See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/225639054

Relationships between geomorphology and vegetation patterns in

subantarctic Andean tundra of Tierra del Fuego

Article in Polar Biology · June 2003

DOI: 10.1007/s00300-003-0499-7

CITATIONS READS

44 444

3 authors, including:

Lisa Brancaleoni Jorge Adrián Strelin

University of Ferrara Instituto Antártico Argentino and CICTERRA UNC
31 PUBLICATIONS 1,001 CITATIONS 69 PUBLICATIONS 1,345 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

IMCONet (FP7 IRSES, Action no. 318718) View project

Cambios climáticos y su respuesta en la criósfera del noroeste argentino desde el Pleistoceno tardío hasta la actualidad View project

All content following this page was uploaded by Jorge Adrián Strelin on 17 December 2014.

The user has requested enhancement of the downloaded file.

Polar Biol (2003) 26: 404-410
DOI 10.1007 /s00300~003-0499-7

ORIGINAL PAPER

Lisa Branealeoni . Jorge Strelin . Renato Gerdol

Relationships between geomorphology and vegetation patterns

in subantarctic Andean tundra of Tierra del Fuego

Received: 3 December 2002/ Accepted: 10 March 2003/ Published online: 15 April 2003
© Springer-Verlag 2003

Abstraet We analysed vegetation patterns in relation to and seasonal cycles of freezing and thawing, snow
geomorphology in subantarctic Andean tundra (Tierra accumulation and soil instability play an important
del Fuego, southern Argentina). Habitat variation ecological role in affecting vegetation patterns (Moore
associated with geomorphology played a major role in 1975). The growing season is quite cold, cloudy and
structuring vegetation. On coarse-grained debris of talus often experiences freezing conditions. The strong winds
cones, the vegetation is dominated by the epilithic fru- can inhibit plant growth, causing physical damage to
ticose lichen Neuropogon aurantiaco-ater, while vascular- buds, fiowers and vegetative organs, thus lowering the
plant cover is very low beca use of insufficient space photosynthetic and reproductive ability in most plants.
available for seedling establishment, water shortage and, Low temperatures affect soil microclimate and hinder
presumably, low nutrient status. Turf-banked lobes on soil development. The soils are skeletal entisols and
moderately inclined slopes subject to downslope creep lithosols with a shallow « 10 cm) organic horizon, often
are characterised by vascular species forming large covered with a pavement of rocks and gravel. In some
cushions (especially Azorella lycopodioides) which pro- areas, such as on scree slopes, coarse debris accumulates
vide suitable microhabitats for several plants. Steep and no appreciable soil has yet developed. The soils are
morainic slopes are extremely poor in species and have unstable and cryoturbation occurs wherever soils with
Marsippospermum reichei and Nardophyllum bryoides as high water content are exposed to multiple freeze-thaw
characteristic species. The latter appears well adapted to cycles, as usually happens in high altitude/latitude re-
withstand mechanical damage owing to its clonal gions (Troll 1960). Solifiuction, and locally gelifiuction,
creeping growth formo Screes represent intermediate are widespread, which implies slow downslope dis-
habitats where the ecotonal species Bolax gummifera placement of the upper soillayers (French 1996).
colonises debris and/or basal rock and may enhance the Increasing attention has been paid in recent years to
establishment of vascular species. analysing the causal relationship between vegetation
patterns and the environmental factors associated with
varying landscape morphology in subantarctic and ant-
are tic regions (Gremmen 1982; Selkirk et al. 1990; Fre-
Introduction not et al. 1998; Bergstrom and Selkirk 1999; Mark et al.
2001). This relationship has, besides a specific relevance
The subantarctic Andean tundra vegetation in Tierra del to ecological studies at the local and regional level, a
Fuego is principally affected by three environmental broader scientific interest with respect to functional
factors: wind exposure, water availability and the plant adaptations to environments in cold-climate ter-
physical nature of the substrate. Above timberline, daily ritories. Both fioristic composition and life-form spectra
in (sub)antarctic plant communities differ dramatically
L. Brancaleoni (18]) . R. Gerdol from those in arctic and northern alpine regions, which
Dipartimento delle Risorse Naturali e Culturali, have been the object of intensive studies since pioneer
Universitá di Ferrara, Corso Porta Mare 2, times (see, for example, Schroter 1926; Aichinger 1933;
44100 Ferrara, Italy
Evmail: [email protected]
Lüdi 1945).
Fax: + 39-0532-208561 In thispaper, we study the fine-scale vegetation pat-
tern in the Andean tundra of Tierra del Fuego. Our aim
J. Strelin
C.A.D.r.C., Instituto Antártico Argentino,
was to analyse relationships between plant distribution
Av. Malvinas Argentinas s/n, and ecological factor s, especially soil texture and mois-
9410 Tierra del Fuego Ushuaia, Argentina ture conditions. We also aimed at exploring whether,
 405

