Content uploaded by Josef Navrátil
Author content
All content in this area was uploaded by Josef Navrátil on Nov 05, 2015
Content may be subject to copyright.
Vegetation gradients in fishpond mires in relation to seasonal fluctuations
in environmental factors
Sezónní kolísání faktorů prostředí a jejich souvislost s gradientem vegetace na rybničních rašeliništích
Jana N a v r á tilová1& Josef N a vr á t i l 2
1Masaryk University, Faculty of Science, Department of Botany, Kotlářská 2, CZ-611 37
Brno, Czech Republic, and Department of Plant Ecology, Institute of Botany, Academy of
Sciences, Dukelská 135, CZ-379 82 Třeboň, Czech Republic, e-mail: janaernestova@seznam.cz;
2University of South Bohemia, Faculty of Agriculture, Vančurova 2904, CZ-390 01 Tábor,
Czech Republic, e-mail:jnav@centrum.cz
Navrátilová J. & Navrátil J. (2005): Vegetation gradients in fishpond mires in relation to seasonal
fluctuations in environmental factors. – Preslia, Praha, 77: 405–418.
The composition of the vegetation of fishpond mires in the Třeboň Basin (Czech Republic) was
studied in relation to temporal fluctuations in certain environmental factors. The water-table depth,
water pH and electrical conductivity at 49 permanent plots were measured at approximately three-
week intervals from March to October 2003. Minimum, maximum, mean, median and variation in
the above-mentioned environmental factors were correlated with vegetation composition. The most
important environmentalfactors explaining the variation in vegetation were mean pH and maximum
water-table level. Median conductivity increased with increase in waterlogging and eutrophication.
Some seasonal trends in the dynamics of these parameters were observed. The lowest conductivity
was in spring, increased continuously throughout summer and peaked in autumn. In contrast, water
level decreased in summer, when evapotranspiration was greatest, and rose in autumn after heavy
rainfall. The pH increased from March to June, then was stable and decreased at the end of summer.
Seasonal trends were generally identical in all vegetation types. The fluctuations in the environmen-
tal factors were so considerable that they may influence the reliability of vegetation environmental
analyses.
K e y w o r d s : Central Europe, electrical conductivity, fen, fluctuation, mire vegetation, water pH,
water table
Introduction
Water-table fluctuations and water quality are of fundamental importance for mire vegeta-
tion (Bragazza 1997, de Mars etal. 1997, Asada 2002, Tahvanainen et al. 2002, Hájková&
Hájek 2004). Water pH and electrical conductivity are the most easily measured parame-
ters of water chemistry (Hájek & Hekera 2004). There are many studies of environmental
factors and vegetation types in mountain and boreal mires (e.g. Malmer 1986, Gerdol
1995, Bragazza & Gerdol 1999, Wheeler & Proctor 2000, Økland et al. 2001, Hájek et al.
2002, Johnson & Steingraeber 2003). In contrast, data on vegetation-environment rela-
tionships are not available for Central-European lowland poor fens at fishpond margins,
which have a specific water regime and whose chemical conditions are closely connected
to the intensively managed fishpond ecosystem.
The mires around the fishponds in the Třeboň basin provide a good opportunity to fill
this gap in mire ecology. The ponds were developed from the thirteenth century onwards
from previously swampy lowlands. The littoral ecosystems of old fishponds contain reeds,
Preslia, Praha, 77: 405–418, 2005 405
tall sedges and fen vegetation. The hydrological conditions in these mires are probably
more determined by man than climatic and geological factors like those in mountain or bo-
real mires. These mires are not purely natural, they were formed and are still influenced by
man. Water-table depth in fens, for example, depends on the water regime in the adjacent
fishpond, which is regulated by water gates. Numerous studies indicate that the distribu-
tion of vegetation in mires depends not only on the mean depth of the water table but also
on its fluctuation (Malmer 1962, Dierschke 1969, Rybníček 1974, Asada 2002).
It has been suggested that water-table fluctuations affect root aeration and the mineral nu-
trition of plants (Ingram 1967). Analyses of seasonal variation in water chemistry during the
growing season, using comparably sampled data sets are important for assessing the sea-
sonal availability of nutrients in surface water (Tahvanainen et al. 2003). Most of the studies
of seasonal variation in mire hydrological conditions have concentrated on ombrogenous
bogs (Damman 1988, Bragazza 1993, Proctor 1994, Bragazza et al. 1998). Little is known
about fluctuating environmental factors in minerogenous fens (Malmer 1962, Proctor 1995,
Vitt et al. 1995, Hájková et al. 2004). It is very difficult to say to what extent the seasonal pat-
terns found in other mires apply to climatically and geologically different regions and to
those with different human impact. Hájek & Hekera (2004) report that major water chemis-
try variables connected with base saturation are stable and thus do not affect the reliability of
vegetation-environment analyses in spring-fed fens. The extrapolation of their results to
lowland fishpond mires is not, however, possible due to the completely different hydrologi-
cal regime and nutrient sources in fishpond mires. The study of seasonal fluctuation in major
ecological factors in mires located around fishponds is therefore needed to provide a more
detailed insight into the role of seasonal fluctuations in Central-European mires in general.
