising various fishing
methods. Some characteristics of the fishery of Ubolratana reservoir are given in Table
268 B. Sricharoendham, U.S. Amarasinghe, R.P.P.K. Jayasinghe & S.M. Aypa

15.1. Sixty seven fish species are represented in the landings but only 21 are considered to
be economically important (see below). The majority of fishers are subsistence-oriented,
while gillnets, long lines, cast nets, scoop nets and lift nets are the major gear types
overall. Use of these gear types however, varies seasonally and is associated particularly
with rainfall. For example, lift netting for catching Clupeichthys aesarnensis is mainly
carried out close to the inlet area, especially during the rainy season. The stretched mesh
sizes of gillnets range from 2.5 to 18 cm but gillnet mesh sizes between 4.5-7.0 cm are
the most frequent in the reservoir. The bigger mesh gillnets (14-18 cm) are effective
during the rainy season, when large-sized fishes are caught. Usually fishers change the
mesh sizes of gillnets seasonally according to the availability of fish. The main fishing
gear used by large-scale fishers is the lift net. The width of a lift net is about 8 m and the
mesh size is 2.5 cm. Two types of lift net can be distinguished according to the mode of
operation, namely fixed and movable lift nets.

Materials and methods

Data on catch and effort and species composition of the landings were collected from the
fisheries by a survey team in each of the three countries. In addition, data available from
reliable official sources were also gleaned, especially for Ubolratana reservoir (Thailand)
and Lake Taal (Philippines). It must be noted that data analyses in the five lacustrine water
bodies were not identical due to differences in the nature of the fisheries in individual
water bodies, and data availability.
In Lake Taal (Philippines), the commercial fisheries data were collected and recorded
by locally hired data collectors using survey forms prepared for the purpose. Data
collection was undertaken simultaneously for five successive days every month from
July 1998 to August 1999 at five major fish landing sites. Fish landed were grouped into
species and the total weight and percentage of each species in the catch were recorded
for each fishing method separately. Based on the catch data from the five landing sites
sampled, the total fish harvest in all eight fish landing sites of the lake was estimated. Fish
consumption by the fishers’ households was estimated on the basis of data collected from
sampled boats and this was added to landing data in computing total annual fish harvest in
the lake. In addition, data on the fishery of Lake Taal, such as number of fishermen, gear
types in different littoral municipalities, average catch per day by different fishing gear,
average number of fishing days per month and total catches during 1993-1998, reported
by Villanueva et al. (1995), de Jesus (1998) and Rapera (2000) were used in the present
synthesis.

Table 15.1: Fishing methods, number of fishing units operated in Ubolratana reservoir, Thailand.
NA = Data not available.
1970 1973 1983 1992 2000
number of fishing households 1,060 1 ,379 3,905 4,870 > 5,000
number of fishing gear
- gillnet 12,320 7,850 14,485 31,160 39,860
- lift net 68 159 1,279 448 790
- cast net 195 67 542 NA 3,050
- dip net/scoop net NA 98 812 2,435 1,700
- hooks NA 20,730 1,865 67,279 64,250
Chapter 15 – Status of the capture fisheries in four Asian reservoirs and... 269

Investigations of fishing activities and fisheries in Ubolratana reservoir (Thailand) were

conducted by means of questionnaire surveys and interviews with fishermen and fish
traders from 1998 to 2000. The commercial catches were recorded at 16 fish landing
sites. Landing statistics were recorded on a daily basis at one landing site. In the other 15
landing sites, data were collected through systematic random sampling once every two
weeks. Species-wise and gear-wise data on total landings of fish were recorded at each
landing site. Previous records on the fishery of Ubolratana reservoir (Mekong Secretariat,
1972; Duangsawasdi & Raikhuntod, 1973; Chantakanon et al., 1993) were also consulted
to investigate trends in the fishery of the reservoir.
In the three Sri Lankan reservoirs, daily data on fish catch and effort were collected
between January 1998 and July 2002. Data collection was performed by fisheries
statisticians employed for this sole purpose, for a minimum of 20 days in a month. This
method was effective for obtaining reliable information (Nissanka et al., 2000; De Silva et
al., 2001; Amarasinghe, 2003). The fish species composition of the landings, total catch in
each boat and the number of boats operated were recorded systematically at each landing
site. Fish landings from Minneriya and Victoria could be recorded accurately when the
fish vendors weighed the fish catches in individual boats at the landing sites. However, in
Udawalawe reservoir, an arbitrary unit called a ‘handful’ is used by fish vendors, which
was observed to vary among species groups (see also Chapters 19, 20). As such, for four
major species groups which are landed in Udawalawe reservoir [i.e., cichlids (indigenous
and exotic), catla, smaller Indian carps (mrigal and rohu) and Labeo dussumieri], the
relationship between actual weight and ‘handful’ was obtained through linear regression
with zero intercept (Zar, 2000). These relationships were then used to estimate actual
landings of different species groups which were sold to fish vendors as ‘handfuls.’
Monthly mean water level data for the Sri Lankan reservoirs were obtained from
the irrigation and hydroelectric authorities. Rainfall data were obtained from the
Meteorological Department. The mean water level in each reservoir during the study
period was calculated and the months with water level higher and lower than the mean
were treated separately. Similarly, months with higher and lower rainfall than the mean
value for the study period were treated as wet and dry months respectively. The study
period was also stratified according to lunar phases. The new moon period was defined as
new moon day ± 3 days whereas full moon period was defined as full moon day ± 3 days.
Consequently, the four lunar periods were new moon period, second quarter, full moon
period and last quarter. Catch per unit effort (CPUE) was estimated on the daily basis as
kg per boat-day for each boat. As daily data of CPUE were not available species-wise,
analyses were performed for CPUE of all species. Frequencies of CPUE were stratified
according to water level (high and low), weather seasons (dry and wet) and lunar phases
(four phases).
As CPUE is known to be log-normally distributed (Gulland, 1983), Ln (CPUE +
1) transformed data were used to investigate the effects of three factors (water level,
rainfall and lunar phase) and their interactions on CPUE using a three-way ANOVA. As
unequal observations for the three factors represented an unbalanced design, the three-
way ANOVA was performed using a general linear model (Ryan & Joiner, 2001).
Inflow data and water level data for Ubolratana reservoir were obtained from
the Electricity Generating Authority of Thailand (EGAT), which is the hydrological
management authority. The influence of these hydrological data on monthly biological
270 B. Sricharoendham, U.S. Amarasinghe, R.P.P.K. Jayasinghe & S.M. Aypa

