SOME ASPECTS OF THE LIMNOLOGY
OF
KANDY LAKE, SRI LANKA
Thesis submitted
for the
Degree of Master of Philosophy
in the
University of Kelaniya, Sri Lanka
by
Samaradiwakara Rajapaksha Mohottalalage Swamapali Samaradiwakara (B.Sc.) Institute of Fundamental Studies Hantana Road Kandy Sri Lanka
December 2003 n
I do hereby declare that the work reported in this thesis was exclusively carried out by me under the supervision of Professor E. I. L. Silva, Institute of Fundamental
Studies, Kandy and Professor U. S. Amarasinghe, Department of Zoology,
University of Kelaniya. It describes the results of my own independent research except where due reference has been made in the text. No part of this thesis has been submitted earlier or concurrently for the same or any other degree.
^tnz£res,' rCrC
Date Signature of the candidate
Certified by:
1 .Name and Signature of the supervisor:
Date:
2.Name and Signature of the supervisor:
Date: Ill
Acknowledgements
I am highly gratitude to my supervisor Prof. E. I. L. Silva, the Project Leader of Ecology and Environmental Biology Project, Institute of Fundamental Studies,
Kandy, for his guiding and continuous encouragement throughout my study. Also I am very much appreciate the invaluable guidance, constructive comments of my internal supervisor Prof. U. S. Amarasinghe, Department of Zoology, University of
Kelaniya. Without their untiring guidance, this thesis would not have become a reality.
I would like to extend my gratitude to Prof. K. Thennakone, Director, IFS for permission granted to continue the work until completion of this thesis at IFS. I wish to thank the other members of IFS staff who offered their support in many ways and
I regret that they could not be acknowledged here individually.
I thank very much to Director, Upper Mahaweli Watershed Management
Project, Dam site, Polgolla for providing me a land use map of the watershed area of
Kandy Lake, and the Municipal Commissioner of Kandy for granting me permission to field sampling form Kandy Lake.
My sincere thank goes to my husband Namal P. Athukorala for his continuous assistance in the field as well as laboratory analysis and all the other supports to make this effort success. I acknowledge the valuable contribution of Dr. IV
(Ms) S. Nathanael for reading the draft and constructive comments with her busy schedule.
I will never forget the kindnesses and helpfulness of Mrs. G.A.R.K. Gamlath,
Ms. I. Thumpela, Mrs. T. Weligodapola, and Mrs. N. R. N. Silva in the laboratory work at IFS. My Special thank goes to Dr. (Mrs.) Yasantha Mapatuna, former
Director (Acting) of the Department of National Museums, Colombo for her encouragement. My sincere thanks are also to Mr. C. Weerasinghe, Mr. R. Perera,
Mrs. Manaram de Silva, Mr. Lalith Kariyawasam, Mr. S. Jayasekara and Mr. U.
Rajapaksha for their kind support in computer works. I wish to thank Dr. C.
Nissanka, Ms. T. Ariyananda, Mr. L. R. Wijepala and L. P. Mahinda for their moral support extended during this study.
Finally, I take this opportunity to dedicate my special gratitude to my late beloved farther, my loving mother, my sisters and brothers and Mr. Sirimanna and family for their encouragements and fullest support in completing this task. To my beloved father and my loving mother and my teachers and to Viduni
.i. VI
Contents
Declaration
Acknowledgements
Dedication
Contents
List of tables
List of figures
List of plates
Abstract
CHAPTER 1: INTRODUCTION
1.1 Freshwater
1.2 Water quality issues
1.2.1. Eutropbication
1.2.2. Organic and inorganic pollutants
1.2.3. Bioaccumulation and bio-magnification
1.2.4. Acidification
1.2.5. Salinization
1.2.6. Stream flow regulation
1.2.7. Sedimentation
1.3 Limnology VII
1.3.1 Limnology in the tropics
1.3.2 Limnology in Sri Lanka
1.4 Human interventions on inland water bodies
1.5 Kandy Lake
1.6 Objectives
CHAPTER 2: MATERIALS AND METHODS
2.1. Study site
2.1.1 Physical settings
2.1.2. Morphology and hydrology
2.1.3 Watershed
2.2. Methods
2.2.1 Laboratory analysis
2.2.2 Phytoplankton primary productivity
2.2.3 Chlorophyll-a analysis
2.2.4 Phytoplankton composition
2.3. Data analysis
CHAPTER 3: RESULTS
3.1 Physicochemical characteristics
3.1.1 Temperature
3.1.2 PH
3.1.3 Total alkalinity vm
3.1.5 Total suspended solids and turbidity
3.1.6 Dissolved oxygen
3.1.7 BOD5
3.1.8 Major cations and anion
3.1.9 Inter-relationships of physico-chemical characteristics
3.2 Nutrients
3.2.1. Dissolved phosphorous and total phosphorous
3.2.2. Nitrogen species
3.3 Photosynthesis and primary production
3.3.1 Light climate
3.3.2 Chlorophyll-a
3.3.3 Photosynthesis
3.3.4 Community respiration
3.3.5 Primary production
3.4 Phytoplankton community
3.4.1 Species composition and abundance of phytoplankton
3.4.2 Spatial distribution of phytoplankton community
3.4.3 Seasonal fluctuations of the plankton community
3.4.4 Phytoplankton species diversity
3.4.5 Vertical distribution of phytoplankton community CHAPTER 4: DISCUSSION
5 Conclusions
6. References X
List of Tables
No Content Page
3.1 Ranges and mean values (±SD) of physicochemical parameters of 48
surface and bottom waters of four sites for first 12 month period.
3.2 Results of one-way ANOVA for inter-site variation of 51
physicochemical characteristics of surface water for first 12 month
period.
3.3 Results of one-way ANOVA for inter-site variation of 52
physicochemical characteristics of bottom waters for first 12 month
period.
3.4 Results of Student's t-test for intra-site variation of physicochemical 53
parameters between surface and bottom waters at station A.
3.5 Relationships of independent and dependent physicochemical 67
parameters examined using least square model.
3.6 Results of one-way ANOVA for inter-site variation of micronutrients 72
in surface water of Kandy Lake during the study period.
3.7 Results of one-way ANOVA for inter-site variation of micronutrients 73
in bottom water of Kandy Lake during the study period.
3.8 Results of Student's t-test for intra-site variation of micronutrients 74
between surface and bottom water at site A of Kandy Lake during the
study period.
3.9 Abbreviations used in the text. 80 XI
3.10 The mean total incoming radiation (TIR) and photosynthesis available
radiation (PhAR) within the period of bottle suspension for
photosynthetic experiment in Kandy Lake.
3.11 Photosynthetic characteristics and related parameters of Kandy Lake.
3.12 Phytoplankton species and their abundance. xn
List of Figures
No Content Page
2.1 Location and bathymetry map of Kandy Lake and sampling 32
stations.
2.2 Landuse map of the watershed of Kandy Lake. 37
3.1 Monthly variation of water temperature during the study period. 54
3.2 Temperature profile at station A of Kandy Lake. 55
3.3 Monthly variation of pH, Electrical conductivity and total alkalinity 57
of surface and bottom water at site A.
3.4 Mean values of total suspended solids and turbidity of surface 60
water in Kandy Lake in to relation low, moderate and high rainfall
seasons.
3.5 Monthly mean values of turbidity of surface water in Kandy Lake 60
during the study period.
3.6 Vertical distribution of dissolved oxygen concentration in Kandy 62
Lake during the study period.
3.7 Monthly variation of dissolved oxygen concentration in surface and 62
bottom water at site A.
3.8 Monthly variation of BOD5 in surface and bottom water at site A of 63
Kandy Lake.
3.9 Maucha's diagram to illustrate the relative ionic proportions of 65 XIII
major cations and anions in Kandy Lake.
3.10 Scatter plots of interrelationships of physico-chemical parameters. 66
3.11 (a) Monthly variation of dissolved phosphorous and (b) monthly 70
variation of total phosphorous in surface and bottom water of
Kandy Lake during the study period.
3.12 Monthly variation of NO 2 - N, NO 3 - N and NH4 - N in the surface 76
and bottom water at station A of Kandy Lake during the study
period.
3.13 Scatter diagram of NH4 - N concentration vs Dissolved Oxygen in 78
the bottom water of site A of Kandy Lake.
3.14 The mean photosynthetically available radiation (PhAR) during 82
incubation for photosynthetic experiments in Kandy Lake.
3.15 Monthly mean percentage of the total incoming radiation (TIR %) 82
during incubation.
3.16 Monthly mean values of Secchi depth (ZSD) of the Kandy Lake 84
during the study period.
3.17 Monthly fluctuation of compensation depth (Zc) at site A of Kandy 84
Lake during the photosynthetic experiments.
3.18 Monthly values of Chlorophyll-a concentration in surface and 86
bottom water at site A of Kandy Lake for a period of two
consecutive years.
3.19 Scatter diagram of Secchi depth vs chlorophyll-a concentration of 86
surface water at site A of Kandy Lake during the dry and lower
rainy months. XIV
3.20 Relationship between total phosphorous and log chlorophyll-a 87
values of Kandy Lake during the study.
3.21 Depth profile of overall mean Chlorophyll-a concentrations of 88
Kandy Lake at site A.
3.22 Depth profiles of photosynthesis and community respiration during 90
the study period.
3.23 Monthly variation of Amax and 0 of Kandy Lake during the study 92
period.
3.24 Scatter diagrams of (a) Amax vs chlorophyll - a and (b) log a vs 92
chlorophyll-a.
3.25 Monthly variation of Chlorophyll-a at the depth of the A max and 0 93
max of Kandy Lake during the study period.
3.26 The monthly variation of (a) area based community respiration 95
(£R) and area based gross primary productivity (£A) and (b) and
daily rates of area based community respiration C££R) and area
based gross primary productivity C££A) m Kandy Lake.
3.27 Relationship between area based gross primary production per hour 96
(£A) and light saturated optimum production per hour (Amax)-
3.28 Relative composition of major taxonomic groups of phytoplankton 99
in the Kandy Lake during the period under investigation.
3.29 Total phytoplankton density of colonies per liter at the each station 105
of Kandy Lake during the first twelve months of the study.
3.30 Seasonal changes of major genera of phytoplankton with the mean 108
chlorophyll-a biomass. XV
3.31 Monthly variation of rainfall and Shannon index for phytoplankton 109
species diversity of Kandy Lake dining the study period
3.32 Vertical distribution of me major phytoplankton genera in different 111
depths along with the chlorophyll-a concentration. XVI
List of Plates
No Content Page
1 Aesthetic value of Kandy Lake. 30
2 The locations of the sampling stations with their landmarks. 40
3 Some phytoplankton species found in Kandy lake. 102
4 Phytoplankton species composition at each sampling stations. 106
5 Observed shifts of phytoplankton composition during the study. 110
I XVII
Abstract
One of the most crucial water pollution problems prevalent in lakes and
reservoirs located in urban environments is cultural eutrophication resulting in
impairment of water for intended uses such as drinking, food processing,
recreation, aesthetic purposes etc., Cultural eutrophication is set off due to
nutrient enrichment (nitrogen and phosphorous) by anthropogenic events rather
than natural processes. A man-made water body, Kandy Lake located in the heart
of the second largest city adjoining a world's famous Buddhist temple in Sri
Lanka was investigated for some aspects of its limnological characteristics at
four sites established along the longitudinal axis in similar distances, monthly
from September 1996 to August 1998. Spatial and temporal variations of
physicochemical characteristics and nutrient concentrations of both surface and
bottom waters at four sites were examined along with the measurements of
photosynthetic rate and phytoplankton community structure in order to
determining the present status of limnology and trophic status of the water body.
Physicochemical characteristics of surface water did not show significant
spatial variation demonstrating a homogeneous condition in Kandy Lake, but marked vertical gradients of some physicochemical parameters were established in the deep basin located towards the dam of the Lake. The seasonal change in surface water temperature of Kandy Lake was between 24.5 °C - 32.0 °C.
Although it was within the tropical range, daytime temperature of Kandy Lake varied within a narrow range with a marked vertical decrease establishing a XVIII
micro temperature/density stratification. At the deep basin, pH in surface water ranged between 7.4 - 8.76 and in the bottom water it was 6.18-7.40 showing alkaline to acidic nature of water column. Dissolved oxygen concentration at the deepest site was characteristic with a clinograde oxygen profile with occurrence of anoxia during some occasions. Calcium and bicarbonate dominant Lake water had specific conductivity around 300 uScm*1 and which is about threefold of inland surface waters found at similar elevations.
The mean ratio of total inorganic nitrogen to total phosphorus was about 16 indicating phosphorus limiting condition but the non-detectable concentrations of dissolved phosphorous was evident of rapid uptake by phytoplankton. The presence of NH4-N, the most dominant nitrogen species and its upper limit (>2 mg l"1) indicates extremely high heterotrophic potential in bottom water under anoxic conditions in the deep basin. The N and P dynamics coupled with the annual rainfall pattern and monsoon bound wind mixing determined the seasonal periodicity of phytoplankton species during the study period. The water^body received sufficient incident light or Total Incoming Radiation and^in turn, photosynthetically active radiation under tropical weather. Underwater light transmission and euphoric depth was high during the rainy season and it decreased significantly under dry weather conditions primarily due to self- shading by phytoplankton. Photosynthetic profiles were characteristic with marked surface photo inhibitions on many occasions resulting in maximum light saturated photosynthesis at sub-surface water. The area based net primary production was positive during the study period when vertical photosynthetic XIX
rate was integrated up to euphotic depth with community respiration for the
calculation of area based primary productivity. This indicates the autotrophic
nature of Kandy Lake during the study period. The relationship between
chlorophyll-a to total phosphorous showed eutrophic conditions of this water
body throughout the study period.
The chlorophyll-a, which showed fluctuations (14-34 ug l"1) under wet and
dry weather condition, had a negative correlation with the Secchi disc visibility.
Thirty-seven phytoplankton taxa commonly found in tropical waters were
recorded in Kandy during the study period with a recently described centric
diatom. Of the phytoplankton assemblage, Pediastrum simplex, a Chlorophyte
commonly found in eutrophic tropical water bodies and Aulacoseira granulata, a
chain forming centric diatom, known to be a climax species in the tropics,
showed a rhythmic oscillation under wet and dry weather conditions. Microcystis
aeruginosa, non-nitrogen fixing form of a bloom forming cyanobacteria,
although present in small numbers, showed a progressive increase towards the
end of the study. Evidently, long-term enrichment resulting from both
anthropogenic and natural processes led to progressive eutrophication of the
water body during the study period. Implementation of either top-down or
bottom-up strategies or a combination of both for the restoration of Kandy Lake
may be possible after a clear understanding on food web interactions and external and internal nutrient loading of this unique water body, is achieved. CHAPTER 1
INTRODUCTION
1.1 Freshwater
About 70% of the earth's surface is covered by water. In terms of volume,
97.20% of it is in the ocean. 2.15% of water remains as frozen water in glaziers.
Available freshwater on land and air is only 0.65% (van der Leeden, et al, 1990).
Though, freshwater is essential for all living beings for their existence, it occupies a relatively small portion on earth compared to seawater. Freshwater is also essential
for agriculture, industrial and mining sectors, domestic and urban uses, recreation and many other human activities, Thus, freshwater is a life supporting, commodity.
Therefore human associations with freshwater systems date back to prehistoric time.
Of the total volume of freshwater on land and air, groundwater comprises 97.54%.
The percentage of surface water is about 1.5% and the balance remains in soil
moisture and seepage (0.8%), and water vapor in the atmosphere (0.16%) (van der
Leeden, etal, 1990).
Surface water is broadly categorized into running water in lotic ecosystems
(rivers, streams, tributaries etc.) and standing water in lentic ecosystems (lakes,
reservoirs, ponds etc.). The primary benefits of standing water bodies for humans are
immense. They provide services such as transportation, navigation, irrigation, energy
production, recreation, aesthetic value and they are also excellent sites for scientific
research and education. They also provide a wide variety of resources such as food,
l natural products, sand, minerals etc. In addition, swamps and bogs are capable of removing and retaining nutrients and toxic substances that could result in pollution.
Ecologically speaking, these surface water bodies are strongly interconnected with their respective watersheds or drainage basins. The global scenario today is that most of these aquatic ecosystems are either polluted or endangered primarily due to anthropogenic activities taking place in their watersheds or within the water bodies.
Therefore, scientific knowledge on aquatic ecosystems is increasingly becoming a crucial factor for sustainable management and conservation of freshwater resources along with socio-economic development. Water with good quality is crucial for safe drinking and sustainable socio-economic development. Though it is costly, contaminated water must be purified before use. Therefore, it is imperative to build up awareness among people to protect water resources from quality deterioration for sustainable use and conservation of aquatic life.
1.2 Water Quality Issues
Surface or ground water can be contaminated either due to point source or
non-point source (diffuse) pollution. Direct discharge of industrial effluent or urban
wastes into waterways is considered as point sources pollution. Agricultural wastes
are also sometimes discharged directly into waterways as return flows whereas
agricultural runoff, drained from rural settlements and dry and wet atmospheric
depositions cause non-point source pollution. When water bodies are enriched with
nitrogen and phosphorous compounds it is known as eutrophication. Acidification is
a phenomenon that results mainly from acid rain. A majority of water bodies in the
2 arid and semi arid regions of the world have been subjected to salinization due to increase in salt content resulting from irrigated agriculture and dry land farming. On the other hand, surface water bodies can also be contaminated with toxic chemicals resulting from industries including mining and clinical wastes. Eutrophic water bodies, which promote growth of cyanobacteria, may also produce toxic chemicals known as cyano-toxins. Heavy load of suspended and bed load sediment transportation into surface water bodies, especially in the case of unplanned reservoirs have resulted in operational difficulties. Deterioration of surface water quality as a result of pollution and sedimentation affects the beneficial uses, aesthetic value and aquatic bio-diversity.
1.2.1 Eutrophication
The most obvious pervasive and persistent water quality problem is that of eutrophication. Lakes and reservoirs have deteriorated through excessive addition of plant nutrients (P and N), organic matter, and silt, which combine to produce increased algae and rooted plant biomass, reduce water quality, and usually decrease lake or reservoir volume. Water bodies under this condition loose much of their beauty, their attractiveness for recreation, and their usefulness and safety as industrial and domestic water supplies. Exotic plants and animals invade eutrophic lakes and reservoirs and infestation of these organisms may become so dense that many uses of the water are curtailed. When located in densely populated areas, lakes and reservoirs may become polluted by inadequate treatment, or non-treatment and by human and animal wastes. This cause eutrophication, and makes the water unfit
for human consumption and increases health risk to water users.
Symptoms of eutrophication such as algal blooms (including surface scum),
low transparency, rapid loss of water volume in reservoirs, noxious odour, tainted
fish flesh impaired potable water supplies, dissolved oxygen depletion, fish kills and
development of nuisance or exotic animal populations can bring about economic
losses in the form of decreased property values, high cost of treatment of raw
drinking water, illness, depressed recreation industry, expenditure for management
and restoration and the need to build new reservoirs. Warm tropical and sub-tropical
water bodies are more susceptible for eutrophication than their counterparts in the
temperate region (Tailing & Lemoalle, 1998). Deep-water anoxia may be more
prevalent in highly eutrophic tropical lakes and reservoirs. Oxygen, which is
saturated in the surface water, declines throughout the water column with the decay
of settling organic detritus. Oxygen is totally depleted in the hypolimnion and H2S
and NH3 become abundant resulting in deep water becoming unsuitable for fish life.
Changes in transparency in eutrophic lakes and reservoirs may be caused by
increasing turbidity due to increasing concentration of suspended organic and
inorganic particles or due to increasing plankton biomass. Increase in mineral materials and inorganic substances is caused by:
a. turbid inflow in fluctuations with high watershed erosion rate; b. re-suspension of bottom sediment by wave action in shallow lakes and
reservoirs;
c. shallow basins of large water bodies;
d. shoreline erosion due to wave action;
e. surface water erosion of unconsolidated shoreline materials.
Depletion of oxygen in the bottom water with the onset of anoxia results in
the mobilization of phosphorus and other elements from the sediment In anoxic
water, nitrogen is commonly found as ammonium and nitrite with most reduced
forms such as Mn+2, Fe+2, H2S and CH4. The presence of NlVmay severely impair
the use of the water particularly as a drinking water source (due to odor, taste,
1 precipitation of metals etc.). These losses can be one of the most detrimental effects
of eutrophicaiton. In many tropical lakes and reservoirs, the onset and progression of
eutrophication result in increased growth of macrophytes, which in turn provide an
increased habitat for waterfowl and snails, which serve as a secondary host for many
human parasites.
1.2.2 Organic and inorganic pollutants
In the case of organic and inorganic pollution of freshwaters, organic matter
which is either allocthonous or autochthonous is largely originated from plant
detritus although some animal debris may also be present (Chapman, 1992). The
rapid decomposition of organic debris by bacteria leads to oxygen depletion, which
can asphyxiate fish and other aquatic organisms. In industrial societies, factories are
5 often directly connected to sewers and discharge metal and organic chemical pollutants to the effluent treatment system. In many case, such elements are highly toxic to aquatic organisms and should be subjected to analysis using bioassays.
Another variety of organic pollutants is toxic organic chemicals in pesticide and fertilizer runoff such as organochlorine and organophosphate compounds and hydrocarbons. They are hazardous to human health since they bioaccumulate in tissues. Among the inorganic pollutants heavy metals are the other major causes of impairment to surface water bodies. Although several have been found in runoff, lead is generally of greatest concern (Moore, 1989). Lead originates primarily from automobile exhaust (Corrin & Natusch, 1997). During rainfall, much of the lead in road dirt is transported as urban runoff. Industrial areas are also likely to generate runoff-containing cadmium, chromium, ferrous and copper, all used in manufacturing and in electroplating.