Although the fishpond mire vegetation has been studied extensively with respect to hydrol-
ogy, there are few studies on seasonal variation (Přibáň & Jeník 2002).
The aim of our study was to characterize the vegetation of fishpond edges and reveal
the seasonal patterns in major environmental factors in relation to vegetation gradients in
fishpond mires.
Materials and methods
Study site
The study site is situated within the Protected landscape area, the Třeboň Basin, in the
south of the Czech Republic. Six localities, fishpond Kukla (48°57'20'' N, 14°53'23'' E),
Příbrazský fishpond (49°02'15'' N, 14°56'14'' E), fishpond Staré jezero (48°58'43'' N,
14°53'52'' E), fishpond Starý Vdovec (49°02'22'' N, 14°50'12'' E), fishpond Velká
Lásenice (49°03'11'' N, 14°57'44'' E) and fishpond Vizír (48°57'43'' N, 14°53'19'' E) were
chosen for recording temporal variations in water level and water chemistry in fens. The
climate is temperate with a mean annual temperature of 7.8 °C, in the coldest month (Janu-
ary) of –2.2°C, and in the warmest month (July) of 17.7°C, and an average annual rainfall
of 627 mm (station Třeboň).
Most of the Třeboň basin is dominated by siliceous deposits with a low concentration
of electrolytes in the soil, and as a consequence poor fens are the most common type of
mires. Fens around the fishponds are characterized by peat deposits of various thickness
(from 10 cm to a few meters) on top of sandy deposits.
406 Preslia 77: 405–418, 2005
Vegetation data
In order to monitor fluctuations of environmental factors in subcontinental minerotrophic
fens, 49 permanent plots were established at the six localities. The distribution of plots was
intentionally not random. The plots were selected to represent all main mire vegetation
types in the study area (as in Podani 1994, Somodi & Botta-Dukát 2004). Species compo-
sition was recorded during the summer of 2003 at each locality in 1 m2plots. The cover of
both vascular plants and bryophytes was recorded using the nine-grade van der Maarel
scale (1979). The height of the vegetation cover was measured and used as an indirect ap-
proximation of the productivity of the vegetation. Plant names are those used by Kubát et
al. (2002), mosses by Kučera & Váňa (2003); the nomenclature of syntaxa follows
Moravec et al. (1995).
Environmental factors
The water-table depth was measured manually in PVC tubes perforated throughout their
length. Water pH and electrical conductivity were measured in situ using portable instru-
ments (PH 114 CM 113, Snail Instruments, Czech Republic). At each of the 49 plots, all of
the above mentioned factors were measured at approximately 20-day intervals from
March to October 2003. This period corresponds to the growing season in Central Europe,
when the water regime has the greatest influence on peat vegetation. The depth of peat was
recorded at each sampling plot using a soil probe.
Data processing
Three related multivariate statistical techniques were used to analyse the data: two-way indi-
cator species analysis (TWINSPAN), detrended correspondence analysis (DCA) and canon-
ical correspondence analysis (CCA). Each approach provides a somewhat different view of
the structure of the data and when employed together these techniques can be used to com-
plement, supplement, and evaluate other analyses (Økland 1996, Lepš & Šmilauer 2003).
Vegetation data from all stations were subjected to two-way indicator species analysis
(TWINSPAN, Hill 1979) to classify the plots into groups of communities. Pseudospecies
cut levels were set at 0, 5 and 25 to suit the dataset composed of percent frequency. Differ-
ences in species number in the different strata were evaluated using the Kruskal-Wallis test.
Gradient analysis was performed using DCA and CCA algorithms of the CANOCO 4.5
package (ter Braak & Šmilauer 2002). The percent frequency of the species was log-trans-
formed and rare species were downweighted. The parameters obtained from consecutive
measurements may have different significance in explaining vegetation gradients. There-
fore, five statistical parameters (mean, median, minimum, maximum and standard devia-
tion) obtained from consecutive measurement of each environmental factor, as well as the
thickness of the peat layer, were used in ordinations.
The vegetation data set was subjected first to DCA, in order to assess the overall varia-
tion patterns in species composition. Ordination site scores were correlated to environ-
mental factors using Pearson’s correlation coefficient. All environmental variables were
plotted onto DCA ordination diagrams as supplementary environmental data for better
ecological interpretation of the axes.
Navrátilová & Navrátil: Vegetation gradients in fishpond mires 407
Subsequently CCA was used to further examine the species-environmental relation-
ships. Sixteen environmental variables in total were subjected to forward selection (ter
Braak & Šmilauer 2002, Lepš & Šmilauer 2003) in order to determine the variables that
best account for the species distribution. The marginal and conditional effects of each of
these explanatory variables on species composition was then tested. The effect of the first
canonical axis was tested by a permutation test (499 permutations were always used). To
test the statistical significance of the second and next canonical axis partial CCA was cal-
culated in which the first axis (or next ones) is partialled out by the covariable. Signifi-
cance was again tested by permutation tests for the first canonical axis.
The seasonal trends in environmental factors and the differences in the factors among
the communities (pH and conductivity also measured in open water) were investigated by
Repeated measurements ANOVA. Data transformation was not required because the data
were normally distributed and homogeneity of variance was assumed.
