Aquatic Ecology 36: 331–340, 2002. © 2002 Kluwer Academic Publishers. Printed in the Netherlands.

331

Epilogue

Understanding the mechanisms of blooming of phytoplankton in Lake Shira, a saline lake in Siberia (the Republic of Khakasia) Andrei G. Degermendzhy1 and Ramesh D. Gulati2 1 Institute

of Biophysics SB RAS, Akademgorodok, 660036 Krasnoyarsk, Russia ([email protected])

2 Netherlands Institute of Ecology, Center for Limnology, Rijksstraatweg 6, 3631 AC Nieuwersluis, The Netherlands

Key words: meromictic lakes, stratification, mathematical models of stratification, trophic scheme, carbon budget, hydrogen sulphide, heterotrophic bacteria, microbial loop, cyanobacteria

Abstract The paper summarises the results of a three-year research study (European Union Grant: INTAS 97-O519) aimed at investigating the planktonic populations and trophic organization of the Lake Shira ecosystem – a saline lake in Khakasia, Siberia. The lake exhibits a stable summer-autumn stratification of the chemical-biological components. The mechanisms responsible for the ‘blooming’ of phytoplankton in the deeper layers were investigated in greater detail, using data from both field and laboratory experiments. The spectra of nutrition were examined to estimate the relationships between the specific growth rates of the hydrobionts and the influence of the limiting factors: light, nutrients. The observed heterotrophic capability of a metalimnetic phytoplankton population might help explain the development in the deeper waters of Lyngbya contorta. The scheme of trophic interactions was put up, based on the assessment of the carbon pools and carbon flows in the pelagic zone of the lake. A mathematical model of the vertical structure of the lake’s plankton populations was constructed, using the ecosystem description and data of vertical turbulent diffusion. The role of light and nutrient limitations and grazing mortality in forming the vertical inhomogeneities, particularly in lowering the depth of the maximal cyanobacterial biomass, has been demonstrated. The theoretical curves for the stratification of chemical and biological parameters have been brought in conformity with the field observations, e.g. for the different patterns of the peaks, and for the biomass maxima of cyanobacteria, purple and green sulphur bacteria, oxygen, and hydrogen sulphide. The calculations revealed that for an adequate assessment of the parameters for the hydrogen sulphide zone it is necessary to introduce flows of allochthonous organic matter. Based on the form of the sulphur distribution curve, the allochthonous input of organic matter and the inflow of hydrogen sulphide from the bottom have been theoretically discriminated for the first time. It has also been ascertained that irrespective of the depth the allochthonous substances limiting bacterial growth, the bacteria are uniformly distributed over depth and can serve as an indicator of the presence of limitation (the effect of autostabilisation in space). Of indisputable interest to limnology are the specific methods developed for understanding the functioning of Lake Shira ecosystem. These include the autostabilisation of the limiting factors, the on-the-spot fluorescent method of determining the three classes of microalgae, the algal mixotrophy and the planktonic population interactions and feedbacks, and development of a more sensitive, bioluminescent method for mapping the nonhomogeneities. Owing to a balanced combination of classical approaches (field observations, in situ data on production-decomposition) and the more recent ones (satellite monitoring, biophysical methods of estimating interactions of populations, mathematical models based on the field and experimental data), many of the structuralfunction relationships in the ecosystem can now be explained, and the models can provide ‘mutual control and mutual agreement’ between the data collected using different approaches.

332 Introduction The task of analysing the consequences of human impact on aquatic ecosystems is a complicated one, mainly because of the complexity of the structure and functioning of most of these ecosystems. It is perhaps relatively easier to work on systems with both a limited number of trophic levels and simple food-web. One of the current priorities of the research on these systems should be to gain insights into the underlying mechanisms of their sustainable existence. Lake Shira, a saline lake, in the Republic of Khakasia, is one such aquatic ecosystem among the saline lakes lying in the heart of the vast arid and dry region of Siberia. As in the other parts of the world, several researchers in Siberia have focused their attention on inland saline lakes of this region, but specially on Lake Shira. The studies on Lake Shira are based on an ecosystem approach with a clear emphasis on the methodological problems. The studies are aimed at integrating and co-ordinating the experimental and field investigations of the lake ecosystems’ components, including their spatial-temporal dynamics. The investigations in progress have also included designing and executing specific experiments with the objective of unravelling the spectra of nutrition and determining the relationships between the hydrobionts, their specific growth rates in relation to the limiting factors, etc. A very important aspect of these investigations, in addition to getting an insight into the structure of the lake ecosystem, was to verify a mathematical model using the observational data and the specific experiments. The adequacy of the model will in turn serve to work out the scenarios for an effective control of the impending environmental threats to the water body’s state.