fish production in the reservoir was investigated. Based on the species composition of
the landings by weight of three-year averages of catch data, 15 fish landing sites were
ordinated by non-metric multi-dimensional scaling (Clarke & Warwick, 1994) using the
Bray-Curtis similarity matrix (Bray & Curtis, 1957) of double square-root transformed
data (Clarke & Gorley, 2001).
All statistical analyses were carried out using the MINITAB (Version 13.32) and
Primer (Version 5.2.2) statistical software.

Results
Lake Taal
The distribution of fishing gear among the 10 littoral municipalities is presented in Table
15.2. Gillnets are the most widely used gear type in all municipalities. Lift nets, fish
pots/shelter and scissor nets are also used in the lake but no quantitative information
is available on their operation. The average daily catch by gear and average number of
fishing days per month are given in Table 15.3. The ring nets, motorised push nets and
beach seines consist of fine mesh net bags and are operated mainly from March to October
(Plate 15.1). As strong lights are also used in association with these fishing methods, they
are very efficient for catching juvenile S. tawilis. Although the catch rates of these fishing
methods are much higher than of others (Tab. 15.3), their areas of operation are restricted
to sheltered lake areas within a few municipalities.
In the five landing sites sampled, estimated fish catches during 60 sampling dates (5
days per month for 12 months) was 74.6 t so that total catch landed and sold during
the year in these sampling sites (assuming that fishing is done daily) was about 454 t.
Considering that there were eight major fish landing areas, total fish landed and sold from
the lake was about 726 t. The amount of fish consumed by the fishers’ households was
estimated to be about 70 t annually (C. de Jesus, pers. comm.) and was not included in
the estimates of total fish landed and sold. Based on all these quantities, the fish harvest
from the lake during the study period can be estimated at around 796 t, giving an average
annual fish yield of about 32.7 kg ha-1. The percentage composition of species groups in
the fish landings during the 1998-1999 period is given in Table 15.4. S. tawilis was the

Table 15.2: Number of fishermen (No), gear types in different municipalities around Lake Taal, July
1995 (Villanueva et al. 1995; de Jesus, 1998; Rapera, 2000).
Municipality No Gill net Ring Motorised Beach Spear Fish Hook and
net push net seine gun corral Line
Agoncillo 916 217 - - 8 58 4 -
Alitagtag 22 47 - - - 6 - 62
Balete 89 84 2 - 3 43 - 24
Cuenca 89 110 - - - - - 7
Laurel 653 104 - - 1 - - -
Lipa 20 - - - - - - -
Mataasna 91 119 - - 10 - 30 -
Kahoy
San Nicolas 704 153 - - - 7 36 5
Sta. Teresita 60 43 - - 12 22 - 8
Talisay 84 48 4 12 5 10 7 -
Tanauan 33 118 - - - 4 - -
Total 2,761 1,123 6 12 39 150 17 106
Chapter 15 – Status of the capture fisheries in four Asian reservoirs and... 271

Plate 15.1: Experimental beach seining in Lake Taal (Photo: David Simon).

dominant species during 1998-1999 with the overall contribution of 58.7%. The peak
contributions of this species to the commercial catch were during August-September
1998 and June-July 1999. The active operation of the motorised push nets, ring nets, and
beach seines during these months, which perhaps coincided with the peak recruitment
season, has contributed greatly to the enhanced catch of this species. The contribution of
S. tawilis during the period December to March was low. The rough waves and strong
winds caused by the northeast monsoon and the relatively cold weather in December to
February could have reduced the efficiency of fishing operations targeting this species. It
is principally the gillnet operators who fish during these months. Silverside (15.4%) and
tilapia (12%) were the other dominant species in the commercial landings. The tilapias
caught were Oreochromis mossambicus and O. niloticus, while the carps consisted of
bighead, common carp and the small cyprinids. The other species caught were milkfish,

Table 15.3: Estimated average catch per day (in kg) by different fishing gear in Lake Taal in 1998
(Rapera, 2000) and average number of fishing days per month (de Jesus, 1998). NA – Information
not available.
gear type catch days
gill net 11 22.4
ring net 198 26.0
motorised push net 351 27.0
beach seine 71 24.0
spear gun 4 25.0
fish corral 8 28.5
hook and line 6 NA
lift net 10 NA
fish pot/shelter 14 NA
scissor net 6 NA
 272
Table 15.4: Monthly variation in percentage species composition of landings and total catch (in kg) in Lake Taal, Philippines (1998-1999).

