Microb Ecol
DOI 10.1007/s00248-014-0378-5
ENVIRONMENTAL MICROBIOLOGY
Ecology of Testate Amoebae in an Amazonian Peatland
and Development of a Transfer Function
for Palaeohydrological Reconstruction
Graeme T. Swindles · Monika Reczuga · Mariusz Lamentowicz · Cassandra L. Raby ·
T. Edward Turner · Dan J. Charman · Angela Gallego-Sala · Elvis Valderrama ·
Christopher Williams · Frederick Draper · Euridice N. Honorio Coronado ·
Katherine H. Roucoux · Tim Baker · Donal J. Mullan
Received: 1 November 2013 / Accepted: 29 January 2014
© Springer Science+Business Media New York 2014
Abstract Tropical peatlands represent globally important
carbon sinks with a unique biodiversity and are currently
threatened by climate change and human activities. It
is now imperative that proxy methods are developed to
understand the ecohydrological dynamics of these sys-
tems and for testing peatland development models. Testate
amoebae have been used as environmental indicators in
Electronic supplementary material The online version of this
article (doi:10.1007/s00248-014-0378-5) contains supplementary
material, which is available to authorized users.
G. T. Swindles () · T. E. Turner · C. Williams · F. Draper ·
E. N. Honorio Coronado · K. H. Roucoux · T. Baker
School of Geography, University of Leeds, Leeds, UK
e-mail: g.t.swindles@leeds.ac.uk
M. Reczuga · M. Lamentowicz
Department of Biogeography and Laboratory of Wetland Ecology
and Monitoring, Adam Mickiewicz University, Poznań, Poland
M. Reczuga
Faculty of Biology, Adam Mickiewicz University, Poznań, Poland
C. L. Raby
Institute of Integrative Biology, University of Liverpool,
Liverpool, UK
C. L. Raby
Institute of Zoology, Zoological Society of London, London, UK
D. J. Charman · A. Gallego-Sala
Geography, College of Life and Environmental Sciences,
University of Exeter, Exeter, UK
E. Valderrama
Putumayo Cdra. 24, Calle Garcia Calderon 246, Iquitos, Peru
D. J. Mullan
School of Natural Sciences and Psychology,
Liverpool John Moores University, Liverpool, UK
ecological and palaeoecological studies of peatlands, pri-
marily in ombrotrophic Sphagnum-dominated peatlands in
the mid- and high-latitudes. We present the first ecolog-
ical analysis of testate amoebae in a tropical peatland, a
nutrient-poor domed bog in western (Peruvian) Amazonia.
Litter samples were collected from different hydrological
microforms (hummock to pool) along a transect from the
edge to the interior of the peatland. We recorded 47 taxa
from 21 genera. The most common taxa are Cryptodifflugia
oviformis, Euglypha rotunda type, Phryganella acropodia,
Pseudodifflugia fulva type and Trinema lineare. One species
found only in the southern hemisphere, Argynnia spicata, is
present. Arcella spp., Centropyxis aculeata and Lesqueresia
spiralis are indicators of pools containing standing water.
Canonical correspondence analysis and non-metric multi-
dimensional scaling illustrate that water table depth is a
significant control on the distribution of testate amoebae,
similar to the results from mid- and high-latitudes peat-
lands. A transfer function model for water table based on
weighted averaging partial least-squares (WAPLS) regres-
sion is presented and performs well under cross-validation
(r2apparent = 0.76, RMSE = 4.29; r2j ack = 0.68, RMSEP =
5.18). The transfer function was applied to a 1-m peat core,
and sample-specific reconstruction errors were generated
using bootstrapping. The reconstruction generally suggests
near-surface water tables over the last 3,000 years, with a
shift to drier conditions at c. cal. 1218-1273 AD.
Introduction
Peatlands are globally important ecosystems in terms of
biodiversity, hydrology and for the role they play in the
carbon cycle [1]. Peatlands store approximately one third
of the soil carbon contained in the terrestrial biosphere,