Results
Fen communities
The vegetation was classified by the third division of TWINSPAN into seven groups (Table 1).
Each community is named according to the dominant or diagnostic species. The “Utricularia
fen” (Type 1) occurs as an initial successional stage on permanently flooded sandy deposits.
Syntaxonomically, this community belongs to the alliance Sphagno-Utricularion characteris-
tically dominated by Utricularia intermedia. The shores of Utrichlaria pools are often occu-
pied by a “Rhynchospora alba community” (Type 2) (alliance Rhynchosporion albae). It oc-
curs on sandy deposits with some peat. The dominant species are Rhynchospora alba,Juncus
bulbosus and Sphagnum denticulatum. Fens dominated by tall sedges such as Carex
lasiocarpa and C. rostrata are referred to as “tall sedge communities” (Type 3) of the
Magnocarition elatae alliance. They are found in the littoral zone of mesotrophic water. The
soil is fen peat. One type of fen with a low electrolyte concentration but higher pH than poor
fens was found in the study area. This “medium-rich fen” vegetation (Type 4) belongs to the
alliance Eriophorion gracilis. Species growing there are more or less confined to rich fens:
Hamatocaulis vernicosus, Sphagnum subsecundum and S. contortum. They grow together
with all the common poor fen species. Such fens develop in stands saturated with mineral-rich
groundwater. Species such as Carex elata and C. lasiocarpa grow together with some of the
above-mentioned poor-fen and intermediate-fen species. The next three TWINSPAN columns
represent poor fen vegetation belonging to the alliance Sphagno recurvi-Caricion canescentis.
It is the most common mire vegetation in the Třeboň basin. Among the bryophytes, Sphagnum
species play a principal role. This group was further divided into three subtypes. The first rep-
resents an intermediate type with raised-bog vegetation (Type 5). The hummock species
Calluna vulgaris and Oxycoccus palustris are present here. In addition, some of the species
typical for pond margins are always present, e.g. Phragmites australis. The species composi-
tion of the “typical poor fen” vegetation (Type 6) is quite uniform. Species such as Carex
rostrata,Eriophorum angustifolium, Sphagnum papilosum or S. fallax often dominate.The
last type (Type 7) is poor fen vegetation associated with willow cars and other wet habitats,
which have an impact on species composition. The peat is slightly mineralized, as indicated by
species such as Polytrichum commune.
408 Preslia 77: 405–418, 2005
Table 1. – Synoptic table of vegetation types obtainedby TWINSPAN classification. The species percentage fre-
quencies (constancies) are shown. Species are sorted according to the decreasing value in the phi coefficient. Di-
agnostic species for particular columns have a phi > 0.20 and are highlighted by frames. Vegetation type:
Sphagno-Utricularion (1), Rhynchosporion albae (2), Magnocaricion elatae (3), Eriophorion gracilis (4),
Sphagno recurvi-Caricion canescentis (5, 6, 7).
Vegetation type 1234567
Number of relevés 55596154
E1
Juncus bulbosus 40 40 . . 17 7 .
Utricularia intermedia 40.80....
Carex lasiocarpa 80 . 100 89 33 13 75
Pinus sylvestris juv. . 80 . 11 67 73 .
Rhynchospora alba 20 40 . . 50 13 .
Epilobium palustre ..80....
Typha latifolia ..40....
Carex acuta ..40....
Galium palustre 20.6044...
Viola palustris . . 40 11 . . 25
Agrostis canina 20 60 80 44 17 20 50
Lysimachia thyrsiflora . 206056 . 20 .
Lysimachia vulgaris . 208089332775
Lythrum salicaria . . 40 56 . . 25
Potentilla palustris 20 . 40 100 . 20 25
Peucedanum palustre . . 40 89 . 7 50
Salix cinerea juv. ...44...
Equisetum fluviatile ...44..25
Carex elata . 204056 . 7 25
Carex canescens . 2040781747 .
Carex nigra . 20204417 7 25
Calamagrostis canescens ...33.1325
Utricularia minor 20 20 . 22 . . .
Carex rostrata . . 60 56 33 33 50
Calluna vulgaris ....17..
Drosera rotundifolia 2020 . 67836025
Phragmites australis 20 . . 11 33 20 .
Oxycoccus palustris . 20 . 33 50 60 50
Hydrocotyle vulgaris .....13.
Molinia caerulea 20 20 20 . 50 47 25
Juncus filiformis ......25
Eriophorum angustifolium 100 100 100 89 83 93 .
Utricularia ochroleuca 20 20 . . . 7 .
Lycopus europaeus 20..11...
Frangula alnus juv. . . . 11 17 . .
E0
Sphagnum denticulatum . 100 . . 33 20 .
Calliergonella cuspidata ..4022...
Sphagnum inundatum . . 40 11 . . 25
Calliergon stramineum . 2080671713100
Sphagnum flexuosum 20.4011. .50
Sphagnum subsecundum ...56...
Warnstorfia exannulata . 204078 . 7 .
Aneura pinguis ...33...
Lophocolea bidentata ...33...
