-
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1434-2944/09/102-0103
Internat. Rev. Hydrobiol. 94 2009 1 103125
DOI: 10.1002/iroh.200811071
GNTER GUNKEL* and CAMILLA BEULKER
Berlin University of Technology, Dept. of Water Quality Control,
Strae des 17 Juni 135, Sekr. KF 4, D-10623 Berlin, Germany;
e-mail:[email protected]
Limnology of the Crater Lake Cuicocha, Ecuador,a Cold Water
Tropical Lake
key words: Andes, atelomixis, caldera lake, Ecuador, Lake
Cuicocha, travertine
Abstract
Cuicocha (3380 m a.s.l.) is a young, a few hundred years old
volcanic lake in the western cordilleras of the Ecuadorian Andes
with some post-volcanic activities, such as emission of volcanic
gases and input of hydrothermal water. Water chemistry is
influenced by the emission of CO2 and weathering of the young
andesitic rocks in the water shed. A calcium cycle exists in the
lake with intensive biological Ca precipitation at the flanks and
formation of travertine crusts, while in the hypolimnion
dissolution of Ca carbonate occurs. The crater lake is
oligotrophic, biodiversity is low; the littoral flora and fauna is
more important than the pelagic species. In the littoral zone, a
small Totora zone occurs, followed by submerged macrophytes down to
35 m water depth. Phyto- and zooplankton occur down into the
hypolimnion. Phytoplankton is strongly influenced by down-welling
of water (atelomixis) and by copre-cipitation with detritial
flocs.
1. Introduction
Volcanic lakes, built up in a caldera, are strongly influenced
by volcanic activities such as gas emissions and hydrothermal water
springs. Thus, different lake types are formed, and have been
classified according to their physical water constraints by
PASTERNACK and VAREKAMP (1997). These authors distinguished
volcanic lakes with different levels of activ-ity, namely cool to
hot acid-brine lakes, reduced to oxidized, acid-saline lakes,
acid-sulphate lakes and bursting to buoyant plume bicarbonate
lakes; only neutral dilute volcanic lakes do not show any activity.
Besides this heterogeneity in water chemistry, a wide range of
mor-phometric characters occur as well as different lake genesis
which influence the limnology of the lakes (LARSON 1989; VAREKAMP
et al., 2000; ARMIENTA et al., 2000; AGUILERA et al., 2006). Some
basic knowledge exists regarding the limnology of crater lakes
(VZQUEZ et al., 2004; SCHABETSBERGER et al., 2004) especially from
the long research program on the Crater Lake in Oregon, which was
mainly concerned with water chemistry, stratification processes and
water currents (LARSON, 1996; LARSON et al., 1996; NELSON et al.,
1996; CRAWFORD and COLLIER, 1997).
Limnology of crater lakes is of high interest, because they are
young lakes, situated in an area with soils in a state of
development. Many of the calderas are deep and have formed lakes
with a depth of a few hundred meters, mostly situated in high
mountain regions. Some
* Corresponding author
-
104 G. GUNKEL and C. BEULKER
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
of these represent a scarce type of cold tropical lake (LFFLER,
1964; CASALLAS and GUNKEL, 2001; ROLDAN and RUIZ, 2001;
SCHABETSBERGER et al., 2004), and a few investigations on this type
of lake have been carried out mainly concerning water mixing
processes (TALLING, 1969; GUNKEL and CASALLAS, 2002, 2002a),
primary production (KINZIE et al., 1998) and occurrence of
zooplankton (GREEN, 1995). The morphometry of caldera lakes is
frequently determined by the steep slopes of the flanks and
non-eroded lake shores. Water chemistry is strongly influenced by
the source of water input (rainwater or hydrothermal water) and by
the weathering of volcanic deposits in the watershed. In some
cases, the continuous input of volcanic gases and/or hydrothermal
water leads to ionic rich water (OHBA et al., 1994; NELSON et al.
1996; RONDE et al., 2002), and a chemocline can build up (WOOD et
al., 1984). Inflow of hydrothermal water and the energy flux from
the volcano strongly influences the stratification processes
(CAMERON and LARSON, 1993; CHRISTENSON, 1994; PASTERNACK and
VAREKAMP, 1997). This can lead to an irregular overturn due to
inverse thermal stratifica-tion, and convergent currents are
observed (GOECKE, 1997; CRAWFORD and COLLIER, 1997). Fauna and
flora are determined by the young age of the lakes, the high
mountain position, and the mixing processes, and it must be assumed
that the complexity of the biocoenosis is reduced.
Crater lakes are poorly investigated worldwide. Since the CO2
eruption of Lake Nyos, Cameroon, followed by intense international
research to analyse the phenomenon of limnic eruptions (LE GUERN
and SIGVALDASON, 1989, 1990; KUSAKABE, 1994; MARTINI et al., 1994),
there have been concentrated efforts to obtain detailed information
on other volcanic lakes with potentially dangerous CO2
accumulation. However, there is still a deficit of knowledge on
volcanic lakes, especially in the Andes of South America, even
though this is a region with a great number of active volcanoes. In
Ecuador, two lakes are known to be active volcanic lakes, the
Quilotoa (AGUILERA et al., 2000; GUNKEL et al., 2008) and the
Cuicocha (von HILLEBRANDT and HALL, 1988; GUNKEL et al., pers.
com.), located in the high Andine region > 3000 m above sea
level (a.s.l.). Both the Quilotoa and Cuicocha volcanoes, have
formed large and deep caldera lakes and currently show
post-volcanic activities in form of volcanic gas emissions and
input of geothermal water. The volcanic gases, mainly CO2, are
accumu-lated in the lake water but without the risk of a limnic
eruption (GUNKEL et al., pers. com.).
Investigations on the Cotacachi-Cuicocha complex carried out by
SAUER (1971), MOTHES and HALL (1991) and GRUPE et al. (2008)
provided a clear picture of the history of the volcano and the
petrogenesis of the erupted lavas, and VON HILLEBRANDT and HALL
(1988) developed a volcanic hazard map of the surrounding
areas.
Lake Cuicocha was the focus of a detailed limnological study
with the objective of a comprehensive characterization of the
volcanic lake and its watershed in respect to petrol-ogy,
morphometry, hydrophysics, water chemistry and limnology. Several
scientific field campaigns were carried out between 2003 and 2006;
primary results have been published regarding macrophytes (KIERSCH
et al., 2004), Al polymerisation (GUNKEL et al., pers. com.) and
the risk assessment of a limnic eruption (GUNKEL et al., pers.
com.).
2. Methods
2.1. Studied Site
Cuicocha is a parasitic volcano of the Cotacachi volcano, which
was active in the Pleistocene period, and is located in the western
cordilleras of the Ecuadorian Andes, situated about 100 km north of
Quito. Lake Cuicocha was formed after the collapse of the Cuicocha
domo and is a young caldera lake (Fig. 1). Today, the caldera has a
diameter of 3.2 km and a maximum depth of 450 m, filled by a lake
to a water depth of 148 m. Recent volcanic activities of the
Cuicocha volcano include emission of volcanic gases and some
hydrothermal water inflow at the lake bottom as well as at the
shore line.
-
Limnology of a Crater Lake 105
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
The water input into the lake is of low quantity, due to the
small catchment area, and to the moderate annual precipitation of
1320 mm (20042006). The caldera has no continuous inflow and water
inflow occurs via a periodic waterfall, two cascades with low flow
rates and the mentioned hydrothermal entries; the lake has no
superficial outflow. The soils consist of volcanic deposits, mainly
andesite with a high SiO2 and Al2O3 content (5761 % SiO2, 1718%
Al2O3; GRUPE et al., 2008) and its weathering products. The soils
are in an early stage of development with a low clay content
classified as andisols covered by paramo vegetation (ZEHETNER,
2003).
2.2. Sampling
Lake Cuicocha was investigated using a regular monitoring
program and sampling was carried out twice a year for four years
(20032006). The twice yearly investigation periods were February to
April (rainy period) and July to October (dry period), from 08/2003
to 04/2006. The data base for the water chemistry included 6
vertical profiles (in 08/2003, 09/2003, 03/2003, 08/2004, 03/2005,
08/2005), and lake profiler diagrams (T, pH, cond., E7) were based
on 10 profiles (in 08/2003, 09/2003, 03/2004, 08/2004, 03/2005,
08/2005, 04/2006 with 2275 data points). Investigations on further
parameters (water physics and stratification stability at the
crater rim and near the islands) were based on 29 lake profiles,
determined during the regular investigation periods.
