Major postglacial summer temperature changes in the central coniferous boreal forest of Quebec (Canada) inferred using chironomid assemblages: MAJOR POSTGLACIAL SUMMER TEMPERATURE CHANGES IN

Abstract

Chironomid head capsules preserved in lake sediments were used to reconstruct 8200 years of summer temperatures in the boreal forest of north‐eastern Canada. Two training sets were used derived from Canadian and Eastern Canadian transfer functions. Both models reconstructed similar climate patterns, but the Canadian model provided temperatures generally 2–3 °C lower than the Eastern Canadian model. Three main thermal changes inferred by chironomids were: (i) the Holocene Thermal Maximum, which occurred between 8 and 5k cal a BP, with temperatures generally higher than today's, maximum temperatures between 8 and 6.5k cal a BP, and an average of + 0.9 °C; (ii) the Medieval Climate Anomaly around 1.1–1.2k cal a BP with an amplitude of + 0.7 °C; and (iii) a colder period reconstructed between the 14th and 19th centuries, corresponding to the Little Ice Age, with summer temperatures on average −0.5 °C lower than the climate normal. For each of these three climatic events, the timing and the amplitude of changes were similar with other published regional, North American and Northern Hemisphere records.

Full text

Major postglacial summer temperature changes in the central
coniferous boreal forest of Quebec (Canada) inferred using
chironomid assemblages
LISA BAJOLLE,
1,2
* ISABELLE LAROCQUE-TOBLER,
3
EMMANUEL GANDOUIN,
4
MARTIN LAVOIE,
5
YVES BERGERON
1,6
and ADAM A. ALI
1,2
1
Institut de Recherche sur la for^
et, Universit
eduQu

ebec en Abitibi-T
emiscamingue, Rouyn-Noranda, QC, Canada
2
Institut des Sciences de l’

Evolution-Montpellier, UMR 5554, Universit
e de Montpellier CNRS-IRD-EPHE, Montpellier, France
3
The L.A.K.E.S Institute, Lyss, Switzerland
4
Aix Marseille Univ, Univ Avignon, CNRS, IRD, IMBE, Aix-en-Provence, France
5
D
epartement de g
eographie and Centre d’
etudes nordiques, Universit
e Laval, Qu
ebec, QC, Canada
6
Centre d’

Etude de la For^
et, Universit
eduQu

ebec 
a Montr
eal, Montr
eal, QC, Canada
Received 28 September 2017; Revised 29 December 2017; Accepted 22 January 2018
ABSTRACT: Chironomid head capsules preserved in lake sediments were used to reconstruct 8200 years of
summer temperatures in the boreal forest of north-eastern Canada. Two training sets were used derived from
Canadian and Eastern Canadian transfer functions. Both models reconstructed similar climate patterns, but the
Canadian model provided temperatures generally 2–3 ˚C lower than the Eastern Canadian model. Three main
thermal changes inferred by chironomids were: (i) the Holocene Thermal Maximum, which occurred between 8
and 5k cal a BP, with temperatures generally higher than today’s, maximum temperatures between 8 and 6.5k cal
a BP, and an average of þ0.9 ˚C; (ii) the Medieval Climate Anomaly around 1.1–1.2k cal a BP with an amplitude
of þ0.7 ˚C; and (iii) a colder period reconstructed between the 14th and 19th centuries, corresponding to the Little
Ice Age, with summer temperatures on average 0.5 ˚C lower than the climate normal. For each of these three
climatic events, the timing and the amplitude of changes were similar with other published regional, North
American and Northern Hemisphere records. Copyright #2018 John Wiley & Sons, Ltd.
KEYWORDS: Chironomidae; Holocene Thermal Maximum; Little Ice Age; Medieval Climate Anomaly; transfer functions.
Introduction
Paleoclimatic records covering the Northern Hemisphere are
still too sparse to obtain a precise representation of past
climate change (Naulier et al., 2015). For example, the
amplitude and duration of the Medieval Climate Anomaly
(MCA) remain of debate (Mann and Jones, 2003; Viau et al.,
2006; Esper and Frank, 2009). The MCA has been character-
ized by notably higher temperatures between 950 and 1200
AD over a large part of the North Atlantic, in southern
Greenland, the Eurasian Arctic and parts of North America
(Lamb, 1965). However, its amplitude, location and extent
are still undefined for most parts of the Northern Hemisphere.
Another period of concern in North America is the Holocene
Thermal Maximum (HTM). In Canada, the HTM seems to
have varied between shorter (7–5k cal a BP) and longer (10–
3k cal a BP) durations and its amplitude was different
depending on the location (Renssen et al., 2012). Thus, it is
impossible to strictly define these periods of interest in North
America since quantitative long-term data are missing in
several regions.
Climate models need quantitative data to be able to
accurately predict future ecosystem functioning. Proxy-
climate-based reconstructions must be realistic and reliable,
and if possible they should be validated by independent
climate reconstructions from other proxies (Birks, 2003).
Climate (temperature, precipitation) records from biological
proxies such as pollen and chironomids preserved in
lacustrine deposits allow adequate long-term information to
quantify past climatic variability (Larocque-Tobler et al.,
2012; Heiri et al., 2015).
Chironomids (Insecta: Diptera: Chironomidae) have been
shown to generate high-resolution independent air tempera-
ture records due to their capacity to respond rapidly to
climatic fluctuation with their short generation time and the
ability of winged adults to move readily from site to site
(Walker and Mathewes, 1987; Larocque-Tobler et al., 2015).
The head capsule of the chironomid larvae is made of chitin,
a substance resistant to degradation, thus allowing the larval
exoskeletons to be preserved for thousands of years. The
distribution of chironomids has been shown to be influenced
by air/water temperatures (Walker and Mathewes, 1987). In
addition, with the help of transfer functions (Birks and Birks,
1998), they can record low-amplitude summer temperature
changes (Brooks, 2006; Larocque-Tobler et al., 2015). For
example, when comparing instrumental data to chironomid-
inferred temperatures, the differences were on average
0.75 ˚C in a Polish lake (Larocque-Tobler et al., 2015) and
0.65 ˚C in a Swiss lake (Larocque et al., 2009).
In the boreal region of north-eastern Canada, Holocene
quantitative reconstructions of past climate change are mostly
based on pollen analysis using the modern analogue tech-
nique (Viau and Gajewski, 2009; Viau et al., 2012). Paleocli-
matic data are available but with contradictory information.
For instance, according to Viau and Gajewski (2009) and
Viau et al. (2012), the MCA was cooler than the present day
(1961–1990 AD) while warmer conditions were recon-
structed in other parts of the country (Naulier et al., 2015).
Thus, doubts persist regarding the recording of this climate
anomaly in north-eastern Canada, as well as for the other
Correspondence: Lisa Bajolle, as above.
E-mail: lisa.bajolle@uqat.ca; bajolle.lisa@gmail.com
Copyright #2018 John Wiley & Sons, Ltd.
JOURNAL OF QUATERNARY SCIENCE (2018) ISSN 0267-8179. DOI: 10.1002/jqs.3022