whilst covering only approximately 3 % of the land and
freshwater surface [2]. There is much concern over the sta-
tus of global peatlands as they are threatened by climate
change, pollution and other human impacts such as peat
cutting, drainage and deforestation. Peatlands can be con-
sidered to be ecohydrological systems as there are linkages
and feedbacks between climate, hydrology, chemistry, vege-
tation and microbial communities [3]. There is often marked
variation across the surfaces of peatlands, leading to micro-
topographic, hydrological and vegetation gradients within a
single site [1].
Tropical peatlands are thought to contain approximately
88.6 Gt of carbon, equivalent to 15–19 % of the global
peatland carbon pool [4, 5]. They support important ecosys-
tems and are found in both lowland and upland areas in
SE Asia, Africa and Central and South America [6–9]. A
wide variety of peatlands have recently been discovered in
the subsiding Pastaza-Marañón basin in Peruvian Amazo-
nia including minerotrophic palm swamps (peatlands that
receive surface runoff and/or ground-water recharge from
surrounding mineral-soil sources) and ombrotrophic domed
bogs (peatlands that receive all water and nutrients from
direct precipitation). These peatlands have been classified
by differences in surface nutrient status, topography and
vegetation communities [8, 10]. Peat thickness is also vari-
able (from <1 to 7.5 m) [8, 10]. These peatlands are
different to those in SE Asia as they have not been heav-
ily disturbed by human activity, and the domed bogs may
therefore be the best remaining examples in the world [10,
11]. Peatlands have also been recently reported from Cen-
tral Amazonia (Brazil), although peat thicknesses are not as
great as in the west [12].
Peatlands in the tropics are vulnerable to destabilisa-
tion through climate-induced changes and human activities
including deforestation, drainage and burning [5, 13, 14].
To fully understand how tropical peatlands may respond
to such drivers of change, knowledge of their developmen-
tal history and past ecohydrological dynamics is needed.
Testate amoebae (Protozoa: Rhizopoda) are a polyphyletic
group of protozoans characterised by morphologically dis-
tinct tests [15]. Testate amoebae are often found in abun-
dance living on the surface of peatlands (comprising up
to 30 % of the microbial biomass), and subfossil remains
are often preserved in peats [15, 16]. The use of tes-
tate amoebae for palaeohydrological reconstruction is well
established for mid-latitude peatlands [16–21] and subarc-
tic/boreal peatlands to a lesser extent [22]. Palaeohydrologi-
cal reconstruction in peatlands is achieved using calibration
models known as transfer functions which are based on
the modern relationship between taxa and the environ-
mental variable(s) of interest [16, 19]. The potential of
testate amoebae as hydrological indicators in tropical peat-
lands has not yet been assessed despite several ecological
G. T. Swindles et al.
studies of testate amoebae in the tropics [23–25]. Hydro-
logical reconstructions from tropical peatlands may prove
particularly useful as relatively little hydrological monitor-
ing data exists, especially from sites in Africa and South
America.
This paper has three aims as follows:
1. To describe the testate amoebae communities in an
Amazonian peatland
2. To determine the most important environmental
parameters that influence the testate amoeba commu-
nities. We test the hypothesis that water table depth is
the strongest environmental control on the distribution
of testate amoebae in an Amazonian peatland
3. To elucidate whether testate amoeba transfer functions
for the reconstruction of hydrological change can be
developed in these environments
Study Site
Aucayacu is a nutrient-poor domed peatland in Peruvian
Amazonia which now operates as an ombrotrophic ‘raised
bog’ system [10]. It is situated on alluvial fan sediments
between a stream of the Pastaza fan and the Tigre River
(Fig. 1). The peatland was initially a nutrient-rich mine-
rotrophic system that gradually became an ombrotrophic
raised bog [11]. Aucayacu represents the deepest peatland
that has been discovered in the Amazon basin (up to 7.5-m
thick) and peat initiation at the site has been dated to c. 8870
cal. BP [11]. The vegetation of Aucayacu is characterised
by ‘pole’ and ‘dwarf’ forest communities.
At the nearby city of Iquitos (Fig. 1), average annual
rainfall of up to 3,000 mm is typical, with the wet sea-
son spanning the months of November to March when
the intertropical convergence zone (ITCZ) has migrated
just south of the equator and is positioned over northern
Peru [26]. Even the dry season from June to September
can experience monthly rainfall totals of up to 170 mm
[27]. Owing to its equatorial position, the altitude of the
midday sun is always close to vertical, leading to nearly
constant monthly temperatures throughout the year. The
average annual temperature at Iquitos is 26 ◦ C, with a
diurnal range of approximately 10 ◦ C (30–32 ◦ C daytime
temperatures and 21–22 ◦ C at night) [27]. The climate of
this region is classed as equatorial under the Köppen climate
classification (Af ).
Methods
A 1.4-km long linear transect from the edge to the inte-
rior of the Aucayacu peatland was established and 100
Ecology of Testate Amoebae in an Amazonian Peatland
Fig. 1 Map showing the
location of the Aucayacu
peatland, Loreto region,
Peruvian Amazonia. Location of
the study site is shown on a
Landsat TM RGB false color
image (NASA Landsat Program,
Orthorectified, WRS-2, Path
007, Row 063, downloaded from
http://earthexplorer.usgs.gov/).
Band 4 was assigned to red,
band 5 was assigned to green
and band 7 was assigned to blue
surface sampling points were specifically selected to
cover the entire microtopographical/hydrological gradient.
A slight change in direction was needed half-way along the
transect to avoid working in an area containing snakes. The
transect was surveyed using a Leica level and staff and the
locations of the sample points recorded using a hand-held
GPS. Litter samples of approximately 5 cm3 were sam-
pled from each point and placed into Ziplock bags. The
size and shape of each microform along with the vegetation
composition (within an area of 5 m2 ), % litter and vegeta-
tion cover were recorded at each location (Supplementary
file 1, 2). A hole was augered at each sampling point and
the water table depth measured at regular intervals until it
equalised before being measured with a metal ruler (Supple-
mentary file 3). The water table measurements were carried
out over a 3-day period to ensure they were internally con-
sistent. pH and conductivity measurements were carried out
on water samples extracted from the boreholes using cali-
brated field meters. One core, 0–1-m depth, was extracted
from the interior of the Aucayacu peatland using a Russian
corer following the parallel hole method [28, 29].
Following courier transport, all samples were returned to
the laboratory and stored in refrigeration at 4 ◦ C prior to
further analysis. Approximately one half of each litter sam-
ple was weighed, oven-dried and re-weighed to determine
moisture content. The samples were subsequently burnt in
a muffle furnace at 450 ◦ C for 8 h to determine loss-
on-ignition [30]. Testate amoebae were prepared using a
modified version of the standard method [31]. Samples of
known volume were sieved through a 300-μm sieve and
no fine-sieving was carried out following [32]. The sam-
ples were stored in deionised water. Testate amoebae were
counted under transmitted light at ×200–400 and identified
using morphology, composition, size and colour to distin-
guish taxa. At least 100 specimens were counted per sample
[33]. The taxonomy uses a morphospecies approach in cer-
tain circumstances, where a designation that includes other
species has been classed as a ‘type’. Testate amoebae were
identified using several standard keys [34–38]. Scanning
electron microscope images were taken using a Hitachi S-
3700N scanning electron microscope (http://www.sem-eds.
amu.edu.pl/). The peat core for the palaeoenvironmental
study was sub-sampled in the laboratory, and samples were
prepared for testate amoebae analysis as outlined above.
Above-ground plant material (e.g. leaf fragments, wood
and seeds) were extracted from the peat samples and AMS
14 C dates at 14 Chrono (Queen’s University Belfast) and
the SUERC Accelerator Mass Spectrometer Laboratory
(East Kilbride, Scotland). All samples were acid-alkali-acid
washed prior to analysis. We looked for spheroidal carbona-
ceous particles (SCPs) in the top 50 cm of the peat core in an
attempt to date the recent century [39]; however, none were
present.
The gradient length of the contemporary data was deter-
mined using detrended correspondence analysis (DCA). As
the data are characterised by a long gradient, canonical
correspondence analysis (CCA) was used to explore the
relationships between testate amoebae taxa and environ-
mental variables. The relative contributions of the envi-
ronmental variables were investigated using a series of
partial CCAs [40], enabling an estimation of how the total
variance is partitioned and the intercorrelations between
variables. Monte Carlo permutation tests were used to deter-
mine the statistical significance of these analyses [41].
Our use of CCA enables direct comparisons with previ-
ous studies of peatland testate amoebae that have relied on
 G. T. Swindles et al.

this technique. A number of environmental variables (plant Results

functional types, distances and heights, peat thickness, %
litter and vegetation cover, number of plant taxa, microform Site Characteristics
area) were considered as indirect factors and were included
as passive (supplementary) variables in the analysis. As The topographic survey illustrates that Aucayacu is a domed
there have been some criticisms of the use of the χ 2 dis- peatland with a steep rand (Fig. 2). Environmental param-
tance in CCA [42, 43], nonmetric multidimensional scaling eters vary within the peatland, and there is clear evidence
(NMDS), [44, 45] was also used to examine the relation-
ship between testate amoebae and environmental variables.
In contrast to CCA, NMDS does not make assumptions
Core
about species distributions over environmental gradients. location (top 1m)
Species data were square root transformed prior to NMDS N TRANSECT 2 S NE TRANSECT 1 SW

ordination and the Sorenson distance measure was used. 0 (a)

The optimum solution was identified through comparison
-1
of final stress values. The analysis was carried out using the

Height (m)
Vegan package in R version 2.15.1 [46, 47]. The Shannon -2
diversity index (SDI) was used to examine the community
diversity [48]. The SDI is defined as follows: -3 River
level

s     -4
 Xi Xi
(b)
SDI = ln (1) 4.4
Ni Ni
pH
4.2
i=1
4.0
3.8
Conductivity (μScm-1)

where Xi is the abundance of each taxon in a sample, Ni 100 (c)

80
is the total abundance of the sample and s is equal to the
60
species richness of the sample. Environments are considered
40
to be healthy if the SDI falls between 2.5 and 3.5, in transi- 20
tion between 1.5 and 2.5 and stressed between 0.1 and 1.5 60 (d)
WTD (cm)

[49, 50]. 36

Detrended canonical correspondence analysis (DCCA) 12

was performed on the dataset to determine the gradient -12
length and therefore to deduce whether linear or unimodal- 100 (e)
Litter (%)

80
based regression methods would best represent the taxon- 60
environment relationships [51]. Transfer functions were 40
20
constructed using several regression models—weighted 0
averaging (WA), tolerance-downweighted weighted aver- 80 (f)
Vegetation (%)

64
aging (WA-Tol), locally-weighted weighted averaging 48
(LWWA), weighted averaging partial least-squares (WA- 32
16
PLS) and maximum likelihood (ML). Models were also
%LOI

developed using the modern analogue technique (MAT) and 95

(g)
weighted modern analogue technique (WMAT). The models 90
% MC

85
were built using C2 [52]. The performance of the mod- 80
els was assessed using r 2 and the root mean square error 75
Microform area (m 2 )