For depth determination and detailed bathymetric mapping of the
lake, a double frequency sonar (50 and 200 kHz; Garmin Fishfinder
250 C) in combination with GPS (Garmin 60CS) was used; the
bathymetric map was developed by the sonar GPS data and digitalised
topographic maps (1 : 15 000, Souris, IRD). The use of sonar
allowed detection of the lake floor as well as recognition of
volcanic gas emissions and resuspended sediments as a consequence
of gas eruptions. Furthermore, the use of sonar facilitated the
positioning of the equipment in a specific water depth and thus
allowed high precision for data registration near the sediment.
Temperature, pH, conductivity, CO2 and redox potential were
determined by a lake profiler (Ocean Seven 316, Idronaut, Italy)
with extremely high accuracy of the probes (with values for
temperature + 0.003 C, for pH 0.01 pH units and for redox potential
1 mV). The CO2, probe was calibrated immediately after utilization
by chemical determinations of CO2/HCO3 and CO32 according to the
German Standard Methods for pKa and pKb analysis (DEV 2005). The
density of the water was cal-culated under consideration of
salinity, local pressure and temperature using the formula of CHEN
and MILLERO (1986) for natural waters and is expressed as (S, T,
Psurface) (= density 1000 kg/m3); salinity was calculated by the
total ionic content. Oxygen concentration within the carbonate
precipitates was deter-mined using optical oxygen sensor of 2 mm
with a Fibox 3 oxygen meter (PreSens, Germany). Water samples were
taken using a Ruttner water sampler (Hydrobios, Germany) at 78 m (9
samples in 178 m; N 01811.4 W 782210.9) and 148 m depth (13 sample
in 1148 m; N 01742.3 W 782126.7; Fig. 2), portions of the water
samples were filtered immediately, using 0.45 m cellulose-acetate
filters for cation and anion analyses. Water samples for chemical
analyses were preserved by HNO3 as well as HCl (both at pH ~ 1),
using HDPE bottles.
Gas sampling was done directly with GC vials at the lake shore
and on Island Yerovi.Samples of suspended material for Scanning
Electron Microscopy was obtained by filtration of
250 mL lake water from the water sampler, immediately after
sampling in the boat, using 0.4 m poly-carbonate filters
(Nuclepore). The wet polycarbonate filters were stored in plastic
bags with taps soaked with formaldehyde (37%) for preservation.
Sediment samples were collected by a sediment gravity corer, 5
cm with a sediment capturer. Sedi-ment sampling was recognized to
be very difficult due to only a thin sediment layer on the stony
floor and due to sediment oversaturation by gases, which led to the
loss of sediments while degassing during lifting of the equipment.
Sediment samples were prepared using an HCl/HNO3 acid digestion
method (VDLUFA 1991) and chemically analysed using the analytical
methods mentioned below. Sedimentation behaviour of seston was
studied using 8 sediment traps, 8 cm , 50 cm length, which were
exposed in the western bay of the lake for two periods, each one
for 2 weeks at depths of 15, 30, 50 and 70 m (N 01815.3 W
782214.9). Samples were preserved with 4% glutaraldehyde solution
for Scanning Electron Microscopy with Energy Dispersive
Spectroscopy (SEM-EDS) analyses.
A SONY HCR-HC16E digital video camera was modified as an
underwater camera and was pro-tected by a purpose-built aluminium
housing and equipped with 4 underwater lamps (NEMO 8C Xenon, 14
watt). With this camera, sediment type and gas emissions with
sediment resuspension were regis-
-
106 G. GUNKEL and C. BEULKER
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
tered as well as the penetration of the sediment core sampler
into the sediment. During two campaigns, divers were used for
sampling and underwater filming, and the experiences of the divers
were helpful for further investigations.
Phytoplankton samples were collected from the water sampler (11
depths in 0145 m), preserved with Lugols solution. Zooplankton was
collected using a plankton net (55 m) with trapping mechanism (a
profile was made by 8 sections from 0145 m), and preserved with
HANEY and HALL solution. Plants were collected using a macrophyte
anchor (KIERSCH et al., 2004) in 1998 within 6 profiles from the
lake surface down to 40 m. Sampling of the invertebrates was done
by a fine sieve and a plankton net at 6 different positions in the
littoral zone, surficial and down to a few meter water depth,
preserved with 4% formaldehyde. Fish control was done by using a
sonar (Garmin Fishfinder 250 C).
2.3. Chemical Analyses
Nitrite was determined using the Merck test (Aquaquant, 0.005
0.1 mg L1 NO2) immediately after the sample was collected. Chemical
analyses for CO2, HCO3 and CO32 were undertaken on the sam-pling
day applying the German Standard Methods for pKa and pKb
determination (DEV 2005). Oxygen concentration was analysed using
the WINKLER method, soluble reactive phosphorus (SRP), ammonia and
nitrate were analysed according to US APHA Methods (1998) in the
Laboratory of Chemistry, University Central, Quito, Ecuador. Water
chemistry of the non-reactive cations and anions were carried out
in the laboratory of the Berlin University of Technology, Dept. of
Water Quality Control, Germany, using the acid-preserved water
samples. Water samples for the determination of total amounts of
ions were digested in an autoclave under acidic conditions using
K2S2O8 (121 C, 1.3 bars for 2 h). Total phosphorus and total
nitrogen were analysed photometrically by flow injection analysis
in accordance with EN ISO 15681-1 (2004) and 11905-1 (1998),
respectively (Foss Tecator FIAstar 5000) with a detection limit of
0.005 g L1.
The cations Ca2+, Na+, Mg+, K+, As2+ and Fe3+ were analysed by
flame AAS (GBC Scientific Equip-ment, Pty. Ltd. Victoria,
Australia), lower concentrations of Li+, Fe3+ and Mn2+ (< 0.1 mg
L1) were ana-lysed by graphite furnace ASS (Varian Spectra A-400).
The anions Cl, SO42 and NO3 were analysed using an ionic
chromatograph (AS 50 Dionex) with CD 20 detector, GD 50 gradient
pump and an AS 11 column for separation. Boron was determined
photometrically in accordance with DIN 38405-D17 (DEV 1981; Dr.
LANGE LCK 307). Analyses of the gases CO2, CO, O2, N2, N2O, and CH4
were carried out using a GC with FID and TCD detectors at the
Leibniz-Institute of Freshwater Ecology and Inland Fisheries,
Berlin, Germany. 14C-analyses of soils were carried out by the
Leibnitz Institute for Applied Geosciences, Hannover, Germany.
2.4. Biological and Microscopic Analyses
The determination of seston (bacteria, precipitations, detrital
flocs) was done by Scanning Electron Microscopy as with Energy
Dispersive Spectroscopy. For SEM-EDS analyses the wet polycarbonate
filters were fixed on a bracket, air dried and then sputtered with
gold or carbon. A SEM-EDS Hitachi S 2700 electron microscope was
used with an acceleration voltage of 20 kV and an IDFix hardware
and software from SAMx for analysis.
Phytoplankton was determined qualitatively using a Zeiss Laboval
4 and quantitatively according to the Utermhl method (DEV 2007)
using a Olympus microscope CK30, about up to 5 transects of the
Utermhl chamber (2 to 50 ml) were counted for quantitative
determination, for the calculation of the biovolume 20-100 cells
were measured in the Zeiss using an ocular with a reticle. The
biovolume was calculated using related geometrical bodies (PADISK
and ADRIAN, 1999). Diversity, abundance and biovolume were
determined for each of the 11 sampling depths.
Determination of macroinvertebrates was carried out by Dr.
CORREOSO, Catholic University of Quito.
-
Limnology of a Crater Lake 107
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
3. Results
3.1. Lake Formation and Morphometry
The strato volcano Cotacachi was active during the middle
Pleistocene period up to 630 000 years ago (INECEL 1983), which
documents the end of volcanic activities. Cui-cocha, a parasitic
volcano of the Cotacachi, began its activity with a series of
eruptions 45003000 years BP followed by the collapse of the dome
and a caldera was formed. Final eruptions took place more or less
simultaneously about 14501330 BP and led to the formation of four
domes within the caldera, nowadays the islands of Yerovi and Wolf
(GRUPE et al., 2008; Fig. 1), and the differentiation of the
caldera into two main basins of different depth. Following the
extinction of the Cuicocha volcano, the caldera was partly filled
with sediment from the catchment area; fine grained and clay
material sealed existing fissures and fractures within the caldera
and after the end of the colmation process, lake water accumulation
began, probably a few hundred years ago.