major climate changes, notably those of weak amplitude,
such as the Little Ice Age (LIA). Consequently, it is important
to increase our knowledge of millennialscale temperature
fluctuations using proxies able to capture low-amplitude
climatic fluctuations. In this context, the main goal of this
study is to present a new Holocene mean August air
temperature reconstruction in the coniferous boreal forest of
north-western Quebec (Canada) based on subfossil chirono-
mid assemblages, allowing a discussion of the duration and
amplitude of the HTM, MCA and LIA for this region.
Methods
Study area and site
The study site is located in central Quebec (Fig. 1), 66 km
north of the town of Chibougamau. ‘Lac Aur
elie’ (unofficial
name; 50˚2501200N, 74˚1304700 W; 440 m a.s.l.) covers an area
of 1 ha and its maximum water depth (10 m) is in the central
part of the lake. The lake is located in the black spruce 
feather moss bioclimatic domain (Saucier et al., 2009)
dominated by black spruce (Picea mariana Mill.) and jack
pine (Pinus banksiana Lamb.). Mean annual temperature from
the closest meteorological station [Chapais 2: 1971–2000,
49˚470N, 74˚510W] is 0.0 1.3 ˚C with an August temperature
average of 14.9 1.4 ˚C. Annual and August precipitation
averages are 961 and 105 mm, respectively (Environnement
Canada, 2017). Further details about the study site character-
istics can be found in El-Guellab et al. (2015).
Coring, LOI analysis and chronology
Two overlapping sediment cores (providing a continuous
record) were taken in March 2010 on a frozen surface at
the center of the lake using a modified Livingstone-type
square-rod piston corer (100 5 cm) (Wright et al., 1984).
Sediments were covered in polyurethane and aluminum
foil for transportation to the laboratory and preserved at
Figure 1. Location map of the study lake ‘Lac Aur
elie’ in central Quebec, Canada.
Copyright #2018 John Wiley & Sons, Ltd. J. Quaternary Sci. (2018)
2 JOURNAL OF QUATERNARY SCIENCE

4 ˚C before they were sliced for analyses. The water–sediment
interface was collected using a Kajak-Brinkhurst (KB) gravity
corer (Glew, 1991). Sub-samples (1 cm
3
) were used for
measuring the organic matter content by loss-on-ignition
(LOI) at 550 ˚C for 4 h (Heiri et al., 2001). Radiocarbon dates
[calibrated at 2 sigma ranges based on the Intcal13.14C data
set (Reimer et al., 2013)] and chronology have already been
published by El-Guellab et al. (2015). A summary is presented
in Table 1.
Chironomid analysis
A total of 180 samples were analysed (Supporting Informa-
tion, Table S1) at a temporal resolution varying between ca. 5
and 165 years (ca. 45 years on average). Chironomid head
capsules were extracted from 2-cm
3
subsamples by soaking
the samples overnight in 10% KOH. The subsamples were
then water-rinsed through a sieve of 100 mm. The remnant
was poured into a Bogorov tray and observed under a Leica
stereomicroscope at a magnification of 20. Individual
subfossil head capsules were picked with fine forceps and
mounted on a microscope slide in a drop of Hydromatrix.
Previous studies suggested that at least 50 head capsules
should be mounted for temperature reconstructions (Larocque
et al., 2001; Heiri and Lotter, 2010), but samples larger than
30 head capsules can also provide inferences below the error
of the model (root mean square error of prediction, RMSEP)
(Larocque et al., 2009; Larocque-Tobler et al., 2016).
Taxonomic identification was made using a Motic micro-
scope at a magnification of 400–1000following the
taxonomic keys of Brooks et al. (2007), Larocque and Rolland
(2006), Oliver and Roussel (1983) and Wiederholm (1983).
Statistical analysis, paleoecological diagrams and
constrained zonation
A chironomid percentage diagram was drawn using C2
software (Juggins, 2003). Detrended correspondence analysis
(DCA) was performed using ade4 and vegan packages from R
v3.2.2 (Borcard et al., 2011) on the n(number of samples) by
p(number of taxa) chironomid matrix of percentages. Data
were square root transformed to stabilize the variance. Rare
taxa (present in only one sample or with a relative abundance
always <2%) were removed from the analysis. The length of
the first DCA axis determines whether the distribution of the
data set along this axis is linear or unimodal (Borcard et al.,
2011). Here, a gradient of 1.95 standard deviation units (SD)
was obtained, suggesting linear techniques such as principal
components analysis (PCA) were appropriate on our data set.
A PCA was then performed using C2 on the same data
matrix used for the DCA. This method was coupled with a
constrained sum-of-squares cluster analysis (CONISS) using
the program ZONE version 1.2 (Juggins, 1991) to highlight
major changes in assemblage composition (Grimm, 1987).
The optimal number of significant zones created was deter-
mined by a broken stick model (Bennett, 1996). Percentage
diagrams of warmer-than-today and colder-than-today taxa
were made using the temperature optimum (Table 2) for each
taxon obtained using the Eastern Canadian calibration set (see
below). Percentages of eutrophic, oligo-mesotrophic, littoral
and profundal taxa were calculated using ecological descrip-
tions from Brooks et al. (2007). To better understand changes
in the lake’s conditions, ratios using these percentages were
calculated. A >1 eutrophic/oligo-mesotrophic ratio suggests
a tendency towards eutrophic conditions, whereas a ratio <1
suggests a tendency towards oligo-mesotrophic conditions. A
littoral/profundal ratio >1 indicates a dominance of littoral
taxa and a dominance of profundal taxa if it is <1.
Transfer functions
Two chironomid-based transfer functions were used to recon-
struct mean August air temperature and evaluate their
reliability in inferring past climate changes. The Eastern
Canadian transfer function was first published by Larocque
(2008) and was modified. Two lakes were added, and 11
lakes had their temperature measurements changed to mean
August air temperature (Lakes A–H) as the earlier version of
the calibration possessed only punctual measurements. In the
new Eastern Canadian model, mean August temperature
varied from 3 to 21 ˚C (instead of 27 ˚C) for a temperature
gradient of 18 ˚C. The new data will be available on the
NOAA website once the paper is published. The calibration
set comprises 75 lakes and 79 taxa. The WAPLS
999-bootstrap transfer function with two components yielded
a correlation coefficient (r
2boot
) of 0.85, an RMSEP of
1.67 ˚C and a maximum bias of 3.05 ˚C.
The second transfer function combined data sets from
Canada (Fortin et al., 2015). It comprises 485 lakes and 78
taxa. Mean August temperatures varied from 0.3 to 15.7 ˚
C (gradient ¼16 ˚C). It contains 52 of the 75 lakes in the
Eastern Canadian transfer function of Larocque (2008). The
correlation coefficient (r
2boot
) is 0.73, the RMSEP is 1.8 ˚C and
the maximum bias is 2.9 ˚C (Fortin et al., 2015).
To verify if the transfer functions could be applied to the
fossil assemblages of Lac Aur
elie, we estimated if fossil
assemblages had modern analogues using the minimum
distance to modern assemblages (Overpeck et al., 1985). If
the minimum distance was within the 1st–5th percentiles, the
assemblages were considered as having ‘good analogues’. If
the distance was above the 20th percentile, the samples
were considered as having no analogues. Furthermore, the
goodness-of-Fit was calculated by passively adding downcore
samples of Lac Aur
elie into a Canonical Correspondence
Analysis (CCA) analysis of the Eastern Canadian transfer
function samples constrained by temperature (Heiri and
Table 1. Radiocarbon dates at different depths. Dates were obtained from terrestrial plant macroremains. Chronology (not show in this study)
was obtained from accelerated mass spectrometry (AMS) and has already been published by El-Guellab et al. (2015).
Lab code Depth (cm)
14
C age (a BP) Calibrated
14
C age range (cal a BP; 2s) Materials
Poz-35983 43–44 2870 30 3007 (2879–3136) Plant macroremains
Poz-35984 111–112 3990 35 4443 (4319–4568) Plant macroremains
Poz-36014 163–164 4750 35 5457 (5329–5586) Plant macroremains
Poz-36016 220–221 6140 40 7047 (6931–7163) Plant macroremains
Poz-36017 236–237 6490 40 7396 (7317–7476) Plant macroremains
Poz-36018 326–327 7460 50 8279 (8185–8373) Plant macroremains
Poz, Pozna
n Radiocarbon Laboratory.
Copyright #2018 John Wiley & Sons, Ltd. J. Quaternary Sci. (2018)
MAJOR POSTGLACIAL SUMMER TEMPERATURE CHANGES IN 3