1.2.3 Bioaccumulation and bio-magnification
These processes are very important in the distribution of toxic substances in lake and reservoir ecosystems. Bioaccumulation and food chain amplification of toxic compounds or bio - magnification determines the exposure and consequent effects on these substances at each trophic level (Moore, 1989). Even in lakes and reservoirs remote from direct industrial discharges or heavy automobile traffic, they may be contaminated by toxic nutrients such as trace metals and dioxins, which can be bio-accumulated in fish. These toxic materials, including dioxins and pesticide residues can be bio-accumulated in fishes including commercially important ones. 1.2.4 Acidification
One of the major issues related to lakes and reservoirs in particular and to
freshwater in general is the progressive acidification associated with deposition of
rain and particulate (wet and dry deposition) enrichment of mineral acids (Bartram &
Balance, 1996). The problem is characteristics of lakes and reservoirs in specific
regions of the world, which satisfies two major critical conditions: Lakes and
reservoirs must have soft water (i.e. low hardness, conductivity and dissolved salts)
and may be subjected to acid rain. All lakes and reservoirs have some acid buffering
capacity due to the presence of dissolved salts from the watershed. However, in
carbonate terrain such as in the areas of crystalline rocks or quartz sandstones, this
buffering capacity is rapidly exhausted and free hydrogen ions create progressive
acidification of the lake. Acidification of lake and reservoir waters together with acid
leaching through watershed soils results in mobilizing of many trace elements which
are toxic to aquatic organisms. In addition, due to high concentration of free carbonic
acid (pH = 4.5 or lower) water becomes corrosive to metals and concrete as a result
of the formation of soluble bicarbonate. The ability to affect the calcium carbonate
component of concrete has led to the term 'aggressive carbonic acid' or 'aggressive
carbon dioxide', which is also termed as free carbon dioxide. The potential impacts
associated with freshwater acidification are decrease in pH and increase in aluminum and heavy metals such as manganese, copper, etc. These are responsible for fish
skeletal deformities and rapid depletion of invertebrate populations. This may subsequently affect higher species (Cresser & Edwards, 1984). 1.2.5 Salinization
Though the process of salt increase or salinization is a natural phenomenon
(Silva, 1988), today it is becoming a widespread water quality issue in arid and semi- arid regions (Maybeck et ai, 1989). It takes place due to changes in water balance, salt-water intrusion, and increased salt due to soil leaching. Salinization is becoming a wide-spread water quality issue which often leads to the loss of usable water resources including lake and reservoir water. A majority of streams and rivers draining through the dry regions are endangered by salinization resulting from irrigated agriculture and dry land farming. In virtually all cases, this is due to the mismanagement of the water for agricultural purposes. Increasing salinity has unfavourable effects on the use of water by man and the ecological sustainability of aquatic ecosystems. Aquatic ecosystems are adapted to particular salinity regimes and any changes can result in adverse impacts on the biota. Economies and societies throughout the world are depending on surface water for irrigation, industry, hydropower, potable supplies and recreation. But human populations and their technologies are expanding so rapidly and as such, protection, management and restoration of freshwaters have become critical, though the problems are poorly recognized and understood.
1.2.6 Stream flow regulation
Though at present, a majority of rivers in the world are regulated by damming or by diversion for the benefit of mankind, it is one of the major reasons
8 for deterioration of water quality. Reservoirs that are constructed in this manner are operated for irrigation water supply or hydropower generation. Since dams, reservoirs and diversion systems allow the modification of natural patterns of stream flow, it creates an artificial lake environment on the stream which changes the biotic from lotic to lentic and can increase water losses to evaporation and ground water recharge (Gordon ef al, 1992). Some impounded rivers suffer from increased downstream erosion and sedimentation (Moore, 1989). Another problem associated with these water bodies are water born diseases (such as schistosomiasis, malaria, sleeping sickness), and resettlement and loss of land used for subsistence agriculture
(Moore, 1989). In addition, reservoirs are highly susceptible to sedimentation.
Changes in the hydrologic characteristics in reservoirs and waterways due to sedimentation have always accompanied intensive agriculture activities. The sediments provide substrates for bacterial growth, and they bind many toxic heavy metals and some toxic organic compounds (Pesticides). They are transported to consumer via the drinking water supply.
1.2.7 Sedimentation
Reservoirs act as major sediment traps within the drainage basins and they will arrest downstream transport of bed-material and suspended load. Headwaters of the drainage basins provide more than 75% of the sediment load in many river systems. A majority of river systems in Sri Lanka are well known for high sediment concentrations during rainy seasons. In particular, Mahaweli, the longest river draining the largest drainage basin in the country is known as one of the highest sediment laden rivers. Due to the physical and geological heterogeneity, different factors dominate the sediment yield in different parts of the watershed.
Geomorphologic characteristics and topography dominate the sediment delivery in the respective drainage basin while precipitation pattern influenced by monsoons result in seasonal variation in sediment transport. In general, sediment data have received less priority compared to river discharge data in terms of density of network, frequency of observations and processing of data in Sri Lanka.
Human activities are responsible for two opposite effects on sediment flux: construction of dams that lead to sediment retention and deforestation, which induces erosion. Although the objective of reservoir construction is to store water, reservoirs act as man made retention systems for sediment transport by rivers. An enormous amount of sediment is retained by reservoirs every year reducing the storage capacity of dams. The nature of sediment trapping by dams is found to change with the age of the reservoir. Long term impacts on dams are reported to be lesser than the impact observed during the initial years of construction either due to engineering improvement or natural or man-made developments in the watersheds.
Land use: Although the major human induced impacts on sediment movement are the land use changes, a greater debate exists regarding such impacts in the regional context Since more than 70% of the population in Sri Lanka is engaged in agricultural activities, clearing of forests for agriculture, chena (= 'slash-and-burn') cultivations and other economic crops is a regular process. Although forest clearing takes place in every habitable area of the island, these activities are intensive in the
10 highlands. Similarly, roads and other constructions for infrastructure development activities in mountainous areas increases sediment transport due to induced landslides, earth slips and socio-economic developments.
Adequate information or data are not available for the assessment of impacts of human activities on the basin-scale delivery of sediments in Sri Lanka. Very few long-term time-series are available for assessing the sediment transport trend of the
Mahaweli River in relation to human activities taking place in the watershed.
Watershed management: Apparently, many of man's activities taking place in respective watersheds have direct or indirect effects upon river systems and associated standing water bodies. In extreme cases, the consequent environmental changes have been widespread affecting the flora and fauna in channels, riparian, flood plain habitats, lakes and reservoirs, and estuaries and lagoons. Many long term and irreversible effects cause changes in genetic diversity, or even in the extinction of certain species. Changes in river hydrology, sediment transport, and channel characteristics as a consequence of human activities have been well documented.
This accounts for extensive soil erosion and massive siltation of lakes and reservoirs.
The complete absence of wastewater treatment in many areas, strongly indicate that the lakes and reservoirs of these areas are also affected by excessive biological production and its consequences. Rapid infilling of major impoundments in the third world nations in particularly is noticeable in view of their needs for irrigation water, potable supplies and flood control. Usually deforestation and cultivation of marginal lands on steep slopes of reservoirs, yield soil losses that will fill some impoundments
U in the long run. Most of these man-made activities are applicable to Sri Lanka with
respect to degradation of reservoirs and tanks. The followings are well documented
as major watershed based human activities, which result in surface water pollution
and degradation of reservoirs and tanks in Sri Lanka.
a. deforestation for cultivation and timber (clear cutting);
b. excessive use of agrochemicals (fertilizer and pesticides);
c. cultivation and construction in steep slopes in the highland;
d. livestock and human settlement with poor sanitary facilities;
e. direct discharge of municipal and clinical wastes;
f. haphazard dumping of solid waste;
g. mining for gems and dolomite and quarrying;
h. infrastructure development;
i. construction of dams in the higher elevations.
Two major problems (sedimentation and eutrophication) related to poor catchment land use are apparent in reservoirs and tanks in Sri Lanka. Upstream reservoirs (e.g. Polgolla, Rantambe) of the most regulated river system, Mahaweli have been subjected to severe sedimentation over the last decades. The rate of sedimentation of downstream irrigation tanks is hitherto unknown. With respect to eutrophication, most of the shallow irrigation reservoirs are considered as eutrophic (Amarasinghe et
ai.y 1983).
12 Unfortunately there is no systematic national monitoring programme in Sri
Lanka to assess the status of the surface and ground water quality and their trends in pollution, This is important with respect to aquatic biodiversity, human health and
other human uses. Water related problems in the country have already been identified as eutrofhication, faecal contamination, organic and sediment loading, heavy metal pollution, salt water intrusion, salinization, increase in organic residues and prevalence of water born diseases (Anon., 2003). However, very little is known on whether Sri
Lanka's surface water resources are presently threatened, already polluted or chronically contaminated (Silva, 1999).
13 Limnology
By definition, limnology is the scientific understanding, which deals with morphometry, hydrology, physical properties and chemical constituents and the biological nature of inland water bodies (Wetzel, 1983). Although, early studies in limnology were primarily confined to ecological diversity, processes and functions in naturaJ: lakes under physical and chemical environment, present approaches are more integrated with associated watershed characteristics (Edmondson, 1994). This is because at present, most of the inland water bodies are being deteriorated to a greater extent on a global scale as a result of human activities taking place in their respective watersheds. Therefore, implementation of mitigation measures for conservation and management of inland water bodies either natural or man-made will be a < iream without having a sound knowledge on their limnology.
13 Limnology in the tropics has only recently developed since the stage of exploration, but the need for application of limnological knowledge is as pressing at tropical latitudes as it is in the temperate zone. However, the extent to which the limnology of temperate latitudes can be applied in the tropics is not always clear
(Tailing & Lemoalle, 1998). General limnological principles are often transferable across latitudes. For example, the growth of algae in a lake or reservoir at any latitude is likely to be limited by the availability of one or more key nutrients. An increase in the supply of these nutrients by humans is likely to change many of the characteristics of lakes and reservoirs. On the other hand, some limnological principles are not so easily applicable across the latitude. For example, seasonal growing cycles the in temperate region are different from those in the tropical region. Because of the relative scarcity of natural lakes, reservoirs are the predominant lake type in many regions of the tropics. This leads to the conclusion that in the tropics as a whole, limnology must be more strongly oriented toward the conservation and management of reservoirs than it has been at temperate latitude
(Lewis, 1996).
13.1 Limnology in the tropics
Although people in all parts of the world have been aware of the quality and biotic resources of their lakes and rivers, limnology emerged as a science in Europe.
It matured and flourished in Europe and North America where institutes of limnology were established in many countries. Early in the last century, scientists from these countries led expeditions to different countries of Asia, Africa and South
14 America. The three Tanganyika expeditions of 1894, 1897 and 1904-5 were evoked largely by a hypothesis finally disapproved. The data of the first Limnological
Sunda Expeditions held in 1928-1929 are the baseline for aquatic science in South
East Asia. Other early examples are: studies in Lake Nyassa in Malawi (Fuelleborn,
1900), lakes in Central America (Juday, 1924). Likewise there were only rudimentary quantitative analyses of plankton communities in Sri Lanka (Apstein,
1907, 1910). Recent studies in South Asia, for example Lake Lanao in the
Phillipines (Lewis, 1973, 1974, 1977, 1978), and Lake Parakrama Samudra,
(Schiemer, 1983) created new limnological finding on processes and functioning of tropical lakes and reservoirs. A very recent example is INCO-DC programme of
European Union funded FISHTRAT project conducted in five water bodies (four reservoirs and one lake) in Sri Lanka, Thailand and The Philippines (Amarasinghe et
al.t 2001). The objectives of the project were to determine ecosystem oriented limnological aspects, status of fish communities and fisheries potential and socio economics of the riparian fishing communities (Schiemer et al., 2001).
1.3.2 Limnology in Sri Lanka
Sri Lanka: Sri Lanka is a tropical island (05° 54'-09° 52'N and 79° 39'- 81° 53'E) located off the southern tip of the Indian peninsula. In surface configuration, Sri
Lanka comprises a highland massif situated in the north center, which is surrounded more or less by an intermediate zone of upland ridges and valleys at a lower elevation. The island is influenced by two wind regimes, the SW monsoon (summer monsoon) from May to September and NE monsoon (winter monsoon) from
15 December to February. The annual rainfall in the southwest is 2500 mm while in the northeast and southeast the annual average is less than 1950 mm. On the basis of rainfall, the island is divided into three major zones: the wet zone, the dry zone and the arid zone. There are marked climatic variations in Sri Lanka, due to the central highlands (2400 m above) being surrounded by an extensive lowland area. The average annual temperatures range from about 18°C in the central highlands to about
26°C in the lowland. Sri Lanka covers an area of 65,610 km2 together with inland waters of 1,510 km2. The Precambrian shield, which covers about 90% of the surface area of the island, consists predominantly of charnockite meta-sedimentry rocks. The rest of the geological forms are Jurassic deposits and limestone of Miocene forms developed in the northeast and northern peninsula, respectively. Until the turn of the
19th century, Sri Lanka was almost entirely covered by natural forest Because of the population pressure on forestland, at the present it is reduced into about 2.0 million ha or 22% of the land area. This remaining forest cover is rich and luxuriant, with a great variety of trees, creepers and shrubs. The population in Sri Lanka is above 18 million.
Surface water. The only surface water source of Sri Lanka is rainfall. Sri Lanka has no natural lakes. The prominent natural feature of surface water in the country is the running water in 103 river systems, which radiate from the central highland to the sea. Seasonal streams, brooks, creeks, and tributaries are the other lentic water
systems. The permanent standing water bodies are essentially man-made and they are irrigation tanks in the dry zone and hydropower reservoirs in the central
highland. There are approximately 3500 reservoirs in existence (Schiemer, 1983).
16 These cover a total surface area of about 1500 km . There are two types of irrigation reservoirs: one is major ancient irrigation tanks and the other one is seasonal tanks.
Ancient irrigation tanks were created during the twelfth century or earlier for rice cultivation. Seasonal tanks are usually shallow and less than 30 ha in surface area and collect water only during certain seasons from their own catchment Under the
Accelerated Mahaweli Programme three major reservoirs were constructed mainly for hydropower generation namely Kotmale, Victoria and Randenegala and several storage reservoirs. Further, ancient reservoirs were augmented through transfer from the Mahaweli River to the reservoirs fed by Kala Oya, Malwathu Oya and Yan Oya.
But at present most of them are multipurpose and provide drinking water, fish production in addition to irrigation supply or hydropower generation.
Some of the rivers inundate flood plains along their lower reaches. These alluvial plains ('villu') may be temporary, lasting only during the flood or it may be permanent as in the lower reaches in the Mahaweli River. These contribute a little to the surface water (125 km2). Considering the major surface water uses of Sri Lanka, it has been estimated that for domestic and industrial purposes around 1230x10** m^ is utilized and irrigation and hydropower consume around 4920x106 m3.
Limnological Studies in Sri Lanka: Though Sri Lanka has no natural lakes, the island is reputed for construction of man-made lakes by damming and diversion of streams and rivers since ancient time. Some of these cascading ancient tank systems look more natural than artificial when they are filled with water. The knowledge on limnology of these ancient water bodies is in its infancy when compared to their
17 existence. The knowledge of taxonomy of freshwater fauna in Sri Lanka stretched
out far beyond the other fields of limnology. Studies on phytoplankton in Sri Lankan
waters date back to the early twentieth century (West & West, 1902; Apstein, 1907,
1910) and were followed by several other studies (Rott, 1983; Rott & Lenzenweger,
1994; Rott et al., in press). Since late 1960s, most of the limnological studies in the
country were focused on taxonomy of aquatic flora and fauna (Fernando, 1965).
Knowledge on taxonomy of fresh water fauna of Sri Lanka is richer than any other
tropical countries (Fernando, 1990).
Although Mendis (1965) studied the limnological characteristics of 21
reservoirs in Sri Lanka no emphasis was paid to water quality characteristics except the measurement of pH, colour and turbidity using less sophisticated methods. A
limited number of water quality parameters for Minneriya, Kaudulla, Kantale and
Parakrama Samudra have been reported for several years by the Environment
Assessment Study Group of the Accelerated Mahaweli Programme (TAMS, 1980).
This information includes the results of the studies conducted by US Operation
Mission (1961), Amarasiri (1973) and TAMS (1980). Several parameters of a systematic study on water chemistry of Kalawewa was conducted in 1978-1979 by
Gunawardena and Adhikari (1981). By examining chemical composition of
Kalawewa water, they calculated the Sodium Adsorption Ratio (SAR) as one of the most important determinants of irrigation water quality. This study concludes that
SAR, which has the threshold limit of 6.0 meq l"1, is less than 2.0 meq l"1 in
Kalawewa and associated irrigation outflows. Since then, no study examined the irrigation water quality of lowland reservoirs in Sri Lanka. Most of the studies in
18 lowland reservoirs have been devoted to limnological investigations. Parakrama
Samudra has been subjected to a comprehensive limnological investigation in 1979 and 1980 by a group of Austrian limnologists in collaboration with Sri Lankan scientists (Schiemer, 1981, 1983). This study reveals the physicochemical nature, nutrient dynamics, spatial patterns in trophic characteristics and the flora and fauna of Parakrama Samudra (Dokulil et al., 1983; Gunathilake, 1980, 1983; Gunathilake
& Senaratne, 1981; Fernando & Rajapaksa, 1983). There were subsequent studies on limnology of shallow lowland reservoirs. For example, Amarasinghe et al., (2001) studied several reservoirs in Malwatu Oya, Kala Oya and Yan Oya basins (e.g.
Nachchaduwa, Nuwerawewa, Kalawewa, Rajangana and Huruluwewa). Silva and
Davies (1986,1987) compared the physicochemical characteristics in several lowland reservoirs in the Mahaweli basin (e.g. Parakrama Samudra, Kaudulla, Giritale, and
Minneriya). A similar study was conducted at Uda Walawe reservoir betweenl988 and 1989 (Chandrasoma et al., 1986). Studie on phytoplankton, chlorophyll and nutrients in several reservoirs situated in the Mahaweli basin revealed the relative importance of nutrients and pH on the dominance of cyanobacteria (Silva &
Wijeyaratne, 1999). Physicochemical characteristics of thirteen lowland reservoirs have been reported by Nissanka (2002) while Silva et al., (2002) determined the photosynthetic characteristics and water chemistry of eleven water bodies, in the lowland dry zone.
Limnological investigation in upper Mahaweli reservoirs was initiated since
1987 on a regular basis by the Mahaweli Authority of Sri Lanka. The surveys comprised of regular monitoring of the spatial and temporal patterns of the following
19 parameters: temperature, dissolved oxygen, pH, electrical conductivity, alkalinity, orthophosphate, hardness, sulphate, sulphide, nitrate, nitrite, ammonia, and turbidity.
Initial finding of these investigations reported thermal stratification, oxygen depletion and accumulation of ammonia and hydrogen sulfide in deep layers
(Piyasiri, 1991, 1992). Increasing conductivity from Kotmale to Rantambe was attributed to nutrient loading, and potential risk of algal' growth in lowland reservoirs was predicted. Concurrent studies have been conducted in Upper
Mahaweli reservoirs primarily on basic physicochemical characteristics, phytoplankton systematics and potential fish production (Silva 1991; De Sirva,
1993a,b). However none of these studies collected information relevant to nutrient enrichment and trophic characteristics (Total-N and Total-P, Chlorophyll-a, or algal bio-volume). More attention was paid on the upper Mahaweli reservoirs with the emergence of a thick cyanobacteria bloom in the uppermost Kotmale reservoir in
1991 (Piyasiri, 1995, 2000; Piyasiri & Perera, 2001). Several hypotheses were forwarded to explain the sudden outbreak of Cyanobacteria bloom in Kotmale
(Duncan et al., 1993; Piyasiri, 1995).
Further, physicochemical characteristics and phytoplankton in the Victoria reservoir was examined along with studies on colonization success of cichlid fish in the reservoir (Nathanael, 2001). A systematic comparative study (FISHSTRAT) was conducted (1998-2000) on Victoria, Minneriya and Uda Walawe reservoirs together with Ubolratana reservoir in Thailand and Lake Taal in the Philippines under the sponsorship of the European Union. The results of this study clearly demonstrate the physicochemical characteristics, trophic status, seasonal plankton dynamics in
20 Victoria, Uda Walawe and Minneriya reservoirs in relation to hydraulic balance
(Silva & Gamlath, 2000; Silva & Schiemer, 2000; Schiemer et al, 2001). No comprehensive studies have been undertaken in Sri Lankan reservoirs on pesticides, trace metals and algal toxins although some of them are used for drinking and except for a few hydropower reservoirs. Also reservoirs contribute immensely to inland fish production, the major source of animal protein of the rural poor.
In the case of taxonomy, recently published work on some rare and interesting phytoplankton species is Sri Lanka by Rott & Lenzenweger, (1994), can be used as a useful guidance for identification of phytoplankton. According to that study, while 50% of identified species are cosmopolitan, only 10% are tropicopolitan. Forty percent of tropical species belong to the rare desmids taxa.
Considering the faunal characteristics, zooplankton plays an important role in the trophic status of inland water bodies. Fernando & Rajapaksa (1983) have investigated zooplankton composition, density, distribution and seasonal changes in
Parakrama Samudra. They have indicated that no endemic species are found in Sri
Lanka. Extensive study on the zooplankton community in Kotmale reservoir has been conducted by Piyasiri & Chandrananda (1998) during a period of twelve
months in 1990-1991. Here, observations on species composition, seasonal variations, vertical and horizontal distribution, and the size class distribution of the
zooplankton community were recorded. According to the study, the vertical
distribution pattern of zooplankton has no temporal variation and the zooplankton
community is evenly distributed throughout the reservoir. 1.4 Human Interventions on inland water bodies
Sri Lanka's economy has been based on agriculture from ancient times and it will continue to be the primary source of the nation's economic activity for many more decades to come. Currently about 29% of the land area is cultivated and about
35% of the labour force is engaged in the agriculture sector. Rice paddy, the major cultivation in Sri Lanka is mainly confined to the dry zone rich in ancient irrigation reservoirs. In addition, 87% of electricity in the country is generated from recently built hydropower reservoirs. Further, some of the ancient reservoirs and newly built water bodies are used as public water supplies after conventional treatment. It has also been estimated that inland water bodies produced about 20% of the nation's fish production before the cessation of government patronage for the inland fisheries sector in 1990 (Amarasinghe, 1998). Large seal human association with standing water bodies in Sri Lanka is results in the amount of anthropogenic influences on them. According to the already available information, it would be accepted that the majority of water bodies in the country are endangered in the form of water quality.
Eutrophication, increase in organic residues, industrial effluents and prevalence of water borne diseases are the already identified water related problems in Sri Lanka.
The reservoirs in Sri Lanka receive a variety of pollutants, especially from non-point sources both within their catchments and in the water transferred from other catchments. Erosion of silt and sediment resulting from land degradation, deforestation, agriculture practices on steep slopes without soil conservation measures, unplanned infrastructure development, gem mining and quarrying
22 contribute to fluvial sediment transportation and deposition in reservoirs in the upper watersheds. The excessive use of NPK fertilizers in the agro industry, and in some recreational pursuits such as golf courts, create high levels of nutrients in the surface and sub-surface runoff, and in particular may contribute to relatively high nitrogen concentrations in stream water. In addition to normal fertilizers, sulphate-containing nitrogenous fertilizers and dolomite are regularly applied to tea plantations. It is estimated that some 50,000 tons of pesticides are applied annually in Sri Lanka
(Pesticide Registrar, personal communication), most of which are highly complex organophosphorous compounds, whilst fungicides contain persistent chemicals and salts of toxic heavy metals.