Nowadays Lake Cuicocha is situated about 3072 m a.s.l. and has a
maximum water depth of 148 m, a mean depth of 72 m; the volume
amounts to 0.28 km3 and the lake surface 3.78 km2 (Table 1). The
lake morphometry is determined by the caldera flanks with steep
inclinations, and in some parts of the crater rim, a water depth of
100 m is reached only 20 m from the shore line. The bathymetric map
(Fig. 2) was developed on the basis of 1250 data points and shows
two lake basins, one with a maximum depth of 148 m extending east
of the islands, and the other with a depth of 78 m, situated in the
western area of the lake.
From the catchment area, water inflow into the caldera occurred
until recently, however, the watershed is small (18.2 km2) with a
surrounding factor of only 5.9. Additional hydro-thermal water has
been entering into the caldera and mainly influences the water
chemistry. The lakes water conductivity clearly points to a
significantly increased ionic content com-pared with the inflow
water of the catchment, and a rough estimate shows in total about
30% hydrothermal water, based on rain water inflow and recent
hydrothermal water quality (see below).
3.2. Lake Stratification and Mixing Processes
Lake Cuicocha is a weakly thermal stratified lake with a
monomictic cycle, and overturn occurs during the windy dry period
from July to October (Fig. 3). The epilimnion stretches down to 40
m. Temperature differences between surface waters (approx. 1618 C)
and the deep water body were small and amounted to about 2 C during
stratified periods. The stabil-ity of the stratification in Lake
Cuicocha was further reduced due to nocturnal cooling and deep lake
mixing, the so called atelomixis, which was analysed in detail in
the nearby Lake San Pablo (GUNKEL and CASALLAS, 2002, 2002a).
During a 24 hour period in March during stratification period (Fig.
4), heating occurred in the upper water body of 50 m, mainly as a
consequence of daily radiation input; and in the deeper water layer
of 70 to 100 m depth a decrease in temperature was registered due
to convection currents which occurred during the night after
cooling of the surface layer.
An input of hydrothermal water was detected in the 78 m lake
basin, and near the lake bottom both temperature and conductivity
increased significantly (T = about 0.04 C; Cond = about 25 S cm1;
Fig. 5), resulting in a decreasing density of (S,T) = 0.00915 kg m3
(P < 0.0001) in the 70 m water depth compared to the 55 m depth.
In the whole lake density of the water increased from the surface (
= 999.0191) to the ground ( = 999.2731 in 78 m depth, = 999.2986 in
148 m depth), but no chemocline was build up and the hydrothermal
water was quickly mixed with the lake water. The inflow of
hydrothermal water at the lake floor could not be quantified. Well
known is the hydrothermal water inflow at the shore line
-
108 G. GUNKEL and C. BEULKER
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
of Island Yerovi, which was used for water chemical analyses, in
2005 an additional inflow of hydrothermal water was observed near
the Island Yerovi in 510 m water depth.
The conductivity profile in the 148 m lake basin showed no
increase, and an inflow of hydrothermal water must be excluded.
In the western basin of Lake Cuicocha, at 78 m depth, a
permanent emission of CO2 took place detected by sonar and few gas
bubbles reached the lake surface. The composition of the gas (see
below) clearly pointed to the volcanic origin. The emission of the
gas led to a resuspension of sediments proved by sonar and
underwater filming and to a partial mixing of the water body by
billows during upraising of gas bubbles.
Figure 1. Lake Cuicocha with domos Yerovi (left) and Wolf
(right).
Table 1. Morphometric data of Lake Cuicocha.
Parameter Lake Cuicocha
Lake coordinates NorthSouth N0 18 49.9 W78 21 40.1 N0 17 32.6
W78 21 28.8WestEast N0 18 22.1 W78 22 33.1 N0 18 03.9 W78 20
52.8Watershed (km2) 18.2Surrounding factor 5.9Lake water level (m
a.s.l.) 3,072Length (m) 3,238Width (m) 2,232Shore line without
islands (km) 10.14 Shore line with islands (km) 14.43Shore line
development without islands 2.9Shore line development with islands
4.2Surficial area (m2) 3,781,012Maximum depth (m) 148Mean depth (m)
72Relative depth (%) 3.3Volume (m3) 282,053,575Volume of the
islands underwater (m3) 103,648,907Volume of Jerovi above water
line (m3) 9,011,699Volume of Wolf (m3) above water line (m3)
29,381,720
-
Limnology of a Crater Lake 109
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
3.3. Water Chemistry and Post-Volcanic Activities
Gas analyses showed that the main composition was CO2 (51%) and
N2 (23%) with smaller amounts of O2 (3.0%) CH4 (1.7%) and CO
(0.3%). Gaseous emissions of boron compounds led to increased BO43
concentrations in the lake water, a typical component of volcanic
gases.
The pH of Lake Cuicocha in the epilimnic water (040 m) was
weakly alkaline, the pH was raised to 8.3 as a consequence of loss
of CO2 from the oversaturated water to the atmos-phere and primary
production. The hypolimnion (> 40 m) exhibited neutral
conditions with a pH down to 7.0 (Fig. 6), caused by CO2 input from
volcanic gases from the lake floor in the western basin (GUNKEL et
al., pers. com.).
The carbon dioxide concentrations were elevated in the whole
water column with the highest concentrations in the deep water
layers > 60 m, a consequence of the hypolimnic gas input in the
shallow basin, combined with high wind-induced horizontal water
cycling between the different parts of the lake. In the deep water,
an oversaturation of CO2 of up to 78 times, compared with the
atmosphere and the local pressure was observed, this cor-responded
to a CO2 concentration of 40 mg L1 or 31 mL CO2 per litre water.
Nevertheless, the in situ CO2 saturation of the hypolimnic water
was low, due to the high pressure of the water column in the deep
water, where great quantities of CO2 are soluble. Taking into
consideration the partial pressure of the gas, leaving the volcano
(xi = 0.39%), the in situ saturation of CO2 in the bottom-near
water layers amounted to only 0.08%.
Figure 2. Bathymetric map of Lake Cuicocha, = sampling points, 1
= cascades, 2 waterfall.
-
110 G. GUNKEL and C. BEULKER
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
Figure 3. Temperature isopleths of Lake Cuicocha, based on 10
profiles.
Figure 4. Temperature profiles and temperature differences over
a 1 day period during lake stratifica-tion period, Lake Cuicocha,
148 m depth position, 30/31.3.2004.
-
Limnology of a Crater Lake 111
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
The dissolved ionic concentration of Lake Cuicocha resulted in a
conductivity of about 810 S cm1; this was elevated compared with
the water inflow of the cascades ( 260 S cm-1) and a nearby rain
water-filled caldera lake, Lake Mojanda (mean 43 S cm1). Thus the
Cuicocha lake water must be influenced significantly by the input
of hydrothermal water.
Lake Cuicocha is a sodium hydrogen carbonate water (60.6 mg L1
Na+, 189 mg L1 HCO3) with significant amounts of magnesium and
calcium as cations (29.8 mg L1 Mg+, 46.8 mg L1 Ca2+) and chloride
as anion (70.5 mg L1 Cl, Table 2). Ions of minor concen-trations
were the cations K+, Li+, Fe3+, Mn2+, Al3+ and the anions SO42,
SiO32, B(OH)4, PO43.
In Lake Cuicocha the trophic level is very low, and a median
phosphorus concentration of 12 g L1 Ptotal was observed. The
nitrogen concentrations were also shown to be low (mean 96 g L1
Ntotal). The N/P ratio (median = 10.0, mass basis) for most data
showed P limita-tion. Due to the volcanic origin of the lake, the
siliceous concentration (median = 21 mg L1 SiO2) was very high and
exceeded the normal range in natural lakes.
Pelagic calcium concentration varied between 39.856.1 mg L1 (5-
and 95-percentile), and the epilimnic Ca saturation index of + 0.75
(030 m) showed a slight oversaturation. Nevertheless, precipitation
of calcium carbonate was registered at the shore line of Lake
Cuicocha. These precipitations were continuously built up by
epilithic algae and cyanobac-teria (see below), leading to CaCO3
crusts of about 1 cm in thickness, which reached down deep into the
epilimnion. Wave action partly destroyed these crusts, forming fine
carbonate debris, which sank to the lake bottom, where dissolution
of the CaCO3 precipitates occurred under conditions of negative Ca
saturation index (S = 20.2 at 45135 m, see Fig. 8). An up-welling
of hypolimnic water by the lake floor heating (see 3.2.) led to an
internal calcium cycle.