Geographical distribution of salt lakes In the foregoing papers in the Special Issue of Aquatic Ecology (Gulati & Degermendzhy, 2002) on the different study aspects of Lake Shira, the Khakasian lakes are introduced with a review paper by Parnachev & Degermendzhy (2002). This paper describes in detail the geographic distribution of the salt lakes in the Khakasian Republic, giving geological conditions of the lake formation. The mineral content of these lakes varies from 2 to 150–200 g l−1 . The major cations (Mg, Na and Ca) and the anions (SO4 and Cl) predominate the amount of dissolved inorganic matter (DIM). The variations of the DIM in the lakes are more

marked during the dry periods. Most such saline lakes are associated with synclinal structures and terrigenous red-coloured, Upper Devonian deposits, with the presence of gypsum layers. Before the project INTAS97-O519 was started in 1999, the literature contained scarcely any data on the chemical composition of the Lake Shira water and on some other mineral lakes in the Republic of Khakasia (Malakhov et al., 1963; Kuskovsky & Krivosheev, 1989). These data were restricted mostly to the contents of the main components in the water and in the sediments. In other words, there were a few if any data on the chemical and mineral composition and contents of the rocks encompassing the lakes and influencing the composition of the ground and surface waters that flow into the lake basins. These investigations have contributed to our knowledge of the water chemistry of these lakes, including the contents of rare and dispersed elements (Banks et al., 2001). In addition, the chemical and biogeochemical stratifications of the water have been monitored in detail by Karnachuk & Parnachov (2001). These authors have schematised and estimated the intensity of the inputs of mineral and organic substances into the lake, with the both temporary and permanent inflows into the lake, including both surface- the groundwater inflows, and atmospheric precipitation.

Stratification and the accompanying phenomena The uniqueness of Lake Shira lies in the combination of a number of important physico-chemical factors: a moderate mineral content of the water and the dominance of sulphates ions, keeping in mind that the sulphate lakes usually have very high mineral content (80 g l−1 ) and are shallow. The lake is stratified with regard to both temperature and oxygen and chemical parameters. The anoxygenic zone close-to-bottom layers contains large amounts of sulphides, phosphates and ammonium. Other chemical elements are uniformly distributed over the lake area (Kalacheva et al., 2002). The oxygen maximum is recorded in the 6-8 m stratum, coinciding with that of the photosynthetic activity. The generalised stratification pattern for the basic parameters of the Shira Lake ecosystem is shown in Figure 1. The lake water can be characterised as ‘sulphate-chloride-bicarbonate’ and ‘sodium-magnesium’. The chemical stratification remained stable throughout the summer-autumn field season and the lake can

333

Figure 1. The vertical profiles of the main physicochemical and biological parameters of the Shira Lake ecosystem during summer stratification (15 July, 2001). Photosynthetic active radiation (PAR) measured using an underwater quantum sensor LI-192SA (LiCOR, USA); water temperature (t), oxygen concentration (O2 ) and redox-potential (Eh) were measured with a Data Sonde 4 (Hydrolab Corp., USA); hydrogen sulphide (H2 S) was determined according to (Standard methods..., 1989) the chlorophyll a content (Chl a) was determined on a double beam UV/VIS spectrophotometer UVICOM 943 (Kontron Inst., Italy) after pigments extraction according to Nusch (1980); wet zooplankton biomass (Zoo B) was calculated by converting zooplankton population density into biomass; luciferase index (LI).