Table 2. Occurrence of taxa in the sediment of Lac Aur
elie (total of 179 samples). Comparison of temperature optima obtained with weighted
averaging for the three transfer functions: (a) Eastern Canadian transfer function (Larocque, 2008); (b) Canadian transfer function (Fortin et al.,
2015) and temperature categories.
Taxa identified in sediment Occurrence in
sediment Optima Category
(a) Eastern
Canada
(b) Canada Colder than today
(15 ˚C)
Warmer than today
(15 ˚C)
Ablabesmyia spp. 131 14.2 4.5 X
Allopsectrocladius spp. 33 15.8 X
Brillia spp. 2 14.4 10.4 X
Chaetocladius spp. 34 13.8 9.2 X
Chironomus spp.
Chironomus anthracinus-type 164 9.7
Chironomus plumosus-type 147 12.1 Merged into
Chironomus spp.
X
Cladopelma lateralis-type 75 14.7 Merged into
Chironomus spp
X
Cladotanytarsus mancus-type 87 13.5 10.2 X
Constempellina spp. 34 12.2 9.3 X
Corynocera oliveri-type 45 14.5 11.2 X
Corynoneura spp. 43 10.4 8.3 X
Cricotopus/Orthocladius 7.3 7.9 X
Cricotopus spp. 96 7.7
Cryptochironomus spp. 48 11.8 Merged with
Orthocladius
X
Dicrotendipes nervosus-type 119 14.9 11.3
Einfeldia spp. 11 14.5 10.0 X
Endochironomus tendens-type 54 12.9 11.8 X
Glyptotendipes pallens-type 36 15.6 11.1 X
Heterotrissocladius spp. 14.7 10.9 X
H. grimshawi-type 17 7.7
H. marcidus-type 17 12.4 Merged in
Heterotrissocladius
X
H. subpilosus-type 6 12.0 Merged in
Heterotrissocladius
X
Labrundinia spp. 2 8.2 Merged in
Heterotrissocladius
X
Lauterborniella spp. 60 Not in model 13.9
Limnophyes spp. 19 22.6 13.4 X
Mesocricotopus spp. 8 17.0 8.6 X
Micropsectra spp. 10.0 8.7 X
Micropsectra bidentata-type 19 6.2
Micropsectra insignilobus-type 44 14.7 Merged in Micropsectra X
Micropsectra radialis-type 47 12.9 Merged in Micropsectra X
Microtendipes pedellus-type 130 9.4 Merged in Micropsectra X
Nanocladius spp. 3 13.5 11.0 X
Orthocladius spp. 39 12.4 10.8 X
Pagastiella spp. 26 13.5 X
Parachaetocladius spp. 2 14.5 11.2 X
Parachironomus varus-type 5 Not in model Not in model
Paracladius spp. 1 13.6 10.4 X
Paracladopelma spp. 2 7.4 4.5 X
Paracricotopus spp. 1 13.8 X
Parakiefferiella spp. 40 11.0 9.6 X
Paratanytarsus spp. 96 10.7 8.4 X
Paratendipes nudisquama-type 13 12.8 Merged in Tanytarsina X
Pentaneurini spp. 83 Not in model 11.9 X
Phaenopsectra spp. 9 14.9 9.3
Polypedilum nubeculosum-type 101 14.9 11.0
Procladius spp. 168 18.5 11.0 X
Psectrocladius septentrionalis-type 55 14.5 9.5 X
Psectrocladius sordidellus-type 141 12.2 10.6 X
Pseudochironomus spp. 63 12.4 10.9 X
Pseudosmittia spp. 7 Not in model 11.9
Rheocricotopus spp. 1 15.5 8.5 X
Sergentia coracina-type 20 Not in model 9.1
Smittia spp. 5 25.8 7.1 X
Stempelinella spp. 21 11.6 Not in model X
Stenochironomus spp. 2 14.6 10.6 X
Stictochironomus spp. 10 Not in model Not in model
Tanytarsina 7.9 7.4 X
Copyright #2018 John Wiley & Sons, Ltd. J. Quaternary Sci. (2018)
4 JOURNAL OF QUATERNARY SCIENCE