Many desk studies have reported that Sri Lankan surface waters have been contaminated with coliform bacteria (State of the Environment, Sri Lanka, 2001).
Surface water bodies located in urbanized or semi-urbanized watersheds with high population densities are lacking in formal sanitation or waste treatment facilities.
Even where sewage treatment facilities are provided, many are malfunctioning.
Cattle grazing and livestock are predominant in some local watersheds. Generally, low lying lands in urban and semi-urban watersheds are used as sites for solid waste disposal, and even formal sanitary landfills in major townships such as Kandy are often poorly designed and constructed.
Although Sri Lanka is not an industrialized country some inflow canals into reservoirs which intercept urban centres (e.g. Mid Canal of Kandy to Polgolla reservoir) carry high concentrations of trace metals and complex chemicals, sources
23 of which can be attributed to automobile exhaust fumes, service stations, and effluents from small scale industries (foundry, electroplating, galvanizing)
(Dissanayaka et al., 1982). Complex chemicals in clinical wastes, or sometimes by products of radio-isotopes, are released without treatment to surface water drains from hospitals, universities, research institutes and rubber factories. Although used on a small scale for research or clinical purposes, no facilities are available for the disposal of wastes of radio-isotopes. The lack of monitoring information for organic and inorganic chemicals and trace metals precludes any assessment of the nature and scale of such pollution, either for drinking water or with respect to impacts on fish and aquatic wildlife.
Algal blooms: Water quality problems associated with algal blooms have been reported in many reservoirs. The most recent bloom has developed in Vendarasan reservoir in
Kantale and two reservoirs have developed algal scums. Although it had previously been assumed that the deeper highland reservoirs in the Mahaweli basin are least susceptible to hyper-eutrophication, a dense thick bloom of Mycrocystis aeruginosa developed over the entire Kotmale reservoir in 1991. The "Kotmale Bloom" had serious impacts on operational activities but disappeared gradually with the onset of the northeast monsoonal rain. In 1993, a filamentous blue-green algae (Anabaena aphanizomenoides) developed in the southern basin of the Parakrama Samudra with the onset of the second inter-monsoon, but disappeared with the release of water to the command area during the following Maha season. Its presence was attributed to development activities in 'System G' of the Mahaweli Scheme. M.aeruginosa bloom has appeared in many terminal reservoirs (Silva & Wijeyaratne, 1999).
24 The occurrence of dense populations of blue-green algae in lowland reservoirs during the low water level period (July to September) is common, but they disappear as water levels increase with the onset of the northeast monsoon. However, the emergence of blooms during high water level in terminal reservoirs (Parakrama Samudra in 1993;
Nuwerawewa in 1999 and Vendarasan in 2003) cannot be readily explained. It has been hypothesized that terminal reservoirs in cascading systems are more susceptible for hypereutropbication (Schiemer & Duncan, 1988). However, some species of cyanobacteria can produce chemicals that are toxic to mammals including humans.
They are found mainly in eutrophic and hypertrophic water bodies where they can form blooms at water temperature > 20°C where there is a high phosphorous concentration.
Further, some organic compounds produced in hypertrophic water bodies are converted into carcinogenic compounds during the process of chlorination.
Aquatic weeds: In 1985, an aquatic fern, giant Salvinia (Salvinia melesta) invaded the Maduruoya reservoir three years after impoundment and about one third of the reservoir was gradually covered by the weed. Its presence had an adverse effect on the irrigation releases and fishing operations in the reservoir. Manual control methods were attempted but were not successful. Eventually masses of Salvinia strands were carried by winds to the edges and bays of the reservoir, where they accumulated when the water level dropped, and were subsequently burned. Since then, the reservoir has been virtually free of Salvinia.
Fish kills: The only large-scale fish mortality reported in Sri Lankan reservoirs in the recent past was that of a population of a small cyprinid fish, Ambfypharyngodon
25 melettinus ("Wew Salaya") in the southern basin of the Parakrama Sumudra
(Dumbutulu Wewa) in June 1998. Although it was postulated that the mortality of this small cyprinid was due to the toxin produced by a cyanobacteria, M. aeruginosa, no scientific evidence was found to prove this. The case became a major issue because it was the first incident where species-specific and site-specific mass mortality of a fish had been reported in a reservoir. In 1980s, there were fish kills in Pimburettewa reservoir and Tisawewa (Professor U. S. Amarasinghe, personal communication).
Similar issues have been reported on Beira Lake (Costa & De Silva, 1978a).
Complaints by users: No complaints have been reported about the taste or odor of water from the reservoir sources used by the National Water Supply and Drainage
Board, but discoloration is frequent during periods of low water level. Colour problems arise mainly as a result of high densities of plankton and suspended particles, which are stirred up from the bottom due to wind action.
1.5 Kandy Lake
The Kandy Lake, the study site of this investigation is the only aesthetic water body in Sri Lanka. It is located in the hill capital, Kandy, the second largest city in Sri Lanka with paramount scenic value. Being situated in the heart of Kandy, adjacent to world famous Buddhist Temple Dhalada Maligawa where the sacred tooth relic of Lord Buddha, is placed it attracts thousands of local pilgrims and foreign tourists. The Lake was built by the last King of Sri Lanka, Sri Wikrama
26 Rajasinghe between 1810 and 1812 using forced labour to add a panoramic view to the aesthetic value of the sacred city (Historical evidance).
The lake was in existence since its construction as a perennial water body and the water from the lake had been used to augment the city water supply during the mid seventies for a short period, but at present the lakes water is not used either for drinking or for other purposes. The sluice gate of the lake is kept closed and excess water spills over during the rainy season. Since it is located adjoining the Dhalada
Maligawa, this small water body has become one of the largest tourist attractions of the country. It has functioned as a supplementary drinking water source for the city before the 1980s. Today, bathing, washing and fishing in the lake is not allowed but people make pleasure trips by motor boats which have been operated for a long period of time.
The first limnological survey was carried out in Kandy Lake on meso and macro fauna and flora with analysis of a few physicochemical characteristics by De
Silva & De Silva (1984) who found no significant seasonal or temporal variation in physicochemical features. During this study period, the phosphorous concentration in the lake water ranged from 5 ug l"1 to 16.2 ug l"1 while conductivity varied around
150 uS cm*1. The pH values were indicated that the lake water was a mostly in neutral condition. Dissanayaka et al, (1982) examined the polluted status of the
Kandy Lake during the dry period in 1980, as a case study of urban aquatic ecosystem in Sri Lanka. In this study water samples were taken from the different location of the lake and drains entering into the Lake, and analyzed for chemical
27 constituents and coliform counts. This study revealed >240 uS cm of conductivity in lake water indicating that the lake receives a large amount of free ions externally.
Water samples analyzed from some inflows had about 70 mg l"1 of nitrate ion concentrations while the nitrate concentration in the open lake water ranged from 5 mg l"1 to 10 mg l"1 and the concentration of phosphorous varied around 1.75 mg l"1.
In addition, they found faecal contamination of lake water with extremely high total and faecal coliform counts. The Kandy Lake was also subjected to an analysis of trace metal contamination by Dissanayaka et al, (1986 a,b). According to them, 66 water samples were highly contaminated with Pb and Cd and the concentrations were around 150 ug l"1 and 77 ug l"1 respectively. In May 1991 and 1996 the
National Water Supply and Drainage Board also investigated the organic pollution levels in the Kandy Lake. They also found a high phosphorous concentrations in offshore water which ranged between 200 ug l"1 and 500 ug l'1, and 140 ug l*1 and
580 ug l"1 in April 1986 and in May 1991 respectively. According to an investigation carried out by the Central Environmental Authority during a period of ten months in
1991, BOD5 levels of lake waters ranged from 1.5 mg l"1 to 5.0 mg l"1 indicating that the lake water was in a low organically polluted condition (Silva, 1996). Analyzing available information and by judging from its appearance, Silva (1996) categorized
Kandy lake as a stagnant eutrophic water body susceptible to hypereutrophication.
However, blooming of nuisance algae or hypereutrophic condition has never been reported since its construction. Environmentalist and nature lovers show a keen interest in this gorgeous water body and have now claimed that Kandy Lake is a severely polluted water body and aquatic life in it is smothered. Therefore it was
28 decided to launch a detailed limnological investigation on Kandy Lake starting in
1996 with the following primary objectives.
1.6 Objectives
• To determine whether there are spatial and temporal variations in physico-
chemical parameters and nutrient concentrations;
• To determine whether there are temporal variations in phytoplankton primary
productivity and in the phytoplankton species composition; and
• To evaluate the trophic status of Kandy lake.
29 Plate 1: The study site, Kandy Lake, enhances the aesthetic value of the
Sacred City Kandy.
30 CHAPTER 2
MATERIALS AND METHODS
2.1 Study site
2.1.1 Physical settings:
The Kandy Lake (7° 17' 12"- 36" N and 80° 38'12"- 48"E) is located at 510 m above mean sea level in the heart of the Kandy city (Figure 2.1). It has a small watershed of 2.87 km2 and it is confined to the southern part of the Kandy city in
Rajapihilla, and southwest slope of the Udawattekelle forest reserve behind the
Maligawa. The northwestern corner of the Lake, in front of the Queens Hotel is the landmark for an intersection of three major main roads of the country (i.e.Colombo-
Kandy, Kandy-Jaffiia, and Kandy-Padiyatalawa). The Lake is skirted by the popular lake round road from the outlet up to the tennis court end for 3.5 km. A small islet is located more or less at the centre of the lake and the ornamental wall surrounding the western and northern boundaries of the lake also contribute to enhance its aesthetic value.
31 Figure 2.1: Location of Kandy Lake with its bathymetric map.The sampling stations (A3.C and D) are also depicted. Contours of water depth are in metres above mean sea level. 2.1.2 Morphometry and Hydrology
The Kandy Lake is 18 ha in surface area within the circumference of 3.05 km. It is a northwestwardly elongated water body which has a maximum fetch of
1.16 km, and lies in a north-west and south-east direction. The mean depth is 8.5 m while the maximum depth is 12.5 m. The depth of the lake is not uniform and the bottom is muddy. Figure 2.1 illustrates the depth contours of Kandy Lake. About
75% of the bottom area of this shallow water body is less than 4 m and the rest is more than 10 m deep. Hence the lake can be divided roughly into two small basins, shallow and deep. The relative depth of the lake is 1.72%. The mean width of the lake is 165.22 m while the maximum width is 389 m. The volume of the Lake at full supply level is 0.3839 MCM. The lake has no prominent littoral zone and the edge of the lake differs from place to place. On the western and northern sides, there is no shore as such and the water level seldom recedes from the embankment. Four major silt traps have been constructed across the inflows, around the lake and several road surface silt traps are also located at various points. The bank around the lake has been planted with large trees, which belong to different families comprising flowering, ornamental and shading plants. They are dominated by Cassia fistula
(Ehala), Mesua nagassarium (Naj, Peltophorum inerm (Mara) etc.
Hydrology: In Kandy Lake, the stream flow is the main contribution to the water input, which shows a marked variation with rainfall. Input through direct rainfall is not a very significant contributor to the water budget since the lake area is relatively small. The outflow is also comparatively low except during the rainy season
33 (October to December). Evaporation losses may be significant especially during the dry season (February to April). No information is available on the importance of seepage input and output. Water budget of the Lake is not monitored hence, calculation of residence time or its inverse, flushing rate is unlikely. The only outlet of the lake is the sluice gate that is located at the southwest corner of the lake. This is usually kept closed. The lake is flushed only during the rainy season or when some constructions are taking place in the lake.
The lake has two major inflows namely Heelpankandura Ela and
Nuwarawela Ela. Headwaters of about 1 km long Heelpankandura Ela are confined to the southeast segment of the watershed. Several small brooks (e.g. Rajawella Ela), most of which are seasonal, merge with the Heenpankandura Ela on its way downstream. A major silt trap, called Rathubokkuwa is located about 300 m upstream of the outfall of the Heelpankandura Ela into the lake. Outfall of the silt trap water flows through an underground canal up to the lake. The second major brook, popularly known as Nuwarawela Ela, which is about 2 km in stream length, originates from a small forest, which is located at the southern most end of the catchment It empties into a "U" shaped bay, which is located at the southeast end of the lake. This bay also acts as a major sediment trap and it is separated from the
Lake by a concrete wall of about 1.5 m high with an opening for water to pass through. Third brook which is also perennial in nature, but carries a very small flow under dry weather empties into the lake near Sangaraja Mawatha via a Hillwood silt trap. Though, these brooks are perennial, high flow conditions can be seen only under wet weather. Besides, their flows have been affected by various infrastructure
34 development and other human activities taking place in the catchment. As a result of these poorly managed land use practices, the Lake receives a considerable amount of silt and sediment from the surroundings.
2.13 Watershed
Geologically, Kandy is located in the Highland Series, which consists of charnokite metasedimentary rocks of Precambrian origin. The predominant rock types in the watershed of the Lake are marble, quartz, quartzite hornblende-biotitic gneiss and granitoid gneiss (Almond, 1994). The soils in this steeply dissected hilly and rolling terrain is moderately drained by reddish brown, fine textured and medium acidic podzolic (Panabokke, 1996). Kandy experiences a year round rainfall with a prominent peak during the second inter monsoon (October -November).
During the second inter-monsoon Kandy receives about 40% of the annual rainfall, which ranges between 1500 mm and 2000 mm. The rainfall during the southwest monsoon is higher than during the northeast and the lowest rainfall occurs during the first inter-monsoon (March- April). The average monthly air temperature of Kandy ranges approximately between 23°C and 27.5°C and the first three months of the year
(January-March) are considered as warmer months.
Land use: The Lake is situated in a small valley flanked by steep hills in the
Hantana Range. The present land use in the watershed is essentially urban housing and government and private sector buildings. Five leading schools are located in the
Lake catchment while there are six hotels and two private nursing homes of different
35 capacity. Land use of Kandy Lake watershed area is illustrated in Figure 2.2. About
90% of the drainage basin shows high-density residential land use and a few typical
Kandyan home gardens which is different from urban housing are also available towards the southwest boarder of the drainage basin. Private houses and other buildings are not connected to a central sewer system. An abandoned paddy field converted into a marshy land is located at the southeastern boundary of the lake where there is a luxuriant growth of Colocascea and Cyperus species. The lake catchment shows a variety of land use. Nearly 40% of the watershed, especially the area just around the lake, has undergone rapid urbanization. There are few a secondary schools, some governmental offices, nursing homes and tourist hotels of different capacities in the immediate vicinity of the lake. Vehicle repairing (garages) and washing activities are also in operation in the vicinity. Although urbanized, there are no minor or major industries within the catchment. The lake itself is circumscribed by major roads with heavy inflow of traffic throughout the daytime.
Being situated at a low elevation amidst a steep urbanized watershed, the lake receives surface runoff and domestic waste water carried by effluent canals. It is estimated that a total number of 30 drains carrying domestic waste water drain into the lake at various points. A territory of the Dhalada Maligawa covers the northern area of the lake. The northeast boundary of the Dhalada Maligawa is bordered by the
Udawatta Kele forest reserve, which has a hydrological value for the Kandy Lake as part of itscatchment. About 30 % of the total area of the lake catchment has a dense natural forest cover in Udawatta Kele and in the upper most section of the southern part of the watershed.
36 Figure 2.2: Landuse map of the watershed area of Kandy Lake.
37 ; 'V, 2.2 Methods
Four sampling stations were fixed at similar distance intervals along the mid line of the water body representing the entire water mass of the lake (Figure 2.1).
The locations of the sampling stations are also showed in Plate 2. Sampling was carried out monthly, for a period of two consecutive years from September 1996 to
August 1998. Water samples were collected using a 1L Rutner Sampler from the surface and just above the bottom at each sampling station. Water temperature was measured in situ using a Barnant-100 thermocouple thermometer. Transparency at each station was determined by means of a weighted white Secchi disk of 20 cm diameter. Lowering the Secchi disk into the water from the side of the boat, measurements were taken at the depth at which it disappeared and reappeared. Net samples of phytoplankton for identification of taxonomic composition were collected from each station by towing a 55 um mesh size Wisconsin plankton net for 15 seconds while the boat was running at low speed. The samples were immediately fixed with Lugol's solution.
In addition, stratified water samples were collected from the surface, 0.5 m, 1
m, 2 m, 3 m and 4 m, at the deepest sampling station (A), for incubation in bottles to
determine photosynthetic and respiration rates using the light and dark bottle
technique. Four 125 ml borosilicate bottles were carefully filled with lake water from
each depth. One bottle was immediately fixed with Winkler reagents A and B for
determination of initial oxygen. Another one was wrapped with aluminum foil and
was inserted into a black light proof bag. The other two were attached
38 perpendicularly to a vertical rope and suspended at respective depths with a black bag using horizontal floats anchored in the lake bottom. Incubation lasted for three hours during peak solar radiation in many instances from 11.00 to 14.00 hours. Total incoming radiation during the exposure was recorded using a light meter (Dickson
Inc. Ltd, USA) and the readings (W m"2) were converted into phosynthetically available radiation (PhAR) expressed in uEm"2 s"1 calibrated with a quantum sensor
(LI-COR, model LI-192S). Rainfall data relevant to the Kandy Lake were obtained from the Department of Meteorology. Water samples from each depth were also collected in clean Nalgene bottles for laboratory analysis of chemical parameters, micronutrients, chlorophyll-a and enumeration of phytoplankton counts. Samples were transported to the laboratory immediately, within 5 minutes after incubation.
Following 3 hr incubation period, all bottles were fixed with Winkler reagents A and
B for the determination of dissolved oxygen in the laboratory. Sampling for water quality at stations B, C, and D were conducted only for one year and sampling at station A was continued for another year along with in situ incubations for determination of photosynthetic and respiration rates.
39
2.2.1 Laboratory analysis
A known volume of each water sample was filtered through Whatman GF/C circles (0.45um pore size and 47mm in diameter) using a Millipore filtering manifold. The filtrate was used to determine micro nutrients (NO2 - N, NO3 - N, NH4
- N and dissolved phosphorous), major cations (Ca2+, Mg2+, Na+, K*) and anions (CI' and S042"). GF/C circles were wrapped in aluminum foil and kept at 4°C in a deep freezer overnight for the extraction of chlorophyll. Fifty milliliters of sub samples of unfiltered water were fixed with Lugol's solution for quantitative assessment of phytoplankton species composition.
Unfiltered samples were also used to determine pH, alkalinity, conductivity, total dissolved solid (TDS), turbidity, suspended solids and total phosphorus. pH measurements were taken using a Genway 3030 pH meter calibrated with two buffer solutions (pH = 7 and pH = 9.2). Total alkalinity, which is the acid neutralizing capacity of lake water was determined by titration with 0.1N HC1 using
Methyl Orange as the indicator. The end point of the titration was also detected at pH 4.5. Specific conductance and TDS were measured using portable ATI Orion conductivity-1 tester (Model 116) and portable TDS-1 tester (Model 112) respectively.
Biotic and abiotic turbidity was measured by using a Nephelometer (Model-
800). A known volume of raw water was filtered through pre-dried and pre-weighed
Satorious membrane filter papers (pore size 0.45 um). The filter papers were then
41 oven dried at 105°C for gravimetric quantification of suspended solids. Dissolved
oxygen fixed in the field and processed for BOD5 as Mn(OH)x, was dissolved in 1:1
H2SO4 and the equivalent amount of iodine liberated was determined by iodometric titration with 0.025 N Na2S2C>3 solution using starch as an indicator (APHA, 1989).
Nutrients: For determination of total phosphorous (dissolved and suspended forms) in lake water, all forms of phosphorous was converted into orthophosphate by persulfate digestion. Then the orthophosphate was determined spectrophotometrically by ascorbic acid method after formation of a phosphomolibdate blue coloured complex by the reacting ammonium molybdate, potassium antimonyl tartrate and orthophosphate in an acid medium (APHA, 1989).
The same method was used to determine dissolved orthophosphate in unfiltered samples.
Nitrite concentration in lake water was determined by a colorimetric method called diozotization. Sulfanilamide and N- (l-naphthyl)-ethylenediamine react with nitrite in low pH (pH 2.0-2.5) and formed a reddish purple azo dye depending on the nitrite concentration in the water sample. The colour intensity was measured using a
6100 visible range spectrophotometer at 543 nm. Concentrations of nitrate were determined by reducing nitrate into nitrite using Cu/Cd reduction column (APHA,
1989). Subsequently, total concentrations of nitrite were determined following the diozotization method described below. Concentration of NH/ was measured calorimetrically by the Indo-phenol Blue method. In this method, indophenol blue
42 colour forms resulting from a reaction between ammonia, hypochlorite and phenol catalyzed by sodium nitroprusside.
Cations and anions: The concentrations of major cations (Na+, K+, Ca2+ and Mg2+) in the lake water was analyzed by means of an Atomic Absorption Spectrometer in the flame emission mode. SO42" was measured by turbidimetric method. Turbid condition was formed by adding a gelatin-barium chloride solution in an acid medium which was and measured spectrophotometrically (APHA, 1989).
Comparing with the absorbance of known concentrations of SO42' solutions at the wavelength of 420 nm, SO42" concentrations in lake water samples were quantitatively determined. Chloride ion concentration was measured using a CI' 1 sensitive electrode (Model 96-17B) coupled with Double Junction Reference
Electrode (Model 90-02) and Expandable ion analizer (EA 920). All the chemical f analyses were completed immediately, as the samples were transported to the laboratory. If this was not the samples were stored in the refrigerator at 20°C.
Biological Oxygen Demand (BODs): BOD5 was estimated by determination of the dissolved oxygen concentrations in air saturated lake water and aerated water samples incubated at 25°C for five days (APHA, 1989).
2.2.2 Phytoplankton primary productivity
The dissolved oxygen concentrations of initial (I), light (L) and dark (D) bottles were determined immediately after incubation in the laboratory using a C
43 modified Winkler titration to estimate respiration, net and gross production using the conventional relationship (Respiration = I -D; Net Primary Production = L-D and
Gross Primary Productivity = Respiration + Net Primary Productivity). The rate of photosynthesis and community respiration per unit volume (O2 mg l"1) per hour were integrated over depth (Excel spreadsheet) to obtain corresponding estimates of area- based gross (EA) and net (ENP) rates and community respiration (ER) per unit time
(O2 mg l'1 h"1). Subsequently, total area-based gross and net primary productivity per day (EEA and EENP) were calculated by multiplying by a factor 10 (Tailing &
Lemoalle, 1998) and also by assuming that community respiration is constant throughout the day. Vertical profile was also used to estimate surface photo- inhibition (Zj„) as a percentage.