20 Large pools (h)

of prediction (RMSEP) with leave-one-out cross valida-
15
tion (jack-knifing) and bootstrapping. The transfer function 10
models were improved through removal of 19 samples 5
with high residual values (>10 cm). A further 11 sam- 0
Pool and Hollow morphology (relative size) (i)
ples were screened out based on the influence of other
(non-hydrological) factors. The best performing model was
the one based on WA-PLS (component 2). This water 1250 1000 750 500 250 0
table transfer function was applied to subfossil data and Distance from river (m)

sample-specific errors of prediction were generated by 999 Fig. 2 Topographic and stratigraphic profile of the Aucayacu peatland
bootstrap cycles [53, 54]. with environmental variables measured along the transects
Ecology of Testate Amoebae in an Amazonian Peatland

of river influence at the edge of the site causing higher pH
and reduced loss-on-ignition (through delivery of minero-
genic material). A silty clay deposit underlies the peat-
land, and there is a small natural levee at the peatland
edge (Fig. 2). The vegetation survey suggests the pres-
ence of at least 87 plant taxa (Fig. 3; Supplementary file
1), the most commonly encountered trees and ‘small trees’
include Alibertia sp. 1, Iryanthera ulei, Virola pavonis,
Zygia sp. 1 and Oxandra euneura. The most commonly
occurring palm trees are Mauritia flexuosa and Oenocarpus
mapora. Understory herbs include Trichomanes pinnatum
and Pariana sp. 1. There are variations in the distribution
of plant functional types (PFTs) and the relative contribu-
tion of different plant families across the site (Fig. 3). For
Rubiaceae
Myristicaceae
Fabaceae
Euphorbiaceae
Arecaceae
example, individuals from the families Myristicaceae and
Arecaceae become more abundant with distance from the
river. Plants from the Euphorbiaceae and Annonaceae fam-
ilies are well-represented in both the peatland margins and
interior, whereas there is a zone dominated by plants of the
Rubiaceae family approximately 500–800 m from the river.
A series of microforms (ecohydrologically distinct sur-
face features within the mire microtope [55]) was encoun-
tered in the Aucayacu peatland including hollows and pools
(the latter characterised by standing water), flat areas (‘lit-
ter flats’), raised ridges and mounds of accumulated litter
(‘litter hummocks’ and ‘ridges’) and raised areas caused
by litter accumulation around the roots of large trees (‘tree
hummocks’). Photos of these microforms are provided in
Supplementary file 2. These microforms range in scale from
<1 to 20 m2 . The pools and hollows become larger in the
interior of the peatland and more aligned to the contours,
5
4 similar to Northern peatlands (Fig. 2, Supplementary file 2).
3
2
The microforms are characterised by contrasting water table
1
0 depths (Fig. 4, Supplementary file 3).
6
4
Ecology of Testate Amoebae
2
0
6 A total of 47 testate amoebae taxa from 21 genera were
4 identified at Aucayacu (Figs. 5 and 6a, b; Table 1). The
2
most common taxa are Cryptodifflugia oviformis, Euglypha
0
2 rotunda type, Phryganella acropodia, Pseudodifflugia fulva
1
type and Trinema lineare. One species found only in the
southern hemisphere, Argynnia spicata, is present [56].
0
12
Arcella spp., Centropyxis aculeata and Lesqueresia spiralis
are indicators of pools with standing water. We recorded one
Number of individuals

8
4
0
Annonaceae 5 48
4
3
2
1
40
0
Trees (black), Palm trees (grey) 15 32
10
Water table depth (cm)

5 24
0
Small plant 8 16
6
4
8
2
0
Lianas 2 0

1 -8

0
Herbs -16
5 n = 19 n = 20 n = 27 n = 19 n = 15
4
TREE HUMMOCK
LITTER FLAT

LITTER HUMMOCK
POOL

HOLLOW

3
2
1
0
0
900

800

700

600

500

400

300

200

100
1400

1300

1200

1100

1000

Distance from river (m)

Fig. 3 Plant families and plant functional types in the Aucayacu Fig. 4 Boxplot of water table depths measured for each microform
peatland (abundance plotted against distance from the river) type
 Fig. 5 Contemporary

The total count and Shannon

in order of water table depth.

Diversity Index are also shown

percentage testate amoebae data
from Aucayacu peatland, ranked

” e”
re tur
” ertu per
r ow " " m
ar or or llu
ap n a
o
“n aj in pe nt
i s i s "m "m ty a i s l che ta poi lyg
- x )
a t a m m e v a v a i a i a de m
a t a i l ic s r r p t a la d . a r
t t s
c tr pu nt “3 “po m
u In
. m ifo a la lis is a po y (c
c t a to a r um lea o ph str rni vifo v l a ta a ed a ty
t d d s a cula a tic ubf u bf i r a r m a n o er a ola spp
e c ula me iro ion de ula ula nat s
s a c c y s i t t h
ia e a is t o cu r n o o o u ta ra n o r e va s s i ta cr h a l s p
a x en is g tu a sp ata ifo ris rd la a at is ia er ym fis ryt ph ar ar pl he di are er de
ar . sp . ud nt ica sc a
is
ae co ec gia gi le p. rig be os yl ia ni ia rb en lita na cta u la cr gr lug t iv e
en v p ca de sp mu xis is is lu lu u sp cr cr de ro st tu r s n op a s ria co gly xis xis om nc ran e D bl
p yx pyx pyx diff d a h a ha ha ha h a ha era era phe p he r es ba lag mi pe tin tub nel h ila hila diff ula lell de olo leu py py a c a e a g a lin
oun n t a
ar no sp ia ia ia na py
o g
iff ia p
gi s oc oc do c u o o no m m m m no
lla lla lla nn nn nn li tro tro tro tro pto pt lu lu yp yp yp yp yp yp op op os lo ue ela ela ela ela ela ela ga di dr n he he n lc er
ce ce ce gy gy gy su en en en en ry ry iff iff gl gl gl gl gl gl ele ele yal ya sq eb eb eb eb eb eb ry ys ys eu xi ua he ac ac igo igo ine ine ine ine ta an at
Ar Ar Ar Ar Ar Ar As C C C C C C D D Eu Eu Eu Eu Eu Eu H H H H Le N N N N N N Ph Ph Ph Ps Py Q Sp Tr Tr Tr Tr Tr Tr Tr Tr To Sh W
15
87
88
7
11
27
4
18
31
74
78
20
23
40
46
72
79
1
50
70
73
91
100
6
19
25
56
95
3
10
26
30
39
47
54
68
71
14
24
28
32
34
37
45
81
96
22
42
49
65
8
38
48
53
93
35

Sample number
85
82
98
52
44
61
94
62
63
29
2
17
75
97
64
9
86
5
55
66
59
99
36
76
16
51
69
57
60
80
43
89
41
58
33
67
13
84
12
77
90
83
21
92

0 0 0 0 0 0 0 0 20 40 20 0 20 20 40 60 0 0 20 40 20 40 0 0 0 20 40 60 80 0 0 0 0 20 40 0 20 0 0 0 0 0 0 0 0 20 40 0 0 0 20 0 0 0 0 0 0 20 20 20 20 0 20 40 100 150 1.5 2.5 8 48