Figure 5. Increases in temperature and conductivity with
decrease of the density near the lake bot-tom in the 78 m lake
basin; (n = 107; variances R2 for temperature = 0.8298,
conductivity = 0.8913,
density = 0.8297).
-
112 G. GUNKEL and C. BEULKER
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
The inflow of hydrothermal water was analysed at the shore of
Island Yerovi. The water was of relatively low temperature (2326
C), rich in sodium (247 mg L1 Na+), calcium (120 mg L1 Ca2+) and
iron (7.4 mg L1 Fe3+) as cations and sulphate (193 mg L1 SO42),
chloride (171 mg L1 Cl) and silicate (110 mg L1 SiO2) as
anions.
3.4. Littoral Zone and Travertine Formation
At the littoral zone of Lake Cuicocha the reed plant Totora
(Schoenoplectus totora) built up a dense but narrow vegetation
zone, followed by a few submerged plants including Myriophyllum
quitense (from 02.3 m depth), Potamogeton illinoensis (28 m), P.
pectina-tus (214 m), Chara rusbyabana (2729 m), Ch. globularis (830
m) and Nitella acuminata (2035 m). As a new proof Drepanocladus
capillifolius, a Bryophyta, Fam. Amblystegiace-ae, was found at
2028 m depth (KIERSCH et al., 2004). The biomass of the submerged
macrophytes was very high and they built up dense stands with very
long plants with several meters lengths on the high flanks.
The invertebrates of the littoral zone were dominated by
Hyalella cf. dentata, Fam. Malacostraca; GONZALEZ and WATLING
(2002). Other species with regular mass develop-ment were Dugesia
sp. and some Gastropoda species, Biomphalaria sp. and Potamopyrgus
antipurvarum (Table 3). The biodiversity was very low and the few
species occurred in high abundance.
Only one fish species was observed and filmed by divers, the
Andean catfish Astroblepus ubidiai (Siluriform), living in the
upper part of the littoral zone was a new proof for this lake. This
fish is a resident species of the high Andes, with a few remnant
populations in this area (VLEZ-ESPINO and FOX, 2005).
The colonization of littoral interfaces by benthic algae and
bacteria was of high ecological importance. A dense biofilm with
intensive photosynthetic activity induced the precipita-tion of
travertine on the caldera flanks and consisted mainly of motile
pennate diatoms (Epithemia argus, Mastogloia smithii, Cymbella
pusilla and Rhoicosphenia abbreviata as dominant species) and one
frequent cyanobacteria species, Calothrix parietina, which
lived
Figure 6. pH isopleths of Lake Cuicocha, based on 10
profiles.
-
Limnology of a Crater Lake 113
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
Table 2. Water chemistry of Lake Cuicocha, median and
percentiles of 20032005; n = 84.
Parameter 5 % Median 95 %
pHepilimnion 7.65 8.14 8.58pHhypolimnion 7.18 7.37
7.86Conductivity (S cm1) 797.2 812.3 818.2Natotal (mg L1) 55.9 60.6
69.0Catotal (mg L1) 39.8 46.8 56.1Ktotal (mg L1) 4.7 5.5 6.4Litotal
(mg L1) 0.100 0.110 0.135Mgtotal (mg L1) 25.3 29.8 41.2Fe dissolved
(mg L1) 0.000 0.006 0.022Fetotal (mg L1) 0.004 0.0120 0.073Mntotal
(mg L1) 0.001 0.005 0.083Altotal (mg L1) 0.003 0.009 0.139B+ (mg
L1) 2.4 3.9 5.8HCO3 (mg L1) 138 189 351Cl (mg L1) 64.0 70.5
74.4SO42 (mg L1) 9.2 13.4 18.3Si dissolved (mg L1) 17.1 20.0
23.3Sitotal (mg L1) 17.0 21.0 23.4Ntotal (mg L1) 0.029 0.096
0.296Ptotal (mg L1) 0.004 0.011 0.027SRP (mg L1 PO4-P) 0.002 0.005
0.013N/P (weight based) 2.5 10.0 28.9O2epilimnion (mg L1) 5.9 6.7
7.0O2hypolimnion (mg L1) 0.8 2.0 3.2
Table 3. Fauna in Lake Cuicocha (rare to mass describes the
relative abundance in a seven grade scale).
Class/Subclass/Family Taxon Abundance
Littoral fauna:Turbellaria Dugesia sp. frequentGastropoda,
Planorbidae Biomphalaria sp. frequent
Gyraulus sp. less frequent Physidae Physa sp. less frequent
Hydrobidae Potamopyrgus antipurvarum massBivalvia, Spaeridae
Sphaerium forbesi less frequentCrustacea, Ostracoda Cypris sp.
frequent Malacostraca Hyalella cf. dentata mass
Insecta, Zygoptera Oxyallagma dissidens less frequentFishes:
Osteichthyes, Siluridae Astroblepus ubidiai rare
Zooplankton:Crustacea, Phyllopoda Daphnia pulex pulex 0.23 n
L1Crustacea, Copepoda Metacyclops mendocinus WIERZEJSKI 0.117 n L1
Malacostraca Hyalella cf. dentata 0.0010.06 n L1
-
114 G. GUNKEL and C. BEULKER
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
at the surface and inside the pore system of the CaCO3 crusts
(Fig. 7). The intensive primary production in this biofilm was
proved using oxygen measurements, which showed a mean O2
oversaturation of 199% (min. = 152%, max. = 216%, n = 13) 5 mm
inside the CaCO3 crusts, resulting in a microenvironment with
increased pH and favoured carbonate precipitation; the CaCO3 crusts
were composed of pure calcium and contained only traces of
magnesium.
3.5. Plankton Community
The phytoplankton community was characterized by Diatomeae
(Aulacoseira granulata, Fragilaria ulva var. acus, Fragilaria sp.),
Crytophyceae (Rhodomonas and Cryptomonas), Euglenophyceae
(Trachelomonas volvocina) and some Chlorococcales species (Oocystis
lacustris, O. naegli, O. parva and Scenedesmus linearis, S.
quadricauda, S. acuminatus,
Figure 7. Calcium carbonate precipitations within the biofilm at
the lake shore, REM picture of the calcium crusts with some diatoms
embedded in CaCO3.
-
Limnology of a Crater Lake 115
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
Table 4. Occurrence of phytoplankton in Lake Cuicocha, abundance
of algae (n mL1), n.d. = not determined. The abundance is given as
the maximum cell number in one of the 11 sampling depths (sampling
depths: 0, 5, 15, 30, 45, 60, 75, 90, 105, 120, 135, 145m), *)
biomass without Mougeotia sp.
and Spirogyra sp.
03/2004 08/2004 03/2005 08/2005
CyanobacteriaeCyanobacteria n.d.
3.306DiatomeaePennalesAsterionella formosa < 4Fragilaria sp.
< 4 72F. ulva var. acus 18 22Navicula sp < 4 < 4Nitzschia
sigmoides 4Pennales n.d. < 4 < 4 91 4CentralesMelosira sp.
< 4Aulacoseira granulata 117 49 108Aulacoseira sp.
108DinophyceaePeridinium sp. < 4 4Gymnodinium sp. 9 <
4Cryptophyceae 4Rhodomonas minuta 9Rhodomonas sp. 90Cryptomonas sp.
9 4 13 31EuglenophyceaeTrachelomonas volvocina 1.054 22 40
9Trachelomonas verrucosa 31 13
4ChlorophyceaeVolvocalesChlamydomonas sp. 81 40
439ChlorococcalesMonoraphidium komarkovae 251 58Oocystis lacustris
157 175 27Oocystis naegeli 9 4 22 9Oocystis parva 117Chlorella sp.
< 4Lagerheimia wratislaviensis 4Planktosphaeria gelatinosa <
4Pediastrum boryanum 144Scenedesmus linearis 36Scenedesmus
quadricauda 332Scenedesmus acuminatus 72Tetraedron minimum
4Botryococcus braunii < 4Dictyosphaerium ehrenbergianum <
4Lagerheimia longiseta 18Neglectella sp. 13Schroederia setigera
< 4CladophoralesCladophora sp. 583ZygnematalesMougeotia sp.