2.23 Chlorophyll-a analysis
The amount of chlorophyll-a was used as a biological indicator;to evaluate the total algal biomass in water samples. The day after sampling, the concentrated chlorophyll samples on filter papers were taken out of the freezer for chlorophyll-a extraction. For the extraction, 99.9+% Methyl alcohol (5 ml) was used and samples were kept in dark conditions for one hour. First the absorbance was recorded spctrophotometrically at 665nm and 750nm. Then chlorophyll-a converted into phaeophytin-a, the degradation product of chlorophyll-a by adding one drop of 0.6N
HC1. Subsequently, the absorbance at the above wavelengths were again recorded.
Eventually, the concentrations of chlorophyll-a of each water sample were computed according to the following equation.
44 pc = (A-Ao) *28.9*Ve/ (Vs*d) where
pc = chlorophyll-a concentration (ug/1)
Ve = total volume of extract (ml)
Vs = volume of filtered sample (1)
A = absorbance before acidification (665-750nm)
A« = absorbance after acidification (665-750nm)
d = path length (cm)
2.2.4 Phytoplankton composition ;
i
Lugol-added net samples were used for identification of phytoplankton species, by observing through a Olympus System Microscope Model BHS and through an inverted microscope (Olympus CK 40). Algal identifications were mainly based on the following publications: Abeywickrama (1979), Rott (1983)1 and Rott &
i i Lenzenwerger (1994).
Species composition of the phytoplankton community was determined by counting of algal species contained in 100ml of lake water samples, which were preserved and sedimented with Luglol's iodine. Counts were made using a binocular i microscope and a Sedgwick-Rafter cell. SPECDIV (Version 1.3, Michael Srawe) I computer program was used to calculate Shannon diversity index in phytoplankton i community.
45 2.3 Data analysis
Temporal and spatial (horizontal and vertical) distribution patterns of some variables were presented graphically. Tables were used to show details of quantitative information. Significance of response and predictor variables was determined employing the least square regression model. Significance of intra-site and inter-site variability were examined using two tail Student's West and one-way
Analysis of Variance (ANOVA) respectively with significance analysis at 5% probability level. In the case of one-way ANOVA, Tukey's pair wise comparisons were used to identify similar and significantly different pairs. MINTAB computer software package was used in all statistical analyses.
46 CHAPTER 3
RESULTS
3.1 Physico-chemical Characteristics
Kandy Lake has no substantial regular surface inflows and outflows
throughout the year. Therefore, the water level remains relatively constant with a
long residential time. Water level was dropped by about one metre from September
to November 1997 by the irrigation department for the construction of a lower level
sluice gate. Since the data on inflows, outflows and evaporation loss were not
available, water budget for the lake or the residence time or it inverse flushing rate could not be quantified. The ranges and mean values (with standards deviation) of the physico-chemical parameters of surface and bottom waters of four sites of Kandy
Lake for the first twelve month study period are given in Table 3.1.
Table 3.2 shows the results of one-way ANOVA employed to determine the inter-site variation of physico-chemical characteristics of surface water during the
study period. Similar results for bottom waters are shown in Table 3.3. Further, results of Student's t-test employed to determine the variability of physico-chemical characteristics of surface and bottom waters of site A are shown in Table 3.4.
47 Table 3.1: Ranges and mean values (± SD) of physicochemical parameters of surface and bottom waters of four sites of Kandy Lake for first twelve month study period.
Site A SiteB Site C SiteD Parameter Surface Bottom Surface Bottom Surface Bottom Surface Bottom Water temperature 27.50 ±1.45 25.82 ±0.69 27.46 ±1.43 27.71 ±1.23 27.63 ±1.33 CO (24.50-32.00) (25.00 - 27.00) (25.10-29.30) (25.90-29.30) (26.00 - 29.70) 1.13 ±0.27 1.06 ±0.26 1.03 ±0.31 0.90 ±0.23 Secchi depth (m) (0.55-1.60) (0.37-1.30) (0.40-1.50) (0.29-1.16) 225 ± 35 252 ±43 209 ±24 217 ±23 207 ±23 202 ±40 208 ±25 209 ±25 EC(uScm') (140 - 290) (190 -385) (180-250) (190 -260) (180-250) (100 - 260) (180-260) (180-260) 122 ±24 142 ±47 122 ±30 127 ±32 124 ±29 124 ±32 122 ±34 124 ±32 TDSOngl1) (80-150) (60 - 220) (50-150) (50-160) (50-150) (50-160) (50-160) (50-160) 8.18 ±0.40 7.04 ±0.36 8.08 ±0.42 7.38 ±0.23 8.14 ±0.39 7.63 ±0.28 8.13 ±0.39 7.88 ±0.44 PH (7.4-8.76) (6.18 - 7.40) (7.54 - 8.64) (6.88 - 7.62) (7.62 - 8.63) (6.94 - 7.98) (7.37 - 8.65) (7.08 - 8.64) Total Alkalinity 1.88 ±0.19 2.19 ±0.44 1.85 ±0.16 1.98 ±0.24 1.86 ±0.15 1.88 ±0.13 1.88 ±0.10 1.80 ±0.24 (meql1) (1.66-2.31) (1.75-3.26) (1.61-2.18) (1.66-2.58) 1.55-2.09) (1.65 - 2.09) (1.75 - 2.09) (1.15-2.14) 9.76 ±4.63 19.01 ± 10.85 12.80 ±5.88 33.29 ±23.29 12.54 ±5.74 26.10 ±12.36 13.53 ±5.16 31.38 ±18.07 TSSCmgl1) (3.00-19.00) (7.00 -51.00) (5.00 - 22.00) (4.00 - 90.00) (3.00 - 20.00) (10.00-46.00) (4.00-22.00) (7.00 - 62.00) 8.26 ±2.16 23.48 ±18.91 10.33 ±4.42 26.17 ±19.04 13.28 ±7.17 20.70 ±16.37 11.54 ±4.74 23.60 ±12.21 Turbidity (NTU) (5.10-11.40) (6.40 - 77.10) (4.70-21.00) (3.40-73.20) (5.20-28.60) (6.60 - 69.36) (6.50-21.80) (8.50-51.50) Dissolved Oxygen 7.59 ±1.78 1.33 ±1.53 7.51 ± 1.60 2.44 ±2.25 7.55 ±1.71 5.52 ±1.47 7.14 ±1.68 6.47 ±1.89 (mgl1) (4.30 - 9.70) (0.00-5.80) (4.50-9.60) (0.00-6.15) (4.50-9.55) (3.30-7.57) (4.10-9.30) (2.60 - 9.00) 0.54 ±0.07 0.52 ±0.07 0.53 ±0.08 0.51 ±0.06 0.52 ±0.06 0.54 ±0.07 0.56 ±0.08 0.56 ±0.07 Na+(meqr') (0.42 - 0.69) (0.40 - 0.62) (0.40 - 0.72) (0.38-0.60) (0.40-0.61) (0.44 - 0.67) (0.44 - 0.75) (0.47 - 0.73) 0.14 ±0.02 0.14 ±0.02 0.14 ±0.02 0.14 ±0.02 0.13 ±0.03 0.14 ±0.03 0.13 ±0.02 0.14 ±0.03 K+ (meql1) (0.11-0.18) (0.12-0.19) (0.11-0.18) (0.11-0.17) (0.07-0.17) (0.11-0.17) (0.11-0.17) (0.10-0.19) 0.73 ±0.19 0.82 ±0.17 0.77 ±0.16 0.79 ±0.17 0.71 ±0.18 0.72 ±0.23 0.71 ±0.20 0.73 ±0.17 Car* (meql"1) (0.36 - 0.96) (0.47-1.03) (0.43-1.03) (0.40-1.07) (0.30 - 0.97) (0.10-0.99) (0.15-0.94) (0.31-1.04) 0.57 ±0.05 0.59 ±0.07 0.58 ±0.07 0.59 ±0.06 0.60 ±0.06 0.59 ±0.05 0.59 ±0.05 0.59 ±0.05 Mgf^meql1) (0.45 - 0.67) (0.41-0.66) (0.48 - 0.74) (0.50-0.69) (0.54-0.73) (0.51-0.71) (0.51-0.72) (0.49-0.71) 0.068 ±0.043 0.078 ± 0.057 0.068 ±0.047 0.068 ±0.058 0.069 ±0.050 0.079 ±0.049 0.078 ±0.056 0.078 ±0.063 2 S04 "(meqr') (0.003-0.126) (0.002 - 0.161) (0.002 - 0.124) (0.001-0.166) (0.001 -0.149) (0.001-0.158) (0.001-0.174) (0.001-0.229) 0.83 ±0.15 0.95 ±0.20 0.83 ±0.13 0.83 ±0.15 0.85 ±0.13 0.86 ±0.14 0.89 ±0.19 0.95 ±0.34 CI-1 (meql1) (0.67-1.18) (0.67-1.36) (0.61-1.06) (0.67-1.12) (0.69-1.15) (0.65-1.09) (0.65-1.21) (0.64-1.93) 2.54 ±1.12 2.51 ±1.02 2.33 ±1.28 2.66 ±1.06 2.32 ±1.22 3.17 ±1.43 2.59 ±1.23 3.18 ±0.78 BOPs(mgr') (0.60-3.94) (1.70-5.10) (0.35-4.21) (1.50-5.30) (0.85 - 4.96) (0.10-5.78) (0.00-4.75) (1.95 - 4.29) There were no significant spatial variations of physico-chemical parameters
of surface water among the four sampling stations. In contrast, bottom water showed
significant inter-site variations for some physico-chemical parameters such as pH,
dissolved oxygen (DO), and alkalinity (Table 3.1 and 3.3). However, such
significant inter-site differences in bottom water was not found for parameters such
as electrical conductivity (EC), total dissolved solids (TDS), turbidity, major cations and anions (except HCO3"). With respect to site A, significant intra-site variations were found for parameters such as pH, alkalinity, EC, DO, total suspended solids
(TSS), turbidity and temperature (Table 3.4).
50 Table 3.2: Results of one-way ANOVA for inter-site variation of physicochemical characteristics of surface water in Kandy Lake during the first twelve month study period. None of the parameter is significantly different at 5% probability level.
Parameter F ratio p - value
Water Temperature 0.11 0.952
Secchi depth 1.57 0.209 pH 0.14 0.933
Alkalinity 0.06 0.993
Electrical Conductivity 0.04 0.990
Dissolved Oxygen 0.70 0.971
Total Dissolved Solids 0.01 0.999
Suspended Solids 0.47 0.703
Turbidity 2.69 0.057
Na+ 0.67 0.578
K+ 0.31 0.820
Ca2+ 0.37 0.774
Mg2+ 0.64 0.591 cr1 0.35 0.786
2 so4 " 0.20 0.897
BOD5 0.16 0.923
51 Table 33: Results of one-way ANOVA for inter-site variations of physicochemical characteristics of bottom water in Kandy Lake during the first twelve month study period.
Parameter F ratio p-value Tukey's Pair-wee Comparison pH 14.33 0.000 ** A°& CD, A°& DD, BD& D 0
Alkalinity 1025 0.000 ** Ab&Cb,AB&DB,Bb&Cb
Electrical Conductivity 3.18 0.330 ns
b Dissolved Oxygen 21.85 0.000 *• Ab& C*jf& Db3b& CVB& DB
Total Dissolved Solids 0.10 0.958 ns
Suspended Solids 1.72 0.156 ns
Turbidity 2.71 0.560 ns
Na+ 1.19 0326 ns
0.07 0.973 ns
Ca2* 0.34 0.796 ns
Mg2* 0.00 0001 ns cr1 0.85 0.475 ns
2 so4 " 1.17 0331 ns
BODS 1.15 0.340 ns
AB - bottom point at site A, BB - bottom point at site B, Cb- bottom point at site C,
Db - bottom point at site D ns - not significant at 5% level
** - significant at 5% level
52 Table 3.4: Results of Student's t-test for intra-site variation of physicochemical parameters between surface and bottom waters at station A.
Parameter t-value p—value pH 4.88 0.000 **
Alkalinity 3.44 0.001 **
Electrical Conductivity 2.13 0.039 **
Dissolved Oxygen 12.73 0.000 **
Total Dissolved Solids 1.41 0.170 ns
Suspended Solids 2.70 0.009 **
Turbidity 8.66 0.000 **
Na+ 0.19 0.850 ns
K+ 0.02 0.980 ns
Ca2* 0.73 0.470 ns
Mg* 0.89 0380 ns cr1 1.72 0.092 ns
SO42" 0.14 0.890 ns
BOD5 2.07 0.046 **
** - significant at 5% level ns - not significant at 5% level
53 3.1.1 Temperature
Air temperature ranged between 23.80°C and 31.50°C and the mean air temperature was 28.16°C (± 2.49) during the study period. The surface water temperature was lowest in December 1996 (24.50°C) and the highest value of
32.00°C was recorded in April 1998 (Figure 3.1). Mean temperature of surface water in Kandy Lake was 27.50°C (± 1.45). Monthly changes in surface temperature did not show a marked seasonal pattern during the study period. Although the gradient was not significant, a gradual decline in temperature was found from surface to the bottom at station A (Figure 3.2).
o e
S-ONDJ-FMAMJ JASONDJ-FMAMJ JA 96 97 98
Figure 3.1: Monthly variation of water temperature (WT) during the study period.
54 Temperature ( C)
Figure 3.2: Temperature profile at station A of the Kandy Lake (horizontal bars
denote standard errors).
Figure 3.3 depicts monthly variations of pH, electrical conductivity (EC), and total
alkalinity in Kandy Lake during the study period, from September 1996 to August
1998.
3.1.2 pH
pH was uniform in the surface water but it was significantly different in the bottom water among the four sampling sites (Table 3.2). pH of surface water ranged from 7.37 to 8.76 and the mean value was 8.10 (± 0.36) for the study period. pH in surface water was significantly higher than the bottom water at site A (t = 4.88, p<0.000, N = 24). Seasonal fluctuation of pH in the surface and bottom water is depicted in Figure 3.3. Surface pH was always higher than that of bottom water, and
55 it was above 7.50 during the study period. Bottom pH varied below 7.5 and it was '..V * _ less than 6.5 in July 1997. pH in the surface and bottom waters of Kandy Lake showed no marked seasonal pattern with rainfall during the study period. However, pH in surface water had a statistically significant relationship with the rainfall (Table
3.5).
3.13. Total alkalinity
Irrespective of the sites, total alkalinity ranged from 1.50 meq l'1 to 2.32 meq l'1 with a mean value of 1.82 meq l"1 (± 0.02) during the study period. Analysis of variance showed no significant difference in total alkalinity of surface water among the four sampling sites (Table 3.2). In contrast, total alkalinity of bottom water was significantly different according to the analysis of variance (Table 3.3). Turkey's pair-wise comparison showed that total alkalinity was significantly different between site A and C, site A and D, and site B and C, but it was not significantly different between site A and B, and site C and D. Mean values of total alkalinity between surface and bottom waters at site A were significantly different according to
Student's t-test (Table 3.3). Seasonal variation of total alkalinity of surface and bottom water at site A is shown in Figure 3.3. Except in March 1998 the total alkalinity of surface waters was above 1.50 meql"1 but exceeded 2.00 meq l"1 only in
September 1996 (Figure 3.3). Total alkalinity of bottom water at site A was more than 1.70 meq l'1 throughout the study period, and it was above 2.00 meq l"1 in several occasions with a highest value of 3.26 meq l"1 in July 1997.
56 6 I i i i I i i I i i i i i i i i i I I i i i i i
4
3 -
3 •
2 •
2 •
1 • t—i—r T—1 1
A-S A-B
Figure 3.3: Monthly variations of pH, conductivity (EC), and total alkalinity in surface and bottom water at site A.
57 3.1.4 Electrical conductivity and total dissolved solids
Electrical conductivity of surface water of Kandy Lake ranged from 140 uS cm"1 to 290 uS cm"1 with a mean value of 215 uS cm"1 (± 30) and it was between 190 uS cm"1 and 385 uS cm'1 for bottom water and the mean was 252 uS cm"1 (± 43) during the study period. EC of surface waters did not show significant inter-site variability according to one-way ANOVA (Table 3.2). A more or less similar trend was found with respect to bottom water (Table 3.3). In contrast, the mean electrical conductivity of surface water was significantly different from bottom water according to the results of the Student's t-test (Table 3.4). Surface conductivity was significantly lower than that of bottom water. Seasonal variation of surface and bottom EC at site A is shown in Figure 3.3. EC was <150 uS cm"1 only in November
1997 and was less than 200 jxS cm"1 in April, May and June in 1997. Surface electrical conductivity was >250 uS cm"1 from May to August 1997. In general, EC of both surface and bottom water of Kandy Lake showed an increasing trend towards the end of the study period (Figure 3.3).
The pattern of monthly changes in TDS was more or less similar to that of EC in
Kandy Lake. TDS ranged between 40 mg l"1 and 160 mg l"1 with a mean value of
118 mg l"1 (±28) during the study period. Spatial distribution of TDS was also more or less similar to that of electrical conductivity (Table 3.2 and Table 3.3). However, the mean value of TDS of surface and bottom waters at site A was not significantly different according to the results of the Student's t- test (Table 3.4).
58 The amount of total suspended solid (TSS) in Kandy Lake varied from 3.00 mg l"1 to 24.70 mg l"1 during the study period with a mean value of 12.10 mg l"1
(±5.73). There was no significant inter-site variation of TSS for both surface and bottom water (Table 3.2 and 3.3). However, the mean values of TSS were significantly different between surface and bottom water at site A (Table 3.4). Figure
3.4 shows the mean values (± SD) of TSS calculated for three seasons defined according to the distribution pattern of annual rainfall (e.g. wet, dry and moderate seasons). Mean value of TSS was relatively higher during the wet season compared to dry and moderately wet months. However, turbidity values did not show such a marked seasonal pattern (Figure 3.4).
Turbidity ranged from 2.60 NTU to 18.20 NTU with a mean value of 9.43
NTU (± 3.36) during the study period. Inter-site distribution of turbidity in both
surface and bottom water was more or less similar to that of TSS (Tables 3.2 and
3.3). Further more, the mean values of surface turbidity were significantly less than
the bottom values at station A, according to the results of Student's t-test (Table 3.4).
Monthly variation of mean values for surface water turbidity is shown in Figure 3.5
and turbidly was less than 10 NTU but more than 5 NTU during most months.
Distribution pattern of turbidity during dry and wet months were not similar to that
of TSS (Figure 3.4). Analysis of variance showed no significant difference in rainfall
defined seasonal pattern in both TSS and turbidity.
i
59 TSS : F-ratio = 1.03 @ p = 0.373 Turtxfity: F - raito = 0.25 @ p = 0.784
0 TSS aTurbity
Wet Moderate Dry
Figure 3.4: Mean values of total suspended solids (TSS) and turbidity at surface
water in Kandy Lake in relation to low, moderate and high rainfall seasons. Vertical
bars denote standards errors.
40-i
30 -I
£ 20 4 .Q $ 10 A
i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i S-0 NDJ-FMAMJ JASONDJ-FMAMJ JA 96 97 98
Figure 3.5: Monthly mean values of turbidity of surface water in Kandy Lake, during the study period
60 3.1.6 Dissolved oxygen
Irrespective of sampling sites, dissolved oxygen (DO) of surface water ranged from 4.02 mg l'1 to 12.72 mg l"1 with a mean value of 7.53 mg I"1 (± 1.91) during the study period and there was no inter-site difference among the four sites
(Table 3.2). In contrast, DO in bottom water was significantly different among the sites because low DO values occurred in the bottom at site A (Table 3.3). At site A, there were instances that DO was not detectable indicating anoxia. Depletion of DO from surface to bottom of the water column at site A was consistent throughout the study period as illustrated by Figure 3.6. The seasonal variation of DO in surface and bottom waters is shown in Figure 3.7. DO concentration in the surface water was more than 6.00 mg l*1 except in September and December 1996, in September 1997 and in May and June 1998 (Figure3.7). Further, DO concentration was more than
10.00 mg l*1 in April, May, June, and August 1997, in January, February, March,
July and August 1998 (Figure3.7).
61 DCMmgr1) 0 5 10
Figure 3.6: Vertical distribution of dissolved oxygen (DO) concentration in Kandy
Lake during the study period (horizontal lines denote standards errors).
A-S A-B 15 i
S-ONUJ-FMAMJ JASON DJ-FMAMJ JA 96 97 98
Figure 3.7: Monthly variation of dissolved oxygen (DO) concentration in surface and bottom water at site A.
62 3.1.7 Biological Oxygen Demand (BODs)
Monthly variation of BOD5 of surface and bottom waters of Kandy Lake is
shown in Figure 3.8. BOD5 of surface water at site A ranged from 0.60 mg l'1 to 3.94 mg l"1 with a mean value of 2.62 mg l"1 (± 0.91) and it was between 1.70 mg l"1 and
5.62 mg l'1 with an average value of 3.31 mg l"1 (± 1.33) for bottom water. The results of Student's t-test showed a significant difference in mean values of BOD5 for surface and bottom water at site A during the study period (Table 3.4). BOD5 of surface water was lowest in December 1996 and the highest value found in August
1997. The lowest bottom values were found in both April and May in 1997 and the highest value in March 1998 for bottom waters. In general, BOD5 of bottom water of the Kandy Lake showed an increasing trend towards the end of the study period, although intermittent fluctuations occurred from March to August 1998.
S-ONDJ-FMAMJ JASONDJ-FMAMJ JA 96 97 98
Figure 3.8: Monthly variation of BOD5 in surface and bottom water at site A of
Kandy Lake.
63 3.1.8. Major cations and anions
Relative proportion of major cations (i.e. Ca2+, Mg2"1', Na+, K*) and anions
2 2 1 (i.e. HC03\ C03 ', SO, ", CI ) are shown in Maucha's ionic phase diagram (Figure
3.8). Cations were dominated by Ca2+ while HCO3" was the most dominant anion in
Kandy Lake water. The overall concentration of Ca2+ in Kandy Lake ranged from
0.106 meq l"1 to 1.070 meq l"1 with a mean value of 0.755 meq l"1 (± 0.001) while
Mg2"1-, the second dominant cation ranged from 0.451 meq f1 to 0.745 meq l'1 with an average value of 0.586 meq l'1 (± 0.056) during the study period. The concentration of Na+ changed within the range of 0.402 - 1.076 meq l"1 with a mean value of 0.548 meq l"1 (± 0.103) while it ranged from 0.078 meq l"1 to 0.30 meq l"1 with an average of 0.143 meq l'1 (± 0.038) for K+ during the study period. The second dominant anion was CI'1 and its concentration ranged from 0.42 meq l'1 to 1.212 meq l"1 with a
2 mean value of 0.802 meq 1"' (± 0.173). The mean concentration of S04 " was 0.078 meq l*1 (± 0.047) and it was within the range of not detectable level to 0.200 meq l"1 during the study period.