%
G. T. Swindles et al.
Ecology of Testate Amoebae in an Amazonian Peatland
Fig. 6 a Light microscope images of testate amoebae from Aucayacu:
A Sphenoderia fissirostris, B Centropyxis aerophila, C Nebela penar-
diana, D Lesqueresia spiralis, E Centropyxis aculeata, F Physochila
griseola, G Quadrulella symmetrica, H Trigonopyxis arcula “3-point
aperture”, I Centropyxis ecornis. b SEM images of testate amoebae
potential new species of Arcella—details of which will be
published elsewhere. The Shannon diversity index values of
the samples range between 0.8 and 2.8.
CCA axes 1 (eigenvalue = 0.149) and 2 (eigen-
value = 0.063) explain 11.7 % of the variance in the testate
amoebae data (Fig. 7). The hydrological variables (moisture
content and water table) and loss-on-ignition are strongly
correlated to axis 1. The associated Monte Carlo permu-
tation test shows that CCA axis 1 is highly significant
(p < 0.001, 999 random permutations), pH and conduc-
tivity are correlated with axis 2. A series of partial CCAs
show that water table depth explains 15.3 % of the variance
in the data (p < 0.002). pH explains 12.9 % (p < 0.002),
moisture content explains 8.9 % (p < 0.025), conduc-
tivity explains 11.8 % (p < 0.030) and loss-on-ignition
explains 10.0 % (p < 0.0470). The strong influence of
from Aucayacu: A Argynnia spicata, B Hyalosphenia subflava ‘major’
(> 60 μm), C Euglypha rotunda type, D Aperture of Euglypha rotunda
type, E Nebela barbata, F Trinema lineare, G Tracheleuglypha den-
tata, H Physochila griseola
hydrological variables is also illustrated by the NMDS ordi-
nation as water table and moisture content are correlated
with NMDS coordinate 1 (Fig. 7).
Transfer Function and Application to Core
The performance statistics for the transfer function models
are shown in Table 2. The best performing transfer func-
tion model (Fig. 8) is based on weighted averaging partial
2
least-squares (WAPLS) component 2 (rapparent = 0.53,
RMSE= 7.70, rj ack = 0.40, RMSEP = 9.13). After the
2
screening of samples (Supplementary file 5), the model per-
2
formance greatly improved (rapparent = 0.76, RMSE=
4.29; rj ack = 0.68, RMSEP= 5.18). The most common
2
subfossil testate amoebae present in the core from Aucay-
acu include Hyalosphenia subflava ‘major’ (> 60 μm),
 G. T. Swindles et al.

Table 1 Taxon codes

Species name Code Authority

Arcella arenaria AA Greef 1866

Arcella nov. sp. AN This paper
Arcella spp. AS n/a
Argynnia caudata AC Leidy 1879
Argynnia dentistoma AD Penard 1890
Argynnia spicata ARS Wailes 1913
Assulina muscorum AM Greeff 1888
Centropyxis aculeata CAC Ehrenberg 1838
Centropyxis aerophila CAE Deflandre 1929
Centropyxis constricta CC Ehrenberg 1841
Centropyxis ecornis CE Ehrenberg 1841
Cryptodifflugia oviformis CO Penard 1890
Cryptodifflugia oviformis “narrow” COL this paper
Difflugia pulex DP Penard 1902
Difflugia spp. DS n/a
Euglypha crenulata ECE Wailes 1912
Euglypha cristata ECI Leidy 1874
Euglypha degraded ED n/a
Euglypha rotunda type ER Wailes and Penard 1911
Euglypha strigosa ES Ehrenberg 1872; Leidy 1878
Euglypha tuberculata ET Dujardin 1841
Heleopera rosea HR Penard 1890
Heleopera sylvatica HSY Penard 1890
Hyalosphenia subflava “major” HMA Cash and Hopkinson 1909
Hyalosphenia subflava “minor” HMI Cash and Hopkinson 1909
Lesqueresia spiralis LS Ehrenberg 1840
Nebela barbata NB Leidy 1874
Nebela lageniformis NL Penard 1890
Nebela militaris NM Penard 1890
Nebela penardiana NP Deflandre 1936
Nebela tincta NTI Leidy 1979; Awerintzew 1906
Nebela tubulata NTU Brown 1911
Phryganella acropodia type PA Hertwig and Lesser 1874; Cash and Hopkinson 1909
Physochila cratera PC Wailes 1912
Physochila griseola PG Wailes and Penard 1911
Pseudodifflugia fulva type PF Archer 1870
Pyxidicula operculata PO Ehrenberg 1838
Quadrulella symmetrica QS Wallich 1863; Schulze 1875
Sphenoderia fissirostris SF Schlumberger 1845
Tracheolocorythion pulchellum TP Penard 1890
Tracheleuglypha dentata TD Deflandre 1929
Trigonopyxis arcula “three-point aperture” TA3 Penard 1912
Trigonopyxis arcula “polygon aperture” TAP Penard 1912
Trinema complanatum TC Penard 1890
Trinema enchelys TE Leidy 1878
Trinema grandis TG Chardez 1960
Trinema lineare TL Penard 1890
Ecology of Testate Amoebae in an Amazonian Peatland

Fig. 7 a CCA of testate

4
DFR (b) NM
amoebae from Aucayacu and MA
environmental variables (water pH
Herb
table depth, moisture content, 4 HAWL
LOI
%V Palm

3
pH, conductivity and loss-on- 11
14
19 DOP
S
Liana

ignition. Abbreviated species

CCA2 (eigenvalue = 0.06284)

15 Tree Cond
MC WTD
Small
codes and sample numbers are 23 %L
shown (see Table 1). b CCA Pool indicator taxa
AA AN TG

2
showing environmental variables 7 18 pH
LS 20 16
with other factors plotted as
ARS TAP 1 13
PC
supplementary variables (plant CAC 22 21 PO
TA3 NTI
functional types, number of AS

1
25 8 HMI
ES ED
plant taxa - S, % litter and LOI 10
CAE NP 2
9
17
PG
vegetation, depth of peatland, 24 PA 88
HSY
5 84
6 HMA 100
distance from river, height above CE NTU
3CC AC 60 99 89
67 77 12
DS

0
ER
water level, microform area). 27 65 73
AD
87TL
98TE 939783
HR TD ECE 85 81 90 92
c NMDS ordination of the CO 72 PF ECI 76 TC55 9694
AM 70QSSF 79 TP
58 3395
MC 31
53 5052
4441ET 669171 86
species and environmental data 39 NL 42
5428 51 26
57
36 63
75
64
8059 69 43
2930
68
37 56 82
74 78 62
45 61 Cond WTD
-1
NB 35
4047 32 46
DP
COL
(a) 49 4834

-2 0 2
CCA1 (Eigenvalue = 0.14885)
0.25
(c)
12

0.2

0.15 16
24
NMDS Coordinate 2

0.1 19
18
pH 13
69
WTD 62
83 868477 72
6
0.05 91 55 87
80
20 15
78
95 829 9496 99 73
100
7566 58
5967 9098 11
63 769392 97 33 65
0 54
57 71 8579
6189 50 39
74 70 30
42 3 4
64 56 2938 26
28 23
68 37
52 49 47 25 7
35 60 44 3441
53 32 8 22
-0.05 46 36
43 81 31 27
51
48 88
45 5 MC 21
Cond
-0.1 1
LOI
40
2
10
-0.15 17 14

-0.1 -0.05 0 0.05 0.1 0.15 0.2 0.25 0.3

NMDS Coordinate 1

H. subflava ‘minor’ (< 60 μm), P. acropodia, Trigonopy- nal changes in the water table reconstruction are mir-
xis arcula ‘polygon aperture’, C. aculeata and C. ovi- rored by principal NMDS axis 1 scores, suggesting that
formis. A count of over 100 specimens was achieved the transfer function is correctly representing the struc-
for all the core samples, and tests were generally well- ture in the subfossil data (Supplementary file 6). The
preserved. However, concentrations were very low and reconstruction suggests near-surface water tables over the
the analysis of several samples took many hours. The last 3,000 years (Fig. 9; Supplementary file 7) with a
transfer function was applied to the subfossil data, and marked shift to drier conditions at c. 50 cm (c. cal. 1218–
there were no missing modern analogues. The directio- 1273 AD).
Table 2 Transfer function performance statistics. The best improved model is WAPLS Component 2