Sphagnum fimbriatum ...33..50
Sphagnum palustre 20 . 20 . 100 7 100
Aulacomnium palustre . . 20 33 50 13 25
Navrátilová & Navrátil: Vegetation gradients in fishpond mires 409
Sphagnum papillosum .....
53 .
Sphagnum fallax 20 40 20 11 50 80 .
Polytrichum strictum .....27.
Sphagnum affine .....13.
Polytrichum commune . 20 . 22 33 47 75
Species present in only one column: E1: Potentilla erecta 3: 20, Utricularia australis 3: 20, Scutellaria
galericulata 3: 20, Cirsium palustre 3: 20, Cardamine amara 4: 11, Eriophorum vaginatum 6: 7, Drosera
intermedia 6: 7, Vaccinium vitis-idaea 6: 7, Betula pubescens juv. 4: 11, Salix aurita juv. 4: 11. E0: Sphagnum
platyphyllum 4: 11, Chiloscyphus polyanthos 4: 11, Drepanocladus aduncus 4: 11, Brachythecium rivulare 4: 11,
Sphagnum magellanicum 4: 11, Hamatocaulis vernicosus 4: 11, Sphagnum contortum 4: 11, Sphagnum obtusum
4: 11, Plagiothecium denticulatum 4: 11, Sphagnum rubellum 6: 7.
410 Preslia 77: 405–418, 2005
-1 4
-1 4
PMEPMD
PMI
PMA
PSD
CME
CMD
CMI
CMA
CSD
WME
WMD
WMI
WMA
WSD
peat depth
height E1
sp. richness
E1
E0
a
xi
s
1
axis 2
Fig. 1. – Ordination diagram of vegetation samples based on DCA with passive environmental variables. WME =
mean water-table depth, WMD = median water-table depth, WMI = minimum water-table depth, WMA = maxi-
mum water-table depth, WSD = standard deviation of water-table depth, PME = mean water pH, PMD = median
water pH, PMI = minimum water pH, PMA = maximum water pH, PSD = standard deviation of water pH, CME =
mean electrical conductivity, CMD = median electrical conductivity, CMI = minimum electrical conductivity, CMA
= maximum electrical conductivity, CSD = standard deviation of electrical conductivity, peat depth = thickness of
peat layer, height E1 = height of herb layer, E0 = cover of moss layer, E1 = cover of herb layer, sp. richness = species
richness. Vegetation types: 쏔=Utricularia fen, 쐽=Rhynchospora alba community, 쑗= tall sedges, 쎲=medium-
rich fen, 왕=Sphagnum palustre poor fen, 왖=Sphagnum fallax poor fen, 쏆=Polytrichum commune poor fen.
The Kruskal-Wallis test showed significant differences (P < 0.001) in the mean species
numbers among communities. The medium-rich fens are different from species-poor
communities like: Utricularia fens, the Rhynchospora alba community and Sphagnum
fallax poor fen.
Gradient analysis
The first two DCA axes are nearly equal inlength (Fig. 1) and explain about 20% of the to-
tal species variability. They also correlate well with environmental data (r1st ax. = 0.918; r2nd
ax. = 0.868). The first ordination axis is correlated significantly (P < 0.01) with mean pH,
with species richness, height of the vegetation and herb layer cover. The second ordination
axis is significantly correlated with variables related to the water regime, mainly maxi-
mum water level, and percent cover of mosses, and less markedly with mean and maxi-
mum electrical conductivity. Thickness of peat layer correlate significantly with both axes,
decreasing with increasing water-table depth and increasing with increasing pH. It is nega-
tively correlated with conductivity (Table 2).
Four canonical axes of CCA with all environmental variables were significant
(P < 0.01), explaining about 26% (first two about 17%) of the total variability in the spe-
cies data. Species-environmental correlation is similar to that in unconstrained ordination
(r1st ax. = 0.892, r2nd ax. = 0.900). The pH parameters and thickness of peat layer were gov-
erned by both the first and second canonical axes, while conductivity and water parame-
ters were governed by the second canonical axis.
Using the forward selection in CCA the four most important variables were selected:
thickness of peat layer, maximum water level, mean pH and median conductivity (Fig. 2).
They explain about 20% of the total variability in species data and a considerable part
(59%) of the variance in the species-environment relations.
Temporal fluctuations in environmental factors
There were marked seasonal fluctuations in water-table depth, water pH and water electri-
cal conductivity in the fishpond fens studied (Fig. 3). In general, seasonal trends were sim-
ilar for all vegetation types. In particular, water level decreased in summer, when
evapotranspiration was greatest, and rose again in autumn after rainfall. The pH increased
from March to June, then was stable and then decreased at the end of summer. Electrical
conductivity was low in spring, then increased continuously throughout summer and
peaked in autumn.
Comparison of environmental factors among communities
Means and standard error of measured environmental parameters in the different vegetation
types are shown in Table 3. Repeated measured ANOVA was significant for both, within-sub-
ject effect (seasonal fluctuation) and also between-subject effect (TWINSPAN clusters) in the
case of all measured factors. According to Tukey unequal N HSD post hoc test, significant dif-
ferences (P < 0.05) were found in pH between open pond water and all fens. Vegetation with
Rhynchospora alba differs in conductivity from medium-rich vegetation and poor fen vegeta-
tion with Sphagnum fallax, which have the lowest conductivity. There were no significant dif-
ferences among vegetation types in water regime according to the Tukey unequal N HSD test.