One-way analysis of Variance showed no significant inter-site difference in major cations and anions in both surface and bottom waters of the Kandy Lake during the study period (Table 3.2 and 3.3). Further, cationic and anionic concentrations of surface and bottom water at site A were also not significantly different according to the results of Student's t-test (Table 3.4). The relative ionic ratio of major cations and anions were as follows; Ca2+: Mg2*: Na+: K+ = 37: 29:
1 2 27:7 and anions and HCO3": CI" : S04 "= 68:29:3.
64 so/ MPT
Figure 3.9: Maucha's diagram to illustrate the relative ionic proportions of major cations and anions in Kandy Lake.
3.1.9. Inter-relationships of physicochemical characteristics of the Kandy Lake
Inter-relationships of independent and dependent physico-chemical parameters examined using least square regression model are given in Figure 3.10 and Table 3.5. Electrical conductivity (EC) was significantly correlated with major cations Na+ and Mg2*, and with major anions and total alkalinity (Table 3.5).
Significant correlations were also found between TDS and total alkalinity (Table
3.5). Further more, pH in surface water was negatively correlated with rainfall and positively correlated with DO. Surface water temperature and dissolved oxygen concentration in the surface water of the Kandy Lake also showed a significant relationship during the investigation period. However, the relationship between turbidity and TSS was not statistically significant.
65 r.0.203 • 0.184 1X0.05
•T 300 1 300 i % I a zoo O 200
10 20 30 40 SO 5 10 15 20 2S n.'n"i (a)
400 r-O.200 p<0.05 p-=0 001
300 B s " 200 a mo 100 15 10 15 20 25 30 3S S 10 cr(mgr') (d) "a" (mar')
r- 0.308 (K0.001
g 300 S S a 2oo
250 "00 150 200 TottlOudnHy (mjr1) (0
400 r - 0.173 f 0.390 • p<0.001 _ 200 • ii
o 200
UJ 100 250 100 150 200 250 100 150 200 TDS(mg r) (ff) Total •IkilinHylmar') (»)
9 r-0 568 p<0.01 1.3
7.5
7 100 200 300 400 500 15 5 10 *fm (<) DOtmgr') I/)
r=" 0.475 r-0 08 13 (ia) 4> A p<0.05 £25 r 8 8 1 15 • • •
28 28 10 15 WT(*C> • r-) Figure 3.10: Scatter plots of interrelationships of physicochemical parameters (a: EC vs Na+, b: Ec vs Ca2*, c: EC vs Mg2*, rf: EC vs CI", t?: EC vs SO*2",/: EC vs Total alkalinity, g: EC
TDS, h: IDS vs Total alkalinity, /: pH vs Rainfall, j: pH vs DO, t DO vs Water temperature, /;
Turbidity vs TSS) of Kandy Lake. Regression lines are indicated only for the
statistically significant relationships at least at 5% probability level.
66 0
Table 3.5: Relationships of independent and dependent physicochemical parameters
examined using least square model. \
Parameter No. of r p-value Significance
observations
EC vsNa+ 190 0.203 <0.05 *
EC vs Ca2+ 190 0.164 >0.05 ns
ECvsMg2+ 114 0.200 <0.05 *
EC vs Cl- 84 0.605 O.001 ***
2 *** EC vs S04 ' 167 0.441 <0.001
EC vs Total Alkalinity 168 0.308 O.001 ***
TDS vs EC 146 0.173 >0.05 ns
TDS vs Total Alkalinity 146 0.390 O.001 *** pH vs Rainfall 24 0.420 O.05 * pH vs DO 24 0.568 O.01 **
DO vs Water Temperature 21 0.475 O.05 *
Turbidity vsTSS 22 0.08 >0.0.05 ns
Ns - not significant at the 5% probability level
* - Significant at the 5% probability level
** - Significant at the 1% probability level
*** - Significant at the 0.1% probability level
67 3.2 Nutrients
3.2.1 Dissolved phosphorus and total phosphorus
Dissolved orthophosphorous (dP) concentration in the surface water ranged from non detectable level to 22 fig l"1 with a mean value of 4.6 ug l"1. dP did not show a significant inter-site variability in both surface and bottom water (Tables 3.6,
3.7). Further more, the concentration of dP in surface and bottom water at site A was also not significantly different according to the results of Student's t-test (Table 3.8).
However, the concentration of dP at site A decreased significantly with increasing pH (r2 =23.2%, F=8.62; PO.01). Significantly negative correlation between dP and dissolved oxygen was also found for the bottom water at site A (r = -0.0569: p<0.05). Monthly changes of dP in surface and bottom waters in Kandy Lake during the study period are shown in Figure 3.10. dP in surface water was less than 5 ug l"1 except in November and December 1996, in May and November 1997 and in April,
May, July and August in 1998. It was more than 10 ug l'1 only in November and
December 1996. The concentration of dP in the bottom water exceeded 5 ug l"1 only during six occasions and it exceeded more than 10 ug l'1 only once during the study period (Figure 3.11).
The concentration of total phosphorous (tP) ranged from 14 ug l"1 to 216 ugl*1
with a mean value of 51 ug l"1 during the study period. Although there was no
significant inter-site variation in tP in both surface and bottom water according to
one- way ANOVA ((Tables 3.6 and 3.7), the results of Student's t-test shows a
68 significant intra-site variability of tP between surface and bottom water at site A
(Table 3.8). Figure 3.11 illustrates monthly values of tP in surface and bottom water at site A. The concentration of tP of surface water at site A was less than 50 ug l"1 except in June, July, October and December in 1997 and more than 100 ug l"1 only in
June 1997 (Figure 3.11). Fluctuation of the concentration of tP in bottom water was more prominent than that of in surface water and it was more than 200 ug l"1 in
November and December 1996, in May and December 1997 and in February 1998
i
(Figure 3.11). Further, tP was less than 100 ug l*1 during nine occasions but less than
50 ug l"1 only during four occasions (Figure 3.11). However, analysis of monthly concentrations of tP showed that neither surface nor bottom water at site A showed a marked seasonal pattern during the study period.
69 (a)
Q. •o
6OO-1
£• 400 • CD
9T 200-
0- 1—I—I—I—I—I I I I I I I I I I I I—I I I I I I I S-ONDJ-FMAMJ JASONDJ-FMAMJJA 96 97 98
-A-S A-B
Figure 3.11: (a) Monthly variation of dissolved phosphorous (dP) and (b) monthly variation of total phosphorous (tP) in surface and bottom water of Kandy Lake during the study period.
70 3.2.2 Nitrogen species (Nitrite, Nitrate and Ammonia)
The concentration of nitrite - nitrogen (NO2 - N) in surface water ranged from non-detectable level to 191.94 ug l'1 with a mean value of 42 ug l*1 (± 48) during the study period. Inter-site variability of the NO2 - N concentration in both surface and bottom waters were not significantly different according to the results of one-way
ANOVA (Table 3.6 and 3.7). Intra-site variability of mean values were also not significantly different for surface and bottom waters at site A according to the results of Student's t-test (Table 3.8). Monthly concentration of NO2 - N in surface and bottom waters are shown in Figure 3.12, and the concentration in surface water was less than 25 ug l'1 except from September to December 1996, in January and
February 1997 and in January, July and August of 1998. It was more than 50 ug l*1 only in January and February 1997 and in August 1998 (Figure 3.12). A more of less similar trend was found in case of the concentration of NO 2 - N in the bottom water, but it was less than surface water except on a few occasions (Figure 3.12).
71 Table 3.6: Results of one-way ANOVA for inter-site variation of micronutrients in surface water of Kandy Lake during the study period. None of values was significantly different at 5% probability level.
Parameter F-value p-value
dP 0.41 0.748
tP 0.51 0.678
N02"-N 0.21 0.946
NO3-N 036 0.781
NH4+-N 0.01 0.998
72 Table 3.7: Results of one-way ANOVA for inter-site variation of micronutrients in bottom water of Kandy Lake during the study period.
Tukey's pair-wise Parameter F-vame p-value Signfficanci X comparison
dP 0.56 0.647 ns
tP 1.10 0359 ns
N02"-N 0.36 0.779 ns
NO3-N 1.40 0255 ns
NH4+-N 11.22 0.000 ** Ab & Bb, Ab & Cb, Ab & Db
ns - not significant at 5% level
** - significant at 5% level
Ab - bottom point at site A, Bb - bottom point at site B, C b- bottom point at site C,
Db - bottom point at site D
1
73 Table 3.8: Results of Student's t -test for intra-site variation of micronutrients between surface and bottom water at site A of Kandy Lake during the study period.
Parameter t-value p-valoe Significance dP 0.71 0.41 ns tP 421 0.0001 **
N02'-N 0.45 0.66 ns
NO3-N 2.01 0.051 **
NH4+-N 5.91 0.000 **
ns - not significant at 5% level
** - significant at 5% level
The concentration of nitrate-nitrogen (NO3 - N) in surface water of Kandy
Lake raged from 3 ug l"1 to 1277 ug l"1 with a mean value of 337 *ig l"1 (± 319) during the study period. The results of one-way ANOVA showed no significant inter-site variability in NO3 - N for surface water (Table 3.6), but it was significantly different for the bottom water (Table 3.7). Intra-site variability of mean monthly concentrations between surface and bottom waters at site A was also significantly different according to the results of Student's t-test (Table 3.8). Figure 3.12 illustrate monthly concentration of NO3 - N in surface and bottom waters at site A during the study period, The NO3 - N concentration in surface water was more than 100 ug l"1 except in September 1996, from May to July in 1997 and from April to July in 1998, while it was more than 500 ug l"1 in November and December 1996, and in February,
74 August and November 1997 (Figure 3.12). An extremely high concentration of 1277
ug l"1 was recorded in November 1997. NO3 - N concentration in bottom water was less than surface values except in December 1996, in January, May and September of 1997 and in June and July of 1998 (Figure 3.12). The highest bottom value of 968 ug l"1 was recorded in December 1996 and the bottom values were extremely low in
June, July and November 1997 and from March to May 1998. Both surface and bottom concentrations of NO3 - N did not show a marked seasonal pattern during the study period.
75
Extremely wide ranges were recorded for both surface and bottom concentrations of ammonia (NH4 - N) in Kandy Lake during the study period.
Surface concentration ranged from not detectable level to 774 ug l"1 with a mean value of 177 ug l"1 (± 150). Although one-way ANOVA showed no significant inter- site variability in NH4 - N concentration of surface water (Table 3.6), it was significantly different in the case of bottom water (Table 3.7). Further, the concentration of NH4 - N in the bottom water at site A was significantly higher than that of surface water (Table 3.8). Figure 3.13 shows a significant negative correlation of NH4 - N with dissolved oxygen concentration (r = -0.552, p<0.006); The changes of monthly mean concentration of NH4 - N in both surface and bottom waters are shown in Figure 3.12 and the surface concentration were always less than the bottom concentrations, throughout the study period. The concentration of NH4 - N in surface water did not exceed 250 jig l"1 except in December 1996, in July, September and
December of 1997 and from May to July in 1998 (Figure 3.12). In contrast, bottom concentration was more than lmgl'1 during twelve occasions during the study period.
Extremely high concentrations of NH4 - N more than 1.5mg T1 was recorded from
February to August and October and November in 1997 and in April and May in
1998. Monthly concentration of NH4 - N in both surface and bottom waters did not show any marked seasonal pattern during the study period.
77 2, • R2 = 0.3051 • • 1.5- • 2. 3-
0.5-
500 1000 1500 2000 2500 3000 DO(mgH)
Figure 3.13: Scatter diagram of NKU - N concentration vs dissolved oxygen (DO) the bottom water of site A of Kandy Lake.
78 33 Photosynthesis and Primary Production
33.1 Light Climate
Abbreviations and their corresponding units, used in the text related to photosynthesis characteristics are listed in the Table 3.9. Table 3.10 shows the mean total incoming radiation (TIR) and corresponding photosynthetically available radiation (PhAR) during each exposure and Figure 3.14 illustrates the mean photosynthetically available radiation (PhAR) during the incubation for photosynthetic experiment in Kandy Lake. TIR ranged from 243 Wm*2 s"1 (in
October 1996 and September 1997) to 1400 Wm"2 s1 (in March 1998). In general
TIR was high during January, February and March and the values were relatively low during the rainy months. The mean PhAR values during the exposure time (ll00
-14°°hrs) was more than 2000 uE m"2 s'1 in February, March and November 1997 and from February to April in 1998. In contrast, PhAR was <500 uE m"2 s'1 in
September and October 1996, August, September and December 1997, and in
January, June, and July 1998. TIR was more than 1000 at 62% occasions of total exposures and it coincided with the dry whether (February to April). The irradiance of TIR was highest around noon and it decreases gradually towards afternoon under clear sky (Figure 3.15). A similar pattern was observed in the case of PhAR which was computed as 45% of the TIR.
79
2 Table3.9: Abbreviations used in the text
Abbreviation Unit Explanation TIR WmV Total Incoming Radiation
PhAR uEm'V1 Photosynthetically Available Radiation
ZSD m Secchi depth
Zeu m Estimated euphotic depth (ZeU = ZSD*2.75)
Compensation depth (where the depth A=R) Zc m
Za m Aphotic depth (non photosynthetic depth)
A mg02l"V photosynthetic rate per unit water volume
1 1 Amax mg02l' h Maximum photosynthetic rate per unit water volume
2 1 ZA g02m" h" Area based gross primary productivity per hour
1 IEA gC^m^d" Area based gross primary productivity per day
ZNP g02m¥ Area based net primary productivity per hour
2 1 IZNP g02m" d" Area based net primary productivity per day
2 1 IR g02m" h" Area based respiration per hour
2 1 IZR g02m" d" Area based respiration per day
1 1 R mg02r h- Respiration per unit water volume
Chl-a I-1 Chlorophyll-a concentration
B mg Chl-a m'2 Mean chlorophyll-a biomass of the water column
2
ZeUB mg Chl-a m" Mean euphotic chlorophyll-a biomass
0max mgmg h Amax/Chl-a
Spi % (Pmax - Psur / Pmax )* 100
, 1 Asur mg 02l' h' Photosynthetic rate at surface
V:0 m ZA/ Amax
80 Table 3.10: The mean total incoming radiation (TIR) and photosynthesis available radiation (PhAR) vrithin the period of bottle suspension for photosynthetic experiments in Kandy Lake.
Date TIR (WmV) PhAR(uE mV1)
12 Sep 96 493 895 14 Oct 96 243 441 13 Nov 96 1013 1840 17 Dec 96 856 1554 15 Jan 97 1092 1984 13 Feb 97 1215 2207 12 Mar 97 1261 2290 10 Apr 97 870 1581 28 May 97 945 1517 12Jun97 950 1726 15 Jul 97 845 1535 28 Aug 97 378 687 25 Sep 97 243 441 17 Oct 97 1069 1942 10 Nov 97 1103 2004 16 Dec 97 328 596 15 Jan 98 316 574 18 Feb 98 1186 2154 HMar98 1400 2544 20Apr98 1281 2326 14May 98 570 1035 15 Jun98 246 447 24 Jul 98 473 858 18 Aug 98 537 975
81 200 4—i—i i i i i i i i i i i i i i i i i i—i—i—i—i—i S-0 NDJ-FMAMJ J ASONDJ-FMAM J J A 96 97 98
Figure 3.14: The mean photosynthetically available radiation (PhAR) during incubation for photosynthetic experiment in Kandy Lake.
i 82 Figure 3.16 illustrates the monthly variation of Secchi depth which was used as a measure of under water light attenuation. The lowest Secchi depth value of 0.55 m at the deepest area was recorded in October 1996, and it was more than 1.00 m except in April, May and October 1997 and in January, July and August 1998. A maximum Secchi depth of 1.60 m was observed in February 1997 and it was above
1.25 m in December 1996, July and December 1997 (Figure 3.16). Secchi depth showed a decreasing trend in Kandy Lake towards the end of the study period.
Euphoric depth (1% of PhAR) which was approximately estimated by multiplying
Secchi depth from factor of 2.75 (Dokulil et a/., 1983), ranged from 1.51 m to 4.40 m with a mean value around 3.00 m during the study period. Figure 3.17 illustrates the monthly fluctuation of estimated euphotic depth during the photosynthetic experiment. Euphotic depth of the Kandy Lake also showed a decreasing trend towards the end of the study
The compensation depth (Zc) where the community respiration was equal to photosynthetic oxygen production ranged from 1.70 m to 5.50 m with an average value of 3.30 m during the study period.