and to what extent, vegetation patterns and habitat a morainic system dating back to the Little Ice Age (the period
factors were related to the landforms, which represent between the sixteenth and nineteenth centuries; Strelin et al.
2001), comprising steep moraine next to the present ice body,
major landscape features in this region. as well as till with angular c1asts superficially covered by
boulders and pebbles, and "push-moraine" with granules,
pebbles and few angular boulders in the interstitial matrix;
Materials and methods talus cones, eroding the fractures of Monte Godoy's western
slope and partially covering older morainic deposits;
Studyarea screes, i.e. debris sheets accumulating on the slopes as an effect
of rock gelifraction and gravity fall;
The study was performed in the upper, cirque-shaped, valley of turf-banked lobes, characterised by surface instability due to
Arroyo Martial, a tributary of Arroyo Buena Esperanza, on the solifluction and gelifluction.
southern slope of Mount Martial (1,250 m). This area lies around
5 km northwest ofUshuaia (54°49'S, 68° I6'W) in Tierra del Fuego,
the southern province of Argentina (Fig. 1). The Martial range lies
east of Cordillera Darwin, in the Fuegian Andes. Here, the oro-
graphic system consists of sedimentary rocks of early-mid Creta- Sampling
ceous age that belong to the Yahgan formation, an andesitic
volcanoclastic apron, coeval with a Pacific volcanic are, which fills Forty-four sampling plots were chosen in the summer of 2000-
a marginal basin floored with oceanic crust (Olivero and Martini- , 2001. The plots were located, each at a different site, on the four
oni 2001). principal geomorphological types described above: turf-banked
The c1imate is cold oceanic with strong winds, principally from lobe, scree, talus cone and moraine. The partitioning of sites among
the southwest, especially in late spring and summer (Tuhkanen the four geomorphological types was approximately proportional
1992). The mean annual temperature at Ushuaia is 5SC, with to the area covered by each of them, but the selection of the indi-
mea n temperatures of 1.6°C in the coldest month (July) and 9.6°C vidual sampling plot was subjective.
in the warmest month (January; World Meteorological Organisa- At each plot, a 20-m-long transect was [Un along the con tour
tion 1962). Frost may occur at any time. Precipitation is fairly lines, In each transect, vegetation was recorded at 20-cm intervals,
evenly spread over the year, with an annual mean total of about by the point-intercept method (Levy and Madden 1933). At each
500 mm, and 40-50 days with snowfall. The mean annual c1oudi- point, the total number of intercepts was counted for every species,
ness is about 70%, with minimum sunny periods in winter (about when lowering a sharp 30-cm-long pin (0 5 mm) perpendicularly
1 h per day in June) and maximum in spring (about 6 h per day in into the ground. The total number of intercepts for each species
October/November; Tuhkanen 1992). At the timberline (about over the 100 points yields an estimate of projected plant area per
600 m elevation), the estimated mean annual temperature is about unit ground area within the transect. The total number of intercepts
2.5°C and the estimated total precipitation about 1,000 mm for all species over the 100 points yields an estimate of total plant
(R. Iturraspe, unpublished work). cover within the transect (Gaodall 1952).
The study area extends from just above the upper limit of the Soil texture was recorded at each of the 100 points by mea-
Nothofagus pumilio and N. betuloides woodlands to ea. 850 m ele- suring the size of the soil particle at the point where the pin hit the
vation. The latter corresponds to the upper limit of vascular-plant ground. The following categories were distinguished: bou1ders
communities, although scattered vascular plants can be found up to 0> 25 cm; cobbles 5-25 cm; gravel 0.2-5 cm; fine material (sand,
1,200 m elevation in favourable microhabitats (Mark et al. 2001). silt and c1ay) < 0.2 cm. In addition, elevation was recorded and
Different geomorphological forms, resulting from glacigenic, peri- slope angle measured with a c1inometer at each plot. A sample of
glacial and fluvial processes, can be distinguished within the Mar- fine material was collected at each plot and immediately weished
tial cirque. Four main geomorphological landform types occur using a portable balance, The samples were brought to the i~bo-
above the timberline: ratory, oven-dried for 24 h at 105°C, and re-weighed for deter-
mining gravimetrically soil-water content as % dry weight.

