LAKE UNDER RESTORATION PROCESS
N. Mazzeo, F. Scasso, J. Gorga, C. Kruk and J. Clemente
INTRODUCTION
Lake Rodó (34º55’S, 56º10’W) is a small (1.3 ha), shallow (maximum depth 2m) and hypertrophic system. This man-made lake, constructed in 1917 is located in a park in Montevideo (1.5 million inhabitants) Uruguay. Since spring 1996, this urban lake has been under restoration to establish a clear water phase. The lake was completely drained and sediments were removed. Groundwater was used to re-fill the lake and as the water source (480 m3 d-1) and it showed a high nutrient concentrations. Moreover, planktivorous fish were removed in winter with a seine net, in an attempt to reduce their abundance.
Our objectives were to analyse the physicochemical and biological changes of a shallow hypertrophic lake under restoration, and to establish the principal factors affecting the water transparency. We also propose more suitable measures for establishment of a clear water phase.
MATERIAL AND METHODS
Water samples were weekly collected from 01/97 to 04/97 and biweekly up to 05/98 from four stations (Fig. 1). Temperature, dissolved oxygen, conductivity, pH and transparency were registered in situ. These parameters except transparency were also registered in situ in associatedpools and the output of groundwater pumping. Water samples for chemical analysis were taken and include alkalinity, suspended solids, soluble reactive phosphorous (SRP), nitrate, ammonium, total nitrogen (TN), total phosphorous (TP), silicate chlorophyll a using hot ethanol (Nusch, 1980). At the lake sediment samples were taken every two months, with a 6 cm diameter Kajak-Corer. The content of organic matter was determined by gravimetry (Häkanson & Jansson 1983). Sediment total nitrogen and total phosphorus were analysed according to Valderrama (1981).
Phytoplankton net tons (25 mm mesh) where taken and preserved (5 % neutralised formalin). Integrated triplicate samples were collected with a 5-l Schindler trap, preserved (acid Lugol) and counted (1-ml Sedwick-Rafter chamber). Zooplankton was collected with a 68 mm mesh sized net and preserved (4% neutralised formalin). Integrated triplicate samples were taken with a 5-l Schindler trap and counted using a Sedwick-Rafter chamber of 1 to 5 ml. Benthic communities were sampled (ten replicates) every two months, with a 6 cm diameter Kajak-corer, sieved through a 500 m m mesh and stored in 95% ethanol. Fish were collected bimonthly using a 20 x 2 m seine net (1 mm mesh size), preserved (10% neutralised formalin). Individuals were counted and weighed. Planktivorous fish were removed in winter with a seine net, in an attempt to reduce their abundance.
RESULTS
Lake Rodó is warm continuously polymictic (according to Lewis, 1983) and turbid system, condition attributed principally to algal biomass. A strong decrease in water transparency was observed during the first three months after the re-filling, and the highest water transparency (1.4 m Secchi depth) was registered in spring (Fig. 2).
The lake was characterised by oxygen oversaturation, neutral or basic pH and high phosphorus, nitrogen and silicate concentrations (Table 1). The TN:TP and Chl a:TP ratios indicated that phosphorus was the primary limiting nutrient. According to silicate concentration and the SiO2/TP ratio, Lake Rodó presented high availability of this element (Table 1).
The conductivity, SRP and nitrate showed clear differences between the sampling sites. The groundwater supply had higher conductivity and nutrient concentrations compared with the lake and pools, and it was quite constant. As a result the concentrations of organic matter, nitrogen and phosphorous showed an increased especially in the sediment surface since December 1997.
The benthos genera registered during the study were Coelotanypus, Procladius, Larsia, Polypedium, Goeldichironomus (Chironomidae), Hellobdella (Hirudinea), and Limnodrilus hoffmeisteri (Oligochaeta) . The benthos density was variable throughout the study period, including a period when no organisms were observed. The total abundance was correlated negatively with the proportion of total phosphorus and organic matter of the sediments (rs=-0.83, p<0.05, rs=-0.85, p<0.05, respectively).