Lotter, 2010), using the CANOCO 4.5 program. If the
distance between downcore and transfer function samples
was above the 10th percentile, the downcore sample was
characterized as not having a good fit to temperature (Andr
en
et al., 2015). Thirdly, a percentage of the fossil taxa present in
the training set was calculated. A reconstruction should be
considered adequate if samples have high percentages of
fossil taxa present in the transfer function. Finally, to interpret
and compare the chironomid-inferred temperature with other
paleoecological climate records, a LOESS regression (span
¼0.2) was applied.
Results
Organic matter content
The 335-cm-long core was composed entirely composed of
gyttja. At the bottom of the core (326327 cm), an age of ca.
8.2–8.4k cal a BP (2 sigma range) was determined (Table 1).
The uppermost 22 cm of the sediment was used completely
for biological analysis, and thus LOI analysis could not be
performed for this section of the core. Before ca. 7.7k cal a
BP, the proportions of organic matter were below 60%.
Between ca. 7.7 and 7.8k cal a BP, the percentages were very
low (<20%) and the sediment was composed of a sandy
layer.
Chironomid analysis
Of the 180 samples analyzed, one (48 cm) did not have any
head capsules. Fifteen samples with fewer than 30 head
capsules were merged for a total of 164 samples. Of those
164, 35 samples had head capsule numbers between 32 and
49.5, and thus 80% of the samples had more than 50 head
capsules. In total, 63 taxa were identified to genus or species
morphotypes. PCA axes one and two had eigenvalues of 0.25
and 0.11, respectively. The 25 most abundant taxa (percen-
tages reaching at least 10%) are shown in Fig. 2(A).
Five significant zones (Ach-1 to Ach-5) were numeri-
cally identified (Fig. 2A). In zone Ach-1 (ca. 8.3–6.7k cal
a BP), the PCA axis one scores were positive. This zone
was dominated by taxa with temperature optima above
13.5 ˚C (Fig. 2B) such as Dicrotendipes nervosus-type,
Procladius spp., Tanytarsus mendax-type, Polypedilum
nubeculosum-type and Endochironomus tendens-type.
Colder-than-today taxa (Fig. 2B) reached 90% at the
beginning of the record (8.2–8.0k cal a BP) and many of
the cold indicators such as Corynocera oliveri-type, Micro-
psectra radialis-type and Heterotrissocladius marcidus-type
(Fig. 2a) had their highest percentages of the record in the
first few samples of this zone. The lake was possibly
oligotrophic during this zone, as the ratio of eutrophic/
oligo-mesotrophic taxa was below 1 (Fig. 2bB). Littoral
taxa dominated after 8.1k cal a BP.
In zone Ach-2 (ca. 6.7–5.5k cal a BP), PCA axis 1 scores
were below 0 (Fig. 2B). Warm indicators had percentages
generally below average except in one sample at ca. 6.2k cal
a BP. Tanytarsus spp. dominated the assemblages with
percentages around 40%. Chironomus plumosus-type, C.
anthracinus-type, T. lugens-type and Procladius spp. were
also well represented in the assemblages. Unstable conditions
were identified, with oscillating eutrophic/oligo-mesotrophic
and littoral/profundal ratios being recorded (Fig. 2B).
Zone Ach-3 (ca. 5.5–4.9k cal a BP) was characterized by
PCA axis 1 scores above 0. The percentages of warm taxa
were above average, and the assemblages were dominated by
D. nervosus-type, T. mendax-type, Procladius spp., P. nube-
culosum-type, C. mancus-type and Pseudochironomus spp.
The ratios of eutro/oligo-mesotrophic taxa were slightly below
1, suggesting mesotrophic conditions. Littoral taxa
dominated.
In zone Ach-4 (ca. 4.9–0.6k cal a BP), PCA scores were all
below 0. This zone was divided into two subzones based on
changes in taxa percentages and PCA axis 2 scores, which
were above 0 after ca. 1.5k cal a BP.
In subzone Ach-4a (ca. 4.9–1.7k cal a BP), the percentages
of warm taxa decreased below average, suggesting conditions
colder than today. The dominant taxon was Tanytarsus spp.,
a taxon which is considered an indicator of colder-than-today
conditions. The lake had a tendency towards oligo-meso-
trophy with ratios generally below 1. Littoral taxa dominated.
In subzone Ach-4b (1.7 k cal a BP to 600 cal a BP),
changes in PCA axis 2 scores were observed and the number
of head capsules was generally between 30 and 70. Both
cold and warm taxa were at average values, suggesting a
slight climatic amelioration to warmer conditions. The main
changes in taxa were the strong decrease in Tanytarsus spp.
Table 2. (Continued)
Taxa identified in sediment Occurrence in
sediment Optima Category
(a) Eastern
Canada
(b) Canada Colder than today
(15 ˚C)
Warmer than today
(15 ˚C)
Tanytarsus lugens-type 145 8.3
Tanytarsus mendax-type 145 13.5 Merged in Tanytarsina X
Tanytarsus pallidicornis-type 74 14.1 Merged in Tanytarsina X
Tanytarsus spp. 173 12.9 Merged in Tanytarsina X
Tanytarsus glabrescens-type 95 12.6 Merged in Tanytarsina X
Tanytarsus with spur on antenna 44 15.8 Merged in Tanytarsina X
Thiennemaniella spp. 16 15.3 Merged in Tanytarsina X
Thiennemanyia spp. 3 Not in model 9.9
Xenochironomus spp. 1 11.7 Not in model X
Zalutschia mucronata-type 49 Not in model Not in model
Number of taxa not included in the
training set
819
Maximum temperature (˚C) 25.8 13.9
Minimum temperature (˚C) 7.3 4.5
Copyright #2018 John Wiley & Sons, Ltd. J. Quaternary Sci. (2018)
MAJOR POSTGLACIAL SUMMER TEMPERATURE CHANGES IN 5

percentages and the increase in both Chironomus types.
These increases in Chironomus types were linked to sharp
increases in eutrophic/oligo-mesotrophic ratios and decreases
in littoral/profundal taxa (Fig. 2B); Chironomus types are
considered as eutrophic and profundal taxa (Brooks et al.,
2007).
During zone Ach-5 (ca. 600 cal a BP to present), PCA axis
1 scores remained below 0 and PCA axis 2 scores continued
to increase. Cold taxa were above average while warm taxa
were below average, suggesting colder conditions. P. sordi-
dellus-type, P. septentrionalis-type, Zalutschia mucronata-
type and Psectrocladius spp. and Cricotopus spp. dominated
for the first time. Trophic conditions had a tendency towards
oligo-mesotrophy with ratios below 1 and littoral taxa were
dominant.
Chironomid-inferred temperature reconstructions
The temperature reconstruction patterns obtained by the two
transfer functions are similar (Fig. 3). However, the Canadian
transfer function provided estimates which were 2–3 ˚C lower
than those of the Eastern Canadian model. The temperature
anomalies obtained from the Canadian model were mostly
inferred as lower than today, which does not fit with known
climate patterns. The Eastern Canadian reconstruction was
thus used to look at details of climate change.
The PCA axis 1 scores and inferred temperatures with the
Eastern Canadian model had a significant correlation (r
Pearson
¼0.62, p<0.05). Low temperatures were recorded before 8k
cal a BP with an average anomaly of 0.8 ˚C (Fig. 3).
Temperatures increased to reach a maximum around ca. 8k
cal a BP and remained high until ca. 6.7k cal a BP with an
average anomaly of 0.9 ˚C. During zone Ach-2 (ca. 6.7–5.5k
cal a BP), temperatures were generally lower than previously
with an average anomaly of 0.5 ˚C. During zone Ach-3 (ca.
5.5–4.9k cal a BP), the inferred temperatures were similar to
today’s with average anomaly of 0.2 ˚C. In zone Ach-4,
temperatures were lower than today (1.4 ˚C in average).
Changes occurred at around 1.2–1.1k cal a BP, with an
average anomaly of 0.7 ˚C, while the zone had lower temper-
atures than today (0.8 ˚C), on average. During zone Ach-5
(ca. 600 cal a BP to present) the average anomaly was
2.1 ˚C.
Discussion
Ecological interpretation
The first axis of the PCA could be interpreted as a climate
gradient probably related to temperature; the correlation
(r
Pearson
) between PCA axis 1 scores and the temperature
reconstruction was 0.62. Many taxa had higher temperature
optima in the Eastern Canadian model than in the Canadian
transfer function (Table 2). Most of the colder-than-today
indicators are similarly characterized in Brooks et al.
(2007).
Changes in PCA axis 2 scores starting at ca. 1.5k cal a BP
with PCA axis 1 scores remaining below 0 suggest that
temperature is not the only factor influencing the changes in
chironomid assemblages through time. In our core, PCA axis
2 scores were correlated with the ratios of oligo-meso/
eutrophic taxa and with the ratio of littoral/profundal taxa.
The effects of nutrients and water depth on chironomids have
been previously shown (Lotter et al., 1997; Larocque et al.,
2006) and models to reconstruct both parameters were
developed (Langdon et al., 2006; Engels et al., 2012). As
most members of the Tanytarsini tribe are both littoral (except
T. lugens-type) and eutrophic, their decrease starting at ca.
Figure 2. (A) Chironomid stratigraphy of the 26 most abundant taxa (percentages reaching at least 10% in more than four samples) and the number of head capsules identified in each sample. (B) Organic matter
content (loss-on-ignition at 550 ˚C). PCA axis 1 and 2 are the sample scores of the first and second axis in a principal components analysis. The vertical line is the PCA 0 score. The warmer- and colder-than-
13.5 ˚C taxa follow their sums of percentages in Table 2. Profundal and littoral taxa were defined using Brooks et al. (2007). Littoral/profundal and eutrophic/oligo-mesotrophic taxa ratios were calculated (see
Methods), the baselines are 1. Ratios >1 suggest oligo-mesotrophic conditions and a dominance of littoral over profundal taxa.
Copyright #2018 John Wiley & Sons, Ltd. J. Quaternary Sci. (2018)
6 JOURNAL OF QUATERNARY SCIENCE