be assumed as a meromictic lake. This assumption needs to be verified by observations at regular intervals throughout the year. The objectives of the future investigations should be: to monitor the composition of the water entering Lake Shira, to investigate the sediments, to characterise the biochemical reactions occurring there, and to get a picture of the history of the lake’s formation. An essential feature of Lake Shira is the densitydependent stratification that depends not only on the temperature (Figure 1) but importantly also on the salinity. The vertical distribution of the temperature in the deep-water zone was deduced using a onedimensional model (Belolipetsky et al., 2002). The water temperature follows the seasonal patterns of air temperature, as influenced by the wind direction and wind force. An outflow being absent, the dynamic processes in the lake occur under the influence of the wind stress. The pattern of wind currents is determined by the geometry of the water body, the direction and strength of the wind, the bathymetry, and the density stratification. Based on the vertically two-dimensional mathematical models, it has been shown that as in the significantly stratified water bodies, in Lake Shira also only the upper layer is brought into movement (Belolipetskii, 2001). The model calculations reveal that for the stratified lakes two types of currents are

possible. The first type of current is due to a weak stratification (or a rather strong wind), which is similar to the flow of homogeneous fluid, when the fluid as a whole is moved by the wind. In bottom layers the compensatory current arises that is directed opposite to the direction of the wind (one-circulation flow, Figure 2, A). The second type of the current is due to the movement of the upper, warmer layers (twocirculation flow, Figure 2, B). Even the small relative variations in the density, of the order of 10−3 , have been demonstrated to lead to a change in the pattern of the current. The satellite remote sensing, a support to simplify the models The satellite remote sensing of the lake’s surface (Shevyrnogov et al., 2002) yields instantaneous information on the possible inter-seasonal nonhomogeneity of the water, which can assist in simplifying the mathematical models. Time series of images of radiation temperature and turbidity based on AVHRR satellite sensor data have been obtained. Morphometric, meteorological and hydrological characteristics of Lake Shira, which were analysed, provided a realistic opportunity to apply the data in prognostic complex for satellite information processing and to develop numerical models of variability of the hydro-

334

Figure 2. The patterns of currents (a) (isolines of the stream function coincide with the trajectories of the movement of the fluid particles) and vertical profiles of temperature at the beginning of the calculation (b): A – one-circulation flow; B – two-circulation flow

logical regime of the lake. The satellite data will be used to solve the important hydrodynamic problem to understand how the non-homogeneity of the surface temperature caused by the wind upwelling of the cold deep water can be recorded and used to calibrate the model of currents.

The mechanisms underlying the lake’s ecosystem development The investigation of the mechanisms responsible for development of the lake ecosystem’s components was the main goal of this interdisciplinary study. The interactions of the components are schematically presented (Figure 3) following Rodrigo et al. (2001). The pelagic zone of the Shira Lake ecosystem exhibits low plankton diversity. The main structural components of the ecosystem are: bacteria, algae (48 spp.), protozoans (43 spp.) and crustaceans (4 spp.).

Phytoplankton dynamics, distribution during summer stratification The phytoplankton in Lake Shira is stratified from late June to September, with the maximum in the lower part of the thermocline at 8-12 m. In these strata the chlorophyll concentration reaches up to 23 µg l−1 and biomass up to 5 mg FW l−1 . Maxima of chlorophyll and biomass of cyanobacteria and green algae are vertically segregated (Figure 3). During the summer period (June to September), a major part of chlorophyll a is present in green algae, but in the cyanobacteria during the ice period. The variable fluorescence proves high photosynthetic activity of algal assemblages in the deeper layers. Epifluorescent analysis disclosed that additional light-harvesting pigments were better developed in cells from the deepest layers. The photoassimilation maximum of the inorganic (14 C) corresponds to the depth maximum of phytoplankton (Lyngbya contorta). At the beginning of summer a major part of phytoplankton in Lake Shira occurs in the epilimnion, at the upper boundary of the photic layer, but at the end of summer at the lower boundary of the photic layer and hy-

335

Figure 3. The diagrammatic representation of the interaction of structural (chemical-biological) components of the Shira Lake ecosystem together with the vertical division of the pelagic zone into the different zones based on stratification, oxygen, etc. The question marks (?) show the uninvestigated processes i.e. gaps in our knowledge of the lake’s ecosystem.