RMSE R2 Ave Bias Max Bias Jack R2 Jack Ave Bias Jack Max Bias RMSEP Boot R2 Boot Ave Bias Boot Max Bias RMSEP

Full model (n = 100)

WA Inverse 8.40 0.44 7.67E-15 29.77 0.36 3.03E-02 31.63 9.01 0.36 4.71E-02 31.90 9.18
WA Classical 12.65 0.44 1.34E-14 16.60 0.37 7.88E-02 20.71 13.04 0.37 8.51E-02 22.75 12.51
WA-Tol Inverse 8.13 0.48 −9.20E-15 29.21 0.35 3.67E-01 31.81 9.12 0.37 2.94E-01 31.81 9.35
WA-Tol Classical 11.76 0.48 −1.71E-14 17.21 0.36 7.66E-01 29.40 12.68 0.38 5.45E-01 28.01 12.07
WAPLS Component 1 8.40 0.44 4.93E-04 29.77 0.36 3.17E-02 31.63 9.01 0.36 4.06E-02 31.80 9.17
WAPLS Component 2 7.70 0.53 −8.80E-05 28.38 0.40 9.55E-02 31.53 8.77 0.40 2.05E-01 31.64 9.13
WAPLS Component 3 7.06 0.61 −1.29E-04 27.64 0.37 1.42E-01 33.42 9.17 0.39 2.20E-01 33.02 9.83
WAPLS Component 4 6.40 0.68 −1.62E-04 23.08 0.33 2.60E-01 33.00 9.94 0.37 2.31E-01 32.54 10.50
WAPLS Component 5 6.14 0.70 −2.14E-04 21.66 0.31 3.78E-01 33.29 10.38 0.35 3.26E-01 32.24 11.18
LWWA (5 closest, squared chord) Inverse 8.25 0.47 −1.12E+00 28.67 0.47 −1.12E+00 28.67 8.25 0.44 −1.01E+00 31.53 9.30
LWWA (5 closest, squared chord) Classical 8.19 0.48 −1.10E+00 28.67 0.48 −1.10E+00 28.67 8.19 0.38 −7.19E-01 31.54 8.08
Maximum likelihood 9.92 0.50 −1.67E+00 18.45 0.35 −9.00E-01 26.30 10.89 0.37 −4.10E-01 26.76 10.50
MAT (5 anaologues) 9.13 0.37 −1.49 28.60
WMAT (5 analogues) 9.11 0.37 −1.49 28.67

Improved model (n = 70)

WA Inverse 4.74 0.71 −3.91E-15 11.26 0.66 1.49E-02 12.53 5.14 0.66 −1.03E-02 12.73 5.29
WA Classical 5.63 0.71 −4.-15 7.56 0.66 3.47E-02 9.19 5.87 0.66 2.19E-02 9.72 5.82
WA-Tol Inverse 4.72 0.71 −3.10E-15 11.96 0.66 −3.62E-01 13.19 5.15 0.67 −2.96E-01 13.42 5.45
WA-Tol Classical 5.59 0.71 −3.65E-15 8.59 0.66 −4.82E-01 10.21 5.56 0.67 −3.60E-01 10.75 5.70
WAPLS Component 1 4.74 0.71 3.31E-04 11.26 0.66 1.44E-02 12.53 5.15 0.66 −1.49E-02 12.73 5.29
WAPLS Component 2 4.29 0.76 1.13E-05 9.06 0.68 8.86E-02 10.84 4.94 0.68 1.39E-01 11.00 5.18
WAPLS Component 3 3.81 0.81 −1.87E-04 7.53 0.65 1.99E-01 11.06 5.24 0.66 1.56E-01 11.18 5.68
WAPLS Component 4 3.60 0.83 −2.58E-04 6.65 0.65 5.01E-02 10.99 5.28 0.65 4.04E-03 10.96 6.00
WAPLS Component 5 3.52 0.84 −2.32E-04 5.70 0.61 −9.44E-02 10.23 5.65 0.63 −1.43E-01 10.08 6.50
LWWA (5 closest, squared chord) Inverse 5.24 0.64 −1.51E-01 12.66 0.64 −1.51E-01 12.66 5.24 0.66 −4.37E-01 12.96 5.68
LWWA (5 closest, squared chord) Classical 5.27 0.64 −1.46E-01 12.69 0.64 −1.46E-01 12.69 5.27 0.59 −1.81E-01 13.33 5.17
Maximum likelihood 5.00 0.76 −5.90E-01 3.96 0.68 −4.60E-01 6.19 5.51 0.69 −1.60E-01 7.10 5.45
MAT (5 anaologues) 5.22 0.65 −0.54 12.30
WMAT (5 analogues) 5.21 0.65 −0.53 12.33
G. T. Swindles et al.
Ecology of Testate Amoebae in an Amazonian Peatland

20
60

ne
(a)

Li
1

10
)
1:

m
(c

cm
th
Model estimated WTD (cm)
45

p
de

0
e
bl
x
de

ta
In

e r
at
y
ys e rsit

W
" e"

2 -10
re ur

ch ea ive

SDI units
tu rt
30

en lin D
er ape

a a on
e r
ta n

n
in go

e e n
l

in in a
p

po ly

Tr Tr Sh
m
ul stri tata "3- "po

1
s o n la la

m
fa ssir de rcu rcu

0
15

gi f ha a a
lu ia p is is

20 0
iff er gly yx yx
od nod eu nop nop
ar s

eu he ac igo igo
is

Ps Sp Tr Tr Tr

0
c
i
i
0

0
a
r2 = 0.53

0
RMSEP = 9.13

ia

d
d

0
o
n = 100

rop
ac
-15

20 40
na ct lla
p e ti n e
a la gan
na
-15 0 15 30 45 60

rd a
a

y
i

eb eb r
N N Ph
Observed WTD (cm)

is

0
e
o rm
32

at nif

0
e
rb ge
(b)
ne

ba la

200
a la
Li

e
eb eb
1

N N
1:

0
e
24

"
or
Model estimated WTD (cm)

in
"m

20 40 60 80
a va
l
u bf
s
16

a
ni
he
sp
" r
jo

o
a

l
ya
"m

H
a

20 40 60 0
8

f l av
b
de da rcu a su
ae co ec gia lex p. egr tun be os nia
en v p ca de xis yxis yxis yxis fflu pu sp a d a ro a tu ra r phe
pe ta
e
d t la

op trop trop tod gia gia lyph lyph lyph op lsos

0

p lu lu g g g le a
y

e
e y
C C C C D D Eu Eu Eu H H

0
0
-8 a
r2 = 0.76

0
is

RMSEP = 4.94
la icta s orm

0
n = 70
i if
op nst orn ov

0
en en en ry iff iff
-16

0
i

-16 -8 0 8 16 24 32
h r

0
i

Observed WTD (cm)

0
r
a ata

0
ta sto ule

Fig. 8 Graph of observed versus model estimated water table depth

c

0
r
m
ud nt a

for a complete dataset and b screened dataset

ar no sp ia ia py
i

20
la lla lla nn nn tro
a

ce ce ce gy gy en
Ar Ar Ar Ar Ar C

0
c.
a .s .

