Phytoplankton Diversity in Lakes of the McMurdo Dry
Valleys, Antarctica
Amy L. Chiuchiolo1, John C. Priscu1, Rachael M. Morgan-Kiss2
1Department
2Delaware
of Land Resources and Environmental Sciences, Montana State University
Biotechnology Institute & College of Marine Studies, University of Delaware
MCM Dry
Valleys
cyanobacteria:
Phornidium sp.
www.geographicguide.com
• Chl-a profiles typically showed maxima in the high light, low nutrient regions
immediately beneath the ice cover and just above the nutrient rich deep waters below
the chemoclines.
Additional samples were
collected and analyzed for
extracted chlorophyll-a for
comparison with results
from the submersible
spectrofluorometer.
Introduction
Long-term research in the McMurdo Dry Valleys, Antarctica has led to questions
concerning the relation between biodiversity and ecosystem function. A major theme
of the latest MCM LTER proposal (2005-2011) is to understand both how the
environment controls the diversity of organisms, and how diversity itself controls the
functioning of the MCM ecosystem. Lakes in this ecosystem are likely hundreds of
thousands of years old and lie at the end of the hydrologic continuum, representing a
repository of past conditions within the MCM. As the key phototrophic organisms in the
MCM ecosystem, phytoplankton have an essential role as primary producers of new
carbon (Fig 1). We present phytoplankton diversity data collected from five lake basins
Chlorophyta:
within the MCM study area (Fig 2).
Chlamydomonas sp.
• The highest chl-a levels occurred in the deeper waters associated with the chemocline
in lakes where a chemocline was present.
• Chloropytes, cryptophytes and chrysophytes were abundant is all lakes;
cyanobacteria were usually absent or present in relatively low numbers.
Fig 4. Principle of under water spectrofluorometry. Algal
Chlorophyll-a is excited with light of five LEDs (emission
wavelength 450 nm, 525 nm, 570 nm, 590 nm, 610 nm).
An iterative gaussian fit weighted with the standard
deviations of the normal spectra facilitates the estimation of
the distribution of the spectral groups. Dissolved yellow
substances measured at 370 nm are used to correct for
background fluorescence in the algal algorithms. Algal group
concentrations are given in μg chl-a/L water sample.
• Chlorophytes dominated the upper freshwater layers in all lakes; chrysophytes and
cryptophytes were relatively more abundant in the deep chl-a layers associated with the
upward diffusion of nutrients through the chemoclines.
Results III
Phytoplankton diversity in the MCM lakes – Cluster Analysis
Results I
Lake Vanda
In situ spectrofluorometer and extracted chl-a: A Comparison
chl-a (µg/L)
0
2
4
6
Fig 5. Depth profile of algal concentration
(sum of all detected algae classes)
determined using the spectrofluorometer
(blue) and extracted chl-a (green)
concentration in East Lobe Bonney.
8
0
Lakes Bonney,
Hoare, Fryxell
Chryptophyta:
Chryptomonas sp, ELB
Fig 2. Location of the MCM Dry Valleys (77 °S, 163°E)
(inset) and the study lakes in the Wright and Taylor Valley.
Depth (m)
10
Chrysophyta:
Ochromonas sp., FRX
30
Fig 1. Dominant groups of
phytoplankton found in the
MCM dry valley lakes.
40
The dry valley lakes contain highly stable water columns resulting from perennial ice
covers, low advective stream inflow, and strong vertical chemical gradients. Each
lake contains unique physical and chemical gradients (Table 1; Fig 3) which create
distinct environments for phytoplankton populations. Temperature, salinity and nutrient
conditions vary widely within each lake, and low light conditions (1-3% of surface
PAR) persist throughout the 6 month growing season.
Size Ice Cover Max Depth
(km2)
(m)
(m)
Fryxell
7.1
6.0
21
Hoare
1.9
3.8
34
E. Lobe
Bonney
W. Lobe
Bonney
Vanda
3.5
4.6
40
1.3
3.9
40
6.7
3.3
Physical Characteristics
stratified water column; brackish, nutrient rich,
anoxic deep water; N&P-deficient
weak density stratification; relatively freshwater,
oxic deep water; P-deficient
highly stratified water column with hypersaline,
nutrient-rich, sub-oxic deep water; P-deficient
Spectrofluorometer chl-a (µg/L)
18
Study Site
Lake
20
12
6
y = 0.7677x + 0.2898
r = 0.80
0
5
10
15
chl-a (µg/L)
0
2
4
6
8
chl-a (µg/L)
10
12
14
3
5
8
10
0.0
0.5
Ice
2
ICE
4
10
Depth (m)
Depth (m)
100
150
15
20
20
25
30
20
40
40
chl-a (µg/L)
4
6
2
10
0
0
0
5
5
8
Conductivity (mS/cm)
Salinity (PSU)
0
50
100
-2
0
2
Temperature (C)
4
-5
0
5
Temperature (C)
10
-10
0
10
Temperature (C)
10
10
15
15
150
Depth (m)
6
20
25
30
0
10
VAN
20
30
40
Ice
50
60
70
80
0
10
20
30
Temperature (C)
Fig 3. Temperature,
salinity and conductivity
profiles for each study
lake.
Methods
Phytoplankton diversity profiles were obtained during the
2004 and 2005 austral spring and summer using a
submersible spectrofluorometer, which
differentiates the four major groups of
phytoplankton in the lakes
(cyanobacteria, Chlorophyta,
Chrysophyta, Cryptophyta) based on
the chlorophyll-a fluorescence excitation
spectra of the light harvesting apparatus
(Fig. 4). All samples were collected from
the deepest portion of each lake.