336 polimnion. The former, probably, formed mainly by the growth of green algae and the latter by the growth of cyanobacteria. Thus, the metabolic pathways determining the ecosystem functioning appear to change from algae-based to bacteria-based. The cause of the ‘switch’ could be the onset of nutrient limitation of algae in the upper, well-illuminated layer. Outbreaks of Botryococcus braunii and Cyclotella tuberculata also trigger the onset of nutrient limitation in the epilimnion. So, because of stratification the epilimnion community of phytoplankton is only indirectly related with the metalimnion (Gaevsky et al., 2002). These data on stratification of phytoplankton populations and the primary production have been used to construct a model (Degermendzhy et al., 2002) for verification of the hypotheses explaining the presence of the depth maximum. The model has confirmed the probability of the depth maximum formation, but so far it cannot discriminate between the different mechanisms (light, nutrients) responsible for the formation of depth maximum. Thus, this is the task for the future investigations. The pronounced water stratification and the deepwater maxima of autotrophic organisms cause an intensive development of heterotrophic bacteria and flagellates at the lower boundary of the photic layer in mixolimnion and chemocline. As a rule, the maximal rates of primary production and bacterial production are found in the chemocline or at the upper boundary of the chemocline. Heterotrophic flagellates at the upper boundary of the chemocline consume a significant part of the bacterial production. The interface between the aerobic mixolimnion and anaerobic monimolimnion is the site of intense biomass production and bacterial mineralisation, which deserves further studies. The data on the vertical distribution of some sulphur-cycle bacteria were used to verify adequacy to a mathematical model of the sulphate-reduction process. The metalimnetic plankton populations of Lake Shira possess heterotrophic capability (Quesada et al., 2002). This was shown for the community dominated by cyanobacteria of several taxa from the deeper layers (12 m). The experiments consisted of measurements of the in situ labeling and utilization of 13 C labelled organic compounds by the community and separation of the bacterioplankton and phytoplankton, based on the size fractionation of the community. The 13 C uptake rates were high, being relatively higher in the bacterioplankton fraction than in the phytoplankton. In addition, the uptake of the organic compounds was

inversely related to the light intensity, using different irradiance regimes. Thus, the photosynthetic organisms can use the organic compounds directly. The advantage to phytoplankton is that at the depth of their usual occurrence in the metalimnion, where the photosynthetic activity is generally limited by light during a considerable part of the day, they can use the available dissolved organic compounds. Hence, another task for the future is to include in the model descriptions of the phytoplankton their heterotrophic growth potential, specification of the limiting factors, the switch to photosynthesis and organic mode of assimilation, etc.

The trophic links & the carbon budget The ecosystem of Lake Shira has a reduced number of trophic links: there are no cladocerans, no cyclopoid copepods, no predatory zooplankton and no macroinvertebrates except Gammarus lacustris that inhabits the littoral zone; and there is a complete absence of ichthyofauna. We used the carbon budget as a tool to examine interactions between the different components of pelagic food web of Lake Shira. The data of carbon pools and fluxes were integrated over time for constructing the carbon budget, using a set of assumptions: 1) the growth yield of all the heterotrophic organisms, except bacteria, was set to 30% (Fenchel, 1988); 2) tentative values of the excretion and egestion rates (sloppy feeding, faeces) of 35% of the ingested carbon were used to calculate protozoa and zooplankton contribution to bacterial substrates (Vadstein, 1989); and 3) the uptake of extracellular organic carbon constituted 30% of primary production (Boulion, 1983). In meromictic Lake Shira, the oxygenic photosynthesis is a more important source of the organic-matter production than the anoxygenic photosynthesis. Furthermore, high rates of dark CO2 assimilation in the chemocline appear to be caused by chemoautotrophy, which represents an important source of organic matter. The estimated total primary productivity is too low to sustain all the bacterial production, implying that the carbon cycle of the lake depends on the inputs of allochthonous material. The microbial loop Our present knowledge of the microbial loop (Figure 4) in Lake Shira is based on the work of Kopylov et al. (2002). Heterotrophic bacterioplankton is the main