B. Sricharoendham, U.S. Amarasinghe, R.P.P.K. Jayasinghe & S.M. Aypa

Aug ‘98 Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul ‘99 Overall
Cardinal fish 1.10 1.93 8.39 7.77 12.69 14.98 1.31 15.54 9.46 10.91 8.13 4.38 5.35
Carps 0.83 1.19 1.01 8.30 8.68 17.16 22.10 7.91 4.69 6.54 4.14 3.22 3.28
Caranx 0.49 0.77 0.86 0.41 18.65 1.99 2.85 1.92 0.47 1.14 1.56 1.74 1.39
Catfish 0.00 0.00 0.00 0.00 0.22 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01
Eel 0.00 0.00 0.00 0.00 0.00 0.45 2.46 0.19 0.00 0.16 0.00 0.00 0.06
Gobiidae 0.29 0.39 0.44 2.15 1.59 2.29 1.01 1.33 1.46 1.31 0.60 0.82 0.71
Milkfish 0.67 0.94 0.49 0.95 1.68 1.82 2.14 0.35 0.03 0.04 0.10 0.16 0.58
Mullet 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01
Scat 0.00 0.00 0.00 0.00 0.00 1.17 0.00 0.04 0.00 0.00 0.00 0.00 0.02
Silver perch 1.08 0.61 1.70 6.65 5.19 4.27 4.49 5.30 3.21 3.32 2.06 1.81 2.10
Snapper 0.08 0.06 0.00 0.00 0.00 0.61 0.00 0.00 0.00 0.26 0.11 0.09 0.07
Snakehead 0.20 0.23 0.48 0.84 2.08 1.04 3.72 0.34 0.04 0.52 0.08 0.32 0.37
Silverside 5.55 20.45 46.38 5.70 1.81 3.75 0.00 46.33 28.43 0.41 0.00 7.75 15.36
S. tawalis 82.24 69.36 30.94 34.65 24.91 15.41 7.01 3.03 36.16 55.02 62.51 71.63 58.72
Tilapia 7.46 4.07 9.31 32.59 22.49 35.05 52.91 17.71 16.03 20.37 20.69 8.07 11.98
Total catch (kg) 27083.4 9797.8 6447.0 1667.5 1953.9 911.8 1105.1 5613.4 5912.4 3529.4 4110.0 6536.4 74637.8
Chapter 15 – Status of the capture fisheries in four Asian reservoirs and... 273

Table 15.5: Capture fisheries yields (metric tonnes) in Lake Taal. Data for 1993-1998 are from
Rapera (2000) and for 1998-1999 are from the present study.
year yield
1993 8,792
1994 2,446
1995 2,767
1996 1,311
1998 1,681
1998-1999 796

mullet, snakehead, eel, goby, scat and snapper. None of the species groups in the landings
except S. tawilis contributed significantly to fish catches in Lake Taal (Tab. 15.4). The
trends in annual fish harvest in Lake Taal (Tab. 15.5) reveal a steady decline.

Sri Lankan reservoirs

Figure 15.1 shows the relationships of actual weight of cichlids which are landed in
Udawalawe reservoir to the arbitrary ’handful’ units used by fish vendors. The relationships
for other major species groups (i.e., C. catla, L. dussumieri, L. rohita) which were also
determined were used to estimate actual landings of different species groups. Annual fish
yields, fishing effort and species composition of the landings in Minneriya, Udawalawe
and Victoria reservoirs are given in Table 15.6. Fish yields in the two lowland reservoirs
(Minneriya and Udawalawe) are much higher than in Victoria. In Udawalawe, exotic
carps are responsible for high yields. In February – March 2000, very high monthly fish
yields (577.1 t in February 2000 and 362.2 t in March 2000) were reported in Udawalawe
compared to monthly fish yields ranging from 12.5t to 77.9 t during the rest of the study
period. During these months of peak catches, over 95% of the landings were formed
by C. catla. Temporal variations in catch rates of individual boats influenced by lunar

Figure 15.1: The relationship between arbitrary measure (‘handful’) of fish weight and actual
weight of cichlids in Udawalawe reservoir.
274 B. Sricharoendham, U.S. Amarasinghe, R.P.P.K. Jayasinghe & S.M. Aypa

Table 15.6: Annual fish yields (in kg ha-1 yr-1) , fishing effort (boat-days yr-1) and species composi-
tion of fish landings in three Sri Lankan reservoirs.
Minneriya Udawalawe Victoria
fish yield 111.60 177.66 32.47
fishing effort 17367 7749 3639

species composition (%)