Navrátilová & Navrátil: Vegetation gradients in fishpond mires 411
Table 2. – Correlation coefficients between environmental variables and DCA ordination scores of the sample
plots along the first and the second axes. ** P < 0.01, * P < 0.05, ns – not significant.
Variable Axis 1 Axis 2
Mean water pH (PME) –0.40** ns
Median water pH (PMD) –0.36*ns
Minimum water pH (PMI) –0.30*ns
Maximum water pH (PMA) –0.35*–0.32*
Mean electrical conductivity (CME) ns –0.45**
Median electrical conductivity (CMD) ns –0.32*
Maximum electrical conductivity (CMA) ns –0.41**
Standard deviation of electrical conductivity(CSD) ns –0.36*
Mean water-table depth (WME) ns –0.63**
Median water-table depth (WMD) ns –0.50**
Minimum water-table depth (WMI) ns –0.54**
Maximum water-table depth (WMA) ns –0.68**
Peat depth –0.38** 0.53**
Height of herb layer –0.40** –0.32*
Species richness –0.62** 0.32*
Cover of herb layer (E1) –0.37** ns
Cover of moss layer (E0) ns 0.66**
412 Preslia 77: 405–418, 2005
-1.5 2.0
-1.5 1.5
PME
CMD
WMA
peat depth
a
xi
s
1
axis 2
Fig. 2. – The samples-environmental variables biplot based on CCA. WMA = maximum water-table depth, PME
= mean water pH, CMD = median water eletrical conductivity, peat depth = thickness of peat layer. For vegetation
types symbols see Fig. 1.
Navrátilová & Navrátil: Vegetation gradients in fishpond mires 413
3-Apr
23-Apr
13-May
2-Jun
22-Jun
12-Jul
1-Aug
21-Aug
10-Sep
30-Sep
-60
-50
-40
-30
-20
-10
0
10
3-Apr
23-Apr
13-May
2-Jun
22-Jun
12-Jul
1-Aug
21-Aug
10-Sep
30-Sep
4.4
4.6
4.8
5.0
5.2
5.4
5.6
5.8
6.0
3-Apr
23-Apr
13-May
2-Jun
22-Jun
12-Jul
1-Aug
21-Aug
10-Sep
30-Sep
20
40
60
80
100
120
140
160
180
Date
Water-table depth
(cm)
Water pH
Electrical conductivity
(µS/cm )
–1
Fig. 3. – Temporal fluctuation in selected environmental variables. Vertical bars denote 0.95 confidence interval.
Measurements were carried out from March to October 2003 at approximately three-week intervals.
Table 3. – Mean values (±standard error, SE) of water characteristics in the different vegetation types. Repeated
measures ANOVA test revealed significant differences (P < 0.05) among vegetations types for the selected water
variables. Means with the same letter do not differ significantly (Tukey HSD multiple comparison test, P > 0.05).
Vegetation type pH Electrical conductivity
(μS/cm) Water-table depth
(cm)
Mean SE Mean SE Mean SE
Utricularia fen 5.48a 0.15 118.7ab 12.4 –6.7a 5.9
Rhynchospora alba community 5.00a 0.13 125.5a 11.1 –15.2a 5.9
Tall sedges 5.36a 0.13 94.6ab 11.1 –14.3a 5.9
Medium-rich fen 5.34a 0.10 65.1b 8.8 –25.9a 4.6
Sphagnum fallax poor fen 5.28a 0.15 85.9ab 12.4 –28.9a 5.4
Sphagnum palustre poor fen 4.94a 0.09 71.1b 7.5 –28.6a 3.4
Polytrichum commune poor fen 5.14a 0.21 57.5ab 17.5 –33.0a 6.6
Open pond water 8.29b 0.29 155.2ab 24.8 ––
Discussion
The role of environmental conditions in plant species composition
The present analysis permitted the identification of the main environmental gradients affecting
the plant species composition of fishpond mires. The first two DCA axes are nearly equal in
length suggesting that the whole dataset is governed by two mains gradients. The first axis cor-
responds to an acidity-alkalinity gradient (from medium-rich fens to poor fens). Accordingly,
pH of surface water was significantly connected with this vegetation gradient. Along the sec-
ond ordination axis, the vegetation of flooded fens was separated from that of the other com-
munities, especially the drier ones (bog-fen-marsh gradient), so the second ordination axis cor-
responds to a water-table depth gradient. The correlation between samples and environmental
variables in CCA is similar to that in unconstrained ordination. This suggests that the selected
environmental variables are responsible for the variation in species composition.