193Spirogyra sp. 108DesmidialesCosmarium sp. < 4 4Abundance (n
mL1) 1094 193 3485 363Biomass (m3 mL1) 784,000 218,000 3,808,000
118,000
247,900*)Depth of abundance maxima 075 m 075 m 130140 m 4575
m
-
116 G. GUNKEL and C. BEULKER
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
Table 4). In general, the abundance (< 1100 cells mL1) and
biomass (< 400 000 m3 mL1) were small, and corresponded with the
oligotrophic character and the Secchi depth of 13.0 14.9 m. On one
occasion the occurrence of Mougeotia and Spirogyra at 140 m depth
led to a maximum biomass of 3 800 000 m3 mL1.
The vertical algae distribution showed continuously occurrence
of algae in the whole water column down to 140 m, and cell number
as well as biovolume did not had any significant maximum in the
epilimnic zone, while other parameters such as E7, conductivity,
pH, CO2 and the Ca saturation index showed an analogous
stratification to temperature (Fig. 8).
Algal species were regularly found in deep water and maximum
abundance occurred down to 75 m, and an extreme maximum abundance
of Mougeotia and Spirogyra occurred once at 130140 m (Fig. 9). The
depth distribution during stratification periods point out clearly
that most species occurred in the whole water column
(Chlamydononas, Scenedesmus linearis, Oocystis lacustris,
Trachelonomas volvocina, T. verrucosa, Aulacoseira sp.), but a few
species were found only in the upper water body down to 80 m such
as Scenedesmus acuminatus, Lagerheimia longiseta and Monoraphidium
komarkovae. Other species, mostly filamentosus forms occurred only
in depth >> 80 m such as Aulacoseira granulata, Pedias-trum
boryanum, Mougeotia sp. and Spirogyra sp. (Fig. 9).
In the plankton some iron-oxidising bacteria such as Siderocapsa
occurred frequently, and Metallogenium-like structures were also
found.
The zooplankton biocoenosis of Lake Cuicocha was formed by two
species, the filter feeder Daphnia pulex pulex and the carnivorous
Metacyclops mendocinus (Table 3). Zoo-
Figure 8. Vertical stratification of physical-chemical water
parameters (redox potential, conductivity, temperature, pH, CO2,
calcite saturation index) and depth distribution of plankton
(phytoplankton cell number and biomass, abundance of Daphnia pulex,
Metacyclops mendocinus and Hyalella cf. dentata)
in the 148 m depth position of Lake Cuicocha, 18.3.2004.
-
Limnology of a Crater Lake 117
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
plankton abundance was low and the maximum number of animals
amounted to 7 n L1, the mean abundance was 1.8 n L1 (Metacyclops
mendocinus) and 0.7 n L1 (Daphnia pulex pulex), respectively. The
vertical distribution of the zooplankton stretched down to a depth
of 140 m (Fig. 8).
In the pelagic water body, the invertebrate found in the
littoral zone, Hyalella cf. dentata was found regularly. Hyalella
occurred in the whole water column with a few animals per litre
(Fig. 8), but was frequent at the lake bottom at 78 m depth, proved
by under water filming; Hyalella was also found at 148 m depth in
the plankton net samples, however, at this depth no intensive
filming was performed.
In the pelagial of Lake Cuicocha no fish were found by use of
sonar, Astroblepus ubidiai occurred only in the littoral zone. A
few decades before, the trout (Oncorhynchus mykiss) was introduced
into Lake Cuicocha as well as into the nearby Lake San Pablo and
Lake Mojanda, however, in Lake Cuicocha the population has broken
down, while in the other lakes, the fish populations have
increased.
3.6. Sediments and Sediment Resuspension
The sediments in Lake Cuicocha formed a very thin layer of up to
1015 cm on volcanic rocks a few decimetres in diameter, proved by
underwater photography (see GUNKEL et al., 2008). The sediments
consisted mainly of mineral compounds and sedimented diatom
frus-tules. These sediments were poor in Ca (1020 g kg1 ds, = dry
substance) but rich in P (2446 g kg1 ds). The sediments were
anaerobic with sulphide formation in the form of
Figure 9. Vertical distribution of some dominate algae species
in the 148 m depth position of Lake Cuicocha (17.03.2005).
-
118 G. GUNKEL and C. BEULKER
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
pyrites, only the surface layer was oxidized down to a few
millimetres. In particular, the sur-face layer in the 78 m deep
area consisted of brown flocs a few millimetres in diameter.
The sediment traps showed extraordinary high sedimentation rates
with the higher rates in deeper exposed traps (13 mm fresh material
within 2 weeks), in contrast to the low abundance of plankton in
the pelagial. The trapped material consisted of the same type of
flocs as were found at the lake bottom, recognizable by a similar
chemical composition. The settled material of both origins was poor
in Ca (12 g kg1 ds) but rich in Fe (74 g kg1 ds) and consisted of
algae, detritus and masses of iron oxidising bacteria (Fig. 10).
Observations with
Figure 10. Sediment flocs in Lake Cuicocha, detritus, diatom
cell is embedded in flocs with iron- oxidising bacteria, SiO2(H2O)n
precipitations and filamentosus fungi structures, free water, 40 m
depth,
18.08.2005, in situ filtration.
Figure 11. Formation of amorphous SiO2(H2O)n in Lake Cuicocha
sediment on Nuclepore filters, 148 m depth, 31.08.2004; EDS
analyses verified the Si composition with traces of Al and Fe.
-
Limnology of a Crater Lake 119
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
sonar and the underwater camera registered resuspension of
sediment and sediment flocs by gas bubbles rising up to 20 m.
Sediment forming processes were supported by two linked water
chemistry processes, the precipitation of polymeric Al(OH3)n and
the formation of amorphous SiO2(H2O)n; due to the atomic
characteristics mixed precipitations are common. In the water as
well as in the sedi-ment, Si precipitation occurred and formed
amorphous structures (Fig. 11). Al precipitation occurred in the
water column due to a decrease in pH to about 7, which meant that
down-welling of epilimnic water leading to oversaturation of the Al
polymer, and gelatinous Al polymers as well as Al microcrystals
like gibbsite were observed (GUNKEL et al., 2008b). Both processes
were frequent and promoted the coagulation of flocs in the water,
formed by detritus, bacteria and living algal cells.
4. Discussion
4.1. Characteristics of Caldera Lakes
Crater lakes are a special type of lake, which are classified by
the lake formation pro-cess (caldera as an intrusion of the magma
chamber of a volcanic dome, maar as explosive craters), without
consideration of any geological or climatic settings. Thus caldera
lakes pos-sess a wide scattering of ecological characteristics. The
physical and chemical parameters of caldera lakes were summarized
by PASTERNAK and VAREKAMP (1997) and VAREKAMP et al. (2000).
Nevertheless caldera lakes are characterized by properties of
ecological interest: caldera lakes are young lakes in a state of
development, mostly located at higher altitudes; only maar lakes
and some marine atoll lakes are found in the plain. The watershed
is nor-mally covered by young volcanic deposits, with weathering
processes and the beginning of soil formation. The watershed is
small, the lakes do not have a regular outflow; caldera lakes have
no connection with the ground water table. Especially in young
caldera lakes the accumulated sediment layer is very small.
Lake Cuicocha is characterized by the young age of about 500
years with weathering of the rocks in the water shed and the lake
succession processes such as rock falls and land slides from the
crater rim, deposition of sediments, colmation and nutrient
accumulation, and water chemistry is strongly influenced by these
processes as well as by post volcanic activities such as gas
emission and hydrothermal water inflow. One effect is the
resuspension of sediments is a process which determines the water
chemistry as well as the development of the phytoplankton
(WEYHENMEYER, 1998).
The colmation of the lake basin is unstable or insufficient due
to the small sediment layer and the post-volcanic activities.
Triggered by an earthquake in 1987, the lake colmation layer was
destroyed and from this time a decrease in the water level of about
30 cm per year occurred. Water percolation leads to a higher risk
of phreatic-magmatic eruption, meaning an explosive eruption, when
the infiltrating water contacts the magma of hot rocks (MASTIN and
WITTNER, 2000; GUNKEL et al., 2008).
Post-volcanic emissions including gases or hydrothermal springs
are frequent in volcanic lakes and affect lake stratification and
water chemistry. In Lake Cuicocha the ionic content of the water
was increased, compared with a nearby inactive caldera lake, Lake
Mojanda, which clearly points to the high significance of the
hydrothermal inflow.