The total number of phytoplankton taxa registered was 150. Since the end of the re-filling process, there were major changes in composition and abundance of the phytoplankton community. It was dominated alternatively by small coccal green algae (Coelastrum sp., Sphaerocystis sp., Ooccystis sp and Scenedesmus sp.), pennate diatoms (Synedra acus), filamentous Cyanobacteria (Limnothrix sp and Anabaena sp.), with a similar begin of the summer 1997-98 with green algae but replaced time by chroococcal Cyanobacteria (Gomphosphaeria sp., Merismopedia sp., Coelosphaerium sp. and Microcystis sp.) and finally by a filamentous genus (Raphidiopsis mediterranea) (Fig 3).
The zooplankton community was dominated by small herbivores, principally rotifers(Keratella tropica) (Fig. 4). The copepods showed the higher abundance at the beginning of the spring (Notodiaptomus incompositus, Metacyclops mendocinus and Tropocyclops prasinus meridionalis. The cladocerans were the less abundant group with their higher abundance in spring and summer 1997-98 (Moina micrura and Diaphanosoma birgei).
The
fish community was represented by four species. Cnesterodon decemmaculatus
(Poeciliidae), was the dominant species with a density of 800900 ind ha-1.
Together with Jenynsia lineata (Jenynsiidae) (13900 ind. ha-1),
both planktivorous species, represented the 96% of the total fish biomass
(246 kg ha-1). The other two species registered were Cichlasoma
facetum and Gymnogeophagus rabdotus (Cichlidae). The
planktivorous fish removal carried out in August and September 1997 accounted
for 20% of the total planktivorous fish biomass, and the removal performed
in November and December 1997, 6%.
DISCUSSION
After sediment removal, diversion of surface run-off and sewage waters and the re-filling of the lake in January 1997, the system rapidly reached an hypertrophic state conditioned principally by the high nutrient input, due to groundwater characteristics: highly contaminated, probably due to wastewater infiltration.
The main physicochemical water changes observed, in relation to previous studies (Sommaruga 1995), were the decrease of total phosphorus (63%) and a significant increase of total nitrogen (300%). The mean TN:TP ratio (by weight) changed from 8.5 to 67.6, transforming the lake from N-limiting conditions into P-limiting conditions.
The mean water transparency increased from 0.20 to 0.56 m (Secchi depth), after the re-filling of the lake, conditioned by the algal community reduction (in almost 50%) and change. Before the beginning of the restoration programme, the phytoplankton was dominated by Planktothrix agardhii. Nowadays, the community is more diverse (150 taxa) while Sommaruga (1995) in a similar period, only found 21. Besides, phytoplankton grazing is now favoured for the presence of more edible microalgae (Gliwicz 1990). The successional processes after re-filling, and the TN:TP and SiO2:TP ratios changes, could have conditioned the structural changes observed in the phytoplankton community before and after the restoration activities. The algal sedimentation promoted the increment of nutrients and organic matter content in the silt. This process would determine the temporal decrease of benthic abundance and prevent colonisation of sediments (citas).
Stable clear water phase (Secchi depth > 1 m) can be expected at TP < 66 m g l-1 and at 45 m g l-1 of chlorophyll a, for mixed phytoplankton shallow lakes. However, high TP levels do not necessary exclude the existence of an alternative clear water state, during spring (Hosper 1997), which sometimes persist for up to one month (Gulati 1983, Sommer et al. 1986, Lampert & Sommer 1997). With TP concentrations higher than 66 m g l-1, during three weeks, between October and November 1997 transparency increased up to 1.4 m. In this clear water phase, the lowest values of chlorophyll a were observed, concomitant with an increase in the mesozooplankton herbivores (principally cladocerans).