- 200 km
Data analysis

Relationships between vegetation and environmental variables

were studied by Redundancy Analysis (RDA) (ter Braak and
Smilauer 1998). The matrix of vegetatian data included the total
'1".),.
number of intercepts for each species (vascular plants, mosses and
~-_.,.-~ ~- lichens) in each of the 44 sampling plots, Species nomenclature
'--_ ,.,_~ Strait of
(7 Le Maire ..•' follows Moore (l983a) for al! vascular species, except for Gaul-
theria pumila (L.f.) D.J, Middleton. Full scientific names of mosses
are: Andreaea regularis C. Müll.; Chorisodontium aciphyllum
~ (Hook. f. & Wils.) Broth.; Polytrichastrum alpinum (Hedw.) G,L.
"Staten Isi.
~ Smith; Racomitrium ptycophyllum (Mitt.) Mitt.; Syntrichia gehee-
biaeopsis (c. Müll.) Zand. Full scientific names of lichens are:
Q. .d Neuropogon aurantiaco-ater Jacq.; genus Pseudocyphellaria Vain.;
Lennoxisi. genus Stereocaulon Hoffm.; Thamnolia vermicularis SW,
The environmental matrix included data for elevation, slope
~·.ilWollaston IsI.
~~
~ ~ f
. Barnavelt [si.
angle, soil-water content and soil texture in each plot with Ofl
scores. The data for soil texture included only percentage of
boulders, in order to avoid computational problems associated
Hermite [si.~ with multicolinearity among variables. Indeed, the percentage of
the four size categories could not vary independently from each
other for obvious reasons.
The vegetation data were synthesised by calculating, for each
Fig. 1 Geographic location of the study are a plot, the following parameters:
406

Figo 2a, b RDA biplot

ordinations of: species centroids 0.5 r-~----------------------------~------------------------------,
and vectors of the
environmental variables (a), Dramus
Cho
Azo sel Be;
and plots (b). Species names are
abbreviated to the first three
• • •
Gau pum Neu eur
0.25
letters of their binomials (see •
Tab\e 2 for full names). The Emp rub

symbols for the Azo {ye

• •
geomorphological types in b Rac pty

are: turf-banked lobe (e); scree •

(.); talus cone (+); moraine
(-). The Roman numerals in b o ~--------------~~--------~p~oa~m~ag~~~~------------~----------------1 Perp¡¡
correspond to the four main Ber bux ••
groups of plots (see text) Chidif •

-0.25
Lyc mag
• Fes con

,Marrei
Nar bry
~
S

-0.5 L.::~
a .L. ---I

-1 -0.5 0.5

ROA Axis 1

p-----------------~----------------------------------~
.
6
_.-.- .,
,.

.•
4 111

~
2

o
... .. I
--.
~~----~--~-=-~.----~~r~.-
r -
.
." .r : - - ......•
••••..

•
,

_.-
••
_
\
.
..~~--------~----------~----------i
~.'.
/'.
..... -.

•••
-.
"

',,' ••
.:

_ 0-
• 0-'

•••._. _ .••. ' I ••.•• j

r •••• ""t .
\
, •
.•. ~./
go..
-2
,
IV
-4

,'ÍI- -"',,
I \
r. r
-6 r I
\
, •
, I

"
--' /

-8
b
-2 -1 o 2 4
RDA Axis 1

species richness, i.e. «-diversity (Whittaker 1972): total num- indicated that soil texture, soil-water content and, to a
ber of species recorded in the plot; lesser extent, slope inclination, all played an important
Shannon index, i.e. fioristic evenness (Shannon 1948); role in affecting plant distribution in the Martial cirque
plant cover: total number of intercepts for all species recorded
in the plot (see above). Table 1 Summary of the ordination parameters

RDA axis

Results 2 3 4

Eigen values 0.199 0.054 0.012 0.004

The biplot ordination of species and plots, as well as the Inter-set correlations
vectors of environmental variables along the first two Elevation 0.38 -0.19 -0.06 -0.20
RDA axes, are shown in Fig. 20 RDA axes 3 and 4 were Slope 0.33 -0.33 0.27 -0.02
Boulders 0.77 0.12 0.03 O
neglected owing to their very low canonical eigenvalues
Soil-water content -0.59 0.22 0.15 -0.10
(Table 1)0 The lengths of the environmental vectors
Table 2 Floristic composition of the plots. The numbers in the table are percentages on the total number of intercepts recorded in each of 44 plots. Four geomorphological types are
indicated as follows: TB turf-banked lobe; S scree; TC talus cone;M moraine