O. mossambicus 7.40 10.89 28.45
O. niloticus 87.00 14.55 38.37
T. rendalli 2.26 0.95 4.25
L. dussumieri - 16.00 -
C. catla - 51.00 -
C. carpio - - 19.40
others 3.32 6.60 9.51

patterns can be seen in all three reservoirs when the frequencies of CPUE during different
lunar phases are considered. Also, it is evident that weather patterns (i.e. dry and wet
seasons) and water level (high and low) influenced temporal variations in catch rates of
individual boats of the three reservoirs. These trends in Minneriya reservoir are shown as
a characteristic example, i.e., lunar pattern (Fig. 15.2 (a) to (d), dry and wet weather (Figs
15.2 (e) and (f) and high and low water levels (Figs 15.2 (g) and (h)).
When an arbitrary value CPUE of 10 kg per boat-day in all three reservoirs is
considered, a significant proportion of boats had lower CPUE than this value during
different seasons. These differences in the frequencies of CPUE vary from reservoir to
reservoir. For example, in Minneriya reservoir where cichlids form the major proportion
of the landings (Tab. 15.6), CPUE of < 10 kg boat-day-1 was observed in a higher number
of boats during the rainy season and during seasons of high water level (Plate 15.2 &
3). On the other hand, in Udawalawe and Victoria, where cyprinids comprise significant
proportions of landings (Tab. 15.6), overall CPUE is found to be higher in most boats
during the rainy season. In Victoria reservoir too, where C. carpio formed a significant
proportion of the landings (19.4%; Tab. 15.6), trends in the influence of rainfall and water
level on the CPUE of individual boats are more or less similar to those in Udawalawe
where catch rates in most boats are high during rainy season. In all three reservoirs,
during the full moon phase, a significant proportion of the boats observed had lower
CPUE (< 10 kg boat-day-1) than during other lunar phases. This was more pronounced
in the two reservoirs where cyprinids formed significant proportions of landings, i.e.,
Udawalawe and Victoria, even during the last quarter. Results of the three-way ANOVA
(Tab. 15.7) indicated that in Minneriya and Udawalawe, CPUE is significantly influenced
by lunar phase, in Udawalawe and Victoria by rainfall and in Minneriya and Victoria
by water level. However, in none of the reservoirs did interactions of these factors have
a significant influence on CPUE at the 5% probability level. In Udawalawe reservoir,
interactions between lunar phase versus rainfall, and rainfall versus water level, had
significant influences on the CPUE at the 10% probability level.
Chapter 15 – Status of the capture fisheries in four Asian reservoirs and... 275

Figure 15.2: Frequencies of boats in different classes of CPUE during four lunar phases (a-d), wet
and dry seasons (e-f) and high and low water levels (g-h) in Minneriya reservoir.
276 B. Sricharoendham, U.S. Amarasinghe, R.P.P.K. Jayasinghe & S.M. Aypa

Plate 15.2: A typical fishing craft manned with two fishers in a Sri Lankan reservoir (Photo: Upali
Amarasinghe).

Plate 15.3: A fish landing site in a Sri Lankan reservoir (Photo: W.S. Weliange).
Chapter 15 – Status of the capture fisheries in four Asian reservoirs and... 277

Table 15.7: Analysis of variance (three-way) of Ln (CPUE + 1) in three Sri Lankan reservoirs.
d.f. Seq. SS Adj. SS Adg. MS F ratio P Sig.
Minneriya
main factors
lunar phase (LR) 3 3.169 0.636 0.212 1.12 0.026 **
rainfall (RF) 1 0.248 0.125 0.125 1.84 0.176 Ns
water level (WL) 1 2.097 1.212 1.212 17.84 0.000 ***
interactions
LR x RF 3 0.183 0.135 0.045 0.66 0.574 Ns
LR x WL 3 0.079 0.181 0.060 0.89 0.447 Ns
RF x WL 1 0.012 0.016 0.016 0.23 0.630 Ns
LR x RF x WL 3 0.200 0.200 0.067 0.98 0.402 Ns
residual 368 25.001 25.001 0.068
total 383 30.987
Udawalawe
main factors
lunar phase (LR) 3 7.104 5.712 1.904 3.17 0.025 **
rainfall (RF) 1 21.151 15.510 15.510 25.83 0.000 ***
water level (WL) 1 0.381 1.386 1.386 2.31 0.130 Ns
interactions
LR x RF 3 3.763 4.462 1.487 2.48 0.062 *
LR x WL 3 1.833 1.419 0.473 0.79 0.502 Ns
RF x WL 1 2.747 2.290 2.290 3.81 0.052 *
LR x RF x WL 3 0.319 0.319 0.106 0.18 0.912 Ns
residual 234 140.483 140.483 0.600
total 249 177.781
Victoria
main factors
lunar phase (LR) 3 15.235 3.739 1.246 2.00 0.113 Ns
rainfall (RF) 1 1.654 2.602 2.602 4.18 0.041 **
water level (WL) 1 24.267 10.215 10.214 16.42 0.000 ***
interactions
LR x RF 3 1.710 1.528 0.509 0.82 0.484 Ns
LR x WL 3 1.977 1.319 0.440 0.71 0.549 Ns
RF x WL 1 0.031 0.045 0.045 0.07 0.789 Ns
LR x RF x WL 3 0.991 0.991 0.330 0.53 0.661 Ns
residual 433 269.415 269.415 0.622
total 448 315.279