1.5k cal a BP identifies trophic levels and water depth as two
of the main drivers of the observed ratios. In zone 4b (1.5k
cal a BP), profundal taxa dominated (as suggested by ratios
above 1), probably due to the increase in both types of
Chironomus. These taxa have hemoglobin (Walshe, 1950),
allowing them to survive short periods of oxygen depletion
(Brooks et al., 2007), suggesting that oxygen availability
might have changed during this period. Percentages of
Polypedilum nubeculoseum-type, associated with macro-
phytes, also increased during this time. The presence of
macrophytes has been shown to influence chironomids
(Langdon et al., 2008) and lower oxygen levels have been
recorded in lakes with a developed macrophyte community
(Rose and Crumpton, 1996). In zone 5, represented by a
further increase in PCA axis 2 scores, the percentages of both
Psectrocladius types and Cricotopus spp. largely increased
with Pentaneurini spp. and Ablabesmyia spp. Psectrocladius
is often associated with macrophytes (Brodersen et al., 2001;
Langdon et al., 2008) and is acidophilic (Pinder and Morley,
1995), as pH is another factor affecting chironomids (Orendt,
1999). Cricotopus,Ablabesmyia and Zalutschia spp. are
associated with vegetation (Brooks et al., 2007), while
Zalutschia spp. and Ablabesmyia occur in acidified lakes
(Brooks et al., 2007).
Based on the changes in the chironomid communities,
macrophytes, changes in oxygen and pH might have influ-
enced the chironomid assemblages since ca. 1.5k cal a BP.
However, low-variability climate changes, such as for the
LIA, have also been reconstructed from our assemblages (see
below), which suggests that, although other factors influenced
the assemblages, the pattern of chironomid-inferred tempera-
ture changes was still adequately reconstructed. Luoto and
Nevalainen (2017) have shown that chironomids can recon-
struct climate effectively even under the influence of eutro-
phication and pollution.
Reliability of the chironomid inferred temperature
reconstructions
The Canadian transfer function of Fortin et al. (2015) used a
very large number of lakes (435) and would, at first, be
considered as the most suitable to reconstruct climate in a
lake in the boreal forest of Quebec. However, only a few of
these lakes have August temperatures above 14 ˚C and the
residuals (fig. 4 in Fortin et al., 2015) suggest that higher
temperatures will be underestimated. As Lac Aur
elie’s current
August temperature is 15 ˚C, which is at the end of the
gradient of the Canadian transfer function, its present-day
Figure 3. Chironomid-inferred August air temperature anomaly (˚C) at Lac Aur
elie. (A) Comparison between the Eastern Canadian transfer
function and the Canadian transfer function of Fortin et al. (2015). The dashed curves indicate the smoothed reconstructions (loess, span 0.2). (B)
Details of the Eastern Canadian transfer function, which illustrate a more reliable record. The vertical lines indicate the zonation obtained from
the chironomid stratigraphy (Fig. 2A). The horizontal bold lines indicate the average anomaly of the chironomid zones.
Copyright #2018 John Wiley & Sons, Ltd. J. Quaternary Sci. (2018)
MAJOR POSTGLACIAL SUMMER TEMPERATURE CHANGES IN 7

temperature should possibly be underestimated with the
model of Fortin et al. (2015). If the temperatures were higher
in the past, the inferences would also be lower than expected.
Furthermore, because the temperature gradient is relatively
small (0.3 to 15.7 ˚C) and contains mostly lakes of low
temperatures, the optimum for each taxon is relatively low
(4.5–13.9 ˚C; Table 2), which explains why the reconstruction
is 2–3 ˚C lower than the one obtained with the other training
set. In addition, Fortin et al. (2015) merged certain taxa to a
lower taxonomic level, as identified in some of the calibration
sets. Consequently, this grouping could induce less realistic
inferred temperature reconstructions. However, it should be
remembered that the pattern of changes is very similar
between each transfer function.
The Eastern Canadian transfer function (Larocque, 2008) is
an extended version of the 52 lakes from Larocque et al.
(2006) that was used in the Canadian model of Fortin et al.
(2015). It contains a few lakes north (3 ˚C) and south (21 ˚C),
thus increasing the temperature gradient (321 ˚C). Although
it comprises fewer lakes (75) than the Canadian transfer
function, the increased gradient might provide more accurate
optima. The temperature optima in the Eastern Canadian
model vary between 7.5 and 20.5 ˚C (Table 2). However, this
training set does not include any lakes with August temper-
atures between 16.8 and 19 ˚C, temperatures which might
have been experienced at Lac Aur
elie in the past, so many of
the fossil samples did not resemble those found in the training
set. To obtain the best coverage (larger gradient and lakes
evenly distributed along the gradient) temperatures that may
have been experienced, lakes with temperatures between 16
and 19 ˚C should be added. Unfortunately, such data were
not available as lakes were not yet sampled within this range.
We assume that the inferences were possibly lower than they
should be. However, as weighted average partial least
squares (WA-PLS) is based on temperature optima for each
taxon found in the fossil record, it is important to include as
many fossil taxa as possible in the training set. This is the
case in the fossil record of Lac Aur
elie, with all samples
having more than 83% of their taxa found in the training set
lakes, and 133 of 179 samples (74%) having fossil taxa
represented in the training set above 95%.
A problem that needs to be considered in evaluating the
accuracy of a model is the absence of modern analogues.
Seventeen of the 164 samples had no modern analogues
(Fig. 3). These samples were mainly in the lower portion
of the sedimentary core (<6.7 cal a BP). Nevertheless,
WA-PLS methods perform well in non-analogue situations
because the estimates are based on modeled taxon temper-
ature optima assuming unimodal responses to temperature
(Birks and Birks, 1998). This allows the model to infer
temperatures outside the range of the calibration set. In
these cases, comparison with other regional paleoclimate
records is essential to evaluate the reliability of the
temperature reconstruction. The applicability of the Eastern
Canadian transfer function to the Lac Aur
elie samples was
further assessed using goodness-offit to temperature. None
of the Lac Aur
elie samples was above the 10th percentile,
so we assume that all downcore samples had good fit to
temperature.
8.2k cal a BP cold event and Holocene Thermal
Maximum
After the retreat of the last remnant of the glacier (Dyke,
2004), the chironomid-inferred temperature in Lac Aur
elie is
2–3 ˚C lower than today in three samples around ca. 8.2k cal
a BP (8282–8175 cal a BP). This might represent the so-called
8.2k cal a BP event (Alley et al., 1997). The Greenland ice
core record indicates that temperatures fell by ca. 3.3 ˚C and
this period lasted 150 cal years (Kobashi et al., 2007; Thomas
et al., 2007). Our results correspond to the ice core record in
both timing (ca. 110 years, limited by our sampling tech-
nique) and amplitude of change (3 ˚C).
The HTM lasted between 11 and 5k cal a BP in the
Northern Hemisphere, but with many regional variations
(Renssen et al., 2009). Based on pollen data from sites in
north-western Quebec, Viau and Gajewski (2009)
highlighted high temperatures between 6 and 2k cal a BP.
Our results show summer temperatures higher than or
similar to today between ca. 8.3 and 4.9k cal a BP, but
with a sharp decline around ca. 6.5k cal a BP. Our
reconstruction matches paleoclimate records obtained from
the Arctic (GISP2 ice core; Kobashi et al., 2010), from
pollen across north-eastern America (Viau et al., 2006) and
from a chironomid record on Baffin Island (Axford et al.,
2009) (Fig. 4A). The increase in temperature (þ2–3 ˚C in
three samples at the beginning of the record) was similarly
recorded in the ice core (þ3 ˚C) and from Baffin Island
(þ4 ˚C) (Fig. 4a). However, the combined pollen records
registered an increase in temperature of about only 1 ˚C
(Viau et al., 2006). This could be due to the merging of
various fossil records across north-eastern Canada. Stacking
of records of different amplitudes at many locations, as per
Viau et al. (2006), decreases the average variation. Another
reason for differences in amplitude is the use of different
models to infer climate. The modern analogue technique
generally provides changes of much lower amplitude than
the weighted average method used here (Birks, 2003). The
amplitude of change has been shown to reach 5 ˚C at the
highest latitudes, between 2.5 and 5 ˚C in our study region,
and smaller amplitudes at lower latitudes (Renssen et al.,
2012). Thus, the amplitude of change reconstructed by
chironomids at Lac Aur
elie seems to be plausible. Further-
more, the decrease in temperature from the late Holocene
(4.5–1.5k cal a BP) could be the result of the cold and wet
Neoglacial period (Viau and Gajewski, 2009).
Medieval Climate Anomaly
Chironomid assemblages showed an increase in temperature
(þ0.7 ˚C) around 1.1–1.2k cal a BP (Figs 3 and 4) probably
corresponding to the MCA recorded by various proxies in the
Northern Hemisphere (Mann et al., 2009). The pollen data
results of Viau et al. (2006) recorded across North America
agree with ours. Viau and Gajewski (2009) also recorded this
warming period in northern Quebec. Rolland et al. (2009)
inferred a warming period between 1160 and 1360 AD on
Southampton Island (Nunavut, Canada) based on chironomid
assemblages. In the central Northwest Territories (Canada),
chironomid-inferred temperature reconstructions have shown
a warming which occurred between 1 and 0.7k cal a BP
(Upiter et al., 2014). Arseneault and Payette (1997) also
observed a warming trend based on tree rings at the treeline
in north-western Quebec. When comparing our record to
other proxy records (Fig. 4B), the period around ca. 1.5–1k
cal a BP is clearly identified as warmer than today with a
concomitant timing (PAGES 2k Consortium, 2013). Using
chironomids, Millet et al. (2009) reconstructed a warming of
1.3 ˚C in the French Alps. Larocque-Tobler et al. (2012)
showed increased temperatures (1–2 ˚C) during the MCA in
two Swiss lakes and the composite records composed of tree-
rings and diatoms inferred an increase of 0.5 ˚C (Trachsel
et al., 2012), exemplifying that combining different fossil
proxies/sites smooths the record.
Copyright #2018 John Wiley & Sons, Ltd. J. Quaternary Sci. (2018)
8 JOURNAL OF QUATERNARY SCIENCE