Plot no. 1 2 3 4 5 6 7 8 9 LO 11 12131415161718192021222324252627282930313233343536 37 38 39 40 41 424344

Geomorphological type TE TE TE M TE TE TE TE M TE TE TE TE TE M S Te S S TE M S S S S S M S S S Te Te M Te Te Te Te Te Te Te Te M M M

Vascular plants
Abrotanella emarginata 1 1 2 1 5 5 2 2 1 7
2 10 1 5 8 3 8 39 1 11 9 19 1
Acaena tenera 1 2 1 1 1 5 11 4
Azorelfa fycopodioides 15 17 8 22 20 17 28 22 25 20 31 2 9 37 12 8 15 9 2 7 8 3 1
Azorella sefago 2 1 12 4 9 6 14 1 3 8 18 2 6
Berberis buxifolia 1 2 1 1 3
Bolax gummifera 11 14 11 2330 28 2 18 26 14 9 2 4 14 1 2
Cerastium arvense 1 1 3 5 2 8 3 10 1
Chiliotrichum diffusum 5 4
Draba magellanica 1 1 1 20
Drapetes muscosus 1 3 4 3 1 2 8 3 2 5 2 5 8
Empetrum rubrum 1 13 10 3 25 33 9 17 4 4 5 1 7 59 43 52 1 3 2123 16 15 20 32 13 23 33 23 7 22 8
Festuca contracta 17 8 22 10 2 2 5 3 9 4 15 2 15 2 4 10 3 30 6 40 5 12 14
Gaultheria pumila 49 46 49 44 39 35 33 36 31 30 18 40 27 27 13 2 18 25 5 5 3 3 5 2
Hamadryas magellanica 3
Leucheria hahnii 2 2 3 1 1 2 1 2 9 3 20 1 10
Luzula antarctica 1 1 5 3 3 3 8 10 6 4 17 3
Lycopodium magellanicum 1 2 3 6 2 1 2 3 7 2 10 4 11 8 9
M arsippospermum reichei 1 5 8 10 3 8 17 2 2 7 1 1 1 5 6071 90
Nardophyllum bryoides 1 2 1 3 1 1 1 2 20 14 10
Nassauvia magellanica 3 2 1 3 10 1 3 19 3 13 4 9 3
Perezia pilifera 2 1 1 1 2 2 3
Poa magellanica 5 1 14 1 25 22 4 5 1 5 18 7 30 15 6 26 13 30 3
Saxifraga magellanica 1 1 4 1 2 9 3 20 5
Senecio alleophyllus 9 4 5 10 2 6 6
Senecio magellanicus 1 1 1 10 3 12 3 39 4 4 3 4
Mosses
Andreaea regularis 1 10 2 1 4 2 17 3 5
Chorisodontium aciphyllum 4 1 2 2 1 1 1 4 2 3 1 6 113 4 7 2 1 8 3
Polytrichastrum alpinum 1 2 3 3 2 4 1 2 1 2 20
Racomitrium ptychophyllum
Syntrichia geheebiaeopsis 2 1 2 1 4 9
Lichens
Thamnolia vermicularis 2 4 2 4 11 6 1 1 2 1 5 2 26 3 4 1 3 2
Neuropogon aurantiaco-ater 1 3 2 1 2 2 5 4 9 1 18 4 3 17 19 32 46 56 50 61 58 62 61 100
Pseudocyphellaria sp. 5 1 3 1 1 4 3 2 27 1 2 5 3 3 18 3 2 4 1 1 1 1
Stereocaulon sp. 2 1 1 1 1 3 4 2 3 1 4 2 3 6