d.f.: Degrees of freedom; Seq. SS: sequential sums of squares; Adj. SS: adjusted sums of
squares; Adj. MS: adjusted mean squares; F ratio: variance ratio test; P: probability level;
Sig.: significance level (*** <0.1% level; ** <5% level; * <10% level; Ns: not significant).
278 B. Sricharoendham, U.S. Amarasinghe, R.P.P.K. Jayasinghe & S.M. Aypa

Table 15.8: Composition of fish in the landings (catch in kg) of Ubolratana reservoir, Thailand dur-
ing 1998-2000.
1998 1999 2000 Average
Catch % Catch % Catch % Catch %
Channa 5,388 0.60 11,023 1.25 12,215 1.31 9,542 1.06
Hampala 4,279 0.48 5,355 0.61 5,735 0.62 5,123 0.57
Glossogobius 22,884 2.56 35,073 3.96 43,587 4.68 33,848 3.75
Notopterus 38,212 4.28 45,264 5.11 58,225 6.25 47,234 5.23
Large Mystus 18,510 2.07 17,741 2.00 22,151 2.38 19,467 2.15
Small Mystus 3,220 0.36 4,841 0.55 6,754 0.72 4,938 0.55
Wallagonia 1,660 0.19 1,122 0.13 472 0.05 1,085 0.12
Ompok 1,058 0.12 761 0.09 1,203 0.13 1,007 0.11
Kryptopterus 26,923 3.01 12,490 1.41 16,066 1.72 18,493 2.05
Morulius 33 0.00 246 0.03 0 0.00 93 0.01
Puntius 3,246 0.36 5,285 0.60 7,872 0.84 5,468 0.61
Small Puntius 67,279 7.53 108,163 12.22 82,061 8.81 85,834 9.50
Osteochilus 44,468 4.98 46,906 5.30 52,572 5.64 47,982 5.31
Probarbus 42,773 4.79 97,180 10.98 131,720 14.14 90,558 10.02
Puntioplites 453,244 50.73 320,441 36.20 305,520 32.79 359,735 39.82
Clupeichthys 37,755 4.23 35,205 3.98 64,959 6.97 45,973 5.09
Pangasius 43,151 4.83 41,947 4.74 27,246 2.92 37,448 4.14
Labeo rohita 281 0.03 168 0.02 144 0.02 198 0.02
Oreochromis 77,279 8.65 92,482 10.45 88,513 9.50 86,091 9.53
Mastacembelus 1,522 0.17 2,819 0.32 2,949 0.32 2,430 0.27
Macrobrachium 321 0.04 596 0.07 1,899 0.20 939 0.10
Total 893,485 100 885,108 100 931,861 100 903,485 100

Figure 15.3: Mean monthly fish catches in Ubolratana reservoir, Thailand (1998-2000).
Chapter 15 – Status of the capture fisheries in four Asian reservoirs and... 279

Figure 15.4: Influence of monthly water level (A) and monthly inflow water (B) on fish catch in
Ubolratana reservoir, Thailand.

Ubolratana reservoir
The number of fishing households in Ubolratana increased from 1,060 in 1970 to 1,379 in
1973, followed by a rapid increase to 3,905 households in 1983 and then 4,870 households
in 1992 (Mekong Secretariat, 1972; Duangsawasdi & Raikhuntod, 1973; Chantakanon et
al., 1993). At the time of this research, there were about 5,000 fishing households in the
reservoir.
Sixty seven species have been recorded in Ubolratana (Bhukaswan, 1985), but only 21
species are found to be economically important. Their contribution to total landings during
1998-2000 is shown in Table 15.8. Estimates of total annual fish harvest at the 14 landing
sites monitored in this study were 889.5 t in 1998, 885.2 t in 1999 and 931.9 t in 2000;
yielding an average of 903.5 t. Approximately 75% of the total fish catch of the reservoir
is landed at these 14 sites. However, these data did not include subsistence consumption
by fishers’ families, which was estimated to be about 460 t per year. Accordingly, mean
280 B. Sricharoendham, U.S. Amarasinghe, R.P.P.K. Jayasinghe & S.M. Aypa

Figure 15.5: (a) MDS ordination and (B) Bray-Curtis similarity of fifteen fish landing sites based
on the species composition.