0
ir a pe

Discussion

0
0
0 0
l

To our knowledge, this is the first study examining the

%
100

ecology of testate amoebae in a tropical peatland. We have

10
20
30
40
50
60
70
80
90
0

Depth (cm)

demonstrated that testate amoebae are sensitive hydrologi-

3372 ± 37
133 ± 37
146 ± 37
786 ± 22
70 ± 37

cal indicators in this Amazonian peatland, suggesting they

C
14

have the potential to be used more widely in Amazonia

and perhaps in the tropics more widely. The multivariate
statistical analysis illustrates the strong hydrological con-
trols on the distribution of testate amoebae, similar to the
research findings from mid- and high-latitude peatlands
[16–22]. pCCA also shows that pH is an important control
on testate amoebae in the Aucayacu peatland—two species
(Trinema grandis, Pyxidicula operculata) are indicators of
higher pH conditions. However, the statistical analysis also
demonstrates that a large proportion of variance in the tes-
tate amoeba data remains unexplained. This may be due to
a combination of inter-correlations between environmental
variables and unmeasured environmental (edaphic/abiotic)
factors. Such factors may include the characteristics of the
Fig. 9 Subfossil testate amoebae diagram from Aucayacu (data are
expressed as percentages). The water table reconstruction and Shannon
diversity index values are shown. Radiocarbon dates are illustrated
canopy (determining the amount of moisture reaching the
peatland surface), litter quality, diversity and decomposi-
tion [25], variations in nutrient status and other unmeasured
geochemical factors. It has also been suggested that short-
term environmental variability may be an important factor
in the community dynamics of testate amoebae [57]. There
is evidence that the Aucayacu stream has an influence on
the SW margin of the peatland (increased pH and decreased