.•.o
-..]
408

Table 3 Partitioning of the sampling plots in the four groups 25

resulting from RDA (I-IV, see Fig. 2) among the four geomor-
phological types •
Turf banked lobe Scree Talus cone Moraine
20
•
en • '" ••
Group 1 10 O O 2 Q)
Ti 15
•
Group II 3 11 1 3 Q) '"
Group Ill O O 10 1
o,
en
'" '" '"
Group IV O O O 3 o 10
• '"
o
Z
•
(Fig. 2a). In contrast, e1evation had a 10w discriminating
5
weight as cou1d be expected based on the modest e1e-
vationa1 range covered by our sampling. Soi1texture had • • • •
a very high positive interset corre1ation and soi1-water O
•
content had a high negative interset correlation, both 600 650 700 750 800 850
with RDA axis l. Slope inclination had a moderate Elevation (m)
negative interset correlation with RDA axis 2 (Table 1).
The plots at the left end of RDA axis 1 (group 1; Fig. 3 Regression of species richness versus elevation for all plots.
Fig. 2b) are characterised by a high frequency and cover The symbols (see Fig. 2) indicate the four geomorphological types
of the cushion forming species Azorella selago, A. lyco- Both were, in turn, associated with significant variations
podioides, Drapetes muscosus, as well as the shrub in most of the environmental variables. Ten out of the 12
G. pumila (Table 2, plots 1-12). The plots in the centre plots in group 1were located on turf-banked lobes, where
of the ordination diagram (group II; Fig. 2b) are char- the substrate was richest in fine material and possessed
acterised by highest frequency of Bolax gummifera, the highest soil-water content (Table 4). Most (61 %) of
which usually forms large cushions, as well as of Draba the plots in group II were located on rather steep screes
magellanica, Saxifraga magellanica and Leucheria hah- with medium-sized substrate (Table 4). All but one ofthe
nii, all forming small cushions. Poa magellanica, Senecio plots in group III were located on moderately inclined
magellanicus, Senecio alleophyllus, Festuca contracta and talus cones where the substrate was coarsest and pos-
Lycopodium magellanicum are also associated with this sessed the lowest soil-water content (Table 4). All three
group of plots (Tab1e 2, plots 13-30). The plots in the plots in group IV were located on steep, medium-sized
right part of RDA axis 1 (group III; Fig. 2b) are char- debris corresponding to the upper part of the moraine.
acterised by very high frequency and cover of the lichen Although elevation was slightly higher for the moraine,
Neuropogon aurantiaco-ater. Other 1ichens such as there were no significant differences among the four
T. vermicularis and Stereocaulon sp., and mosses such as geomorphological types as regards elevation (Table 4).
Chorisodontium aciphyllum and Andreaea regularis, Both floristic parameters (species richness and even-
although much less abundant, are also most frequent in ness), as well as plant cover (intercepts), exhibited sig-
this group. In contrast, the total cover and frequency of nificant differences among the four geomorphological
vascular plants are much lower than in groups 1 and II types, with species richness and evenness being highest
(Table 2, plots 31-41). The 3 plots located at the nega- on turf-banked lobes and screes. Plant cover was highest
tive end of RDA axis 2 (group IV; Fig. 2b) are very poor on turf-banked lobes (Table 4). Although species rich-
in species and characterised by Marsippospermum reichei ness varied significantly in relation to geomorphological
and Nardophyllum bryoides as the dominant species types, it also showed an overall but non-significant
(Table 2, plots 42-44). Empetrum rubrum was, overall, decline with altitude (Fig. 3).
the most abundant vascular species in the area investi-
gated, being recorded in most of the plots (Table 2).
The contingency table relating the four groups of Discussion and conclusions
plots with the four geomorphological types (Table 3)
showed a high1y significant association between floristic Our study has demonstrated that all investigated envi-
grouping and geomorphology (/9= 782.29; P< 0.001). ronmental factors, especially soil-water content and

Table 4 Means (± 1 SE) of the

environmental variables and Turf-banked lobe Scree Talus cone Moraine P(F)
fioristic and plant-cover
parameters for each of the four Elevation (m) 645±49 734±14 736±14 751 ± 12 0.06
geomorphological types. The Slope (0) 11.1 ± 1.9 b 18.4±2.6 ab 14.7±3 ab 21.2±2.7a 0.04
number of intercepts are based Boulders (%) 2.4 ± 1.7b 12.5±2.1 ab 23.4±4.1 a 12.2± 3 ab < 0.001
on 100 point samples. The Soil-water content (% dry weight) 62.1 ±9a 20.8 ± 6.7b 10.8 ± 2.9b 9.8 ± 5.2b < 0.001
means within a row followed by No. of species 15.3±I.4a 13.1 ± 1.2 ab 9.4 ± 1.3 be 6.4± 1.3c <0.001
the same letter do not differ at Shannon index 1.8±0.lab 1.9±0.la 1.3±0.2b 1.4 ± 0.2ab 0.02
P<0.05 No. of intercepts 281 ±39a 136±33b 128±21b 59±26b <0.001
 409