annual fish yield in Ubolratana reservoir during 1998-2000 was estimated to be 40.5 kg
ha-1. Monthly fish catches reveal seasonal fluctuations (Fig. 15.3) with high fish offtake
during the low water level period. As can be seen from Figure 15.4, the water inflow
rate has a positive influence on monthly fish yield while water level influences monthly
fish yields negatively. Monthly catches also increased steadily from January to May,
due to an increase of fishing effort from the part-time fishers who are less engaged in
paddy cultivation during this period. As fishing with lift nets is prohibited during the
fish-spawning season from mid-May to mid-September, fish harvests decrease. For the
15 landing sites (see Chapter 3), results of the cluster analysis performed on the basis of
species composition by weight of the 3-year average of catch data using a Bray-Curtis
Chapter 15 – Status of the capture fisheries in four Asian reservoirs and... 281

similarity matrix of species composition by weight and their MDS ordination are shown
in Figure 15.5. The results showed that there were three main clusters. All the sites of
the first cluster (sites 1, 2, 4 and 5) are located near the outflow of the reservoir, where
the highest landing (over 18% of the total landings) was reported. The second cluster
consisted of sites 3, 7, 8, 9, 10, 11, 12 and 14, and most of these sites (except sites 3
and 14) were located in the Nam Choen arm of the reservoir (see Chapter 3). The other
three sites (sites 6, 13 and 15) are in the Nam Pong arm of the reservoir (see Chapter 3).
This analysis indicates that there are spatial differences in species composition of the
landings.
When the percentage of each main group of fish from the commercial catch is examined
(Tab. 15.8), it can be seen that Puntioplites proctozysron formed the largest proportion of
the landings in 1998-2000, averaging 39.9%. The second highest commercial catch was
minor carp which accounted for 10% of the landings. Oreochromis niloticus was the third
highest, accounting for 9.5%.

Discussion
In inland lakes and reservoirs, seasonal variations in the species composition of commercial
landings are influenced by biological factors such as recruitment seasonality, socio-
economic factors (e.g., consumer acceptability of various fish species), and implementation
of certain management strategies such as gear restrictions, closed seasons and closed areas
(Welcomme, 2001). In Lake Taal, for example, seasonal variation in species composition
(Tab. 15.4) appears to occur due to recruitment seasonality of S. tawilis, which influences
significantly the total commercial landings. In this lake, proportion of S. tawilis landed
had a positive influence on the total commercial catch.
In the three Sri Lankan reservoirs, seasonal variations in CPUE, which are influenced
by rainfall, water level and lunar pattern appear to have significant implications for
fisheries management. Influences of these factors on catch efficiencies are possibly due to
the changes in behavioural patterns of dominant fish species during various lunar phases
and rainy seasons and aggregation of fish during the seasons of low water level, resulting
in increased catch efficiencies. In Udawalawe, where cyprinids dominated landings, lunar
phase and rainfall had significant influences on CPUE. Water level was a major factor
influencing catch rates in Minneriya and Victoria.
Some evidence is found on the influence of lunar pattern on fish catches. For example,
Jutagate et al. (2003) reported that in Sirinthorn reservoir, Thailand, fishers start using
liftnets to catch Clupeichthys aesarnensis in the late new moon phase. Mahon & Hunte
(2001) indicated that the mechanism for the effects of the lunar phase on fish catches in
trap fisheries in coral reefs remains unclear. However, interactions of the three factors did
not have a significant influence at the 5% probability level in any of the three reservoirs.
On the other hand, reduced catch efficiencies in a large proportion of boats during dry
season, low water periods and full moon phase might bring about low income for some of
the fishers. Collective agreements in management decisions within fishing communities
are difficult when the income derived from the fishery shows high individual variations
(Amarasinghe & De Silva, 1999). Although most fisheries are characterised by highly
variable daily catches, this aspect has received less attention. The present analysis indicates
that individual variations in catch rates should also be considered as an important factor
282 B. Sricharoendham, U.S. Amarasinghe, R.P.P.K. Jayasinghe & S.M. Aypa

to be considered in defining co-management strategies. Van Oostenbrugge et al. (2001)

have shown that, due to highly variable daily fish catches, fishers in a coastal fishery in
the Moluccas, Indonesia, were seriously constrained in optimising the outcome of their
fishery through the spatial allocation of effort. As daily catch rates of individual boats
vary seasonally, forcing some fishers to rely on alternative sources of income to meet their
day-to-day expenses, this aspect needs serious attention in defining effective management
strategies for the artisanal fisheries in lakes and reservoirs.
In Ubolratana, monthly fish yield is influenced by the hydrological regime, particularly
inflow and water level. It is known that during rainy seasons with high inflow rates into
reservoirs, rheophilic fish species move towards riverine habitats (De Silva, 1983; Silva
& Davies, 1986). The positive influence of inflow on monthly fish catches might be due
to increased catchability of rheophilic species. Increased catchability of fish concentrated
in the water body during low water level is known to bring about high yields in reservoirs
(Amarasinghe & Pitcher, 1986; Amarasinghe, 1987). It is likely that a similar situation
exists in Ubolratana, where a unique socio-economic influence is the increased fish harvest
from January to May, due to increased fishing effort from the part-time fishers who are less
engaged in paddy cultivation during this period. The influence of management strategies
on fish harvests is also evident in Ubolratana. As fishing with some gear types, especially
liftnets, is prohibited during the fish-spawning season from mid-May to mid-September,
fish catches decrease. Multivariate statistical analyses (i.e., Bray-Curtis Similarity and
MDS) also indicated the spatial differences in species composition in fish landings. Since
these differences are attributed to differences in fish behaviour with increased seasonal
flow, increased catchability of some fish species during the seasons of low water level
and/or seasonal changes in fishing intensity, which are essentially reflected by seasonal
and spatial differences in species composition, should be taken into account in developing
management plans for the Ubolratana fishery.
The present analysis also indicates that in Sri Lankan reservoirs with low species
richness, introduced cichlids produce high yields. However, in Ubolratana reservoir,
although introduced cichlids are present, their contribution is only about 9.5% of the
total landings. As species richness is high in Ubolratana, food and other resources are
efficiently utilised by the constituent species in the fish assemblage so that exotic cichlids
cannot sustain high biomasses. In Lake Taal too, tilapias formed about 12% of the total
landings. The extensive pelagic zone in Lake Taal is not conducive to cichlids sustaining
high biomasses. The pelagic zone occupied by the indigenous clupeid, S. tawilis, shows
better performance in Lake Taal than cichlids. However, it must be noted that an extensive
pelagic zone is not the only factor favouring colonisation by clupeids in lakes and
reservoirs but ample availability of their preferred food zooplankton (Duncan 1999).
Chapter 15 – Status of the capture fisheries in four Asian reservoirs and... 283