4.2. Physical Characteristics of Lake Cuicocha: Mixing
Processes
Besides wind-induced water cycling, lake mixing processes of
Lake Cuicocha are deter-mined by post-volcanic activities like the
emission of volcanic gases, hydrothermal underwa-
-
120 G. GUNKEL and C. BEULKER
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
ter springs and a heat flux via the lake floor. Gas emissions
and hydrothermal water springs at the lake floor lead to a mixing
of the lake due to billows with upraising gas bubbles or divergent
currents of heated water. These processes can be detected only by
the application of special monitoring (use of divers, sonar,
underwater camera and a lake profiler).
Thus the stability of the thermal stratification of Lake
Cuicocha is diminished, and atelo-mixis occurs, that means daily or
frequent periodic mixing down into the hypolimnion e.g. by
nocturnal cooling in periods of low density gradients (TALLING,
1969). Increased losses of phytoplankton by sedimentation take
place, and the atelomixis leads to a more oligotrophic status of
the lake (TAVERA and MARTNEZ-ALMEIDA, 2005), who postulated that
atelomixis was the driving force in determining phytoplankton
composition. In the nearby Lake San Pablo, atelomixis was proved
and led to hypolimnic phytoplankton maxima (GUNKEL and CASALLAS,
2002, 2002a).
4.3. Water Chemical Characteristics of Lake Cuicocha
Water chemistry of the caldera lake is determined by rain water
interacting with vol-canic deposits in the catchment area and
represents a new (young) terrestrial-aquatic linked ecosystem. The
export of nutrients and other ions from the watershed, being
composed of volcanic deposits, lava, volcanic rocks or ashes, which
undergo an intensive mineralization and soil forming process
determine the ionic composition. Typical for the Andean zone is the
high export of Al and Si as characteristic compounds of the
andesites, while the content of Ca must be regarded as low. Lake
Cuicocha, only a few hundred years old, reflects these conditions:
Al and Si are increased, Al polymerisation and microcrystal
formation (GUN-KEL et al., pers. com.), amorphous SiO2(H2O)n
precipitation as a process of sediment build-ing. AlSi
polymerisation processes promote the formation of flocs and it must
be assumed that the stability of flocs is also increased.
Lake Cuicocha is an oligotrophic lake, due to low P and N input
from the watershed with young soils, the andisols (ZEHETNER et al.,
2003) and due to no accumulation of nutrients in the young lake.
These nutrients are further reduced by intensive precipitation with
iron oxides being a consequence of hydrothermal water inflow.
In Lake Cuicocha CO2 accumulation with pH change and changes in
the lime carbonic acid equilibrium occur.
The water of Lake Cuicocha has a low Ca and Mg concentration,
however, in the epilim-nion, the calcite saturation index regularly
shows a positive value, meaning oversaturation, whereas the
hypolimnion values decrease to levels < 0. Travertine formation
occurs as a biological CaCO3 precipitation in an extended biofilm
of a few millimetres thickness with high abundance of diatoms and
median abundance of cyanobacteria. It is of interest that only
traces of magnesium occurs in the precipitates. The productivity of
the biofilm, which stretched down to more than 15 m is very high
(sampling at deeper areas of the crater flanks was not possible due
to diving limitations). Part of the travertine crusts are eroded by
wave action and form a fine carbonate sand in the littoral zone,
which sink to the lake floor, where it is dissolved. This Ca cycle
Ca precipitation in the epilimnic zone, sedimentation of Ca sands,
re-dissolution in the hypolimnic zone and introduction of dissolved
Ca in the epilimnion by regular partial lake mixing is the basis of
the intensive development of the biofilm with travertine
formation.
Similar travertine formation is a frequent process at the
crop-out of hydrothermal springs, where the water flow is a source
of Ca, Mg and corresponding anions. In lakes it has also been
observed at a small rate, induced by photosynthetic pH changes
(GOLUBI, 1969). In crater lakes intensive travertine formation is
known e.g. from Lake Quilotoa, Ecuador (up to 50 cm thickness, own
observations) and the Ries crater, Germany (RIDING, 1979), but in
Lake Quilotoa an abiotic formation process with extreme Ca
over-saturation and aragonite
-
Limnology of a Crater Lake 121
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
precipitates can also become of high significance, and under
these conditions CaMg-car-bonate precipitates occur.
In the hypolimnion the Ca concentration is in the range of
undersaturation, due to decreas-ing pH caused by increased CO2
concentrations. This Ca undersaturation in the hypolimnion leads to
re-dissolving of biogenic precipitated Ca, resulting in a very
small amount of Ca in the lake sediments of < 1%. Consequently
the sediment building rate is reduced and the lake colmation
process is hindered and slowed, and the stability of the colmating
layer must be regarded as low.
4.4. Biocoenosis of Lake Cuicocha
The phytoplankton community is characterized by a low diversity,
which corresponds to the nearby glacial Lake San Pablo (Lake San
Pablo: 31 species, Cuicocha: 37 species), but only 8 species occur
in both lakes (GUNKEL and CASALLAS, 2002). Most of the species in
Lake Cuicocha are wide spread, endemic species of Latin America
were not found, espe-cially of the group of diatoms
(LANGE-BERTALOT, pers. com. after sample revision).
The dominant species in 2004/2005 are Aulacoseira granulata,
Fragilaria ulva var. acus, Trachelomonas volvocina, Oocystis
lacustris, O. parva and Clamydomonas sp., which have to be
classified as meso- to eutraphent species (trophic value 33.5) with
low to median saprobe character (saprobic value 2), these species
are cosmopolite with a wide ecological tolerance (LUB 2005). Only
Monoraphidium komarkovae is a more oligotraphent species (trophic
value 1.5). The more eutrophic species in Lake Cuicocha do not
correspond to the high water transparency and the low phosphorous
concentrations, but it must be considered, that the algae biomass
is low and the continuity is less.
Of high interest is the vertical distribution of the algae with
a significant distribution below the epilimnic zone, the calculated
euphotic depth is 1820 m (TILZER, 1988), but most of the algae we
found in depth > 20 m; this is a phenomenon already investigated
in the nearby Lake San Pablo, and atelomixis as deep nocturnal
mixing caused the temporary residence in the aphotic zone (GUNKEL
and CASALLAS, 2002, 2002a). A significant part of the population is
even in more than 80 m depth, some species such as Aulacoseira
granu-lata, Pediastrum boryanum, Mougeotia sp. and Spirogyra sp.
had a maximum distribution in 130148 m. This paradox seems to be a
consequence of the intensive flocculation processes due to Al and
Si precipitation, the flocs trap algae cells and promote
sedimentation processes and lead to an accumulation of algae near
the sediment.
The biocoenosis of the zooplankton is characterized by some
ubiquists which are highly abundant, however, the biodiversity is
strictly reduced. Only one filter-feeding Phyllopoda, one
carnivorous Copepoda and one Malacostraca form a pelagic food web
with low phyto-plankton and bacterioplankton at the primary trophic
level. Top predation by fish is missing, a catfish being the only
inhabitant of the littoral zone. A very restricted food web was
also observed in the nearby Lake San Pablo, a glacial lake of 38 m
depth (CASALLAS and GUNKEL, 2001). Without doubt high mountain
lakes in this tropical area show infrequent colonization by aquatic
organisms, and the fauna and flora are characterized by introduced
species (direct introduction of fish, and by ballast water from
water planes). Tropical Middle America is a barrier for migration
to cold water species from North to South America, and inversion of
species occurred only from the south of South America via the Andes
mountain range.
Little is known about the occurrence of Hyalella cf. dentata, a
species of the new world, and even the description of some Hyalella
species must be revised (GONZALEZ and WATLING, 2002). Hyalella cf.
dentata found in many South American lakes, even in salt lakes, and
in some North American lakes (GONZALEZ and WATLING, 2002). However,
information on the habitat and ecology of this organism is
insufficient. Although Hyalella cf. dentata is a her-bivorous
littoral species, its occurrence in the pelagial and at the lake
bottom was regularly
-
122 G. GUNKEL and C. BEULKER
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
registered at 148 m, and at the 78 m depth a high abundance was
observed. However, due to the extension of the lake, no leaves or
similar particulate organic material can accumulate in the pelagial
and on the lake floor. It can be assumed, that filamentous algae
such as the desmids Mougeotia and Spirogyra, observed in deep water
in high biomass (Table 4), play a role in the feeding behaviour of
these organisms.