Correlation between vegetation and environmental parameters permitted further clari-
fication of the influence of the environmental factors on vegetation differentiation. The
presence of tall sedge vegetation correlated with high water level, high pH and high elec-
trical conductivity. This vegetation was also the tallest, indicating a higher nutrient avail-
ability in tall sedge communities typically located in the littoral of meso- (eu-) trophic
ponds. In contrast to this, the moss cover increases in poor fen vegetation, as indicated by
the presence of Sphagnum species. The vegetation with the highest species richness occurs
in stands with the highest water pH, quite low electrical conductivity, and little
eutrophication due to man. In this habitat the fluctuation in environmental factors is also
very low. In contrast to this, pH, conductivity and water level fluctuate more in poor fen
vegetation. A very similar result was obtained for Carpathian fens, where water level fluc-
tuation, as well as seasonal variability in water chemistry, were larger in poor than in rich
fen microhabitats (Hájková et al. 2004). The species richness is generally lower in poor
than in rich fens (e.g. Hájková & Hájek 2003) due to the larger species pool of calcicole
species in Central-Europe (Chytrý et al. 2003). Our results suggest another explanation for
this difference in species richness – a pauperization of regional poor-fen flora due to
marked fluctuations in water level, which causes extinction of some obligate fen species
not adapted to changing water level. A periodical flooding by nutrient-rich pond water
seems to be a major factor affecting the occurrence of rare species in poor fens.
414 Preslia 77: 405–418, 2005
Seasonal variation in selected environmental factors
Fluctuation in the environmental variables measured is very conspicuous in fishpond
mires. For example, difference in water level from March to August is about 45 cm, differ-
ence in pH between spring, autumn and summer is about 1 pH unit, and conductivity dou-
bled from March to October. The fluctuation in time is, in some cases, bigger than the dif-
ferences among communities. The fluctuation in environmental factors is due to fluctua-
tions in water level related to evapotranspiration and precipitation. The evapotranspiration
is high in summer and as precipitation in summer 2003 were extremely low, the water level
fell. It is more difficult to explain the fluctuation in pH. Many different factors influence
the complex acid-base balance in mire waters, including hydrology, bedrock, soil quality,
weathering rate, nutrient uptake by plants, cation and anion exchange, decomposition, re-
dox reactions and atmospheric deposition (Shotyk 1988). The cation exchange by Sphag-
num is an important primary source of acidity in many cases (Clymo 1987, Vitt 2000). The
low pH at the beginning and end of the vegetation season may have been caused by Sphag-
num activity. The activity of Sphagnum species has a large impact on the organic acidity of
mire water (Tahvanainen et al. 2002). Sphagnum species, which are not noticeably limited
by low temperatures, acidify mire water mainly in spring and autumn, when they are not
limited by herb layer.
The significant autumnal increase in conductivity might be explained by decreasing
water level (Baumann 1996, Hájková et al. 2004). However, in the fishpond mires studied
the conductivity continued to increase even after the autumnal rains caused the water level
to rise. The water in fishponds mires accumulates after rains in contrast to spring fens
where the rainfall run off is accelerated and the ions are eluted. The dry and hot climate as-
sociated with the water table decrease in summer 2003 probably caused a higher biologi-
cal activity in the peat resulting in the release of chemical elements into the interstitial wa-
ter, which became more mobile after heavy rainfall and influenced conductivity in the
sampling device (Mörnsjö 1969)
In conclusion, the vegetation of fishpond mires is particularly affected by the chemical
and hydrological water conditions. These conditions are not static, but fluctuate markedly
during the growing season and have a significant role in affecting vegetation types.
Conservation note
The intensive fish-production (fertilizing, fish feeding) together with inputs from the
catchment area (agriculture, pollution and nutrient inputs) has caused the eutrophication
of the fishponds (Pechar et al. 2002) over the last 30 years. One of the wide spread meth-
ods used in current fish farming is to retain an extremely high water table in the ponds.
However, optimal hydrology for fens may not be optimal for fish breeding (Lamers et al.
2002). The nutrients in the eutrophic pond water enrich the fen areas, which are usually
distant from the pond edges. Only the vegetation of Utricularia fens, tall-sedge fens and
Sphagnum fallax poor fens can survive in stands influenced by eutrophic pond water.
These vegetation types are more resistant to overgrowing by plant species confined to
euthrophicated stands. The influence of pond water often causes tall sedges and shrubs to
invade low-sedge poor fen vegetation and accelerates the succession towards more pro-
ductive vegetation types.
Navrátilová & Navrátil: Vegetation gradients in fishpond mires 415
Acknowledgements
We thank Michal Hájek and two anonymous reviewers for many helpful comments, and Tony Dixon and Dana
Truffer Moudra for language revision. The research was supported by grant projects nos. FRVS 553/2004,GACR
524/05/H536, MSM 0021622416 and AV0Z 6005016.