References
Amarasinghe, U.S. 1987. Status of the fishery of Pimburettewa wewa, a man-made lake in Sri Lanka.
Aquacult. Fish. Manage. 18: 375-385.
Amarasinghe, U.S. 1998. Reservoir fisheries management in Sri Lanka: Achievements, mistakes
and lessons for future. Int. Rev. Hydrobiol. Special Issue 83: 523-530.
Amarasinghe, U.S. 2003. New approaches to inland capture fisheries statistics in Sri Lanka. in:
New approaches for the improvement of inland capture fishery statistics in the Mekong Basin.
Proceedings of the Ad-hoc expert consultation, Udon Thani, Thailand 2-5 September 2003,
RAP Publication 2003/01, FAO Regional Office for Asia and the Pacific, Bangkok, Thailand:
116-121.
Amarasinghe, U.S. & De Silva, S.S. 1992. Empirical approaches for evaluating the efficiencies of
different fishing methods in tropical, shallow reservoirs: a Sri Lankan case study. in: De Silva,
S.S. (ed) Reservoir Fisheries of Asia. International Development Research Centre, Ottawa,
Canada: 217-227.
Amarasinghe, U.S. & De Silva, S.S. 1999. The Sri Lankan reservoir fishery: a case for introduction of
a co-management strategy. Fish. Manage. Ecol. 6: 387-399.
Amarasinghe, U.S. & Pitcher, T.J. 1986. Assessment of fishing effort in Parakrama Samudra, an
ancient man-made lake in Sri Lanka. Fish. Res. 4: 271-282.
Bayley, P.B. 1988. Accounting for effort when comparing tropical fisheries in lakes, river-floodplains
and lagoons. Limnol. Oceanogr. 33: 963-972.
Bhukaswan, T. 1985. The Nam Pong Basin (Thailand). in: Petr, T. (ed) Inland Fisheries in Multiple-
purpose River Basin Planning and Development in Tropical Asian Countries. FAO Fish Tech.
Pap. 265, FAO, Rome, Italy: 55-90.
Bray, J.R. & Curtis, J.T. 1957. An ordination of the upland forest communities of southern
Wisconsin. Ecol. Monogr. 27: 325-349.
Chantakanon, B., Harnpicharnchai, L. & Kusyakorn, N. 1993. The study on fisheries socio-economics
in the vicinity of Ubolratana Reservoir in 1992. Khon Kaen Inland Fisheries Development
Centre, Department of Fisheries, Bangkok, Thailand. 52pp (in Thai).
Clarke, K.R. & Gorley, R.N. 2001. PRIMER v5: User manual/Tutorial. PRIMER-E Ltd, Plymouth,
U.K. 91pp.
Clarke, K.R. & Warwick, R.M. 1994. Change in marine communities: an approach to statistical
analysis and interpretation. Plymouth Marine Laboratory, Plymouth, U.K. 144pp.
de Jesus, C.C. 1998. Socio-economic dimensions of the Philippine Lake Taal. Paper presented at the
first FISHSTRAT workshop held on 4-18 April 1998, Kandy, Sri Lanka.
De Silva, S.S. 1983. Reproductive strategies of some major fish species in Parakrama Samudra
reservoir and their possible impact on the ecosystem – a theoretical consideration. in: Schiemer,
F. (ed) Limnology of Parakrama Samudra, Sri Lanka. A case study of an ancient man-made
lake in the tropics. Dev. Hydrobiol. 12, Dr. W. Junk Publishers, The Hague, The Netherlands:
185–191.
De Silva, S.S. 1988. Reservoirs of Sri Lanka and their fisheries. FAO Fish. Tech. Pap. 298: 128pp.
De Silva, S.S. 1996. The Asian inland fishery with special reference to reservoir fisheries: A
reappraisal. in: Schiemer, F. & Boland, K.T. (eds) Perspectives in Tropical Limnology. SPB
Academic Publishing, Amsterdam, The Netherlands: 321-332.
De Silva, S.S. & Amarasinghe, U.S. 1996. Reservoir Fisheries in Asia. in: De Silva, S.S. (ed)
Perspectives in Asian Fisheries - A volume to commemorate the 10th Anniversary of the Asian
Fisheries Society. Asian Fisheries Society, Manila, Philippines: 189-216.
De Silva, S.S., Amarasinghe, U.S., Nissanka, C., Wijesuriya, W.A.D.D. & Fernando, M.J.J. 2001.
Use of geographical information systems as a tool for predicting fish yield in tropical reservoirs:
case study on Sri Lankan reservoirs. Fish. Manage. Ecol. 8: 47-60.
284 B. Sricharoendham, U.S. Amarasinghe, R.P.P.K. Jayasinghe & S.M. Aypa