substrate texture, were closely associated with geomor- cushion plants are rich in "nurse plants" that modify the
phology in determining the mesoscale plant distribution microenvironmental conditions, enabling the survival of
in Martial cirque. Water availability, being in turn also associated species (Nuñez et al. 1999; Cavieres et al.
related to substrate texture and wind action, has been 2002). For example, the occurrence of the shrub G.
found to exert a strong influence on vegetation patterns in pumila in association with cushion plants is probably
mountainous regions (Morefield 1992; Velásquez 1994). related to the formation of a favourable microhabitat
Substrate stability has long been known to affect moun- protecting the shrub from the desiccating effect of wind.
tain vegetation, especially above the timberline (Schroter On turf-banked lobes, the ground often exhibited a
1926). Although rates of substrate movement in moun- distinct patterning along the contour lines. The micro-
tainous regions have very rarely been measured in the scale distribution of plant cover in this habitat has been
field (Gardner 1979), there is evidence that downslope described in detail by Mark et al. (2001): vascular plants,
displacement of unstable debris is the result of complex mostly consisting of cushion-forming species, are con-
interactions among slope inclination, substrate texture centrated on the steeply sloping terrace risers while the
and soil-water content (Cannone and Gerdol 2003). treads are almost bare. Our study showed that the
The steepest part of the moraine represents a stressful mesoscale vegetation pattern on turf-banked lobes was,
habitat for plants, as suggested by the very low species overall, homogeneous, in spite of microscale patterning.
richness and plant cover. Environmental stress in this Cushion plants, when successfully established, give rise
habitat is determined by gravitational movement (creep) to stable islands providing suitable conditions for vas-
of drained surface material rather than by downslope cular-plant colonisation. In contrast, the intervening
flow of saturated soil. The vegetation is characterised by areas represent a much harsher microhabitat where
a high frequency of M. reichei and Nardophyllum bryo- plant seedlings are likely to be damaged by frost heave,
ides. M. reichei is a small rush, not possessing any solifluction and/or wind action.
apparent adaptation to mechanical disturbance. How- The talus cones, principally formed of coarse-grained
ever, its short life-cycle, associated with annual pro- debris, rich in boulders, were characterised by dry con-
duction of abundant capsules (R. Gerdol, personal ditions, as a probable effect of rapid drainage and wind
observation), allows the Marsippospermum population exposure. Due to the scarcity of fine material in the
to persist if seeds continuously germinate in small substrate, this habitat presumably was very poor in
pockets of fine material, although most individuals will nutrients as well. Although mechanical disturbance did
die shortly after fruiting. In contrast, Nardophyllum not appear to represent a limiting factor in this habitat,
bryoides is a clonal species with a high ability to spread little space is available for colonisation by vascular
vegetatively. The creeping clonal growth form has been plants. Indeed, the vegetation on talus cones was mainly
found to represent an effective adaptation of arctic-al- formed of acrocarpous mosses and, especially, epilithic
pine scree species to withstand mechanical disturbance lichens. The latter appear well adapted to withstand the
through elongation and regeneration (Somson 1984). environmentally harsh conditions in this habitat for two
Turf-banked lobes are, nonetheless, rather unstable main reasons. First, epilithic lichens, particularly the
as a result of solifluction or frost creep, although slope fruticose lichen, Neuropogon aurantiaco-ater, which
inclination on turf-banked lobes is much lower than on represents the most frequent species on dry fellfield in
the upper portion of the moraine. The plant cover in this (sub)antarctic regions (Smith 1996), have been reported
habitat corresponds to the so-called "feldmark" vege- to tolerate extreme nutrient poorness (Gremmen 1982).
tation (Moore 1983b) characterised by plant species Second, epilithic lichens and erra tic lichens such as
forming large cushions, especially Azorella lycopodioides T. vermicularis have been found to exploit water from
and, secondarily, Azorella selago. Cushion-plant-domi- melting snow (Hovenden et al. 1994; Kappen et al. 1998)
nated feldmark is widespread on windswept slopes and and frost (Pérez 1997) to maintain optimal water content
summits in the subantarctic islands, especially Falkland in their thalli (Kappen et al. 1996). The stable boulders
(Moore 1968), Marion, Prince Edwards (Gremmen on talus cones are situated mainly on the east-facing
1982) and Macquarie (Selkirk et al. 1990) Islands. slopes of Monte Godoy, where snow accumulates due to
Feldmark slopes are, overall, affected by downward flow the winds blowing from the southwest. Snow drift
of saturated soil. Selkirk (1998) reported rates of probably represents the major source ofwater for plants
downward movement ranging from 4 to 14 mm year " in this habitat (Coronato 1996).
on feldmark in subantarctic Macquarie Island. The Environmental stress resulting from water scarcity,
cushion growth form appears well adapted to withstand substrate instability or both, was lower overall on scree
soil instability associated with solifluction or gelifluction. compared to the other geomorphological types investi-
Indeed, cushions of Azorella are able to dam up the soil gated. In this habitat, ecotonal species such as B. gum-
moving downslope. In addition, the clump-cushion mifera can grow on medium-grained debris or, more
growth form provides a warm and humid microenvi- frequently, directly on the basal rock. Small cushion
ronment favourable for seedling emergence and plant plants such as Saxifraga magellanica, Draba magellanica
establishment. The cushion thus becomes a safe site and Abrotanella emarginata also occur on scree, creating
(Harper et al. 1961) available for colonisation by other microhabitats where grasses (F. contracta and Poa
species. In the feldmark habitats, patches colonised by magellanica), herbs (Nassauvia magellanica and Leuche-
410