Souhrn
Na vybraných rybničních rašeliništích Třeboňské pánve (ČR) bylo studováno složení vegetace ve vztahu k sezón-
nímu kolísání faktorů prostředí. Od března do října byla v třítýdenních intervalech prováděna měření výšky vodní
hladiny, pH a konduktivity na 49 trvalých plochách. Se složením vegetace byly následně korelovány minimum,
maximum, průměr, medián a odchylka od průměru měřených faktorů. Nejdůležitějšími faktory vysvětlujícími va-
riabilitu vegetace byly: průměr pH (koreluje signifikantně s 1. osou DCA), a maximální výška hladiny vody (ko-
reluje signifikantně s 2. osou DCA). Medián konduktivity koreloval s oběma osami a zvyšoval se s rostoucím
stupněm zamokření a současně vzrůstající eutrofizací stanovišť. V kolísání sledovaných parametrů byly zjištěny
určité sezónní trendy. Nejnižší konduktivita byla na jaře a zvyšovala se postupně během léta, s maximem na pod-
zim. Voda naproti tomu klesala během léta, kdy byla zvýšená evapotranspirace a začala růst až na podzim po vy-
datnějších deštích. Hodnota pH se zvyšovala od března do června, od konce léta pak klesala na počáteční hodnoty.
Tyto sezónní trendy byly u všech vegetačních typů podobné. Kolísání měřených faktorů prostředí bylo tak
výrazné, že by mohlo ovlivnit spolehlivost vegetačně-stanovištních analýz.
References
Asada T. (2002):Vegetation gradients in relation to temporal fluctuation of environmental factors in Bekanbeushi
peatland. – Ecol. Res. 17: 505–518.
Baumann K. (1996): Kleinseggenriede und ihre Kontaktgesselschaften im westlichen Unterharz (Sachsen-
Anhalt). – Tuxenia 16: 151–177.
Bragazza L. (1993): Seasonal changes in water chemistry in a bog on the southern Alps. – Suo 44: 87–92.
Bragazza L. (1997): Sphagnum niche diversification in two oligotrophic mires in the southern Alps of Italy.– Bry-
ologist 100: 507–515.
Bragazza L., Alber R. & Gerdol R. (1998): Seasonal chemistry of pore water in hummocks and hollows in a poor
mire in the southern Alps (Italy). – Wetlands 18: 320–328.
Bragazza L. & Gerdol R. (1999): Ecological gradients on some Sphagnum mires in the south-eastern Alps (It-
aly). – Appl. Veg. Sci. 2: 55–60.
Chytrý M., Tichý L. & Roleček J. (2003): Local and regional patterns of species richness in central European veg-
etation types along the pH/calcium gradient. – Folia Geobot. 38: 429–442.
Clymo R. S. (1987): Interactions of Sphagnum with water and air. – In: Hutchinson T. C. & Havas M. (eds.), Effect of
atmospheric pollutants on forests, wetlands and agricultural ecosystems, p. 513–529, Springer-Verlag, Berlin.
Damman A. W. H. (1988): Spatial and seasonal changes in water chemistry and vegetation in an ombrogenous
bog. – In: Verhoeven J. T. A., Heil G. W. & Werger M. J. A. (eds.), Vegetation structure in relation to carbon
and nutrient economy, p. 107–119, Acad. Publ., The Hague.
de Mars H., Wassen M. J. & Olde Venrink H. (1997): Flooding and groundwater dynamics in fens in eastern Po-
land. – J. Veg. Sci. 8: 319–328.
Dierschke H. (1969): Groundwasser-Ganglinien einiger Pflanzengesillschaften des Holtumer Noores östlich von
Bremen. – Vegetatio 17: 372–383.
Gerdol R. (1995): Community and species-performance patterns along an alpine poor-rich mire gradient. –
J. Veg. Sci. 6: 175–182.
Hájek M., Hekera P. & Hájková P. (2002). Spring fen vegetation and water chemistry in the western Carpathian
flysch zone. – Folia Geobot. 37: 205–224.
Hájek M. & Hekera P. (2004): Can seasonal variation in fen water chemistry influence the reliability of vegeta-
tion-environment analyses? – Preslia 76: 1–14.
Hájková P. & Hájek M. (2003): Species richness and above-ground biomass of poor and calcareous spring
fens in the flysch West Carpathians, and their relationships to water and soil chemistry. – Preslia 75:
271–287.
Hájková P. &Hájek M. (2004): Sphagnum-mediated successional pattern in the mixed mire in the Muránska
planina Mts (Western Carpathians, Slovakia). – Biologia 59: 63–72.
416 Preslia 77: 405–418, 2005
Hájková P, Wolf P. & Hájek M. (2004): Environmental factors and Carpathian spring fen vegetation: the impor-
tance of scale and temporal variation. – Ann. Bot. Fenn. 41: 249–262.
Hill M. O. (1979): TWINSPAN – a FORTRAN program for arranging multivariate data in an ordered two-way ta-
ble by classification of individuals and attributes. – Ecology and Systematics, Cornell University, Ihtaca.
Ingram H. A. P. (1967): Problems of hydrology and plant distribution in mires. – J. Ecol. 55: 711–724.
Johnson J. B. & Steingraeber D. A. (2003): The vegetation and ecological gradients of calcareous mires in the
South Park valley, Colorado. – Can. J. Bot. 81: 201–219.
Kubát K., Hrouda L., Chrtek J. jun., Kaplan Z., Kirschner J. & Štěpánek J. (eds.) (2002): Klíč ke květeně České
republiky. – Academia, Praha.
Kučera J. & Váňa J. (2003): Check- and Red List of bryophytes of the Czech Republic (2003). – Preslia 75: 193–222.