East Lobe Bonney
15 Dec 05
35
40
5
10
30
West Lobe Bonney
18 Dec 05
40
Chlorophyta
cyanobacteria
Cryptophyta
0
Chrysophyta
Fig 7. Vertical profiles of chl-a contributed by
the major algal groups along with total chl-a. All
data obtained by in situ spectral fluorescence
measurements. Blue areas represent ice cover
depth. The brown horizontal line represents the
chemoclines (note that ELB has 2 distinct
chemoclines whereas no distinct chemocline is
present in Hoare).
0.5
1
0
total algae
10
20
30
40
50
60
Lake Vanda
29 Nov 04
Shannon-Weiner Index H’
FRX
Whole lake
1.248
Surface waters
0.903
Deep waters
1.085
HOR
ELB
WLB
VAN
1.011
0.504
0.701
0.980
0.624
0.549
0.559
0.569
1.046
0.693
0.690
0.607
•The deep chl layers had higher diversity than the upper chl layers in all lakes.
15
25
chl-a (µg/L)
Lake
•Lake Fryxell showed the highest phytoplankton diversity of all MCM lakes studied
reflecting the relatively high abundance of all algal groups. East and West Lobe of
Bonney showed the lowest diversity (2 algal groups dominated the entire water
column of Lake Bonney).
20
35
Phytoplankton diversity in the MCM lakes - Shannon-Weiner Diversity Index
Fig 9. Phytoplankton diversity based on in vivo pigment fluorescence. Pi values for Shannon-Weiner
Diversity Index H’ were calculated using the proportion of fluorescence for each group to total chl-a
averaged over all depths for “whole lake,” and at selected depths in the near-surface and deep
chlorophyll layer (FRX: 6 and 9m, HOR, ELB, WLB: 6 and 13m, VAN: 5 and 58m).
chl-a (µg/L)
8
Depth (m)
4
Depth (m)
2
Temperature (C)
10
Lake Hoare
11 Dec 05
25
30
0
Depth (m)
20
Lake Fryxell
7 Dec 05
0
30
8
15
12
18
15
6
10
8
10
20
4
5
14
WLB
10
2
Depth (m)
50
Depth (m)
Conductivity (mS/cm)
Salinity (PSU)
0
ELB
10
HOR
FRX
0
200
0
0
16
5
5
10
100
0
0
0
Depth (m)
0
1.0
16
0
Temperature
Conductivity (mS/cm)
Salinity (PSU)
Depth (m)
0
Salinity
Results IV
Results II
6
Conductivity
20
Extracted chl-a (µg/L)
In situ spectrofluorometer algal group profiles
highly stratified water column w/ strong
chemocline at ~65m, temps >20oC below
chemocline; N&P deficient
Conductivity (mS/cm)
Salinity (PSU)
•The 48 m chl-a peak in Lake Vanda formed its own branch, as a result of its almost
complete dominance by chrysophytes.
• The depth profile and correlation analysis show that the in situ spectrofluorometer data
are closely associated (p<0.01) with extracted chl-a. The slope of the relationship is
0.77 when all data are included indicating that the spectroflurometer on average yields
about 25% lower chl-a than that extracted in 90% acetone.
Table 1. Physical characteristics of dry valley lakes: Fryxell (FRX), Hoare (HOR), east lobe
Bonney (ELB), west lobe Bonney (WLB), Vanda (VAN).
Conductivity (mS/cm)
Salinity (PSU)
• Phytoplankton relationships in the surface waters were similar among lakes (they
clustered together) and differed from the phytoplankton relationships observed in the
deep waters. The exception was the sample from the deep chl-a maxima in Fryxell
(F9) dominated by chrysophytes, which clustered with the surface populations.
0
highly stratified water column with hypersaline,
nutrient-rich, anoxic deep water; P-deficient
80
Fig 8. Phytoplankton diversity based on in vivo pigment spectrofluorescence. Cluster analysis
performed using the ratio of chl-a for each algal group to total chl-a for depths representing the
specific chlorophyll maxima within each lake. Label denotes the lake (F=FRX, H=HOR, E=ELB,
W=WLB, V=Vanda) and depth included in the analysis.
Fig 6. Correlation comparing
total chl-a concentration
determined using the
spectrofluorometer with
extracted chl-a
concentration. Data from
Lakes Fryxell, Hoare,
Bonney and Vanda during
the 2004 and 2005 austral
spring and summer seasons.
1.5
Conclusions
• Phytoplankton diversity within the lakes of the McMurdo Dry Valleys reflects both
contemporary and environmental legacies of past events. Phytoplankton in surface
waters receive new nutrients from intermittent glacial stream inflow, release from the ice
cover, and internal recycling, and are exposed to relatively high irradiances, whereas
those in the deep chl-a layers receive upward diffusing nutrients from ancient nutrient
pools that formed as the lakes evolved, and are extremely light limited. These conditions
would likely lead to isolated deep water phytoplankton that interchange little between
lakes or with surface populations, and live on legacy nutrients (Priscu, 1995). Similarities
observed within the surface waters of each lake suggest that these populations are
responding to contemporaneous conditions, or that there is interchange of surface
organisms between lakes.
• Differences between deep water phytoplankton populations of Lake Vanda and the
Taylor Valley lakes implies that differences in the evolution of these lakes produced
different geochemical properties of deep waters. Differences between surface water
populations may also be a result of subtle differences in climate between the two valleys,
which has led to geochemical variations among surface waters of the lakes (Lyons,
2000).
Acknowledgements: Funding was provided by NSF-OPP 0423595 (LTER) and NSF-MCB 0237335 (MOB). Special
thanks to Raytheon Polar Services and MCM LTER team members B. Arnold, D. Rogers, J. Moore, M. Sabacka and
B. Pope for field support.