337

Figure 4. Carbon flux in the oxic and oxic-anoxic zones (0–14 m) in Lake Shira during July and August (1999–2001). Boxes represent biomass (mgC m−2 ), or net production (mgC m−2 d−1 ). Solid lines and black arrows represent uptake rates (mgC m−2 d−1 ). Question marks indicate hypothetical pathways.

constituent of the planktonic microbial community. At some depths, the numbers of bacteria associated with detrital particles are considerable, as well as there are those that form micro colonies. Both could serve as a source of nutrition for the zooplankton. Heterotrophic flagellates are dominant among the protozoa. Ciliates are a minor component of the planktonic microbial community. Grazing of protozoa was derived from the in situ uptake rates of the fluorescently-labelled bacteria. Heterotrophic flagellates graze about 50% of net bacterial production, heliozoans 1.6% and ciliates 0.01% of it. Within the microbial loop a rather linear food chain consisting of bacteria and heterotrophic flagellates appears to be a potential, major pathway transferring bacterial production to higher levels of the planktonic foodweb (Figure 4). Thus, the question that

future research should be able to answer is: is there a significant material-energy flow from heterotrophic flagellates to large zooplankton? If the answer is yes, the microbial loop is an important player in the energy source for the higher trophy levels. If not, the microbial loop can be considered as a sink, ‘removing’ carbon and energy out of the planktonic system. Cyanobacterial remains are invariably encountered in the digestive vacuoles of the heterotrophic flagellates. From the latter’s clearance rate values, we derived that they graze down roughly 12% of primary productivity. Evidently the functional role of deep-water accumulations of heterotrophic flagellates in the chemocline and in the adjacent strata is to use the energy of bacterial photosynthesis and chemosynthesis and to interact with the communities of the oxic and H2 S zones.

338 The zooplankton and the pelagic foodweb The pelagic zooplankton of Lake Shira was comprised of Brachionus plicatilis and Hexarthra oxiuris (Rotatoria), and Arctodiaptomus salinus (calanoid copepoda). Arctodiaptomus was dominant in terms of its estimated, annual, average biomass of 90% of the total zooplankton (Figure 4). In July-August, however, when the rotifers made up 44% of the total biomass, the rotifers seemed to play a significant role in the lake’s planktonic community. A flow-through technique was employed to study the algal diet spectrum of the calanoid Arctodiaptomus in the surface layer of the lake and in the phytoplankton biomass maximum zone at 10-m depth. Daily consumption rates of this calanoid differed by an order of magnitude in these layers, being 1 and 11µg ind.−1 d−1 , respectively. The diet spectrum of Acrtodiaptomus included the algae Dictyosphaerium tetrachotomum, Oocystis submarina, Microcystis sp. and Lyngbya contorta. The cyanobacterial taxa, Lyngbya contorta and Microcystis sp., accounted for > 50% of the daily ration of the calanoid. Nevertheless, the negative Ivlev electivity coefficients indicate that Arctodiaptomus prefers none of these two taxa. The low measured uptakes rates of food suggest that the energy expenditure is compensated by other hitherto unknown resources, presumably microzooplankton and detritus. The overall consumption rates of the biomass of blue-green algae by Arctodiaptomus varied between 60 and 600 mg C m−2 d−1 , equivalent to 6–61% of the primary production (Tolomeyev, 2002). In Lake Shira, the amphipod Gammarus lacustris mainly inhabits the littoral, sublittoral and upper aphytal zones extending up to 13-m depth of the lake on stony-sandy soil and silted sand. The young and adult animals are segregated in their habitats: the former inhabit the submerged or semi-submerged vegetation in the littoral, and the latter the sublittoral and upper aphytal. The growth studies in the laboratory on Gammarus indicate that the heterogeneous assemblage of lake plankton is a more important food source for the amphipod than a selected algal diet. The average specific growth rate under optimal conditions amounted to 0.039 ± 0.007 d−1 , corresponding to a length increment of 0.095 mm d−1 . Gammarus can consume the dominating zooplankters, but during summer the animals inhabiting the littoral zone mostly feed on the freshly depositing seston. The cells of Botryococcus braunii were observed to be egested intact by the amphipods, the egested matter showing an increase in