Lamers L. P. M., Smolders A. J. P. & Roelofs J. G. M. (2002): The restoration of fen in the Netherlands. –
Hydrobiologia 478: 107–130.
Lepš J. & Šmilauer P. (2003): Multivariate analysis of ecological data using CANOCO. – University Press, Cam-
bridge, UK.
Malmer N. (1962): Studies on mire vegetation in the Archaean area of south-western Götaland (south Sweden) I.
Vegetation and habitat conditions on the Akhult mire. – Opera Bot. Soc. Bot. Lund. 7:1–322.
Malmer N. (1986): Vegetation gradients in relation to environmental conditions in north-western European
mires. – Can. J. Bot. 64: 375–383.
Moravec J., Balátová-Tuláčková E., Blažková D., Hadač E., Hejný S., Husák Š., Jeník J., Kolbek J., Krahulec F.,
Kropáč Z., Neuhäusl R., Rybníček K., Řehořek V. & Vicherek J. (1995): Rostlinná společenstva České
republiky a jejich ohrožení. Ed. 2. – Severoč. Přír., Příl. 1995: 1–206.
Mörnsjö T. (1969): Studies on vegetation and development of a peatland in Scania, south Sweden. – Opera
Botanica 24: 1–187.
Økland R. H. (1996): Are ordination and constrained ordination alternative or complimentary strategies in gen-
eral ecological studies? – J. Veg. Sci. 7: 289–292.
Økland R. H., Økland T. & Rydgren K. (2001): A Scandinavian perspective on ecological gradients in north-west
European mires: reply to Wheeler and Proctor. – J. Ecol. 89: 481–486.
Pechar L., Přikryl I. & Faina R.(2002): Hydrobiological evaluationof Třeboň fishponds since the end of the nine-
teenth century. – In: Květ J., Jeník J. & Soukupová L. (eds.), Freshwater wetlands and their sustainable future,
Man and the biosphere series, Vol. 28, p. 31–61, UNESCO, Paris & The Panthenon Publishing Group,
Lancaster.
Podani J. (1994): Multivariate data analysis in ecology and Systematics. – SPB Academic Publishing, The Hague.
Proctor M. C. F. (1994): Seasonal and short-term changes in surface-water chemistry on four English
ombrogenous bogs. – J. Ecol. 82: 597–610.
Proctor M. C. F. (1995): Hydrochemistry of the raised bog and fens at Malharm Tarn National Nature Reserve,
Yorkshire, UK. – In: Hughes J. M. R. & Heathwaite A. L. (eds.), Hydrology and hydrochemistry of British
wetlands, p. 273–289, John Wiley & Sons, Chichester.
Přibáň K. & Jeník J. (2002): Climatic and hydrologic setting of the wet meadows. – In: Květ J., Jeník J. &
Soukupová L. (eds.), Freshwater wetlands and their sustainable future, Man and the biosphere series, Vol. 28,
p. 211–231, UNESCO, Paris & The Panthenon Publishing Group, Lancaster.
Rybníček K. (1974): Die Vegetation der Moore im südlichen Teil der Böhmisch-Mährischen Höhe. – Vegetace
ČSSR, A6, Academia, Praha.
Shotyk W. (1988): Review of the inorganic geochemistry of peat’s and peatland waters. – Earth-Science Rewiews
25: 95–176.
Somodi I. & Botta-Dukát Z. (2004): Determinants of floating island vegetation and succession in a recently
flooded shallow lake, Kis-Balaton (Hungary). – Aquat. Bot. 79: 357–366.
TahvanainenT., Sallantaus T., Heikkilä R. & Tolonen K. (2002): Spatial variation of mire surface water chemistry
and vegetation in north-eastern Finland. – Ann. Bot. Fenn. 39: 235–251.
Tahvanainen T., Sallantaus T. & Heikkilä R. (2003): Seasonal variation of water chemical gradients in three bo-
real fens. – Ann. Bot. Fenn. 40: 345–355.
ter Braak C. J. F. & Šmilauer P. (2002): CANOCO reference manual and CanoDraw for Windows user's guide:
software for canonical community ordination Version 4.5. – Microcomputer Power, Ithaca.
van der Maarel E. (1979): Transformation of cover-abundance values in phytosociology and its effect on commu-
nity similarity. – Vegetatio 39: 97–114.
Vitt D. H. (2000): Peatlands: ecosystems dominated by bryophytes. – In: Shav A. J. & Goffinet B. (eds.),
Bryophyte biology, p. 312–343, Cambridge University Press, Cambridge.
Navrátilová & Navrátil: Vegetation gradients in fishpond mires 417
Vitt D. H., Bayley S. E. & Jin T. L. (1995): Seasonal variation in water chemistry over a bog-rich fen gradient in
continental western Canada. – Can. J. Fish. Aquatic Sci. 52: 587–606.
Wheeler B. D. & Proctor M. C. F. (2000): Ecological gradients, subdivisions and terminology of north-west Euro-
pean mires. – J. Ecol. 88: 187–203.
Received 4 November 2004
Revision received 13 June 2005
Accepted 15 August 2005
418 Preslia 77: 405–418, 2005