ria hahnii) and small petridophytes (Lycopodium magel- Kappen L, Schróter B, Green TGA, Seppelt RD (1998) Microcli-
lanicum) can in turn settle. Festuca contracta, in partic- matic conditions, meltwater moistening, and the distributional
pattern of Buellia frigida on rock in a southern continental
ular, possesses a well-developed root system that allows Antarctic habitat. Polar Biol 19:101:106
this species to withstand substrate upheave by cryotur- Levy EG, Madden EA (1933) The point method of pasture anal-
batíon (Heilbronn and Walton 1984). E. rubrum is also yses. N Z J Agric 46:267-279
abundant on scree, as well as in other types of habitat, Lüdi W (1945) Besiedlung und Vegetationsentwicklung auf den
jungen Seitenmoránen des grolíen Aletschgletschers. Ber Geo-
provided the environment is not extreme. Indeed, this bot Inst Eidg Tech Hochsch Stift Rübel ZuerichI944:35-112
species is adapted to varying levels of soil-water content Mark AF, Dickinson KJM, Allen J, Smith R, West CJ (2001)
(Collantes et al. 1989) and is, furthermore, able to ex- Vegetation patterns, plant distribution and life forms across the
pand by creeping either over gravel or other plants alpine zone in southern Tierra del Fuego, Argentina. Aust Ecol
26:423-440
(Veblen and Ashton 1979).
Moore DM (1968) The vascular flora of the Falkland Islands. Br
Antarct Surv Sci Rep 60:1-202
Acknowledgements This research was supported by a grant from Moore DM (1975) The alpine flora of Tierra del Fuego. An Inst
the ltalian E.N.E.A. (Project P.N.R.A.) to R. Gerdol, and a post- Bot Cavanilles 32:419-440
graduate fellowship from Ferrara University to L. Brancaleoni. We Moore DM (1983a) Flora of Tierra del Fuego. Anthony Ne1son
acknowledge the C.A.D.r.C. staff, particularly Dr. G.M. Pastur, England Missouri Botanical Garden
for technical assistance. We thank Dr. N. Goodall (Ushuaia) for Moore DM (1983b) Flora of Fuego-Patagonian Cordilleras: its
suggestions, Dr. C. Matteri (Buenos Aires) for moss determinations origins and affinities. Rev Chil Hist Nat 56:123-136
and Professor P.L. Nimis (Trieste) for lichen determinations. Spe- Morefield JD (1992) Spatial and ecological segregation of phyto-
cial thanks are due to Professor A.F. Mark for valuable comments geographical elements in the White Mountains of California
on an earlier version of the paper. and Nevada. J Biogeogr 19:33-50
Nuñez C, Aizen MA, Ezcurra C (1999) Species associations and
nurse plant effects inpatches of high-Andean vegetation. J Veg
Sci 10:357-364
References
Olivero EB, Martinioni DR (2001) A review of the geology of the
Argentinian Fuegian Andes. J S Am Earth Sci 14:175-188
Aichinger E (1933) Vegetationskunde der Karawanken. Fischer, Pérez FL (1997) Geoecology of erratic lichens of Xanthoparmelia
Jena vagans in an equatorial Andean Paramo. Plant EcoI129:11-28
Bergstrom DM, Selkirk PM (1999) Bryophyte propagule banks in a Schroter C (1926) Das Pfíanzenleben der Alpen. Albert Raustein,
feldmark on Subantarctic Macquarie Island. Arct Antarct Alp Zürich
Res 31:202-208 Selkirk JM (1998) Active vegetation-banked terraces on Macquarie
Braak ter CJF, Smilauer P (1998) CANOCO for Windows: Soft- Island. Z Geomorphol 42:483-496
ware for Canonical Community Ordination (version 4). Centre Selkirk PM, Seppelt RD Selkirk DL (1990) Subantarctic Mac-
for Biometry, Wagenigen quarie Island: environment and biology. Cambridge University
Cannone N, Gerdol R (2003) Vegetation as an ecological indicator of Press, Cambridge