This investigation did not include any microbiological studies,
however, some interesting effects were observed. Iron-oxidizing
bacteria were frequent, mostly of the genus Side-rocapsa as well as
Metallogenium-like structures (KLAVENESS, 1977), which were
classified by EMERSON et al. (1989) as manganese-oxidizing fungi.
These bacteria, mainly the filamen-tous forms of Metallogenium-like
structures promote the build-up of flocs and increase their
stability. Flocs are cycled in the water body by sediment
resuspension and in turn strongly influence water quality. The role
of flocs in chemical and biological processes in lakes must be
regarded as very important, however, available knowledge is still
insufficient (WEYHEN-MEYER, 1998). In Lake Cuicocha flocs played a
significant role in the occurrence of algae, serving as a substrate
for adhesion and as a nutrient source in the oligotrophic lake.
The lack of fish within the pelagial may have been caused by the
high CO2 concentration, however, it should be noted that in the
nearby high Andean lakes only a few species such as a hybrid
(Carassius carassius, C. auratus) called carp, trout (Oncorhynchuss
mykiss) and pike (Micropterus salmoides) occur. Fish biodiversity
is extremely low in these lakes and little is known about the
native biocoenosis (e.g., Atroblepus ubidiai), because trout, carp
and pike were introduced into all these lakes.
The littoral macrophytes had a high biomass, and invertebrates
in the littoral zone were frequent, whereas the planktic community
was of low abundance and biomass. This increased importance of the
littoral biocoenosis has also been observed in other tropical high
moun-tain lakes, and one reason for this seems to be the increased
losses of pelagic species due to deep diurnal mixing processes
(GUNKEL and CASALLAS, 2002, 2002a). In Lake Cuicocha this process
is forced by the formation of flocs and the coagulating effect of
polymeric Al precipitation, by the formation of amorphous
SiO2(H2O)n and filamentous Metallogenium like structures, and by
the stabilization of flocs by the abundant occurrence of
Siderocapsa colonies.
5. Acknowledgements
This study is part of a cooperation project of the Berlin
University of Technology, Germany, Dept. Water Quality Control with
support from the Dept. of Environment and Marine Science and
Technology (VWS), Dipl. Geol. B. GRUPE, and the Central University
of Quito, Facultad de Ingeniera en Geologa, Minas, Petrleos y
Ambiental (FIGEMPA), Ecuador, Ing. F. VITERI. It was financed by
the Deutsche Forschungsgemeinschaft, Germany (DFG, German
Scientific Society) and the Ministry for Economic Development and
Cooperation, Germany (BMZ). Logistic support was provided by the
National Park of Cotacachi, Ibarra, Ecuador. Dr. P. CASPER,
Leibniz-Institute of Freshwater Ecology and Inland Fisheries,
Germany, carried out gas analyses. Dr. GABRIELE GUNKEL, Berlin,
Germany, carried out the phytoplank-ton analyses. Dr. CORREOSO,
Catholic University of Quito carried out the systematic
identification of the invertebrates. 14C-analyses of soils were
carried out by the Leibnitz Institute for Applied Geosciences,
Hannover, Germany.
6. References
AGUILERA, E., G. CHIODINI, R. CIONI, M. GUIDI, L. MARINI and B.
RACO, 2000: Water chemistry of Lake Quilotoa (Ecuador) and
assessment of natural hazards. J. Vol. Geotherm. Res. 97:
271285.
AGUILERA, X., ST. DECLERCK, L. D. MEESTER, M. MALDONADO and F.
OLLEVIER, 2006: Tropical high Andes lakes: A limnological survey
and an assessment of exotic rainbow trout (Oncorhynchus mykiss).
Limnologica 36: 258268.
-
Limnology of a Crater Lake 123
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
ARMIENTA, M. A., S. DE LA CRUZ-REYNA and J. L. MACAS, 2000:
Chemical characteristics of the crater lakes of Popocatpetl, El
Chichon, and Nevado de Toluca volcanoes, Mexico. J. Vol. Geotherm.
Res. 97: 105125.
CAMERON, W. and G. L. LARSON, 1993: Limnology of a caldera lake
influenced by hydrothermal proc-esses. Archiv Hydrobiol. 128:
1338.
CASALLAS, J. and G. GUNKEL, 2001: Algunos aspectos limnolgicos
de un lago altoandino, el Lago San Pablo, Ecuador. Limnetica 20:
2946
CHEN, C. T. and F. J. MILLERO, 1986: Precise thermodynamic
properties for natural waters covering only the limnological range.
Limnol. Oceanogr. 31: 657662.
CHRISTENSON, B. W., 1994: Convection and stratification in
Ruapehu Crater Lake, New Zealand: Impli-cations for Lake Nyos-type
gas release eruptions. Geochemical Journal 28: 185197.
CRAWFORD, G. B. and R. W. COLLIER, 1997: Observation of a
deep-mixing event in Crater Lake, Oregon. Limnol. Oceanogr. 42:
299306.
DEV, 1981: Bestimmung von Borat Ionen. Deutsche
Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung.
Loseblattsammlung. Wiley. 6 p.
DEV, 2005: DIN 38409-H7. Bestimmung der Sure- und Basekapazitt.
Deutsche Einheitsverfahren zur Wasser, Abwasser- und
Schlammuntersuchung. Loseblattsammlung. Wiley.
DEV, 2007: Anleitung fr die Zhlung von Phytoplankton mittels der
Umkehrmikroskopie (Utermhl-Technik). DIN EN 15204. Deutsche
Einheitsverfahren. VCH, Weinheim, Loseblattsammlung. 56 pp.
EMERSON, D., R. E. GAREN, and W. C. GHIORSE, 1989: Formation of
Metallogenium-like structures on manganese-oxiding fungus. Arch.
Microbiol. 151: 223231.
EN ISO 11905-1, 1998: Wasserbeschaffenheit Bestimmung von
Stickstoff Teil 1: Bestimmung von Stickstoff nach oxidativem
Aufschlu mit Peroxodisulfat (ISO 11905-1:1997); Deutsche Fassung EN
ISO 11905-1. Beuth Verlag GmbH Berlin, 18 pp.
EN ISO 15681-1, 2004: Wasserbeschaffenheit Bestimmung von
Orthophosphat und Gesamtphosphor mittels Flieanalytik (FIA und CFA)
Teil 1: Verfahren mittels Flieinjektionsanalyse (FIA) (ISO
15681-1:2003); Deutsche Fassung EN ISO 15681-1. Beuth Verlag
Berlin, 29 pp.
GOECKE, K., 1997: Basic morphometric and limnological properties
of Laguna Hule, a caldera lake in Costa Rica. Revista de Biologia
Tropical 4445 (3-1 PART B): 537548.
GOLUBI, S., 1969: Cyclic and noncyclic mechanisms in the
formation of travertine. Verh. Internat. Verein. Limnol. 17:
956961.
GONZALEZ, E. R. and L. WATLING, 2002: Redescription of Hyalella
azteca from its type locality, Vera Cruz, Mexico (Amphipoda:
Hyalellidae). J. Crustracean Biol. 22: 173183.
GREEN, J., 1995: Altitudinal distribution of tropical planctonic
Cladocera. Hydrobiologia 307: 7584.
GRUPE, B., G. GUNKEL, F. VITERI, C. BEULKER and C. VON
HILLEBRANDT, 2008: The Cuicocha/Cotacachi volcano complex
(Ecuador): Recent history and secondary caldera manifestation.
Journal of South American Earth Sciences (subm.).
GUNKEL, G. and J. CASALLAS, 2002: Limnological of an equatorial
high mountain lake, Lago San Pablo, Ecuador: Limitation of lake
productivity by deep diurnal mixing. Proc. Int. Ass. Theoret.
Applied Limnol. 28: 18181823.
GUNKEL, G. and J. CASALLAS, 2002a: Limnology of an equatorial
high mountain lake Lago San Pablo, Ecuador: The significance of
deep diurnal mixing for lake productivity. Limnologica 32:
3343.
GUNKEL, G., 2003: Limnologa de un lago tropical de alta montaa,
Lago San Pablo-Ecuador: Caracte-rsticas de los sedimentos y tasa de
sedimentacin. Revista Biologa Tropical 51: 381390.