their potential photosynthetic activity (Gladyshev et al., 2000). The phytoplankton daily consumed by the amphipods amounted to c. 0.5% only of the primary production. Thus, Gammarus is not an active consumer of allochthonous organics, but belongs to the group of sestonophages (Yemelyanova et al., 2002). Amphipods are known to transport biological material between coastal and central parts of the lake (Wilhelm et al., 1999). We need to pay more attention to this last aspect in our Gammarus research in Lake Shira. The chemical interactions manifested by the migratory activity of the aquatic populations in Lake Shira have not been sufficiently studied yet. The modelling of these interactions will give a new impetus to the development of mathematical models (Zadereev & Gubanov, 2002). Arctodiaptomus, when transferred to the water previously inhabited by Gammarus, wherein the Gammarus had apparently released some chemicals, exhibited changes in its vertical distribution and positioning. The average population depth of Arctodiaptomus decreased as the inoculation density of Gammarus increased. In addition, with food concentration in the experimental medium corresponding to the maximum of algal concentration in Lake Shira, this calanoid tended to situate itself higher in the experimental vessels than in the controls without food. Future research should substantiate the anecdotal data on possible chemical communication within the Lake Shira zooplankton community. The pelagic zooplankton (rotifers and Arctodiaptomus) of Lake Shira significantly suppressed the phytoplankton biomass, consuming up to 60% of its daily production. The dominance in the pelagic zooplankton by the copepod Arctodiaptomus salinus, a ‘coarse’ filterer that cannot capture and ingest individual bacteria and heterotrophic flagellates, would imply that in Lake Shira the microbial ‘loop’ does not operate in direct conjunction with the foodweb formed by the pelagic zooplankters. That the microbial loop is not clearly regulated by the top-down factors is further corroborated by the absence in the lake of cladoceran zooplankton, e.g. Daphnia spp. which, as ‘fine’ filters feeders, can consume the bacteria and flagellates (protozoa). In the absence of fish, however, the causes for the simplicity of zooplankton diversity, including the absence of major elements of the pelagic zooplankton, such as daphnids, could be attributed to phenomena related to meromixis and to salinity.

339 The planktonic interactions The scheme of trophic interactions (Figure 4) is a preliminary attempt to understand the interactions in the planktonic community of Lake Shira. We hope that the future investigations will concentrate on the gaps in our knowledge, indicated by several question marks in the energy flow scheme, as well as confirm the preliminary data that allowed us to produce this energy-flow chart.

The anoxic zone and modelling of biological stratification In the anoxic zone, biomass of heterotrophic bacteria accounted for 93% of the total microbial biomass, pigmented microorganisms 6.1%, heterotrophic flagellates 1.5% and fungi 0.1%. High concentration of anaerobic heterotrophic flagellates in the monimolimnion is the distinctive feature of Lake Shira. The predominant flagellates are members of the genera Hexamita and Trepomonas. The estimation of the trophic interactions between the various groups of anaerobic bacteria and anaerobic flagellates can be the basis for a more adequate evaluation of the functioning of anaerobic communities. The model The mathematical model (Degermendzhy et al., 2002) demonstrates that the stratification of phytoplankton and other components of the lake ecosystem (e.g. the sulphur cycle) are based on the field and experimental data. The latter comprise a new means of evaluating the interaction coefficients (IC) in microbial communities interacting through the physicalchemical environmental factors. This method differs from the classical one suggested by E.P. Odum and is based on estimating the acceleration of the population growth in response to the disturbance of the size of this population or of a different one. Redefining the IC allows evaluating the experimental (actual) values and theoretical values of the coefficients for the hypothetical interaction layout. A comparison of theoretical and experimental values of IC enables us to assess the adequacy of the hypothetical interaction scheme in the microbial communities. The dominance of the negative experimental IC values, which has been experimentally shown, is indicative of the negative feedback in bacterio- and phytoplankton links of