Figure 4. (Continued).
Copyright #2018 John Wiley & Sons, Ltd. J. Quaternary Sci. (2018)
MAJOR POSTGLACIAL SUMMER TEMPERATURE CHANGES IN 9

During this period, PCA axis 2 scores suggested that factors
other than climate (macrophytes, oxygen, pH) affected the
chironomid assemblages at Lac Aur
elie. However, because the
MCA seems to have been plausibly reconstructed, it appears that
it did not completely affect the relationship with temperature.
Little Ice Age
The LIA is a colder-than-today period which occurred
between 1300 and 1850 AD (Matthews and Briffa, 2005) and
is defined in north-eastern Canada between the late 1500s to
the late 1800s (Payette and Delwaide, 2004). During the LIA,
the cooling with the largest amplitude occurred in northern
Quebec (Viau and Gajewski, 2009). The temporal resolution
of our record decreases during this period (about 40–80
years). The main features of this period are recognized with
the start of the cooling being recorded at around 1390 AD
and punctuated by a short warming period. Three colder
periods are generally seen in high-resolution records at
around 1650, 1779 and 1850 AD (Mann and Jones, 2003;
Mann et al., 2008). At Lac Aur
elie, the two lowest temper-
atures inferred were temporally close to these high-resolution
records at ca. 1663 and 1900 AD. However, the colder-than-
today temperatures of 1 ˚C (average 0.5 ˚C) were inferred in
only three samples and should be considered with caution.
This amplitude is consistent with high-resolution chironomid
records from Switzerland (i.e. Larocque-Tobler et al., 2012),
but larger than dendrochronological records in northern
Europe (Moberg et al., 2005) and combined proxy records
from Switzerland (Trachsel et al., 2010).
In conclusion, despite methodological limits to the
transfer function, good correspondences between our cli-
mate reconstruction and those obtained at other sites and
from diverse proxies suggests that chironomids are useful to
detect temperature changes at this site. Main climate events
such as the HTM, the Neoglacial period, the MCA and the
LIA have been recorded by chironomids at Lac Aur
elie.
Hence, further high-resolution regional studies, based on
chironomid analyses, should be conducted to improve our
understanding of past natural climate changes in boreal
forest ecosystems.
Supplementary Information
Table S1. Chironomid counts (percentage) for each level Lac
Aur
elie core.
Acknowledgements. This research was funded by the Natural
Sciences and Engineering Research Council of Canada (NSERC), the
Centre National de la Recherche (France), and the Institut