83 CD 00 0> CD CD j j WOZQ )il5 E 08 o 1.2 1-6 A 2J Figure 3.16: Monthly mean values of Secchi depth (Zsd) of Kandy Lake during the study period. CD CD WOZO-!iU.2<2t) 2- rl 3- 4- 5- Figure 3.17: Monthly fluctuation of estimated euphotic depth (Zeu) at site A of Kandy Lake during the photosynthetic experiments. 84 0 3.3.2 Chlorophyll-a The content of chlorophyll-a (chl-a), an index of phytoplankton biomass in the surfaces water of Kandy lake ranged from 14 ugl"1 to 34 ugl"1 with a mean value of 23 ugl"1 (± 4.66) during the study period. During dry months (i.e. February - March), chl-a was highest whilst the lowest concentration recorded was during wet months (in December - January). A significant difference in chl-a concentration was found between dry months and wet months (t = 3.01, p<0.01, df = 21), although chl- a concentration showed a negative relationship with the monthly rainfall. Pearson correlation was statistically non significant (r = 0.316; p > 0.05). Monthly concentrations of chl-a in surface and bottom water at site A are shown in Figure 3.18. In surface water, the lowest value of 14 ug l'1 was found in December 1996 and it was below 20 ug l'1 only in December 1996, July 1997 and April and May 1998. The highest value of 34 ug l"1 was recorded in January 1998 and values above 25 ug l"1 concentrations were found in November 1996, in June 1997 and from January to March and August 1998. In the case of the bottom water at site A, the lowest value of 5 ug l"1 was recorded in October 1996, in May 1997 and in August 1998, and it was less than 10 p.g l*1 only on seven occasions. The highest value of 45 Ug l"1 was found in February 1998 and it was also the highest value determined for Kandy Lake during the study period. In general, chl-a concentration in bottom water i was less than the values for surface water. However, in February and June 1998 the concentrations in bottom water exceeded the values of surface water. A significant negative relationship was found between secchi depth and chl-a for Kandy Lake only during dry and lower rainy months. Figure 3.19 is a scatter diagram of secchi 85 depth and chl-a of surface water of site A. Figure 3.20 shows the relationship between the integrated depth (surface to four meter depth) mean concentration of total phosphorous and log chl-a values. Figure 3.18: Monthly values of Chlorophyll-a (Chl-a) concentration in surface and bottom water at site A of Kandy Lake for a period of two consecutive years. 40-i y =-0.1938x +47.282 R2 = 0.2797 O) 30-1 • • • 1 20 A 10 75 85 95 105 115 125 135 Secchi depth (m) Figure 3.19: Scatter diagram of Secchi depth vs chlorophyll-a (Chl-a) concentration of surface water at site A of Kandy Lake during the dry and lower rainy months. 86 100 • g • (0 10 CO o 20 30 40 50 60 70 80 tP (ug I"1) Figure 3.20: Relationship between total phosphorous (tP) and log chlorophyll-a (Chl-a) values of Kandy Lake during the study. Figure 3.21 depicts the vertical distribution pattern of mean chl-a values at site A. The maximum chl-a concentration of 27 ugl"1 (± 9) occurs at two metre depth while the minimum was 14 ug l"1 (± 9) at the bottom layer. Although the chl-a concentration increased from surface to 2 m depth, it decreased at 3 m and then increased slightly again at a mean depth of 4 m demonstrating a bimodal vertical distribution pattern. 87 Chi-a (jigl"1) 10 20 30 °1 "te— 2- »5-H _ 4- E I 6- ° 8- 10- 12- t 14- Figure 3.21: Depth profile of overall mean Chlorophyll-a (Chl-a) concentrations of Kandy Lake at site A. Vertical line denotes the standard errors of the mean. 333 Photosynthesis Twenty four depth profiles (depth vs oxygen production per hour) are illustrated in Figure 3.22. A majority of the depth profile exhibits surface inhibition and in most cases light saturated maximum photosynthetic rate occurred (Amax) at 0.5 m below the surface. Surface photoinhibition was more than 30% under dry weather conditions but lesser values were computed under overcast conditions. The highest value of Amax of 1.705 mg O21"1 h"1 was determined for the surface layer in 1 1 July 1998 and the lowest Amax was 0.13 mg 021* h" in August 1997 at 0.5m depth. 1 1 Mean value of Amax was 0.54 mg O21* h" (±0.40) during the study period. Specific photosynthetic rate or the rate per unit chl-a content (0) was lowest (4 mg mg'1 chl-na h"1) in August 1997. This was observed under heavy rainy condition. However, the highest 0 value of 83 mg^chl - a h"1 was observed in July 1998 when Amax was also 88 highest. Figure 3.23 and Table 3.11 shows monthly values of Amax and 0 for two consecutive years. These values do not show a particular seasonal pattern but there is a general increasing trend in Amax towards the end of the study period. The Amax 1 1 value was more than 0.50 mg 02 l" h' during eight occasions of which five occasions occurred during the second year of study. Further, Amax was more than 1.00 mg O21"1 h"1 during the last two occasions of the study. Similarly high 0 values were also more prominent during the second year compared to the first year of the study. 0 value was more than 25 mg"1 chl-a h"1 only on two occasions during the first year while there were five occasions of 0 more than 25 mg mg"1 chl-a h'1 including an unusually high value of 83 mg mg'1 chl-a h"1 during the second year of the study. Figure 3.24 (a) and (b) scatter plots between Amax vs chl-a and log 0 vs chl-a. A positive relationship between Amax and chl-a (Figure 3.23 a) was found only for the first twelve months of the study (r=0.78, p<0.05). However, the overall relationship between Amax and chl-a was not statistically significant The relationship of log 0 vs chl-a was negative but not significant at at 5% probability level. (Figure 3.24b). Monthly values of 0 and chl-a at the depth of the Amax for two consecutive years are shown in figure 3.25. 89 mg02l"V 0 0.5 1 1.5. 0 0.5 1 1.5 0 0.5 1 1.5 0 0.5 1 1.5 _j 1— 7 —I L_ _J —I l_ —I I..,. I 12/9/96 14/10/96 13/11/96 17/12/96 0 0.5 1 1.5 0 0.5 1 1.5 0 0.5 1 1.5 0 1 J 1 1 -fl 1—' ' 1 I 1 1S/1/97 J 13/2/97 J 12/3/97 I 10/4/97 0 0.5 1 1.5 0 0.5 1 1.5 0 0.5 1 1.5 0 0.5 1 1.5 -~i 1 1 12/6797 12/6/97 15/7/97 21/8/97 0 0.5 1 1.5 0 0.5 1 1.5 0 0.5 1 1.5 0 0.5 1 1.5 1l- —1 1 _l 1 23/9/97 17/10/97 10/11/97 16/12/97 Continued. 90 Figure.3.22: Depth profiles of photosynthetic (outer line) and community respiration (inner line) during the study period. 91 Figure 3.23: Monthly variation of Amax and 0 of Kandy Lake during the study period. Figure 3.24: Scatter diagrams of (a) vs chl-a and (b) log 0 vs chl-a 92 S-ONDJ-FMAMJ J A SON DJ-FMAM J J A 96 97 98 Figure 3.25: Monthly variation of Chlorophyll-a at the depth of the A max and 0 max of Kandy Lake during the study period. 33.4 Community respiration Respiration per unit water volume in Kandy Lake during the study period 1 1 1 1 ranged from 0.020 mg 021" h" to 0.298 mg 021" h" with a mean value of 0.070 mg O21"1 h"1. Community respiration was also generally higher during dry month while relatively low values were observed during the rainy seasons. Vertical respiration profiles are shown together with photosynthetic profiles in Figure 3.22. Although respiration was more or less similar in the water column down to a 4.5 m depth there were instances of depth specific higher respiration rates (Figure 3.22). The average value of area base community respiration (£R) was 0.311 g O2 m"2 h"1 (±0.059) and it ranged between 0.076 g O2 m"2 h"1 and 1.19 g O2 m"2 h*1. Rate of community respiration during each incubation is given in Table 3.11. Community respiration was more than 0.03 mg O2 I'1 h"1 only on one occasion during the first year of the 93 study, while it was less than 0.03 mg O21"1 h"1 only once during the second year of the study. Community respiration was relatively high and more than 0.20 mg Chf'h"1 towards the end of the study period (Table 3.11). 33.5 Primary Production Figure 3.26 (a) depicts monthly variation of area based gross primary production and corresponding community respiration of Kandy Lake during the period under investigation. Area based values were obtained by depth integration of photosynthetic rate and community respiration. The area based gross primary 2 1 2 1 production rate QTA) ranged from 0.21 lg 02 m' h" to 2.775 g 02 m' h" with a mean value of 0.98 g O2 m"2 h"1 (± 0.62). £A was less than 0.300 g O2 m"2 h"1 in September and October 1996 and more than 0.500 g O2 m'2 h"1 during the rest of the months except in September and December 1997 (Table 3.11). Further, £A was more than 1.00 g O2 m* h on nine occasions during the period under investigation. £A was less during the north east monsoonal month during both years. Net primary production (£NP) calculated per unit area using £A and £R values ranged from 2 1 2 1 2 1 0.073 g 02 m" h" to 1.715 g O2 m" h" with a mean value of 0.67 g 02 m" h" (± 0.51). Figure 3.26 b and Table 3.11 further shows a positive net primary production in Kandy Lake during the study period. Net productivity also shows a more or less similar trend to gross primary productivity. Seasonal fluctuation of daily production C££A) and community respiration (££R) are shown in Figure 3.26 (b). A gradual increase in daily gross primary production is shown from September 1996 to July 1997, but values were relatively low during August and September 1997. An 1 I 94 increasing trend in daily production is shown again during 1998 with the highest daily value of 27.75 g O2 m"2 d"1 in July 1998. Figure 3.27 shows the relationship between light saturated maximum photosynthetic rate and area based gross primary production. This positive relationship is highly significant (pO.OOl) and the ratio of optimal photosynthesis per unit volume to column photosynthesis per unit area (V: O) ranged between 0.074 and 3.7 (Table 3.11). Figure 3.26: The monthly variation of (a) area based community respiration(£R) and area based gross primary productivity Q]A) and (b) and daily rates of area based community respiration (££R) and area based gross primary productivity (£Z A) in Kandy Lake. 95 2.0-. y = 0.5471x + 0.0032 1.6- R2 = 0.743. • o 1.2- E 0.8- i 0.4- 0.0- —1 1 1 1— i —i 0.0 0.5 1.0 1.5 2.0 2.5 3.0 SA (gOam'V1) Figure 3.27: Relationship between area based gross primary production per hour O) and light saturated optimum production per hour (Amax) (r = 0.86, p< 0.001). 96 Table 3.11: Photosynthetic characteristics and related parameters of Kandy Lake. Date Chl.-a 0 R 14 V:0 12/09/96 20 0.162 8 0.038 0.288 0.155 0.133 36% 1.78 14/10/96 23 0.210 9 0.037 0.215 0.142 0.073 0% 1.02 13/11/96 27 0.328 12 0.033 0.672 0.129 0.543 51% 2.05 17/12/96 13 0.257 19 0.034 0.634 0.139 0.495 45% 2.47 15/1/97 23 0.347 15 0.033 0.725 0.132 0.593 35% 2.09 13/2/97 23 0.427 19 0.038 1.087 0.145 0.942 20% 2.55 12/3/97 25 0.625 25 0.037 1.688 0.150 1.537 31% 2.70 10/4/97 30 0.493 16 0.035 1.257 0.142 1.115 6% 2.55 28/5/97 35 0.863 25 0.032 1.840 0.125 1.715 4% 2.13 12/6/97 26 0.443 17 0.033 0.892 0.137 0.755 0% 2.01 15/7/97 25 0.569 23 0.040 1.350 0.158 1.192 21% 2.37 21/8/97 26 0.113 4 0.034 0.223 0.132 0.091 24% 1.97 23/9/97 20 0.676 34 0.020 0.498 0.076 0.422 0% 0.74 17/10/97 25 0.485 20 0.092 0.763 0.377 0.386 0% 1.57 10/11/97 20 0.272 14 0.065 0.608 0.246 0.362 0% 2.24 16/12/97 25 0.311 13 0.062 0.384 0.254 0.130 0% 1.23 15/1/98 34 0.276 8 0.148 0.851 0.601 0.24? 2% 3.08 18/2/98 37 0.304 8 0.084 0.569 0.325 0.245 13% 1.87 11/3/98 32 1.08 14 0.129 1.691 0.489 1.202 36% 3.70 20/4/98 22 0.636 29 0.102 1.521 0.412 1.108 39% 2.39 14/5/98 14 0.402 28 0.067 0.828 0.271 0.557 0% 2.06 15/6/98 25 0.486 20 0.126 0.567 0.479 0.088 0% 1.17 24/7/98 21 1.705 83 0.298 2.775 1.190 1.586 0% 1.63 18/8/98 31 1.521 49 0.249 1.680 1.062 0.618 38% 1.10 97 3.4 Phytoplankton community 3.4.1 Species composition and abundance of phytoplankton Thirty seven species of phytoplankton belonging to eight taxonomic groups were identified to the generic level from Kandy Lake. Some phytoplankton species found during the study period are shown in plate 3. A list of reported species with their presence or absence during each sampling occasion is shown in Table 3.12. Among these taxonomic groups, the Family Chlorophyceae was the more diverse and contributed 41% to the total population by species number. Cyanophyceae contributed 24% while contribution of Diatomophycea and Euglenophyceae to the phytoplankton community were 11% and 8% respectively. Zygnemaphyceae and Xanthophyceae were represented by only two species and only one species was found from each of the Dinophyceae and, Chrysophyceae contributing minority of the phytoplankton community in the Kandy Lake. Figure 3.28 depicts the relative composition of major taxonomic groups of phytoplankton recorded from Kandy Lake during the period under investigation. 98 41% a Cyanophyceae m Dinophyceae • Diatomophyceae • Chrysophyceae • Chlorophyceae m Zygnemaphyceae • Xanthophyceae • Euglenophyceae Figure 3.28: Relative composition of major taxonomic groups of phytoplankton in Kandy Lake during the period under investigation. 99 Table 3.12: Phytoplankton species composition and their abundance in Kandy Lake. 1996 1997 1998 Taxa Remarks S O N D J F M A M J J A s O N D J F M A M J J A Cyanophyceae Aphanizomenon sp. 1 1 1 V p p p 1 p C r Aphanocapsa sp. 1 1 1 1 I 1 1 D 1 Chroococcus sp. 1 1 V 1 1 1 1 1 1 1 1 1 1 1 1 1 B Coelosphaerium sp. 1 1 1 1 1 1 1 1 1 V I 1 V 1 1 1 B Merismopedia punctata 1 P 1 1 1 P P P s s p p 1 1 P p 1 1 p s p 1 s p A r Microcystis aeruginosa P P P 1 P P 1 1 1 p 1 1 1 1 1 p p 1 p s 1 p A Microcystis flos-aquae 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 D 1 * pd B 1 r Microcystis incerta 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 p 1 p B •JL D Microcystis wesenbergi 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 • r, . B I Dinophyceae Peridinium sp. 1 V 1 1 1 D Diatomophyceae Aulacoseira granulata d d d d d d s s P p s s s s d d s s s d d s d s A Cyclotella sp. 1 1 1 1 V 1 1 1 1 1 V 1 1 V 1 1 B Urosolenia sp. 1 1 1 1 1 1 1 1 1 1 1 1 V 1 V 1 1 1 B Synedra acus 1 1 1 1 V 1 1 1 V 1 V 1 1 1 1 1 1 1 V 1 B Chrysophyceae Mallomonas sp. V V V V V V V V D Cblorophyceae Ankistrodesmus sp. V 1 1 1 1 1 V V V 1 1 p 1 1 p 1 B Botryococcus sp. V 1 1 1 1 1 1 V 1 1 1 1 p 1 1 1 C Chlamydomonas sp. V V V V 1 D .(continued) Coelastrum indicum 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 B Coelastrum reticulatum 1 1 p 1 p p 1 p p p p 1 p 1 1 1 p 1 p p p d P P A Coenococcus sp. 1 p 1 1 1 p 1 1 1 1 1 1 1 1 B Kichneriella sp. V V V V V D Monoraphidium sp. V V V V V V p 1 p 1 1 C Oocystis sp. V V V V 1 D Pediastrum duplex V V E Pediastrum simplex s s s s s s d d d d d d d d s s d d d p p P P P A Pediasturm tetras V V E Scenedesmus sp. V V 1 p V V V 1 1 1 1 1 1 1 1 1 1 B Tetraedron minimum V V V V V D Volvocalese species V V V E Zygnemaphyceae Closterium sp. 1 1 1 V 1 1 V 1 V V V p P V 1 1 P V V 1 1 1 1 1 B Staurastrum sp. 1 V V p 1 V V V V V V V 1 p 1 1 1 B Xanthophyceae Isthmochloron gracile V V V V V E Pseudostaurastrum sp. V V E Eugenophyceae Euglenophyte V V E Phacus sp. V V V E Trachelomonas sp. V V V E d - dominant, s - sub dominant, p - occur in large numbers, 1 - occur in low numbers, v - occur in very low numbers, A - very abundant throughout the study period, B - common throughout the study period but in comparatively low numbers, C - occur occasionally, D - rare in occurrence, E - very rare in occurrence Plate 3: Some phytoplankton species found in Kandy Lake during the study. 1. Microcystis aeruginosa 7. Ankistrodesmus sp. 2. Microcystis flos-aquae 8. Closterium sp. 3. Pediasturm simplex 9. Staurastrum sp. 4. Microcystis wesanbergi 10. Coelastrum sp. 5. Coelastrum indicum 11. Aulacoseira granulata 6. Merismopedia punctata 102 During the study period the phytoplankton community in the Kandy Lake was dominated by Pediastrum simplex and Aidacoseira granulata which belong to the chlorophyta and diatoms respectively. In some^instances, two cyanobacteria species namely Microcystis aeruginosa and Merismopedia punctata and a chlorophyta species (Coelastrum reticulatum) also played a dominant role in the phytoplankton community of Kandy Lake while others occurred in small numbers. The most predominant species were a flat star-like colony forming form of Chlorophyceae: Pediastrum simplex, Coelastrum reticulatum and a filamentous form of a centric diatom namely Aulacoseira granulata and a colony forming form of Cyanophyceae: Microcystis aeruginosa and Merismopedia punctata The colony sizes and the shape of Microcystis aeruginosa, which has gas vacuoles, varied from each other. In the Chlorophyceae, Pediastrum simplex was the most dominant compared to the other fifteen species of the family. Six species out of fifteen were present in considerable numbers throughout the study period (Table 3.12), while others appeared occasionally in large numbers, particularly Chlamydomonas sp. Kirchneriella sp., Oocystis sp., and Tetraedron minimum were rare and Pediastrum duplex, P. tetras and some unidentified volvocalace species were very rare and found only two or three times during the study. Species composition of Diatomophycea was limited and the most dominant genus was Aulacoseira with A. granulata, and a sub species of A. granulata (as described by Rott, E., 1983). The other three taxa recorded from the Family Diatomophycea were the genus Cyclotella, Urosolenia and Synedra acus. They occurred throughout the study but in 103 relatively low numbers. Of the Cyanophyceae, the genus Microcystis and the genus Merismopedia were the most dominant with the other six genera found in Kandy Lake. In addition to M aeruginosa, M. flos-aquae, M. incerta and M. wesenbergi also occurred intermittently in small numbers in Kandy Lake throughout the study period. Merismopedia punctata periodically became the subdominant species of Cyanophyceae. Other species found in the family Cyanophyceae namely Chroococcus sp. and Coelosphaerium sp. were common throughout the study period, but occured in low numbers, while the genus Aphanizomenon occasionally became common in low numbers, and the genus Aphanocapsa was found rarely. Though the Zygnemaphyceae was not a dominant phytoplankton group in the Kandy Lake, genus the Closterium and the genus Staurastrum were found throughout the study period and occasionally became common and contributed significantly to the phytoplankton community. With respect to the species diversity, Euglenophyceae and Xanthophyceae were poorly represented but their contribution to the total phytoplankton assemblage of the lake could be important. Only one species of the genus Mallomonas belonging to the Family Chrysophyceae and genus Peridinium belonging to the Family Dinophyceae were found occasionally in small numbers in Kandy Lake during the period under investigation 3.4.2 Spatial distribution of phytoplankton community Species composition of phytoplankton was uniform throughout the lake and there was a prominent spatial pattern. Plate 4 shows the species composition at each station on the same sampling day. This plate shows that there was ho marked 104 difference in species composition among the four stations. The total phytoplankton density at each station during each month during the first twelve months of the study of Kandy Lake is shown in Figure 3.29. ,00 I 1 Figure 3.29: Total phytoplankton density of colonies l"1 at each station of Kandy Lake during the first twelve months of study. 105 Plate 4: Phytoplankton species composition at each sampling stations. 106 3.43 Seasonal fluctuation of the plankton community Seasonal changes of phytoplankton composition along with the mean chl-a concentration are depicted in Figure 330. A shift in the phytoplankton composition was observed from an Aulacoseira granulata to a Pediastrum simplex dominated assemblage and vice versa during the study period. In the last four months of the study, it was observed that while Aulacoseira granulata was the dominant phytoplankton species, a shift from the genus Pediastrum to a subdominant Cyanophyceae Merismopedia puctata and vice versa was observed. A progressive development of Merismopedia punctata and Microcystis aeruginosa in the phytoplankton assemblage was significant towards the end of the study. Although, the relationship between rainfall and the abundance of Aulacoseira granulata (r=0.024; p>0.05) and Pediasturm simplex (r=0.158; p>0.05) were positive, it was not significant. An inverse relationship was found between rainfall and the genus Microcystis (r=-0.104) but it also was not significant at 5% probability level. A significant relationship was found between the abundance of Pediastrum simplex and the chl-a concentration (r = 0.454; p<0.05). Plate 5 is shown that the observed shift of phytoplankton composition during the study period. 3.4.4 Phytoplankton diversity Shannon diversity index ranged from 0.255 to 0.632 during the study period and it shows a marked relationship to monthly rainfall (Figure 3.31). Diversity 107 increases following the northeast monsoonal rainfall and decreases to a greater extent during the prolonged dry period. S-ONDJ-FMAMJ JASONDJ-FMAMJ J 96 97 98 gAulacoseira granulata ^Pediastrum simplex rjMicrocystis sp, ^Coelastrum sp. ffMerismopedia sp. ^Staurastrum ^Others Figure 3.30: Seasonal changes of major genera of phytoplankton with the mean chl- a biomass. 108 Figure 3.31: Monthly variation of rainfall and Shannon index for phytoplankton species diversity of Kandy Lake during the study period. 109 c Plate 5: Observed shirt of phytoplankton composition during the study period. A: Dominance of Aulacoseira granulata in the phytoplankton assemblage. B: Dominance of Pediastrum simplex in the phytoplankton assemblage. C: Dominance of Microcystis aeruginosa in the phytoplankton assemblage. 110 3.4.5 Vertical distribution of the phytoplankton community With respect to depth stratification of phytoplankton, most of the Microcystis sp. were largely confined to the upper layer from the surface to 5 m and, Pediastrum and Aulacoseira were mostly absent between 2 m to 3 m depths. Densities of various plankton species in different strata along with the chl-a concentration at different depths are illustrated in Figure 3.32. All three dominant genera were found in the bottom of the lake on each sampling occasion in small number. Figure 3.32 clearly shows that there is a relationship between vertical distribution of phytoplankton and chl-a concentration. Data are not available below 5 to 11.5 m to investigate the continuous vertical pattern. 