Lake Shira and of the negative inter-population (phytoplankton on bacterioplankton) interactions. Further investigations are needed to provide a precise outline of the interactions (Adamovich et al., 2002). The proposed new experimental method to determine the IC and its theoretical interpretation can be indispensable for constructing the schemes of the growth regulation for natural planktonic communities, if neither the substances that influence the populations nor the degrees of their limitation or inhibition are known. The basic model describes the interaction mechanisms involving: 1) the plankton populations in aerobic and anaerobic zones, involving the cycling of carbon and sulphur; 2) the primary production, limited by light and phosphorus and inhibited by light; and 3) the elucidation of the kinetic characteristics of plankton populations. The role of light and nutrient limitation and grazing mortality, in forming the vertical heterogeneities, particularly in lowering the depth of the maximal cyanobacterial biomass, has been demonstrated. The theoretical curves for the stratification of chemical and biological parameters have been brought in conformity with field observations, e.g. for the different patterns for the peaks, and the biomass maxima of cyanobacteria, purple and green sulphur bacteria, oxygen, and hydrogen sulphide. The calculations revealed that for an adequate assessment of the parameters for the hydrogen sulphide zone it is necessary to introduce flows of allochthonous organic matter. Based on the form of the sulphur distribution curve, the allochthonous input of organic matter and the inflow of hydrogen sulphide from the bottom have been discriminated theoretically for the first time (Degermendzhy et al., 2002). Also, it has now been theoretically ascertained that irrespective of the depth where the allochthonous substances limit bacterial growth, they are uniformly distributed over depth and can serve as an indicator for the presence of limitation (the effect of autostabilisation in space). Thus, the task is to find such organic or nutritive substances, or both. The investigations of the Lake Shira ecosystem are far from having been completed. The stratification of the lake’s components seems to be of a finer structure and this can be seen in the luciferase index variations (Figure 1, LI) demonstrating a fundamentally different vertical distribution (Vetrova et al., 2002) compared with the Eh values. Thus, a target sampling can be performed more precisely, i.e. including layers with the extremes of the index variations, by using ‘horizontal’ samplers. The production-decomposition ratios still

340 need to be determined to complement the list of the producers, to determine many kinetic characteristics essential for mathematical models and to make an inventory of all the allochthonous inputs to the lake. On the other hand, it is safe to assume that the stable stratification of the lake plankton will serve as a basis for the construction of new models. This model will provide the means to control the stratification. Thus, if we have the model and know the mechanisms responsible for the stratification of the lake ecosystem’s components, we will be able to identify the ‘switches’ i.e., inputs, limiting factors and more that will allow us to regulate the stratification characteristics, viz. the forms of the vertical distribution curves, the depth location of the maxima, etc. Lastly, of indisputable interest to limnology are the specific methods developed for understanding the functioning of Lake Shira ecosystem: the autostabilisation of the limiting factors, the on-the-spot fluorescent method of determining the three classes of microalgae, the determination of algal mixotrophy, the estimation of the planktonic population interactions and feedbacks and to develop a more sensitive, bioluminescent method for mapping the nonhomogeneities, etc.

References Adamovich VV, Zotina TA and Degermendzhy AG (2002) Interaction coefficients in the Lake Shira algal-bacterial community. Aquat Ecol 36: 261–270 Banks D, Parnachev VP, Frengstad B, Holden W, Karnachuk OV and Vedernikov AA (2001) The hydrogeochemistry of the Altaiskii, Askizskii, Beiskii, Bogradskii, Shirinskii, Tashtipskii and Ust’ Abakanskii regions, Republic of Khakassia, Southern Siberia, Russian Federation. Norges geologiske undersokelse rapport 2001.006, Trondheim. Belolipetskii VM (2001) Numerical modelling of wind currents in stratified reservoirs. Water Res 2: 133–137 Belolipetskii VM, Genova SN, Gavrilova LV and Kompaniets LA (2002) Mathematical models and computer programmes for the investigation of hydrophysical processes in Lake Shira. Aquat Ecol 36: 143–152 Boulion VV (1983) Pervichnaya produktsiya planktona vnutrennikh vodoyomov (Primary production of the plankton in inland water bodies). Leningrad:Nauka -130 pp Degermendzhy AG, Belolipetsky VM, Zotina TA and Gulati RD (2002) Formation of the vertical heterogeneity in the Lake Shira ecosystem: the biological mechanisms and mathematical model. Aquat Ecol 36: 271–297 Gaevsky NA, Zotina TA and Gorbaneva TB (2002) Vertical structure and photosynthetic activity of Lake Shira phytoplankton. Aquat Ecol 36: 165–178 Gladyshev MI, Emelianova AY, Kalachova GS, Zotina TA, Gaevsky NA and Zhilenkov MD (2000) Gut content analysis of Gam-