loss-on-ignition; Fig. 2) which may affect the testate amoe-
bae communities there. There is also a possibility that
occasional high-magnitude river flooding events affect the
peatland interior, although there is no evidence for this.
There appear to be differences in the ecology of certain
species compared to the findings from mid- and high-
latitude peatlands. For example, H. subflava, Difflugia pulex
and T. arcula are not unambiguous dry indicators as reported
from mid-latitude Sphagnum peatlands [19, 21, 58]. How-
ever, the ecology of these taxa may be complex since in
more recent studies they have also been observed in wet
fen environments as well as dry bog hummocks in subarctic
and boreal peatlands [22]. The large abundance of D. pulex
and H. subflava in this tropical peatland is particularly inter-
esting as there have been some problems finding modern
analogues for these taxa in temperate peatlands [15, 16, 18],
and it has also been suggested that these two taxa are char-
acteristic of highly variable conditions [57]. However, C.
aculeata is consistently a wet indicator in our study and tem-
perate peatlands [59]. There is morphological variability of
certain taxa reported here such as the marked differences in
the aperture of T. arcula. Ogden and Hedley (1980) describe
the highly variable aperture shape of T. arcula [35], which
was also noted by Bobrov et al. (1995) in populations from
Russia and Canada including the occurrence of three-point,
four-point and almost polygonal/circular apertures [60].
Palaeohydrological reconstructions based on testate
amoebae may prove particularly useful for examining the
developmental history of tropical peatlands. Previous pollen
and stratigraphic data suggest that Amazonian peatlands
undergo major vegetation transitions in their developmental
history [61]. Testate amoebae may provide important infor-
mation about the role of changing hydrology across such
ecological transitions. However, poor preservation of tests
may hinder this for older sections of the subfossil record
[62–64].
Relatively little is currently known about the ecohydro-
logical dynamics of tropical peatlands. Peatland develop-
ment models [65, 66] modified for tropical ecosystem PFTs,
productivity and decomposition (e.g. [67]) may shed light
on the long-term ecohydrological and C dynamics of these
systems. Of particular interest is how peatlands respond to
climatic shifts [68]. Testate amoebae-based reconstructions
may therefore prove useful for testing the hydrological out-
puts of such models and understanding peatland responses
to changing climate. In the case of Amazonian peatlands,
changing flooding regime through time and river chan-
nel migration may also affect the peatland development
trajectories [11].
Previous palaeoenvironmental studies of lakes and
swamps in Amazonia have suggested distinct phases of cli-
mate changes during the Holocene. For example, there is
compelling evidence for a period of increased precipitation
G. T. Swindles et al.
in several areas of Amazonia at c. 700–1300 cal. BP [69–
73]. Peatlands have the potential to provide high-resolution
archives of environmental and climatic change during this
period and could add substantial temporal and spatial detail
to the existing picture. Further work is needed to identify
a clear climatic signal in the Amazonian peatland archives
as their accumulation is controlled by a complex interac-
tion of biological, geochemical, hydrological and climatic
factors. Future studies should focus on (i) the generation of
modern and palaeoecological data from other suitable sites,
(ii) the development of robust core chronologies, (iii) high-
resolution sampling, (iv) multiproxy approaches, (v) inter-
and intra-site comparison studies and (vi) the comparison
of peat-based reconstructions to independent palaeoclimatic
data. Furthermore, hydrological monitoring data will help
understand the sub-annual and inter-annual hydrological
dynamics of Amazonian peatlands.
There are major concerns about the effects of climate
change and human activity on tropical peatlands as they
are globally important carbon sinks [4, 5]. Future cli-
mate change scenarios for north-east Peru, derived from
an average of 21 climate models and expressed as relative
changes from the 1961–1990 baseline climate to the year
2100, project an average annual increase in temperatures of
3.0 − 3.5 ◦ C and a 5–10 % increase in precipitation across
the region [27]. Agreement between models is generally
low (particularly for precipitation), but nonetheless a con-
sistent pattern of warming and increasing precipitation is
projected [27]. As the Amazonian basin is of special inter-
est as a biodiversity hotspot [74], further research is needed
to examine the sensitivity of Amazonian peatlands to cli-
mate change and loss of biodiversity from human impacts.
Our study suggests that testate amoebae may prove to be a
particularly useful tool in this endeavour by enabling us to
reconstruct past hydrological conditions which are crucial
to the functioning of these ecosystems.
Conclusions
1. We present the first analysis of testate amoebae from a
tropical peatland (Aucayacu, Peruvian Amazonia). We
recorded 47 testate amoebae taxa from 21 genera in
surface litter samples. The most common taxa are C.
oviformis, E. rotunda type, P. acropodia, P. fulva type
and T. lineare. Arcella spp., C. aculeata and L. spiralis
are indicators of pools with standing water.
2. Multivariate statistical analysis illustrates that water
table depth is the most important control on the dis-
tribution of testate amoebae in the peatland explaining
15.3 % of the variance in the data (p < 0.002). pH is
the next most important variable explaining 12.9 %
(p < 0.002). A transfer function model for water table
Ecology of Testate Amoebae in an Amazonian Peatland
based on weighted averaging partial least-squares
(WAPLS) regression is presented and performs well
2
under cross validation (rapparent = 0.76, RMSE=
4.29; rj ack = 0.68, RMSEP= 5.18).
2
3. The transfer function was applied to a 1-m peat core
and sample-specific reconstruction errors were gen-
erated using bootstrapping. The reconstruction gener-
ally suggests near-surface water tables over the last
3,000 years, with a marked shift to drier conditions
at c. cal. 1218–1273 AD. Testate amoebae may prove
very useful for reconstructing the hydrological dynam-
ics of tropical peatlands in Amazonia and elsewhere.
Acknowledgments This work was funded by a Royal Society
research grant to GTS (grant no. 481831). Radiocarbon dates were
provided by the UK NERC Radiocarbon Laboratory allocation num-
ber 1681.1012 to DJC and AGS. We thank Outi Lähteenoja for advice
on accessing the Aucayacu peatland and Ricardo Farroñay Peramas
and Denis del Castillo Torres of the Instituto de Investigaciones de la
Amazonı́a Peruana in Iquitos for assisting with fieldwork planning.
Aristidis Vasques is acknowledged for piloting the boats and helping
us run the field campaign. Many thanks to the villagers of Bellavista
and Malvinas for assistance in the field (especially Lucho Freyre and
David Huayaban). Scanning electron micrographs (SEM) were taken
in The Scanning Microscopy and Microanalysis Laboratory, Faculty
of Geographical and Geological Sciences, Adam Mickiewicz Univer-
sity. We kindly thank Monika Lutynska for technical support. Monika
Reczuga would also like to thank Katarzyna Marcisz for assisting with
the identification of testate amoebae.
References
1. Charman DJ (2002) Peatlands and environmental change. Wiley-
Blackwell
2. Holden J (2005) Peatland hydrology and carbon release: why
small-scale process matters. Philos Trans R Soc A Math Phys Eng
Sci 363:2891–2913
3. Belyea LR, Baird AJ (2006) Beyond the limits to peat bog growth:
cross-scale feedback in peatland development. Ecol Monogr
76:299–322
4. Page SE, Rieley JO, Banks BG (2008) Global and regional impor-
tance of the tropical peatland carbon pool. Glob Chang Biol
17:798–818
5. Moore S, Evans CD, Page SE, Garnett MH, Jones T, Freeman C,
Hooijer A, Wiltshire AJ, Limin SH, Gauci V (2013) Deep insta-
bility of deforested tropical peatlands revealed by fluvial organic
carbon fluxes. Nature 493:660–663
6. Anderson JAR (1964) The structure and development of the peat
swamps of Sarawak and Brunei. J Trop Geogr 18:7–16
7. Joosten H (2009) The global peatland CO2 picture. Wetlands
international. Ede:33p
8. Lähteenoja O, Ruokolainen K, Schulman L, Alvarez J (2009a)
Amazonian floodplains harbour minerotrophic and ombrotrophic
peatlands. Catena 79:140–145
9. Lähteenoja O, Ruokolainen K, Schulman L, Oinonen M (2009b)
Amazonian peatlands: an ignored C sink and potential source.
Glob Chang Biol 15:2311–2320
10. Lähteenoja O, Page SE (2011) High diversity of tropical peat-
land ecosystem types in the Pastaza-Marañón basin, Peruvian
Amazonia. J Geophys Res 116:G02025
11. Lähteenoja O, Reategui Y, Rasanen M, del Castillo D, Oinonen
M, Page SE (2012) The large Amazonian peatland carbon sink in
the subsiding astaza-Marañón foreland basin, Peru. Glob Chang
Biol 18:164–178
12. Lähteenoja O, Flores B, Nelson B (2013) Tropical peat accumula-
tion in Central Amazonia. Wetlands 33:495–503
13. Page SE, Siegert F, Rieley JO, Boehm HDV, Jaya A, Limin S
(2002) The amount of carbon released from peat and forest fires
in Indonesia in 1997. Nature 420:61–65
14. Miettinen J, Shi C, Liew SC (2012) Two decades of destruction in
Southeast Asia’s peat swamp forests. Front Ecol Environ 10:124–
128
15. Mitchell EAD, Charman DJ, Warner BG (2008) Testate amoe-
bae analysis in ecological and paleoecological studies of wetlands:
past, present and future. Biodivers Conserv 17:2115–2137
16. Woodland WA, Charman DJ, Sims PC (1998) Quantitative esti-
mates of water tables and soil moisture in Holocene peatlands
from testate amoebae. The Holocene 8:261–273
17. Lamentowicz M, Mitchell EAD (2005) The ecology of testate
amoebae (Protists) in sphagnum in North-western Poland in rela-
tion to peatland ecology. Microb Ecol 50(1):48–63
18. Charman DJ, Blundell A, ACCROTELM Members (2007) A new
European testate amoebae transfer function for palaeohydrological
reconstruction on ombrotrophic peatlands. J Quat Sci 22:209–221
19. Swindles GT, Charman DJ, Roe HM, Sansum PA (2009) Envi-
ronmental controls on peatland testate amoebae (Protozoa: Rhi-
zopoda) in the North of Ireland: implications for Holocene palaeo-
climate studies. J Paleolimnol 42:123–140
20. Turner TE, Swindles GT, Charman DJ, Blundell A (2013) Com-
paring regional and supra-regional transfer functions for palaeohy-
drological reconstruction from Holocene peatlands. Palaeogeogr
Palaeoclimatol Palaeoecol 369:395–408
21. Amesbury MJ, Mallon G, Charman DJ, Hughes PD, Booth RK,
Daley TJ, Garneau M (2013) Statistical testing of a new testate
amoebae transfer function for water-table depth reconstruction on
ombrotrophic peatlands in Atlantic Canada and far north-eastern
United States. J Quat Sci 28:27–39
22. Lamarre A, Magnan G, Garneau M, Boucher E (In Press) A testate
amoeba-based transfer function for paleohydrological reconstruc-
tion from boreal and subarctic peatlands in northeastern Canada.
Quat Int
23. Bobrov AA (2001) Findings of the tropical group testate amoebae
(Protozoa: Testacea) at the far East (Sikhote Alin reserve). Biol
Bull Russ Acad Sci 28:401–407
24. Krashevska V, Bonkowski M, Maraun M, Scheu S (2007) Tes-
tate amoebae (protista) of an elevational gradient in the tropical
mountain rain forest of Ecuador. Pedobiologia 51:319–331
25. Krashevska V, Maraun M, Scheu S (2012) How does litter quality
affect the community of soil protists (testate amoebae) of tropical
montane rainforests? FEMS Microbiol Ecol 80:603–607
26. Martinez R, Ruiz D, Andrade M, Blacutt L, Pabon D, Jaimes
E, Leon G, Villacis M, Quintana J, Montealegre E, Euscategui
CH, Jorgensen PM (2011) Synthesis of the climate of the Trop-
ical Andes. In: Herzog SK, Martinez R, Tiessen H (eds) Cli-
mate change and biodiversity in the Tropical Andes. MacArthur
Foundation, Inter-American Institute of Global Change Research
(IAI) and Scientific Committee on Problems of the Environment
(SCOPE), vol 348. Sao Jose dos Campos, Paris, pp 97–109. ISBN:
978-85-99875-05-6
27. Met Office (2011) Climate: observations, projections and impacts.
Peru, Met Office. Exeter
28. Jowsey PC (1966) An improved peat sampler. New Phytol
65:245–248
29. De Vleeschouwer F, Chambers FM, Swindles GT (2010) Coring
and sub-sampling of peatlands for palaeoenvironmental research.
Mires Peat 7:1–10