1 AJaxeaiasp Fttastnrnsp Atoogsfcip CM-a (PGR ) 0 10000 2000O 30000 0 10000 20000 30000 0 2000 4300 6000 12 20 23 Figure 332: Vertical distribution of the major phytoplankton genera at different depths along with the chlorophyll-a concentration. ill CHAPTER 4 DISCUSSION A limnological study on Parakrama Samudra, a large shallow impoundment in the dry zone during 1979-1980 (Schiemer, 1983), is one of the most comprehensive investigations carried out in Sri Lanka. Although Kandy Lake has been subjected to several studies (Dissanayaka, et al., 1982; De Silva & De Silva 1984; Dissnayaka, et al, 1986b) a comprehensive limnologcal study has never been carried out in Kandy Lake or other such urban water bodies in Sri Lanka. Therefore, it is anticipated to discuss the results of the present study to fill this long existing lacuna. Stagnant water bodies with long retention time are susceptible for cultural eutrophication and subsequent out-breaks of algal blooms. By examining available information on water quality, Silva (1996) predicted a progressive eutrophication in Kandy Lake. He also hypothesized a sudden out break of Microcystis bloom with the onset of the Southwest monsoon in 1999 as a result of a sudden drop in water level and wind action (Silva, 2003). This small water body was horizontally homogeneous with respect to most of the physical and chemical parameters during the study period, indicating proper horizontal mixing throughout the year. Spatial homogeneity is common in most of the shallow reservoirs in Sri Lanka but, when the basins are deep, vertical stratification occurs (Duncan et al., 1993). Studies conducted in Parakrama Samudra (Schiemer, 1983), Udawalawe, Minneriya and Victoria 112 reservoirs (Silva & Gamlath, 2000; Silva & Schiemer, 2001) clearly demonstrated spatial homogeneity in shallow water bodies and vertical stratification in the reservoirs with deeper basins. However, minor thermal gradients may be established even in shallow reservoirs under calm conditions, which can break with the prevailing wind (Bauer, 1983). Although, Kandy Lake is a shallow water body in comparison to the other reservoirs in Sri Lanka, the deeper basin shows a permanent stratification throughout the study period. This has resulted in a significant vertical gradient in some physico-chemical parameters such as pH, temperature, alkalinity and dissolved oxygen from surface to bottom. Stratification of deep basins may be attributed to several characteristics such as basin morphology, fetch, inflow and outflow, and occurrence of wind breakers in the surrounding landscape. The surface temperature was relatively low from December to March compared to dry months (March - April), but the seasonality was not very prominent during the study period. However, a vertical gradient of temperature in the deep basin was significant although it was not established as a thermocline. Slight seasonal changes of surface temperature in Kandy Lake may be attributed to climatic seasonality to a greater extent The relatively low temperature range compared to lowland reservoirs may be due to its location in the mid altitude around 500 m amsl. Lake water was alkaline throughout the study period: and the surface water was more alkaline than the bottom water. There were instances that bottom water in the deeper basin became acidic. In general, Sri Lankan reservoirs show marked seasonality in pH in relation to the annual rainfall pattern and acidic pH has been 113 reported even in downstream irrigation tanks during the Northeast monsoon (Silva & Davies, 1987). pH of the surface water in Kandy Lake also has dropped significantly just after rainfall (Silva, unpublished data). When photosynthetic organisms consume free carbon dioxide, the bicarbonate concentration will increase in the water while microbial activities under anoxic condition may results in acidic pH (Wetzel, 1983). This indicates that microbial processes occurring in the bottom layer of the deep basin of the Kandy Lake were quite different form shallow bottom waters. Sharaff (2003) found a fair number of acid producing facultative bacteria species in the bottom waters of the deep basin. Seasonal and spatial trends in bicarbonate alkalinity were more or less similar to that of pH in the Kandy Lake, but the pattern at site A was reversed, and the bottom water was more alkaline than the surface water in terms of bicarbonate concentration. Electrical conductivity (EC) of Kandy Lake was relatively high during the study period, compared to the other water bodies located at a similar elevation (Silva & Gamlath, 2000). The lowest value coincided with the highest rainfall and it increased significantly under dry weather perhaps due to high evaporation loss. EC is significantly related to the concentration of bicarbonate ions and relatively high EC in bottom water of the deeper basin may also be attributed to relatively high concentration of bicarbonate. However, question arises as to why electrical conductivity of Kandy Lake was relatively high compared to the water bodies located at similar altitude. This may be attributed to the high amount of wastewater discharged into the lake and the eutrophic nature of the water body. In general, urban water bodies contain more cations and anions as a result of human activities. Spatial 114 and temporal trends in total dissolved salts in Kandy Lake were more or less similar to that of electrical conductivity. Electrical conductivity is a function of total cations and anions, and temperature. Silva (2003) derived a highly significant linear relationship between electrical conductivity and total dissolved salts for irrigation reservoirs in Sri Lanka. Therefore, EC of lake water is a good predictor of total dissolved salts. In the case of Kandy Lake, relationship between EC and TDS was not significant and cannot be explained. Total suspended solids in the lake were relatively high during the rainy season and there was no marked difference in TSS during the dry and moderately rainy seasons. However, this pattern was not in agreement with the seasonal pattern of turbidity. Turbidity was relatively high during the moderately rainy season compared to dry and wet periods. The particles suspended in the water could be either inorganic or planktonic. The content of inorganic particles are extremely high in the inflow area during the wet season due to sediment loading and they settle down to the bottom within a few days. It is interesting to note that there was no significant relationship between turbidity and TSS in Kandy Lake during the study period. This situation cannot be readily explained. Perhaps this is due to analytical or experimental errors. Kandy Lake showed intermittent fluctuations in dissolved oxygen in both surface and bottom water within a wide range. Although a horizontal gradient was not prominent during the study, a vertical gradient at the deep basin was prominent throughout the study period. The vertical gradient could be a result of high 115 photosynthetic rate at the surface layer, poor mixing and high consumption of oxygen by anaerobic bacteria. This is a common phenomenon in eutrophic tropical water bodies (Tailing & Lemoalle, 1998). It is worthwhile mentioning that stratification of the deep basin collapsed only during a few occasions under prevailing high wind during the peak southwest monsoon. Since the bottom layers of the deep basin was anoxic or contained less oxygen during most of the year, only Chaoborus, a dipterans larva was found in the bottom of the lake. The stratification which occurred in Kandy Lake cannot be compared with temperate water bodies since there was no distinct vertical zonation such as the epilimnion, metalimnion and hypolimnion. However, anaerobic processes which occur in the bottom layer of the deep basin perhaps significantly affect the biogeochemical processes and nutrient dynamics of the entire lake ecosystem. Of the major cations, calcium was the most dominant cation and magnesium concentration was higher than the sodium ions. The lowest concentration of cation was potassium. Among the anions, bicarbonate is very common in tropical freshwaters. However, with repect to composition and ratio of anions, the chloride ion was subdominant in Kandy Lake while sulphate was least abundant among the anions during the study period. Most of the surface waters in Sri Lanka are rich in sodium, and the concentration of calcium and magnesium are more or less similar but potassium has the lowest concentration (Dissanayaka & Weerasooriya, 1986). Even though, the bicarbonate concentration was higher compared to chloride and sulphate a 116 substantial concentration of chloride is present in surface waters of Sri Lanka (Silva & Gamlath, 2000). The ionic composition and relative ionic ratios in lowland dry zone reservoirs are also more or less similar to mat of the highland water bodies but concentrations are relatively higher. Watershed geochemistry has been shown to be the most important determinant of water chemistry (Douglas, 1968; Webb & Walling 1974). Gibbs (1970) indicated that rock dominance, precipitation and evaporation-crystallization processes explain the chemical composition of epicontinental waters. Altitude can also influence the water chemistry due to variation in water balance (Vitousek, 1977). However, nothing is definitive, spatial and temporal variation in water chemistry conditioned by the overall framework with the hydrogeochemical system functions but, within a particular climatic region, rainfall, watershed characteristics, vegetation and soil become more significant (Gower, 1980). Relatively higher concentration of calcium and magnesium compared to sodium in Kandy Lake may be attributed to watershed characteristics especially in urbanized areas without waste treatment facilities or geochemical characteristic of the watershed. Calcium and magnesium are higher in Sri Lankan waters when it drains a watershed dominated by dolomite. In general, sodium should be higher when the water body is receiving urban wastewater. The concentration of the upper limit of dissolved phosphorous in Kandy Lake was slightly higher than the reported values for Sri Lankan reservoirs (Gunathilake, 1983; Silva & Gamlath, 2000) but the mean value is within the range. Undetectable concentration of dissolved phosphorous indicates a rapid turnover within the water body. Eutrophic water bodies with higher algal biomass uptake dissolved 117 phosphorous from the nutrient pool very rapidly (Wetzel, 1983). However, mobilization of dissolved phosphorous from the anoxic bottom is also extremely rapid. Silva (2003) hypothesized that rapid nutrient uptake by non-nitrogen fixing cyanobacteria when water level was reduced has resulted in an outbreak of Microcystis aeruginosa bloom in 1999 in the Kandy Lake. Extremely high concentration of total phosphorous was found in the bottom waters of the lake, which indicates accumulation of organic matter in the bottom. This may be from sinking of dead algal biomass as well as faeces of the fish. When fish biomass is extremely high in non-harvesting water bodies their contribution to the nutrient pool is colossal. Relatively low concentrations of total phosphorous in the surface layer compared to the bottom of the water body may be attributed to rapid uptake by phytoplankton. Dynamic fluctuation of phosphorous species without a particular seasonal trend is an indication of the biogeochemical processes occurring in the lake. It is more significant that it impacts on rainfall bound inflows or otherwise the lake may be receiving high loads of phosphorous containing waters intermittently through its waste water drains located around the water body. The dominance of NH4 - N of the nitrogen species in Kandy Lake is an indication that organic nitrogen is converted into ammonia through the process of ammonitrification. This transformation is carried out by all the microorganisms. Sharaff (2003) reported the occurrence of nitrification and denitrifying bacteria in high densities in Kandy Lake. Planktonic algae prefer to uptake NH4 - N as then- nitrogen source over NO2 - N or NO3 - N (Brezonik, 1972; Zevenboom & Mur, 1981). When concentration of NH4 - N or other nitrogen species are high in the 118 water body, it results in colonization of non-nitrogen fixing cyanobacteria such as genus the Microcystis, Merismopedia etc. Relatively high concentration of nitrogen species and their intermittent fluctuation in Kandy Lake during the study period indicates organic pollution of the water body. This may be either from allochthonous or autochthonous organic loading. Seepage of autochthonous NO3 - N into the water body may be extremely high from the surrounding watershed, which has schools, nursing homes and hotels without having central water treatment facilities. It has been observed that some hotels release their waste, even night soils to the lake directly during the night through underground canals. Further, a slight increase in total phosphorous and total nitrogen in the surface water of the lake was found following rainfall although there was no clear seasonal pattern in nutrient concentration. This suggests that wastewater generated in the watershed also contributes a substantial amount of nutrients to the water body. Furthermore, the droppings of aquatic birds such as roosting cormorants, wading pelicans and visiting birds (e.g. egrets) and the bats residing on trees around the lake will certainly contribute to the nutrient loading of the lake. In addition, the foliage parts (e.g. leaves) fallen from the trees around the lake would also append to the lake as a nutrient source. Kandy Lake is also affected by recycling of nutrients within the water body which enhances microbial activity. Particularly, the: feeding behaviour of cichlid fish e.g. tilapia, the most dominant fish in the water body stirs up bottom sediment as they browse on the bottom substrate. In addition, vertical gradient of nutrients in the deep basin of the lake may be attributed to the chemical gradient (e.g. pH, DO, temperature) and also to the distribution pattern of the microbial population. This study clearly shows a progressively increasing trend in the densities of non- 119 nitrogen fixing cyanobacteria towards the end of the study. Although dissolved silica also plays an important role in relation to nitrogen and phosphorous with respect to phytoplankton dynamics and species composition, no information has been collected on dissolved silica from Kandy Lake during this study. Many studies have demonstrated the response of the diatom population to dissolved silica concentration or it relative proportion to total nitrogen and phosphorous (Werner, 1977; Paasche, 1980). Aulacoseira granulata showed a rhythmic oscillation with Pediasturm simplex in Kandy Lake during this study. Perhaps dissolved silica or its relative ratios to total phosphorous or total nitrogen may play an important role with respect to phytoplankton composition in Kandy Lake. Dokulil et al (1983) reported the average total incoming radiation (TIR) of 24.4 ± 1.9 (7.7%) M J m"2d"' during their studies on photosynthetic primary production of Parakrama Samudra. They also found a relatively narrow range of Photosynthetically Active Radiation (PhAR) of 2150 - 2350 uE m"2 s"1 and 1600 - 2150 uE m"2 s"1 during incubation periods in 1979 and 1980 respectively. This fraction is around 45% of TIR as reported by Tailing (1965) for equatorial lakes in Africa. The range of PhAR at Kandy Lake during the study period was much wider than the reported values by Dokulil et al (1983) and the incidence of low mean values of PhAR during the incubation were also much higher. Relatively low incidence of PhAR on Kandy Lake could be attributed to the cloud cover, which is common at higher altitudes in the wet zone of the country. Silva et al. (2002) reported a relatively wide range of PhAR (888 - 2241 uEm"2 s"1) during the northeast monsoon when they conducted photosynthetic experiments at eleven irrigation 120 reservoirs located islandwide. Irrespective of monsoon driven seasons, incidence of PhAR on Kandy Lake showed intermittent fluctuations. However, PhAR impinging on Kandy Lake was sufficient for light saturated maximum photosynthesis (Amax) in the lake except for a few instances. Secchi depth values an indirect approximation of euphotic depth in Kandy Lake indicate a relatively high vertical extinction coefficient. Although there were instances of extremely low Secchi depth values at the beginning of the study, secchi depth showed a decreasing trend towards the end of the study period. Extremely low secchi depth recorded in October 1996 can be directly attributed to enormous sediment loading into the lake following heavy rains. A more or less similar situation but in lesser magnitude occurred in October 1997 and August 1998. Apparently, a slight gradient of $ecchi depth was established from inflow to outflow when sampling was conducted following a rainy day. Underwater light extinction was also affected by phytoplankton densities of the water body. This is evident from the negative relationship between Secchi depth and chlorophyll-a content of Kandy Lake during dry and lower rainy months. Low Secchi depth values during dry months may be attributed to high growth of phytoplankton and the overall decreasing trend during the study period is an indication of progressive eutrophication. The variation in light extinction in Kandy Lake may be attributed to increased light absorption and scattering by dense populations of phytoplankton. Euphotic depth of the deep basin of Kandy Lake was confined up to 4.4 m when it is calculated using the relationship derived by Dokulil et al. (1983) for 121 Parakrama Samudra (Z eu = ZSD * 2.75). However, the results show that Z eu was less than 3.0 m during most of the time period under investigation. Recent studies conducted in Kandy Lake on light condition and photosynthetic primary production showed a stronger relationship between ZSD and Z eu as Z eu = ZSD * 2.5 (Silva, unpublished data). Accordingly, in the deep basin of the lake, about 8 m of the water column remained unproductive as the aphotic zone (Za) during the period under investigation. Relatively low euphotic depths estimated for Kandy Lake during this study may be a combined product of absorption and self-shading by phytoplankton and scattering by inorganic particles (Silva et al, 2002). A significant relationship between ZSD and chlorophyll-a occurred only in dry and lower rainy months is evident that Kandy Lake had more or less similar densities of phytoplankton in the euphotic zone which can be affected by sediment loading resulting from heavy rains. Relatively high chlorophyll-a content in the water body during dry months indicates high growth of planktonic algae, as a result of photosynthetic efficiency, whereas a decrease in chlorophyll-a, during rainy months is essentially a result of dilution and biomass loss through the outflow. The lake spills over only during the rainy season. The concentration of chlorophyll-a in the lake was above eutrophic level throughout the study period according to the classification of Dillon & Rigler (1974) for water bodies. This situation in Kandy is quite different from other water bodies such as deep upland reservoirs and shallow irrigation lowland reservoirs in Sri Lanka. Schiemer et al, (2001) showed a trophic shift in lowland irrigation reservoirs from rainy season to dry season with respect to chlorophyll-a content, while chlorophyll-a content in the deep upland reservoirs remains constant 122 throughout the year. This indicates a consistent eutrophic nature of Kandy Lake during the period of investigation leading to a hypereutrophic condition. The occurrence of chlorophyll-a in the aphotic zone but in lesser amounts indicates wind mixing circulation of algal biomass. Certain cyanobacteria for example, the genus Microcystis is capable of sinking down to the lower layers in the night and becomes buoyant during the daytime. Further, a higher abundance of Merismopedia species was observed in the bottom layers of the Kandy Lake. Furthermore, certain phytoplankton species are physiologically adapted to stay under dark conditions for a prolonged period of time (Reynolds, 1984). Surface inhibition, essentially a light induced diel phenomenon (Harris, 1978) was characteristic of a majority of the photosynthetic profiles derived from the light and dark bottle experiment in the Kandy Lake. The occurrence of subsurface maxima or maximum photosynthesis around noon (Dokulil et al., 1883; Silva & Davies, 1986, 1987; Silva et al., 2002) during this study suggests that surface photoinhibition is a light induced ecophysiological phenomenon, which is common in tropical freshwaters. Subsurface PhAR intensity should exceed 900 uEm"2 s"1 for surface inhibition to occure (Dokulil et al. 1983) but it varies among the reservoirs and the lower and upper limits are 200 uEm'2 s"1 and 2200 uEm"2 s~l respectively (Harris, 1978). However, there are instances when the surface maximum occurred when subsurface PhAR intensities were well above the photoinhibition threshold resulting from the ecophysiological adaptation of the phytoplankton population (Vincent, 1979; Ward & WetzeL 1980). Such instances were evident with vertical profile derived from the Kandy Lake during this study. Although Silva & Davies 123 (1987) have shown a monsoon bound seasonal pattern in surface inhibition in dry zone irrigation reservoirs such a pattern was not prominent in the Kandy Lake. However, the percentage surface inhibition was markedly higher under dry weather conditions. A wide range of Amax optimal light saturated rate of photosynthesis may be attributed to several factors such as PhAR, nutrients, algal biomass, etc. Non significant relationships between chlorophyll-a and Amax is a good indicator of multifactorial effects on light saturated photosynthetic efficiency. The upper values are within the range reported for Parakrama Samudra under extremely low water level in 1979 by Dokulil et al. (1983). The lower limits were within the range reported by several workers for different types of surface water bodies in Sri Lanka (Silva & Davies, 1986, 1987; Silva et al, 2002). An increasing of Amax in Kandy Lake during the period under investigation indicates a progressive transfer towards hypereutrophicaton. Specific photosynthetic rates (0max) and the rate of Amax per unit chlorophyll- a were within the range reported for other tropical freshwater bodies (Tailing & 1 Lemoalle, 1998) except an extremely high value of 83 mg 02 mg Chl-a "'h' found during this study for corresponding chlorophyll-a value of 21 jag 1"'. Silva et al, (2002) reported 0max value of 46 mg mg"1 h"1 for Chandrikawewa for a corresponding chlorophyll-a value of 6 ug l"1. The variation of the intrinsic shape of the photosynthetic profiles of the Kandy Lake may be attributed to several factors. The shape of the profile is primarily determined by light climate in shallow water bodies. When stratification of 124 algal biomass is confined mainly to the upper layer, photosynthetic rate may be modified resulting in irregular shapes of vertical profiles (Tailing, 1965; Dokulil et al, 1983). Silva & Davies, (1986) reported bi-model photosynthetic profiles for Parakrama Samudra and Pimburettewa reservoirs and state that they occurred during afternoon hours. Bi-modal photosynthetic profiles, in thermally stratified lakes in the Parakrama Samudra have been explained in terms of changes in the light regime and ecophysiological adaptations of phytoplankton. Bi-modal photosynthetic profiles found in Kandy Lake may be a result of either ecophysiological adaptation of phytoplankton or changes in light regime which may also be a result of experimental errors. The variation in compensation depth where photosynthetic rate is balanced by the respiration was compatible with the euphotic depth in Kandy Lake and the decreasing nature of both euphotic and compensation depths towards the end of the study indicate an increasing nature of eutrophic status. The range of oxygen consumption in the dark exposure, the community respiration and its magnitude relative to photosynthetic rate was wide in the Kandy Lake compared to the range established for tropical waters (Thornton, 1990; Tailing & Lamoalle, 1998). In a situation, dominated by autotrophs, the potential value of respiration to light saturated photosynthesis in the tropics is within 0.05 - 0.1 mg 02 l"1 h'1 and corresponding absolute specific rate 0max is 1 mg mg"1 h"1 (Tailing, 1965; Ganf, 1974; Lemoalle, 1983). Silva et al, (2003) observed exceptionally high oxygen consumption and negative net primary production in some Sri Lankan reservoirs with high algal biomass. In Kandy Lake the potential value of respiration 1 to Amax ranged from 0.02 to 0.174 mg 021"' h' during the period under investigation. 