marus lacustris from a Siberian lake using biochemical and biophysical methods. Hydrobiologia 431:155–163 Fenchel T (1988) Marine plankton food chains. Ann Res Ecol Syst 19: 19–38 Kalacheva GS, Gubanov VG, Gribovskaya IV, Gladchenko IA, Zinenko GK and Savitsky SV (2002) Chemical analysis of Lake Shira water (1997-2000). Aquat Ecol 36: 123–141 Karnachuk OV and Parnachov VP (2001) Biogeochemical characteristic of the water column of Lake Shira. In: Problems of geography in Siberia. Issue 24, Tomsk , 172–182 Kopylov AI, Kosolapov DB, Degermendzhy NN, Zotina TA and Romanenko AV (2002) Phytoplankton and bacterial production and protozoan bacterivory in a stratified, brackish-water Lake Shira (Khakasia, Siberia). Aquat Ecol 36: 205–217 Kopylov AI, Kosolapov DB, Romanenko AV and Degermendzhy AG (2002) Structure of planktonic microbial food web in a brackish stratified Siberian lake. Aquat Ecol 36: 179–204 Kuskovsky VS, Krivosheev AS (1989) Mineral lakes of Siberia. Nauka, Novosibirsk Malachov FM, Skornyakov VA and Zizarin GV (1963) Hydromineral resources of the health resort ‘Lake Shira’. In: Materials of investigations of medical muds, mud lakes and deposits, Moscow, 51–151 Nusch EA (1980) Comparison of different methods for chlorophyll and phaeopigment determination. Arch. Hydrobiol. Beih. Ergeben. Limnol 14: 14–36. Parnachev VP and Degermendzhy AG (2002) Geographical, geological and hydrochemical distribution of saline lakes in Khakasia, Southern Siberia. Aquat Ecol 36: 107–122 Quesada A, Jüttner F, Zotina T, Tolomeyev AP and Degermendzhy AG (2002) Heterotrophic capability of a metalimnetic plankton population in saline Lake Shira (Siberia, Khakasia). Aquat Ecol 36: 219–227 Rodrigo MA, Miracle MR and Vicente E (2001) The meromictic Lake La Cruz (Central Spain). Patterns of stratification. Aquatic Sciences 63: 406–416 Standard Methods for the examination of water and wastewater (1989) Ed. By Lenore S. Clesceri, Arnold E. Greenberg, R. Rhodes Trussels. Washington DC 20005. Shevyrnogov AP, Kartushinsky AV and Vysotskaya GS (2002) Application of satellite data for investigation of dynamic processes in inland water bodies. Aquat Ecol 36: 153–163 Tolomeyev AP (2002) Phytoplankton diet of Arctodiaptomus salinus (Copepoda, Calanoida) in Lake Shira (Khakasia). Aquat Ecol 36: 229–234 Vadstein O, Harkjerr BO, Jensen A, Olsen Y and Reinertsen H (1989) Cycling of organic carbon in the photic zone of eutrophic lake with special reference to the heterotrophic bacteria. Limnol Oceanogr 34: 840–855. Vetrova EV, Kratasyuk VA and Kudryasheva NN (2002) Bioluminescence characteristics of Lake Shira water. Aquat Ecol 36: 309–315 Wilhelm FM, Hudson JJ and Schindler W (1999) Contribution of Gammarus lacustris to phosphorus recycling in fishless alpine lake. Can J Fish Aquat Sci 56: 1679–1686 Yemelyanova AYu, Temerova TA and Degermendzhy AG (2002) Distribution of Gammarus lacustris Sars (Amphipoda, Gammaridae) in Lake Shira (Khakasia, Siberia) and laboratory study of its growth characteristics. Aquat Ecol 36: 245–256 Zadereev YS and Gubanov MV (2002) The effect of chemicals released by Gammarus lacustris on the depth distribution of Arctodiaptomus salinus in laboratory conditions. Aquat Ecol 36: 257–260