surface instability in rock glaciers. Aret Antaret Alp Res (in press) Shannon CE (1948) A mathematical theory of communication. Bell
Cavieres L, Arroyo MTK, Peñaloza A, Molina Montenegro M, Syst Tech J 27:379-243
Torres C (2002) Nurse effect of Bolax gummifera cushion plants Smith R1L (1996) Terrestrial and freshwater biotic components of
in the alpine vegetation ofthe Chilean Patagonian Andes. J Veg the Western Antarctic Peninsula, In: Ross EM, Hofmann EE,
Sci 13:547-554 Quetin LB (eds) Foundations for Ecological Research West of
Collantes MB, Anchorena J, Koremblit G (1989) A soil nutrient the Antarctic Peninsula. Antarct Res Ser 70: 15-59
gradient in Magellanic Empetrum heathlands. Vegetatio Somson P (1984) Structure des organes hypogés de quelques
80:183-193 espéces lithophiles pyrénéennes en relation ave e la dynamique
Coronato A (1996) Desarollo de circos glaciarios en el sector des pierriers. Ber Geobot Inst Eidg Tech Hochsch Stift Rübel
sudoccidental de los Andes fueginos (Argentina). XIII Cong- Zuerich 51:78-117
reso Geológico Argentino y III Congreso Argentino de Hi- Strelin J, Casassa G, Rosqvist G, Holmlund P (2001) Holocene
drocarburos Actas 4:347 glaciations at Glaciar Ema Valley, Monte Sarmiento, Tierra del
French HM (1996) The periglacial environment, 2nd edn. Long- Fuego. Resúmenes del simposio: Cambios vegetacionales y
man, London climáticos durante el último ciclo glacial-interglacial a lo largo
Frenot Y, Gloaguen JC, Cannavacciuolo M, Bellido A (1998) de Chile continental. La Serena, Chile
Primary succession on glacier forelands in the subantarctic Troll C (1960) The relationship between the climates, ecology and
Kerguelen Islands. J Veg Sci 9:75-84 plant geography of the southern cold temperate zone and of the
Gardner JS (1979) The movement of material on debris slopes in tropical high mountain. In: Pantin CFA (ed) A discussion ofthe
the Canadian Rocky Mountains. Z Geomorphol NF 23:43-57 southern cold temperate zone. Proc R Soc Lond B 152:529-532
Goodall DW (1952) Some considerations in the use of point quad- Tuhkanen S (1992) The climate of Tierra del Fuego, from a vege-
rats for the analysis of vegetation. Aust J Sci Res Ser B 5:1-41 tation geographical point of view and its ecoclimatic counter-
Gremmen NJM (1982) The vegetation of the Subantarctic Islands parts elsewhere. Acta Bot Fenn 145:1-64
Marion and Prince Edward. Junk, The Hague VebIen TT, Ashton DH (1979) Successional pattern above tim-
Harper JL, Clatworthy JN, McNaughton JH, Sagal GS (1961) The berline in South-Central Chile. Vegetatio 40:39-47
evolution of closely related species living in the same area. Velásquez A (1994) Multivariate analysis of the vegetation of
Evolution 15:209-227 volcanoes Tláloc and Pelado, Mexico. J Veg Sci 5:263-270
Heilbronn TD, Walton DW (1984) Plant colonization of active1y Whittaker RH (1972) Evolution and measurement of species
sorted stone stripes in the subantarctic. Arct Alp Res 16:161-172 diversity. Taxon 21:213-251
Hovenden MJ, Jackson AE, Seppelt RD (1994) Field photosyn- World Meteorological Organisation (1962) Climatological normals
thetic activity of lichens in the Windmill Islands oasis, Wilkes for climate and climateship Stations for the period 1931-60.
Land, continental Antarctica. Physiol Plant 90:567-576 WMO/OMM no. 117, TP.52. WMO, Geneva
Kappen L, Schroter B, Scheidegger C, Sommerkorn N(, Hestmark
G (1996) Cold resistance and metabolic activity oflichens below
o-e Adv Space Res 18:119-128

View publication stats