Duangsawasdi, M. & Raikuntod, V. 1973. The fisheries survey in the Ubolratana Reservoir. Annual
Report, The Extension Fish Production in Ubolratana Reservoir Unit, Freshwater Fisheries
Division, Department of Fisheries, Bangkok, Thailand: 73-89 (in Thai).
Duncan, A. 1999. Pelagic fish and fisheries in Asian and African lakes and reservoirs. in: van
Densen, W.L.T. & Morris, M.J. (eds) Fish and Fisheries of Lakes and Reservoirs in Southeast
Asia and Africa. Westbury Publishing, Otley, U.K.: 347-382.
Fernando, C.H. 1967. The beach seine fishery in Ceylon’s freshwaters. IPFC Curr. Aff. Bull. 50:
1-21.
Fernando, C.H. & Holčik, J. 1982. The nature of fish communities: A factor influencing the fishery
potential and yields of tropical lakes and reservoirs. Hydrobiologia 97: 127-140.
Fernando, C.H. & Holčik, J. 1991. Fish in reservoirs. Int. Rev. Gesamten Hydrobiol. 76: 149-167.
Gulland, J.A. 1983. Fish Stock Assessment: A Manual of Basic Methods. FAO/Wiley Series on
Food and Agriculture, Volume 1. John Wiley & Sons, Chichester. 223pp.
Jutagate, T., De Silva, S.S. & Mattson, N.S. 2003. Yield, growth and mortality rate of the Thai river
sprat, Clupeichthys aesarnensis, in Sirinthorn Reservoir, Thailand. Fish. Manage. Ecol. 10:
221-231.
Mahon, R. & Hunte, W. 2001. Trap mesh selectivity and the management of reef fishes Fish and
Fisheries 2: 356-375.
Mekong Secretariat, 1972. Socio-economic aspects of fisheries operations in the Nam Pong
Reservoir, 1970-1971. Eco. Comm. for Asia and Far East. 84pp.
Moreau, J. & De Silva, S.S. 1991. Predictive fish yield models for lakes and reservoirs of the
Philippines, Sri Lanka and Thailand. FAO Fish. Tech. Pap. 319: 42pp.
Nguyen, T.T.T. & De Silva, S.S. 2005. Freshwater finfish biodiversity – an Asian perspective. 1.
Current status. Aquaculture Asia 10(1): 11-17.
Nissanka, C., Amarasinghe, U.S. & De Silva, S.S. 2000. Yield predictive models for the Sri Lankan
reservoir fishery. Fish. Manage. Ecol. 7: 425-436.
Rapera, C.L. 2000. The economics of open catch fishery in Taal Lake. Professorial lecture for
the Metro Manila Commission Professorial Chair, College of Economics and Management,
University of the Philippines, Los Baños, College, Laguna, Philippines.
Ryan, B. & Joiner, B.L. 2001. Minitab Handbook. 4th edn. Duxbury, California. 464pp.
Shiklomanov, I.A. 1993. World freshwater resources. In: Gleick, P.H. (ed) Water in Crisis: A Guide
to the World’s Fresh Water Resources. Oxford University Press, New York, USA: 13-24.
Silva, E.I.L. & Davies, R.W. 1986. Movement of some indigenous riverine fish in Sri Lanka.
Hydrobiologia 137: 265-270.
Siriweera, W.I. 1994. A study of the economic history of pre-modern Sri Lanka. Vikas Publishing
House Pvt Ltd. New Delhi, India. 182pp.
van Oostenbrugge, J.A.E., van Densen, W.L.T. & Machiels, M.A.M. 2001. Risk aversion in allocating
fishing effort in a highly uncertain coastal fishery for pelagic fish, Moluccas, Indonesia. Can. J.
Fish. Aquat. Sci. 58(8): 1685-1691.
Villanueva, L.S., Luistro, A.P. & Calabig, C.S. 1995. Assessment of Taal Lake capture fisheries with
emphasis on the exploitation of Harengula tawilis. Research report submitted to Dept. of Agri.,
Reg. Office No. IV, Quezon City, Philippines. 9pp.
Welcomme, R.L. 2001. Inland Fisheries: Ecology and Management, Fishing News Books,
Blackwell Science, Oxford, U.K. 358pp.
Zar, J.H. 2000. Biostatistical Analysis, 4th edn. Prentice Hall, New Jersey, USA. 663pp.