30. Schulte EE, Hopkins BG (1996) Estimation of soil organic mat-
ter by weight-loss-on-ignition. In: FR Magdoff et al. (eds) Soil
organic matter: analysis and interpretation. SSSA Spec. Publ. 46,
Madison, WI
31. Hendon D, Charman DJ (1997) The preparation of testate amoe-
bae (Protozoa: Rhizopoda) samples from peat. The Holocene
7:199–205
32. Payne R (2009) The standard preparation method for testate amoe-
bae leads to selective loss of the smallest shells. Quat Newsl
119:16–20
33. Payne R, Mitchell E (2009) How many is enough? Determining
optimal count totals for ecological and palaeoecological studies of
testate amoebae. J Paleolimnol 42:483–495
34. Charman DJ, Hendon D, Woodland W (2000) The identification
of peatland testate amoebae. Quat Res Assoc Tech Guid 9:147p
35. Ogden CG, Hedley RH (eds) (1980) An atlas to freshwater testate
amoebae. Oxford University Press, London
36. Mazei Y, Tsyganov AN (2006) Freshwater testate amoebae.
KMK, Moscow
37. Meisterfeld R (2000) Arcellinida. The illustrated guide to the
protozoa, 2nd edn, pp 827–859
38. Meisterfeld R (2000) Testate amoebae with filopodia. The illus-
trated guide to the protozoa, 2nd edn, pp 1054–1083
39. Swindles GT (2010) Dating recent peat profiles using spheroidal
carbonaceous particles (SCPs). Mires Peat 7:1–10
40. Borcard D, Legendre P, Drapeau P (1992) Partialling out the
spatial component of ecological variation. Ecology 73:1045–1055
41. Dale B, Dale AL (2002) Application of ecologically based sta-
tistical treatments to micropalaeontology. In: Haslett SK (ed)
Quaternary environmental micropalaeontology. Arnold, London
42. Rao CR (1995) A review of canonical coordinates and an alterna-
tive to correspondence analysis using Hellinger distance. Qüestiió
19:23–63
43. Legendre P, Gallagher E (2001) Ecologically meaningful trans-
formations for ordination of species data. Oecologia 129:271–
280
44. Kruskal JB (1964) Nonmetric multidimensional scaling: a numer-
ical method. Psychometrika 29, 115–129
45. McCune B, Grace JB (2002) Analysis of ecological communities.
MJM Press
46. Oksanen J, Blanchet FG, Kindt R, Legendre P, Minchin PR,
O’Hara RB, Simpson GL, Solymos P, Henry M, Stevens H,
Wagner H (2012) vegan: Community Ecology Package. http://
CRAN.R-project.org/package=vegan
47. Core Team R (2012) R: A language and environment for statis-
tical computing. R foundation for statistical computing, Vienna,
Austria, http://www.R-project.org/
48. Shannon CE (1948) A mathematical theory of communication.
Bell Syst Tech J 27:379–423,623–656
49. Magurran AE (1988) Ecological diversity and its measurement.
Princeton University Press, Princeton
50. Patterson RT, Kumar A (2000) Assessment of arcellacea (the-
camoebian) assemblages, species and strains as contaminant
indicators in variably contaminated James Lake, North Eastern
Ontario. J Foramin Res 30:310–320
51. Birks HJB (1995) Quantitative palaeoenvironmental reconstruc-
tions. In: Maddy D, Brew JS (eds) Statistical modelling of qua-
ternary science data. Technical guide 5. Quaternary Research
Association, Cambridge
52. Juggins S (2003) C2 user guide. Software for ecological and
palaeoecological data analysis and visualisation. University of
Newcastle, Newcastle Upon Tyne
G. T. Swindles et al.
53. Birks HJB, Line JM, Juggins S, Stevenson AC, ter Braak CJF
(1990) Diatoms and pH reconstruction. Phil Trans R Soc B
27:263–278
54. Line JM, ter Braak CJF, Birks HJB (1994) WACALIB ver-
sion 3.3: a computer program to reconstruct environmental vari-
ables from fossil assemblages by weighted-averaging and to
derive sample-specific errors of prediction. J Paleolimnol 10:147–
152
55. Ivanov KE (1981) Water movement in Mirelands. Academic
Press, London
56. Deflandre G (1936) Etude monographique sur le genre Nebela
Leidy. Annales de Protistologie 5:201–286
57. Sullivan ME, Booth RK (2011) The potential influence of short-
term environmental variability on the composition of testate
amoeba communities in Sphagnum peatlands. Microb Ecol 62:80–
93
58. Turner TE, Swindles GT (2012) Ecology of testate amoebae in
moorland with a complex fire history: implications for ecosystem
monitoring and sustainable land management. Protist 163:844–
855
59. Booth RK, Zygmunt JR (2005) Biogeography and comparative
ecology of testate amoebae inhabiting Sphagnum-dominated peat-
lands in the Great Lakes and Rocky Mountain regions of North
America. Divers Distrib 11:577–590
60. Bobrov AA, Yazvenko SB, Warner BG (1995) Taxonomic and
ecological implications of shell morphology of three testaceans
(Protozoa: Rhizopoda) in Russia and Canada. Archiv für Protis-
tenkunde 145:119–126
61. Roucoux KH, Lawson IT, Jones TD, Baker TR, Coronado ENH,
Gosling WD, Lähteenoja O (2013) Vegetation development in
an Amazonian peatland. Palaeogeogr Palaeoclimatol Palaeoecol
374:242–255
62. Wilmshurst JM, Wiser SK, Charman DJ (2003) Reconstructing
Holocene water tables in New Zealand using testate amoebae:
differential preservation of tests and implications for the use of
transfer functions. The Holocene 13:61–72
63. Swindles GT, Roe HM (2007) Examining the dissolution char-
acteristics of testate amoebae (Protozoa: Rhizopoda) in low
pH conditions: implications for peatland palaeoclimate studies.
Palaeogeogr Palaeoclimatol Palaeoecol 252:486–496
64. Mitchell E, Payne R, Lamentowicz M (2008) Potential impli-
cations of differential preservation of testate amoeba shells for
paleoenvironmental reconstruction in peatlands. J Paleolimnol
40:603–618
65. Frolking S, Roulet NT, Tuittila E, Bubier JL, Quillet A, Talbot
J, Richard PJH (2010) A new model of Holocene peatland net
primary production, decomposition, water balance, and peat accu-
mulation. Earth Syst Dyn 1:1–21
66. Morris PJ, Belyea LR, Baird AJ (2011) Ecohydrological feed-
backs in peatland development: a theoretical modelling study. J
Ecol 99:1190–1201
67. Kurnianto S (2013) Modeling carbon accumulation dynamics in
tropical peat swamp forests (abstract), New frontiers in tropi-
cal biology: the next 50 years (A Joint Meeting of ATBC and
OTS)
68. Swindles GT, Morris PJ, Baird AJ, Blaauw M, Plunkett G (2012)
Ecohydrological feedbacks confound peat-based climate recon-
structions. Geophys Res Lett 39:L11401
69. Bush MB, Colinvaux PA (1988) A 7,000-year pollen record from
the Amazon lowlands, Ecuador. Vegetatio 76:141–154
70. Frost I (1988) A Holocene sedimentary record from Anañgucocha
in the Ecuadorian Amazon. Ecology 69:66–73
Ecology of Testate Amoebae in an Amazonian Peatland
71. Liu KB, Colinvaux PA (1988) A 5,200-year history of Amazon
rain forest. J Biogeogr 15:231–248
72. Behling H, Berrio J, Hooghiemstra H (1999) Late Quaternary
pollen records from the middle Caquetá river basin in cen-
tral Columbian Amazon. Palaeogeogr Palaeoclimatol Palaeoecol
145:193–213
73. Correa-Metrio A, Cabrera KR, Bush MB (2010) Quantifying
ecological change through discriminant analysis: a palaeoeco-
logical example from the Peruvian Amazon. J Veg Sci 21:695–
704
74. Hoorn C, Wesselingh FP, ter Steege H, Bermudez MA, Mora
A, Sevink J, Sanmartin I, Sanchez-Meseguer A, Anderson CL,
Figueiredo JP, Jaramillo C, Riff D, Negri FR, Hooghiemstra H,
Lundberg J, Stadler T, Sarkinen T, Antonelli A (2010) Amazonia
through time: andean uplift, climate change, landscape evolution,
and biodiversity. Science 330:927–931