The structure of the zooplankton community was similar in composition and temporal variation to the one registered by Sommaruga (1995). The cladoceran species reached their maximum abundance in spring and summer and it was higher than that observed before the emptying of the lake and it was not accompanied by increase in the water transparency, as was registered in spring 1997.
The increase of the transparency in spring, could have been influenced by the 20 % winter reduction of planktivorous fish stock. The composition of fish community before 1997 is not known, except for the presence of Cichlasoma sp. and schools of Cnesterodon decemmaculatus (Sommaruga 1995). The species composition and abundance of fish stocked are unknown, but the community became rapidly characterised by a high fish biomass, with dominance of planktivores and the absence of piscivores.
The highly successful reproductive strategies of C. decemmaculatus and Jenynsia lineata (Lorier & Berois 1995, Siccardi 1955) coupled with a short generation time and a low predation pressure, have favoured the quick re-colonisation and high biomass reached in the lake based on the introduced individuals in a short period. The lower relative abundance of these Cichlids could be due to the short time from the re-colonisation of the lake, and to a longer generation time. In the future, Cichlasoma might exercise an effective predation pressure on the young of the planktivorous species.
The analysis of nutrient concentrations at the four selected sampling stations, allowed us to identify a falling gradient from groundwater to pool 2 (groundwater Õ lake Õ pool 1 Õ pool 2). Based on this we have designed a re-circulation system, pumping water from pool 2 towards the lake. The groundwater supply will be only automatically maintaining the water level. A re-circulation system has been built and to increase their efficiency, aquatic floating plants (Pistia stratiotes and Eichhornia crassipes) had been introduced into the pools and are being periodically removed.
At the present time the lake Rodó offers good chances of restoring through biomanipulation due to the low abundance of filamentous cyanobacteria (Meijer & Hosper 1997). The presence of abundant populations of planktivorous fish maintains a high predation pressure on the zooplankton and leads to an absence of large-bodied herbivores. The most important ways actually implemented or improved to establish are the reduction of nutrient inputs and decrease of the small planktivorous fish populations of the lake.
REFERENCES
GLIWICZ,
Z. M. 1990. Why do cladocerans fail to control algal blooms? Hydrobiologia
200/201:83-97
GULATI,
R. D. 1983. Zooplankton and its grazing as indicators of trophic status
in Dutch lakes. Environ. Monit. Assess. 3:343-354.
H?KANSON,
L & M. JANSSON. 1983. Principles of lake sedimentology. Springer Verlag,
NY, 316p.
LAMPERT, W & U. SOMMER. 1997. Limnoecology: The Ecology of Lakes and Streams. Oxford University Press, New York 382p.
LEWIS
1983. Lorier,
E. & N. Berois. 1995. Reproducción y nutrición embrionaria
en Cnesterodon decemmaculatus (Teleostei: Poeciliidae). Rev. Brasil. Biol.,
55(1):27-44.
MEIJER
, M-L. & H. HOSPER. 1997. Effects of biomanipulation in the large and
shallow Lake Wolderwijd, The Netherlans. Hydrobiologia 342/343:335-349.
NUSCH, E. A. 1980. Comparison of different methods for chlorophyll and phaeopigments determination. Arch. Hydrobiol. Beih. Ergebn. Limnol. 17:14-36.
SICCARDI, E. 1955. La reproducción de los Cyprinodontiformes. Ichthys I(4):137-144.
SOMMARUGA, R. 1995. Microbial and classical food webs: a visit to a hypertrophic lake. FEMS Microbiology Ecology 17(4):257-270.
SOMMER,
U., Z. M. Gliwicz, W. Lampert & A. Duncan. 1986. The PEG-model of seasonal
succession of planktonic events in freshwaters. Arch. Hydrobiol. 106:443-471.
VALDERRAMA, J. C. 1981. The simultaneous analysis of total N and P in natural waters. Mar. Chem. 10:1009-1022.