125 Ganf & Blazka (1974) reported high oxygen consumption by zooplankton in Lake George. Oxygen consumption by heterotrophic organisms takes place in waters with considerable input of organic matter. Similar trends were reported in flood plain water of the Amazon River (Wissamar, et al., 1981) and in Parakrama Samudra during low water level (Dokulil et at., 1983) which was attributed to consumption by heterotrophs. High heterotrophic oxygen consumption in Kandy Lake is evident from relatively higher BOD5 values reported during this study. Further, zooplankton densities are relatively low in this water body (M. Ekanayaka, Personal communication). Negative net primary production was not observed in Kandy Lake when oxygen consumption was integrated up to the estimated euphotic depth. If it was calculated for the entire water column, certainly there will be negative rates of net primary production. Negative production suggests that nutrient limitation for phytoplankton growth be replaced by the light limitation (Carignan & Planas, 1994). Khondker & Parveen (1993) reported high photosynthetic capacity during a seasonal (monsoonal) low abundance of phytoplankton from an Oxbow lake in Bangladesh. Net primary production was high in Kandy Lake as reported for India (Ganapati & 2 Sreenivasan, 1970). The highest X£NP in Kandy Lake was 17.2 g 02 m" d"' that is much higher than the value reported for Sri Lankan water bodies by previous authors (Dokulil et al, 1983; Silva & Davies, 1986 and 1987; Silva et al. 2003). A significant relationship between area based gross primary production (£A) and optimum light saturated production rate was found in the present study as reported from East African Lakes (Tailing, 1965; Mukankomeje, et al.; 1993). Guta and 126 Medda lakes in Panama (Gliwicz, 1976), Lake Mcllwaine in Rhodesia (Robarts, 1979), Lake Chad in West Africa (Lemoalle, 1979, 1981), Lake Xolotlan in Nicaragua (Erikson, et al., 1991), Dhanmondi Lake in Bangladesh (Khondker & Parveen, 1993) and eleven irrigation reservoirs in Sri Lanka (Silva et al., 2003). V/O ratio (Rhode, 1958), gross primary production per unit area, light saturated photosynthetic rate in Kandy Lake was within the range of previously reported values for Sri Lankan fresh water bodies (Dokulil et al., 1983; Silva & Davies, 1987; Silva etal., 2003). Species composition of freshwater phytoplankton in Sri Lankan is fairly well known (Silva & Wijeyaratne, 1999). However, seasonal variation in species abundance, biomass or bio-volume, spatial distribution patterns and the factors governing spatial and temporal distribution patterns are poorly understood. Kandy Lake was dominated by phytoplankton taxa belong to the family Chlorophyceae. Of the Chlorophyceae, the most dominant species was Pediastrum simplex, a common species found in most of the standing water bodies in Sri Lanka (Rott, 1983; Rott & Lenzenweger, 1994). In seasonal abundance Pediasturm simplex became a sub dominant species in the phytoplankton assemblage during certain months of the study and re-appeared as the dominant species in the lake. Of the other chlorophyta, only Coelastrum reticulum was found either in very low numbers or in large numbers and this species could be considered as a common species but found in low numbers. Scenedesmus species also appeared in the seventeen months throughout the study period. It is interesting to note that both Pediastrum duplex and Pediastrum tetras in the Kandy Lake were very rare species. Of the diatoms Aulacoseira 127 granulata was the dominant throughout the study period. A. granulata was also a common and climax species in tropical water bodies. This species is also found in most of the lowland fresh water bodies in Sri Lanka. It was noticed that A. granulata became subdominant when P. simplex dominates and vice versa. This rhythmic oscillation of A. granulata and P. simplex could have some links to the rainfall pattern, which was not prominent It has been reported that the occurrence of diatoms has direct links also to the concentration of dissolved silica or ratio of total nitrogen to dissolved silica. Since the concentration of dissolved silica was not determined during this study, no attempt is made to explain this rhythmic abundance of A. granulata in relation either to dissolved silica or to the total nitrogen to dissolved silica ratio. The relative abundance of other species of diatoms namely Cyclotella sp., Urosolenia sp., Synedra sp. occures in small numbers compared to A. granulata and cannot be readily explained. Recently, Rott et al. (in press) identified Urosolenia species found in Kandy Lake as U. denticulata (equal to freshwater counterpart of genus Rhizosolenia). U. denticulata is not a common species in Sri Lankan water bodies and it was described from Victoria and Minneriya reservoirs in Sri lanka. Of the, cyanobacteria, Merismopedia punctata become a subdominant species in the phytoplankton assemblage during a few instances. Although Microcystis aeruginosa also was found throughout the study period, it never appeared as a subdominant species but was found in large numbers intermittently. In addition, to M. aeruginosa, other species of the genus Microcystis found in Kandy Lake were 128 also common but not abundant (Dr. L.P. Jayathissa, pers. Comm.) reported that there are four strains of Microcystis aeruginosa in Kandy Lake of which two strains were toxic. Silva et al. (2003) showed a clear relationship between the occurrence two species of rotifers and Microcystis aeruginosa. Population densities of Brachionus calyciflorus declined and Keratella tropica became the dominant rotifer species. The abundance of total cladocerans and toxicogenic cyanobacteria, Microcystis aeruginosa were high under relatively acidic condition. B. calcyciflorus re-established itself progressively with decreasing population of M. aeruginosa and increasing densities of a filamentous diatom A. granulata and the colony forming green algae P. simplex. Desmids belonging to the family Zygnemaphyceae were almost absent in Kandy Lake except for the occasional presence of only two genera Closterium and Staurastrum in small numbers. However, Closterium species appeared thrice in large numbers whereas Staurastrum species was found in large numbers twice during the northeast monsoon. Desmids are very abundant in newly built upland reservoirs, which have a clear water column for several metre and also appeare in shallow lowland reservoirs with the filling of waters during seasonal inter monsoons. In these reservoirs an obvious phytoplankton shift has been observed from high water to low water (Rott, et al,. in press), but desmids in deep upland reservoirs remain throughout the year. The occurrence of Euglinophytae was confined only to the dry season under relatively low water level but in small numbers. 129 A uniform distribution of phytoplankton species in Kandy Lake may be attributed to spatial homogeneity of this small water body. The temporal variation species diversity found during the study period could be a result of high dilution effects occurring during the Northeast monsoon. Phytoplankton density in Kandy Lake was low during the rainy season compared to the dry season. The occurrence of Microcystis species in the uppermost layers during the daytime may be attributed to their capacity for buoyancy with the help of gas vacuoles. There may be some physiological adaptation which helps Pediastrum simplex and Aulacoseira granulata to remain in the subsurface layer, but within the euphotic zone. Because of anoxic layer in the deepest basin, except anaerobic micro-organisms other organisms cannot survive in a major art of the deep basin. Further, the occurrence of Chaoborus, a Dipteran larvae, which is found only under anoxic condition have been reported from the bottom sediments of the Kandy Lake (personal communication, Mr. Malida Ekanayake). 130 5. CONCLUSIONS Prolonged stagnant nature of Kandy Lake, which receives substantial inflows only during the northeast monsoon, has lead to it becoming a more or less closed lentic ecosystem with insignificant flushing. Mixing of the entire water body has been prevented by its morphometric features and location resulting in the establishment of permanent stratification in the deeper basin. Stratification has lead to building up of a nutrient rich sediment layer in the deeper basin, which mobilizes nitrogen and phosphorous species under anoxic conditions. No physico-chemical gradient was established along the inflow-outflow axis, but vertical gradients were prominent from surface to bottom in the deep basin and physico-chemical attributes were relatively large compared to the water bodies located at similar altitudes indicating human activities in the watershed and other catchment characteristics have significantly affect this water body. A progressive trophic shift towards more eutrophic conditions within a period of two years showed the magnitude of eutrophication within a short period in a non-harvesting urban water body. Phytoplankton composition and the relative abundance, concentration of micro - nutrients, photosynthetic ratio and the nature of the primary productivity profile are characteristics of a eutrophic water body, which is enriched by allochthonous and autochthonous processes. The overall results of this study 131 demonstrate that although it looks more productive, Kandy Lake is a partially dead water body polluted mainly by organic wastes. 132 6. References Abeywickrama, B.A., 1979. The genera of the freshwater algae of Sri Lanka. Pare 1- UNESCO Man and the Biosphere National Committee for Sri Lanka, Special Publication 6. National Science Council Sri Lanka, Colombo, 103p. Almond, D.C., 1994. Solid-Rock Geology of the Kandy Area Sri Lanka. Institute of Fundamental Studies, 74p. Amarasinghe, U. S., 1998. Reservoir Fisheries Management in Sri Lanka: Achievements, Mistakes and Lessons for Future. International Riviews of Hydrobiology (Special Issue) 83,523-530. Amarasinghe, U.S., H.H. Costa & M.J.S. Wijeyaratne, 1983. Limnological and fish production potential of some reservoirs in Anuradhapura District, Sri Lanka Journal of Inland fisheries 2:14-29. Amarasinghe, U.S., C. Nissanka & S.S.De Silva, 2001. Fluctuation in water level in shallow irrigation reservoirs: Implications on fish yield estimates and fisheries management, In De Silva, S.S. (ed.), Reservoir and Culture-based fisheries: Biology and Management. ACIAR Publication No 98, Canberra, Australia: 101-110. Amarasiri, S.L., 1973. Water quality of major irrigation tanks in Sri Lanka. Tropical Agriculture. CXXTX 19-25. Anon, 2003. Dam Safety and Reservoir Conservation Programme (Status Report): Activity 13: Catchment Land Use, Peripheral, Reservoir and Riverine Impacts, Mahaweli Authority of Sri Lanka 62p. 133 APHA, 1989. Standard Methods for Estimation of Water and Waste Water, 17th Edition, American Public Health Association, Washington DC. Apstein, C, 1907. Plankton in Colombo Lake, See auf Ceylon Zoo. J 6 (Abt, Syst.) 25: pp 201-244 Apstein, C, 1910. Das Plankton des Gregory, See auf Ceylon. Zooi.jb.Abt.Syst. 29:661-679 Bartram, J. & Balance, R., (eds). 1996. Water Quality Monitoring. A practical guide to the design and implementation of freshwater quality studies and monitoring programmes. Chapman and Hall, USA. Bauer, K. 1983. Thermal stratification, mixis and advective currents in Parakrama Samudra Reservoir, Sri Lanka. In: F. Schiemer (ed.), 27 -34 pp. Limnology of Parakrama Samudra - Sri Lanka: a case study of an ancient man-made lake in the tropics, Developments in Hydrobiology 12. Dr. W. Junk Publishers, The Hague. Brezonik, P.L., 1972. Nitrogen: sources and transformations in natural waters. In Allen, H.E. & J.R. Kramer (eds.), Nutrients in Natural Waters, New Yolk: Wiley, 1-47. Carignan, R. & D. Planas, 1994. Recognition of nutrient and light limitation in turbid mixed layers. Three approaches compared in the Parana floodplain (Argentina). Limnol. Oceanogr. 39:439-443. Chandrasoma, J., G. Muthukumarana, K. Pushpakumara & A. Sreenivasan, 1986. Limnology and fish production in Udawalawe reservoir, Journal of Inland Fisheries, Sri Lanka 3:68-88. 134 Chapman, D. (ed.), 1992. Water Quality Assessment A guide to the use of biota, sediments and water in environmental monitoring. Chapman and Hall, London. Corrin, M.L. & D.F.S. Natusch, 1997. Physical and chemical characteristics of environmental lead. In: W.R. Boggess (ed.), 7-31 pp. Lead in the environment, National Science Foundation, Washigton, DC. Costa, H.H. & S.S. De Silva, 1978a Hydrobiology of Colombo Lake II. Seasonal variations in physico chemical factors. Spolia Zeylanica. 32:26-41. Cresser, M. & A. Edwards, 1984. Acidification of fresh waters. Cambridge, USA. De Silva, P.K. 1993a. Physico-chemical characteristics, phytoplankton and the fishery of Victoria. Ceylon Journal of Science (Biological Science) 34:29-39. De Silva, P.K. 1993b. Some aspects of the limnology and the fishery biology of Kotmale and Randenigala reservoirs in Sri Lanka. Sri Lankan Agricultural Science. 30: 71-85. De Silva, P.K. & de Silva, K.H.G.M., 1984. An ecological study of the meso and macro fauna of the littoral region of Lake Kandy in Sri Lanka. In: Arch. Hydrobiol., 1984,102,53. Dillon, P. & Rigler, R.H. 1974. A test of a simple nutrient budget model predicting the phosphorous concentration in lake water. J. Fish. Res. Bd. Can. 31:1771-1778. Dissanayake C.B., S.V.R. Weerasooriya, & S.H.G. de Silva, 1982. The environmental chemistry of Mahaweli Rive. Sri Lanka Int. J. Env. Studies 78:207-293. 135 Dissanayaka, C.B., & V.R. Weerasooriya, 1986. Environmental chemistry of Mahaweli River, Sri Lanka. Int Nat J. Env. Studies. 28:207-233. Dissanayaka, C.B., A.M. Rohan Bandara, S.V.R. Weerasooriya, 1986b. Heavy metal abundance in the Kandy Lake: An environmental case study from Sri Lanka. Geol. Wat. ScL 10(2): 81-82. Dokulil M., K, Bauer & I. Silva, 1983. An assessment of the phytoplankton biomass and primary productivity of Parakrama Samudra, a shallow man- made lake in Sri Lanka. In: F. Scheimer (ed.) 49-76. pp. Limnology of Parakrama Samudra - Sri Lanka: a case study of an ancient man-made lake in the tropics, Developments in Hydrobiology 12. Dr. W. Junk Publishers, The Hague. Douglas, I., 1968. The effects of precipitation chemistry and catchment area lithology on the quality of river water in selected catchments in Eastern Australia. Earth Science Journal, 2:126-44. Duncan, A, A. Gunathilake & F. Schiemer, 1993. Limnological Aspects of Landscape Management in Sri Lanka. In: W. Erdelen, C.Peru, N. Ishwaran and CM. Maddumabandara (eds.), 381 - 395 pp Ecology and Landscape Management in Sri Lanka, Proceedings of an International and Interdisciplinary Symposium, Colombo, Sri Lanka, 12- 26 March 1990, Margraf Verlag, Germany. Edmondson, W.T., 1994. What is limnology? In: Margalef, R. (ed.) 1994. Limnology Now. Elsevier, The Netherlands. Erikson, R., E. Hooker & M. Meija, 1991. The dynamics of photosynthetic activity in Lake Xolotlan (Nicaragua), Verh. Int. Ver. Limnol. 24:1163-1169. 136 Fernando, C.H., 196S. The development of Ceylon's fisheries II. The role of inland waters in relation to the development of Ceylon's fisheries and a note on the pearl oysters fishery. Bulletin of the Fisheries Research Station Ceylon 17:291-297. Fernando, C.H., 1990. The fresh water invertebrate fauna of Sri Lanka. In: Zoological survey of Sri Lanka: freshwater faunal and fisheries of Sri Lanka. C.H. Fernondo (ed.) Natural resources, Energy and Science Authority of Sri Lanka, pp 1 - 447. Fernando, C.H., & R. Rajapaksha, 1983. Some remarks on long-term and seasonal changes in the zooplankton of Parakrama Samudra. Pp. 77-84. 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. Fuelleborn, F., 1900. Uber Untersuchungen in Nyassa-see und in den seen im nordlichen Nyassaland. Verhandlungen der Gesellschaft fur Ergkunde, Berlin 28: 332-338. Ganapati, S.V., A. Sreenivasan, 1970. Energy flow in natural aquatic ecosystems in India. Arch. Hydrobiol. 66:458-98. Ganf, G.G., 1974. Phytoplankton biomass and distribution in a shallow eutrophic lake (Lake George, Uganda). Oecologia 16:9-29. Ganf, G.G. & P. Blazka, 1974. Oxygen uptake, ammonia and phosphate excretion by zooplankton of a shallow equatorial lake (Lake George, Uganda). Limnol. Oceanogr. 19:313-325. 137 Gibbs, R.J., 1970. Mechanism controlling world water chemistry, Science 170: 1088-1090. GliwicZj Z.M., 1976. Plankton photosynthetic activity and its regulation in two neotropical man-made lakes. Ploska Arch. Hydrobiol. 23: -61-93. Gordon, N.D., T.A. McMahon, B.L. Finlayson, 1992. Stream Hydrology. An Introduction for Ecologists. John Wiley and Sons Ltd., England. Gower, A.M. (Ed), 1980. Water quality in catchment ecosystems, John Wiley & Sons, Chichester: 12p. Gunathilake A., 1980. The chemistry of Parakrama Samudra In: Schiemer, F. (ed.) Parakrama Samudra Limnology Project, Interim report Pp 35-54. Gunathilake, A., 1983. Phosphorus and Phosphate dynamics in Parakrama Samudra based on diurnal observations. In F. Schiemer (ed). Limnology of Parakrama Samudra Sri Lanka, Dr. W. Junk publishers 35-47. Gunathilake, A. & C. Senaratne, 1981. Parakrama Samudra Sri Lanka Project A study of a tropical lakes ecosystem. The chemical environment with special reference to nutrients. Verb. Inteneat, Veren. Limnol. 21:1000-1006. Gunawardena H.D. & A.M.K.R. Adhikari, 1981. Studies on the quality of irrigation waters in Kalawewa area. J. Nat. Sci. Council. Sri Lanka 9: 121- 148. Harris, G.P., 1978. Photosynthesis, productivity and growth: the physiological ecology of phytoplankton. Arch. Hydro-biol. Beih. Ergebn. Limnol. 10: 1- 171. Juday, C, 1924. The productivity of Green Lake, Wisconsin. Verh. IntVer.Limnol. 2: 357-360. 138 Khondker, M. & L. Parveen, 1993. Daily rate of primary productivity in hypertrophic Dhamondi Lake. In Tilzer, M.M. & Khondker, M. (eds), Hypertrophic and Polluted Freshwater Ecosystems: Ecological bases for water resources management Bangladesh Department of Botany, University of Dhaka, 181-191. Lemoalle, J., 1979. Activite photosynthetique du phytoplankton du Lac Tchad. OSRTOM Raport pour la Limnologie Afri-caine Nairobi, 16-23 decembre 1979. Lemoalle, J., 1981. Photosynthtic production and phytoplankton in the euphotic zone of some African and temperate lakes. Rev. Hydrobiol. Trop. 11:31-38. Lemoalle, J., 1983. Phytoplankton Production. In Carmouse, J.P., J.R. Durand & Leveque (eds), Lake Chad. Ecology and productivity of a Shallow Tropical Ecosystems. Monogr. Biologicae 53, The Hague, Junk Publishers: 357-384. Lewis, W.M. Jr., 1973. A limnological survey of Lake Mainit, Philippines. Int. Revue ges. Hydrobiol. 58:801-818. Lewis, W.M. Jr., 1974. Primary production in the plankton community of a tropical lake. Ecol. Monogr. 44: 377-409. Lewis, W.M., 1977. Feeding selectivity of a tropical Chaoborus population. Freshwat Biol., 7: 311-325. Lewis, W.M., 1978. Dynamics and succession of the phytoplankton in a tropical lake; Lake Lanao, Philippines, J. Ecol,. 66: 849-880. Lewis, W.M., 1996. Tropical Lakes: How latitude makes a difference. In: Schiemer, F. & K.T. Boland, (eds), Perspective in Tropical Limnology, SPB Academic Publishing, Amsterdam, 43-64. 139 Maybeck, M., D. Chapman, R. Helmar (eds), 1989. Global Freshwater Quality, A First Assessment. Blackwell references, Oxford, 306 pp. Mendis, A.S., 1965. Preliminary study of twenty-one Ceylon lakes. 2. Limnology and fish production. Bulletin of the Fisheries Research Station, Sri Lanka, 18:7-16. Moore, W.J., 1989. Balancing the needs of water use. Springer - Verlag. USA. Mukankomeje, R, P.D. Plisner, LP. Descy & L. Massault, 1993. Lake Muzahi, Ruwanda limnological features and phytoplankton production. Hydrobiologia. 257:107-120. Nathanael, S. 2001. Some aspects of the biology and the fishery of exotic cichlids in Victoria, A deep reservoirs in Sri Lanka. Ph D. Thesis University of Peradeniya 282 p. Nathanael. S & E.I.L. Silva, 1993. Future trends of freshwater resources in Sri Lanka - An urgent need for long term research Sri Lanka Agricultural Science 3: 79-85 Nissanka C, 2002. Empirical models for fish yield prediction in selected reservoirs in the dry zone of Sri Lanka, Ph. D. Thesis University of Kelaniya. 256p. Panabokke, C.R., 1996. Soil and Agro-Ecological Environment in Sri Lanka. Natural Resources Series 2. Natural Resources, Energy and Science Authority. Sri Lanka: 220pp. Paasche, E., 1980. Silicon. In Morris, I. (ed.), The physiological Ecology of Phytoplankton, Oxford: Blackwell, 259-84. 140 Piyasiri. S. 1991. Limnology Project at Mahaweli Reservoirs: 1. Some Physical properties of Kotmale, Victoria and Randenigala Reservoirs. Vidyodaya J. Sci. 3(1): 45-61 Piyasiri. S. 1992. Limnology Project at Mahaweli Reservoirs: 1. Some Physical properties of Kotmale, Victoria and Randenigala Reservoirs. Vidyodaya J. Sci. 4(1): 155-166 Piyasiri, S. 1995. Eutrophication and algae problem in Kotmale Reservoir, Sri Lanka, K.H. Timotius and Goltenboth (eds.) Tropical Limnology Vol.2 Satyas, Waccana University Press, Salatiga, Indonesia Piyasiri, S., 2000. Increasing Eutrophy in Kotmale Reservoir, Sri Lanka; A five years study. Verb. Internal. Verein. Lunnol (in press) Piyasiri, S & W.P.N. Chandrananda, 1998. Studies on the population structure of > zooplankton in the Kotmale reservoir. J. Nam. Sci. Coun. Sri Lanka. 26: 59- 76. Piyasiri, S. & N. Perera, 2001. Role of Oreochromis Hybrids in Controlling Mycrocystis aeruginosa Blooms in Kotmale Reservoir, In: S.S. De Silva (ed). Reservoir and Culture-Based Fisheries: Biology and Management ACIAR Proceedings, Canberra, Australia. 98:137 - 148. Reynolds, C.S., 1984. The ecology of freshwater phytoplankton. Cambridge University press, Cambridge. Robarts, R.D., 1979. Underwater light penetration chlorophyll-a and primary production in a tropical African Lake (Lake Mcllwaine, Rhodesia). Arch. Hydrobiol. 86:423-444. 141 Rhode, W., 1958. Primarproduktion and Seentypen. Verh. int. Ver. Limnol. 13: 121-141. Rott, E., 1983. A contribution to the phytoplankton species composition of Parakrama Samudra, an ancient manmade lake in Sri Lanka, in Schiemer, F. (ed.), Limnology of Parakrama Samudra-Sri Lanka, A case study of an ancient man-made lake in the tropics. Dev.Hydrobiol.12:209-226. Rott, E. & R. Lenzenweger, 1994. Some rare and interesting plankton algae from Sri Lanka. Biologia, Bratislava, 49/4: 479-500 Rott, E., H. Kling & McGregor, (in press). ALM and SEM study of Urosolenia including the description of tropical and temperate new taxa. Diatom Research. * Rott, E., E.I.L.Silva, E.Enriquez & S. Ingmamjitr, (in press). Phytoplankton community-structure (species composition, diversity, chlorophyll, key variables) in five SE Asian water bodies Dev. Hydrobiol. Schiemer F., 1981. Parakrama Samudra (Sri Lanka) project, a study of a tropical lake ecosystem. An interim review, Verh. Internal. Verein. Limnol 21: 987- 993. Schiemer F., 1983 (ed). Limnology Parakrama Samudra Sri Lanka: A case study an ancient man made lake in the tropics. Development in Hydrology, Dr. W. Junk Publishers, The Hague, the Netherlands. Schiemer F. & A. Duncan, 1988. The significance of the ecosystem approach for reservoir management, In: S.S. De Silva (ed), 183-194 pp. Reservoir Fishery Management and Development in Asia. Proceeding Series, International Development Research Center, Ottawa, Canada 142 Schiemer, F. & U.S. Amarasinghe, J. Frouzova, B. Sricharoendhaam and E.I.L. Silva 2001. Ecosystem structure and dynamics - a management basis for Asian reservoirs and Lakes In: S.S. De Silva (ed). Reservoir and Culture- Based Fisheries: Biology and Management ACIAR Proceedings 98: 111 - 125. Sharaff, F.F., 2003. Investigation on the state of pollution and the bacterial consortia of surface water in Kandy Lake. M.Sc. dissertation. Department of Microbiology, University of Kelaniya. 119. Silva, E.I.L., 1988. Irrigation and lotic ecosystems. Ph.D. thesis, University of Ceylon. Silva, E.I.L. 1991. Limnology and fish yield of newly built standing water bodies in Sri Lanka. Verh. Internat Verein, Limnol 24:1425-1429. Silva, E.I.L. 1996. Water Quality of Sri Lanka. A review of twelve water bodies, IFS, Kandy Sri Lanka. 141.pp Silva E.I.L 1999. Status of surface water quality in Sri Lanka. SCOPE 82: 111- 124. Silva, E.I.L., 2003. Emergence of a Microcystis bloom in an urban water body, Kandy Lake, Sri Lanka. In: Current Science., 2003,85, 723 - 725. Silva, E.I.L. & R.W. Davies, 1986. Primary productivity and related parameters in three different types of inland waters in Sri Lanka. Hydrobiologia 137: 239-249. Silva, E.I.L. & R.W. Davies, 1987. The seasonality of monsoonal primary productivity in Sri Lanka. Hydrobiologia 150:165 -175. 143 Silva E.I.L. & G.A.R.K. Gamlath, 2000. Catchment characteristics and water quality of Three Reservoirs (Victoria, Minneriya and Udawalawe) in Sri Lanka. Sri Lanka J. Aquat. Sci. 5:55 - 73. Silva E.I.L. & F. Schiemer 2000. Hydraulic Changes of three Reservoirs (Victoria, Minneriya, and Udawalawe) in Sri Lanka. Sri Lanka J. Aquat Sci. 5: 75-86. Silva E.I.L. & F. Schiemer, 2001. Human Factor: Fourth Dimension of Limnology in the Tropics. In De Silva, S.S. (ed,), Reservoir and Culture based Fisheries: Biology and Management. ACIAR Publication No 98, Canberra, Australia: 111-125. Silva E.I.L & M.J.S. Wijeyaratne 199^. The Occurrence of Cyanobacteria in the Reservoirs of the Mahaweli River Basin in Sri Lanka. Sri Lanka J. Aquat V Sci. 4:51-60 Silva, E.I.L., U.S. Amarasinghe, S.S. De Silva, C. Nissanka & F. Schiemer 2002. Some aspects of photosynthetic characteristics in a set of perennial irrigation reservoirs located in five river basins in Sri Lanka. Hydrobiologia 485:19-33 Silva, E.I.L., M. Ekanayaka & K.M.B.C. Karunathilaka, 2003. Seasonal abundance of species of rotifers (Brachionus calyciflorus and Keratella tropica) in a tropical urban water body, Kandy Lake in Sri Lanka. J. Aquat Sci., (in press) State of the Environment, Sri Lanka, 2001. United Nation Environment Programme, Regional Resource Center for Asia and Pacific (UNEP-RC.AP), ATT, Thailand 106 p. 144 Tailing, J.F., 1965. The photosynthetic activity of phytoplankton in East African lakes. Int. Rev. ges. Hydrobio. 50:1-32. Tailing, J.F. & J. Lemoalle, 1998. Ecological dynamics of Tropical Inland Waters. Cambridge University Press: 441pp. TAMS, 1980. Environmental Assessment: Accelerated Mahaweli Development Program, vol HI Aquatic Environment H: 2-8. Thornton, J. F. (ed.), 1990. Lake Mcllawaine, The eutophication and the recovery of a tropical African man-made lake. Mongr. Biologicae 49. The Hague: Junk. U.S. Operation Mission, 1961. Mahaweli Ganga Trans-basin Diversion Scheme. Vol. Ill Reconnaissance Land Classification and Soil Survey Report Irrigation Department Colombo, van der Leeden, F., F.L.Troise, D.K. Todd, (eds), 1990. The Water Encyclopedia. 2nd edition, Lewis Publishers. USA. 60p Vincent, W.F., 1979. Mechanisms of rapid photosynthetic adaptation in natural phytoplankton communities. J. Phy-col. 15:429-433. Vitousek, P.M., 1977. The regulation of element concentration in mountain streams in the northeastern United States, Ecological Monograph, 47: 65-87. Ward, A.K. & R.G. Wetzel, 1980. Photosynthetic response of blue-green algal populations to variable light intensities. Arch. Hydrobiol. 90:129-138. Werner, D., 1977. Silica metabolism. In Werner, D. (ed.), The Biology of Diatoms, Oxford: Blackwell: 110-149. Webb, B.W. & D.E. Walling, 1974. Local variation in a background water quality, Science of the total environment. 3:141-53. 145 West, W. & G.S. West, 1902. A contribution to the fresh water algae of Ceylon. Trans. Linn. Soc. 2 Ser. Bot. 6:123 - 215. Wetzel, R.G., 1983. Limnology (Second Edition). Saunders college publishing, Philadelphia. 767p. Wissamar, R.C., J.E. Richie, R.F. Stallard & J.M. Edmond, 1981. Plankton Metabolism and carbon processes in the Amazon river, its tributaries and floodplain waters, Peru-Brazil, May-June 1977, Ecology 62:1622-1633. Zevenboom, W. & L.R. Mur, 1981. Simultaneous short-term uptake of nitrate and ammonium by Oscillatoria agardhii grown in nitrate- or light-limited continuous culture. J. gen. Microbiol. 126:355-63. 146