GUNKEL, G., C. BEULKER, B. GRUPE and F. VITERI, 2008: Hazards of
volcanic lakes: Analysis of Lakes Quilotoa and Cuicocha, Ecuador.
In: P. FABIAN and P. LAGOS (eds). The role of Geophysics in Natural
Disaster Prevention. Advances in Geosciences 14: 2933.
INECEL, 1983: Sntesis de los Estudios de Aprovechamiento de los
Recursos Geotrmicos en el Area de Chalupas. Instituto Nacional
Electrificacion. Unpublished Interim Report, pp. 112.
KIERSCH, B., R. MHLECK and G. GUNKEL, 2004: Las macrofitas de
algunos lagos alto andinos del Ecuador y su posible utilizacin como
bioindicatores de eutrofizacin. Revista Biologia Tropical 52:
829837.
KINZIE, R. A. III, A. T. BANASZAK and M. P. LESSER, 1998:
Effects of ultraviolet radiation on primary productivity in a high
altitude tropical lake. Hydrobiologia 385: 2332.
-
124 G. GUNKEL and C. BEULKER
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
KLAVENESS, D., 1977: Morphology, distribution and significance
of the manganese-accumulating micro-organism Metallogenium in
lakes. Hydrobiologia 56: 2533.
KUSAKABE, M. (ed), 1994: Geochemistry of Crater Lakes. Special
Issue. Geochem. J. 28: 137306.LARSON, G. L., 1989: Geographical
distribution morphology and water quality of caldera lakes. A
review. Hydrobiologia 171: 2332. LARSON, G. L., 1996:
Development of a 10-year limnological study of Crater Lake National
Park,
Oregon, USA. J. Lake Reservoir Management 12: 221229.LARSON, G.
L., C. D. McINTIRE, M. HURLEY and M. W. BUKTENICA, 1996:
Temperature, water chemistry,
and optical properties of Crater Lake. J. Lake Reservoir
Management 12: 230247.LE GUERN, F. and G. E. SIGVALDASON, 1989: The
Lake Nyos event and natural CO2 degasing I. Special
issues. J. Volcanol. Geothermal. Res. 39: 23.LE GUERN, F. and G.
E. SIGVALDASON, 1990: The Lake Nyos event and natural CO2 degasing
II. Special
issue. J. Volcanol. Geothermal. Res. 42: 4.LFFLER, H., 1964: The
limnology of tropical high-mountain lakes. Verh. Internat. Verein.
Lim-
nol. XV: 176193.SCHAUMBURG, J., M. COLLING, L. SCHLSSER, B. KPF
and F. FISCHER, 2005: kologische Typisierung
des Phytoplanktons. Informationsbericht und Datenbank.
Bayerische Landesamt fr Wasserwirtschaft, Heft 3/05, Mnchen.
MARTINI, M., L. GIANNINI, F. PRATI, F. TASSI, B. CAPACCIONI and
P. IOIZZELLI, 1994: Chemical characters of crater lakes in the
Azores and Italy: the anomaly of Lake Albano. Geochem. J. 28:
173184.
MASTIN, L. G. and J. B. WITTER, 2000: The hazards of eruptions
through lakes and seawater. J. Vol. Geotherm. Res. 97: 195214.
MOTHES, P. and M. L. HALL, 1991: El paisaje interandino y su
formacin por eventos volcanicos de gran magnitude. Estudios de
Geografa 4: 1938.
NELSON, P. O., J. F. REILLY and G. L. LARSON, 1996: Chemical
solute mass balance of Crater Lake, Oregon. J. Lake Reservoir
Management 12: 248258.
OHBA, T., J. I. HIRABAYASHI and K. NOGAMI, 1994: Water, heat and
chloride budgets of the cater lake, Yugama at Kusatsu-Shirane
volcano, Japan. Geochemical Journal 28: 217231.
PADISK, J. and R. ADRIAN, 1999: Biovolumen. In: W. v. TMPLING
and G. FRIEDRICH (eds.). Biolo-gische Gewsseruntersuchung. Gustav
Fischer, Jena. 334368. pp.
PASTERNACK, G. B. and J. C. VAREKAMP, 1997: Volcanic lakes
systematics I. Physical constraints. Bul-letin Volcanology 58:
528538.
RIDING, R., 1979: Origing and diagenesis of lacustrine algal
bioherms at the margin of the Ries crater, Upper Miocene, southern
Germany. Sedimentology 26: 645680.
ROLDAN, G. and E. RUIZ, 2001: Development of Limnology in
Colombia. In: R. G. WETZEL, and B. GOPAL, (eds). Limnology in
Developing Countries 3: 69119.
RONDE, C. E. J., P. STOFFERS, D. GARBE-SCHNBERG, B. W.
CHRISTENSON, B. JONES, R. MANCONI, P. R. L. BROWNE, K. HISSMANN, R.
BOTZ, B. W. DAVY, M. SCHMITT and C. N. BATTERSHILL, 2002:
Discov-ery of active hydrothermal venting in Lake Taupo, New
Zealand. J. Vol. Geothermal Res. 115: 257275.
SAUER, W., 1971: Geologie von Ecuador. Beitrge zur Geologie der
Erde 11. Borntraeger, Berlin, 316 pp.
SCHABETSBERGER, R., G. DROZDOWSKI, I. DROZDOWSKI, C. C. JERSABEK
and E. ROTT, 2004: Limno-logical aspects of two tropical crater
lakes (Lago Biao and Lago Loreto) on the island of Bioko
(Ecuatorial Guinea). Hydrobiologia 524: 7990.
SCHMID, M., M. HALBWACHS, B. WEHRLI and B. WEST, 2005: Weak
mixing in Lake Kivu: New insights indicate increasing risk of
uncontrolled gas eruption. Geochemistry, Geophysics, Geosystems
6/7: 111.
TALLING, J. F., 1969: The incidence of vertical mixing, and some
biological and chemical consequences in tropical African lakes.
Verh. Internat. Verein. Limnol. 17: 9981012.
TAVERA, R. and V. MARTNEZ-ALMEIDA, 2005: Atelomixis as a
possible driving force in the phytoplank-ton composition of
Zirahzn, a warm-monomictic tropical lake. Hydrobiologia 533:
199208.
TILZER, M. M., 1988: Secchi disk chlorophyll relationships in a
lake with highly variable phytoplank-ton biomass. Hydrobiologia
162: 163171.
US APHA, 1998: Standard methods for the examination of water and
wastewater. US American Public Health Ass., 20. ed.
-
Limnology of a Crater Lake 125
2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.revhydro.com
VAREKAMP, J. C., G. B. PASTERNACK and G. L. ROWE, 2000: Volcanic
lake systematics II. Chemical Contraints. J. Volcanol. Geotherm.
Res. 97: 161179.
VZQUEZ, G., M. E. FAVILA, R. MADRIGAL, C. M. OLMO, A. BALTANS
and M. A. BRAVO, 2004: Limnol-ogy of crater lakes in Los Tuxtlas,
Mexico. Hydrobiologia 523: 5970.
VDLUFA, 1991: Die Untersuchung von Bden. Bd. 1, 1. + 2., ISBN
3-922712-42-8. VDLUFA Verlag, Speyer.
VLEZ-ESPINO, L. A. and A. G. FOX, 2005: Demographic and
environmental influences on life-history traits of isolated
populations of the Andean catfish Astroblepus ubidiai.
Environmental Biology of Fishes 72: 189204.
VON HILLEBRANDT, C. and M. L. HALL, 1988: Mapa de los Peligros
Volcnicos Potenciales Asociados con el Volcn Cuicocha. Esc. 1 :
50.000. Escuela Politcnica Nacional, Quito. Instituto Geogrfico
Militar.
WEYHENMEYER, G. A., 1998: Resuspension in lakes and its
ecological impact a review. Arch. Hy drobiol., Spec. Issues.
Advanc. Limnol. 51: 185-200.
WOOD, R. B., R. M. BAXTER and M. V. PROSSER, 1984: Seasonal and
comparative aspects of chemical stratification in some tropical
crater lakes in Ethiopia. Freshw. Biol. 14: 551574.
ZEHETNER, F., W. P. MILLER and L. T. WEST, 2003: Pedogenesis of
volcanic ash soils in Andean Ecua-dor. Soil Sci. Soc. Am. J. 67:
17971809.
Manuscript received March 6th, 2008; revised July 11th, 2008;
accepted August 19th, 2008