Ecologie &
Environnement through the GDRI ‘For^
ets Froides’. We thank the
French University Institute for its support. Our thanks to R. Julien and
L. Bremond for their participation in fieldwork. We greatly appreciate
the contribution of Dr Andrew Scott Medeiros for providing the
complete data of the Fortin et al. (2015) training set. We thank B.
Fr
echette for her advice and E. Chaste for technical support.
Abbreviations. DCA, detrended correspondence analysis; HTM,
Holocene Thermal Maximum; LIA, Little Ice Age; LOI, loss-on-
ignition; MCA, Medieval Climate Anomaly; PCA, principal component
analysis; RMSEP, root mean square error of prediction; WA-PLS,
weighted average partial least squares.
References
Alley RB, Mayewski PA, Sowers T et al. 1997. Holocene climatic
instability: a prominent, widespread event 8200 year ago. Geology
25: 483–486.
Andr
en E, Klimaschewski A, Self AE et al. 2015. Holocene climate
and environmental change in north-eastern Kamchatka (Russian
Far East), inferred from a multi-proxy study of lake sediments.
Global and Planetary Change 134: 41–54.
Arseneault D, Payette S. 1997. Reconstruction of millennial forest
dynamics from tree remains in a subarctic tree line peatland.
Ecology 78: 1873–1883.
Axford Y, Briner JP, Miller GH et al. 2009. Paleoecological evidence
for abrupt cold reversals during peak Holocene warmth on Baffin
Island, Arctic Canada. Quaternary Research 71: 142–149.
Bennett KD. 1996. Determination of the number of zones in a
biostratigraphical sequence. New Phytologist 132: 155–170.
Birks HH. 2003. The importance of plant macrofossils in the
reconstruction of Lateglacial vegetation and climate: examples
from Scotland, western Norway, and Minnesota, USA. Quaternary
Science Reviews 22: 453–473.
Birks HJB, Birks HJB. 1998. DG Frey and ES Deevey Review 1:
numerical tools in palaeolimnology-Progress, potentialities, and
problems. Journal of Paleolimnology 20: 307–332.
Borcard D, Gillet F, Legendre P. 2011. Numerical Ecology with R.
Springer: New York.
Brodersen KP, Odgaard BV, Vestergaard O et al. 2001. Chironomid
stratigraphy in the shallow and eutrophic Lake Sobygaard,
Denmark: chironomid-macrophyte co-occurrence. Freshwater Biol-
ogy 46: 253–267.
Brooks SJ. 2006. Fossil midges (Diptera: Chironomidae) as palae-
oclimatic indicators for the Eurasian region. Quaternary Science
Reviews 25: 1894–1910.
Brooks SJ, Langdon PG, Heiri O. 2007. The identification and use of
Palaearctic Chironomidae in palaeoecology. QRA Technical Guide
No. 10. QRA: London.
Dyke AS. 2004. An outline of North American deglaciation with
emphasis on central and northern Canada. Developments in
Quaternary Sciences 2: 373–424.
El-Guellab A, Asselin H, Gauthier S et al. 2015. Holocene variations
of wildfire occurrence as a guide for sustainable management of
the northeastern Canadian-Boreal forest. Forest Ecosystems 2:1.
Engels S, Cwynar LC, Rees ABH et al. 2012. Chironomid-based water
depth reconstructions: an independent evaluation of site-specific
and local inference models. Journal of Paleolimnology 48: 693–709.
Environnement Canada. 2017. Normales Climatiques au Canada,
1971–2000, service M
et
eorologique du Canada. Available at:
climate.weatheroffice.ec.gc.ca/climate_normals/.
Esper J, Frank D. 2009. The IPCC on a heterogeneous Medieval
Warm Period. Climatic Change 94: 267–273.
Fortin M-C., Medeiros AS, Gajewski K et al. 2015. Chironomid-
environment relations in northern North America. Journal of
Paleolimnology 54: 223–237.
3
Figure 4. (A) Comparison of various temperature reconstructions during the past 9k cal a BP. The data were obtained from the National Oceanic
& Atmospheric Administration (NOAA) World Data Center for Paleoclimatology. (a) Chironomid-inferred August mean temperature anomaly,
span of 0.2 (black line) and 95% confidence interval (light grey shading) at Lac Aur
elie for the past 8.5k cal a BP related to today’s study area
temperature (15.0 ˚C) (this study). (b) Pollen-inferred North American mean July temperatures for the past 9k cal a BP (Viau et al., 2006). (c)
Chironomid-inferred July temperatures at Lake CF8 (Baffin Island, Canada) for the past 9k cal a BP (Axford et al., 2009). (d) GISP2 ice-core
inferred Greenland temperatures for the past 9k cal a BP (Kobashi et al., 2010). (B) Comparison of different temperature reconstructions during the
past 2k cal a BP. (e) Chironomid-inferred August mean temperature anomaly, span of 0.2 (black line) and 95% confidence interval (light grey
shading) at Lac Aur
elie related to today’s temperature of at Chibougamau (15.0 ˚C) (this study). (f) Northern Hemisphere Arctic temperatures from
1961 to 1990 (PAGES 2k Consortium, 2013). (g) Multi-proxy inferred extra-tropical Northern Hemisphere (90˚–30˚N) decadal mean temperature
anomaly (dark line) related to 1996–1990 with 2 standard deviation error bars (light grey shading) (Ljungqvist, 2010). (h) GISP2, ice-core inferred
Greenland temperature (Kobashi et al., 2010).
Copyright #2018 John Wiley & Sons, Ltd. J. Quaternary Sci. (2018)
10 JOURNAL OF QUATERNARY SCIENCE

Glew JR. 1991. Miniature gravity corer for recovering short sediment
cores. Journal of Paleolimnology 5: 285–287.
Grimm EC. 1987. CONISS: a FORTRAN 77 program for strati-
graphically constrained cluster analysis by the method of incre-
mental sum of squares. Computers and Geosciences 13: 13–35.
Heiri O, Ilyashuk B, Millet L et al. 2015. Stacking of discontinuous
regional palaeoclimate records: chironomid-based summer temper-
atures from the Alpine region. The Holocene 25: 137–149.
Heiri O, Lotter AF. 2010. How does taxonomic resolution affect
chironomid-based temperature reconstruction? Journal of Paleolim-
nology 44: 589–601.
Heiri O, Lotter AF, Lemcke G. 2001. Loss on ignition as a method for
estimating organic and carbonate content in sediments: reproducibility
and comparability of results. Journal of Paleolimnology 25: 101–110.
Juggins S. 1991. Zone Version 1.2. DOS Program for the Zonation
(constrained Clustering) of Palaeoecological Data. ECRC UCL:
London.
Juggins S. 2003. C2 Program. University of Newcastle.
Kobashi T, Severinghaus JP, Barnola J-M et al. 2010. Persistent multi-
decadal Greenland temperature fluctuation through the last millen-
nium. Climatic Change 100: 733–756.
Kobashi T, Severinghaus JP, Brook EJ et al. 2007. Precise timing and
characterization of abrupt climate change 8200 years ago from air
trapped in polar ice. Quaternary Science Reviews 26: 1212–1222.
Lamb HH. 1965. The early medieval warm epoch and its sequel.
Palaeogeography, Palaeoclimatology, Palaeoecology 1: 13–37.
Langdon PG, Holmes N, Caseldine CJ. 2008. Environmental controls
on modern chironomid faunas from NW Iceland and implications
for reconstructing climate change. Journal of Paleolimnology 40:
273–293.
Langdon PG, Ruiz Z, Brodersen KP et al. 2006. Assessing lake
eutrophication using chironomids: understanding the nature of commu-
nity response in different lake types. Freshwater Biology 51: 562–577.
Larocque I. 2008. Nouvelle fonction de transfert pour reconstruire la
temp
erature 
a l’aide des chironomides pr
eserv
es dans les
s
ediments lacustres. Institut national de la recherche scientifique,
Centre Eau, Terre & Environnement.
Larocque I, Grosjean M, Heiri O et al. 2009. Comparison between
chironomid-inferred July temperatures and meteorological data ad
1850–2001 from varved Lake Silvaplana, Switzerland. Journal of
Paleolimnology 41: 329–342.
Larocque I, Hall RI, Grahn E 2001. Chironomids as indicators of
climate change: a 100-lake training set from a subarctic region of
northern Sweden (Lapland). Journal of Paleolimnology 26:
307–322.
Larocque I, Pienitz R, Rolland N 2006. Factors influencing the
distribution of chironomids in lakes distributed along a latitudinal
gradient in northwestern Quebec, Canada. Canadian Journal of
Fisheries and Aquatic Sciences 63: 1286–1297.
Larocque I, Rolland N, INRS-ETE. 2006. Le guide visuel des
chironomides sub-fossils [i.e. sub-fossiles] du Qu
ebec 
al’
^
Ile
d’Ellesmere. Institut national de la recherche scientifique, Eau, terre
et environnement: Qu
ebec.
Larocque-Tobler I, Filipiak J, Tylmann W et al. 2015. Comparison
between chironomid-inferred mean-August temperature from
varved Lake
_
Zabi
nskie (Poland) and instrumental data since 1896
ad. Quaternary Science Reviews 111: 35–50.
Larocque-Tobler I, Filipiak J, Tylmann W et al. 2016. Corrigendum to
‘Comparison between chironomid-inferred mean-August tempera-
ture from varved Lake Zabinskie (Poland) and instrumental data
since 1896 ad’. [Quat. Sci. Rev. 111 (2015) 35-50]. Quaternary
Science Reviews 140: 163–167.
Larocque-Tobler I, Stewart MM, Quinlan R et al. 2012. A last
millennium temperature reconstruction using chironomids pre-
served in sediments of anoxic Seebergsee (Switzerland): consensus
at local, regional and Central European scales. Quaternary Science
Reviews 41: 49–56.
Ljungqvist FC. 2010. A new reconstruction of temperature variability
in the extra-tropical Northern Hemisphere during the last two
millennia. Geografiska Annaler A 92: 339–351.
Lotter AF, Birks HJB, Hofmann W et al. 1997. Modern diatom,
cladocera, chironomid, and chrysophyte cyst assemblages as
quantitative indicators for the reconstruction of past environmental
conditions in the Alps. I. Climate. Journal of Paleolimnology 18:
395–420.
Luoto TP, Nevalainen L. 2017. Quantifying climate changes of the
Common Era for Finland. Climate Dynamics 49: 2557–2567.
Mann ME, Jones PD. 2003. Global surface temperatures over the past
two millennia: global surface temperatures over the past two
millennia. Geophysical Research Letters 30.
Mann ME, Zhang Z, Hughes MK et al. 2008. Proxy-based
reconstructions of hemispheric and global surface temperature
variations over the past two millennia. Proceedings of the
National Academy of Sciences of the United States of America
105: 13252–13257.
Mann ME, Zhang Z, Rutherford S et al. 2009. Global signatures and
dynamical origins of the Little Ice Age and Medieval Climate
Anomaly. Science 326: 1256–1260.
Matthews JA, Briffa KR. 2005. The ‘Little Ice Age’: re-evaluation of an
evolving concept. Geografiska Annaler: Series A, Physical Geogra-
phy 87: 17–36.
Millet L, Arnaud F, Heiri O et al. 2009. Late-Holocene summer
temperature reconstruction from chironomid assemblages of Lake
Anterne, northern French Alps. The Holocene 19: 317–328.
Moberg A, Sonechkin DM, Holmgren K et al. 2005. Highly variable
Northern Hemisphere temperatures reconstructed from low- and
high-resolution proxy data. Nature 433: 613–617.
Naulier M, Savard MM, B
egin C et al. 2015. A millennial summer
temperature reconstruction for northeastern Canada using oxygen
isotopes in subfossil trees. Climate of the Past 11: 1153–1164.
Oliver DR, Roussel ME. 1983. The insects and arachnids of Canada.
Part 11. The genera of larval midges of Canada. Diptera:
Chironomidae. Minist. Supply Serv.
Orendt C. 1999. Chironomids as bioindicators in acidified streams: a
contribution to the acidity tolerance of chironomid species with a
classification in sensitivity classes. International Review of Hydro-
biology 84: 439–449.
Overpeck JT, Webb T, Prentice IC. 1985. Quantitative interpretation
of fossil pollen spectra: dissimilarity coefficients and the method of
modern analogs. Quaternary Research 23: 87–108.
Payette S, Delwaide A. 2004. Dynamics of subarctic wetland forests
over the past 1500 years. Ecological Monographs 74: 373–391.
Pinder LCV, Morley DJ. 1995. Chironomidae as indicators of water
quality with a comparison of the chironomid faunas of a series
of contrasting Cumbrian tarns. In Insects in a Changing Environ-
ment, Harrington R, Stork NE (eds). Academic Press: London;
272–293.
Reimer PJ, Bard E, Bayliss A et al. 2013. IntCal13 and Marine13
radiocarbon age calibration curves 0-50,000 years cal BP. Radio-
carbon 55: 1869–1887.
Renssen H, Sepp€
a H, Crosta X et al. 2012. Global characterization of
the Holocene Thermal Maximum. Quaternary Science Reviews 48:
7–19.
Renssen H, Sepp€
a H, Heiri O et al. 2009. The spatial and temporal
complexity of the Holocene thermal maximum. Nature Geoscience
2: 411–414.
Rolland N, Larocque I, Francus P et al. 2009. Evidence for a warmer
period during the 12th and 13th centuries ad from chironomid
assemblages in Southampton Island, Nunavut, Canada. Quaternary
Research 72: 27–37.
Rose C, Crumpton WG. 1996. Effects of emergent macrophytes on
dissolved oxygen dynamics in a prairie pothole wetland. Wetlands
16: 495–502.
Saucier J-P., Grondin P, Robitaille A, et al. 2009.

Ecologie foresti
ere.
In Man. For. Seconde

Edition Ouvrage Collect.

Editions Multi-
Mondes Ordre Ing. For. Qu
e. Qu
e.2: 165–315.
PAGES 2k Consortium. 2013. Continental-scale temperature vari-
ability during the past two millennia. Nature Geoscience 6:
339–346.
Thomas ER, Wolff EW, Mulvaney R et al. 2007. The 8.2 ka event
from Greenland ice cores. Quaternary Science Reviews 26: 70–81.
Trachsel M, Grosjean M, Larocque-Tobler I et al. 2010. Quantitative
summer temperature reconstruction derived from a combined
biogenic si and chironomid record from varved sediments of Lake
Silvaplana (south-eastern Swiss Alps) back to ad 1177. Quaternary
Science Reviews 29: 2719–2730.
Copyright #2018 John Wiley & Sons, Ltd. J. Quaternary Sci. (2018)
MAJOR POSTGLACIAL SUMMER TEMPERATURE CHANGES IN 11

Trachsel M, Kamenik C, Grosjean M et al. 2012. Multi-archive
summer temperature reconstruction for the European Alps, ad
1053–1996. Quaternary Science Reviews,46: 66–79.
Upiter LM, Vermaire JC, Patterson RT et al. 2014. Middle to Late
Holocene chironomid-inferred July temperatures for the central
Northwest Territories, Canada. Journal of Paleolimnology 52: 11–26.
Viau AE, Gajewski K. 2009. Reconstructing millennial-scale, regional
paleoclimates of boreal Canada during the Holocene. Journal of
Climate 22: 316–330.
Viau AE, Gajewski K, Sawada MC et al. 2006. Millennial-scale
temperature variations in North America during the Holocene.
Journal of Geophysical Research 111.
Viau AE, Ladd M, Gajewski K. 2012. The climate of North America
during the past 2000years reconstructed from pollen data. Global
and Planetary Change 84–85: 75–83.
Walker IR, Mathewes RW. 1987. Chironomids, lake trophic status,
and climate. Quaternary Research 28: 431–437.
Walshe BM. 1950. The function of haemoglobin in Chironomus
plumosus under natural conditions. Journal of Experimental Biol-
ogy 27: 73–95.
Wiederholm T. 1983. Chironomidae of the Holarctic region: keys and
diagnoses. Part 1: Larvae. Entomol Scandinavica. Supplement:1–457.
Wright HE, Mann DH, Glaser PH. 1984. Piston corers for peat and
Lake Sediments. Ecology 65: 657–659.
Copyright #2018 John Wiley & Sons, Ltd. J. Quaternary Sci. (2018)
12 JOURNAL OF QUATERNARY SCIENCE