Lake responses to historical land use changes in northern Spain: The contribution of non-pollen palynomorphs in a multiproxy study

Abstract

Environmental changes discussed in an earlier work using pollen, sedimentology, ostracods and charcoal proxies were evaluated with new non-pollen palynomorph data. The limits of non-pollen palynomorph biozones are consistent with the main environmental changes reported earlier in the Pre-Pyrenean Lake Estanya. The non-pollen palynomorph diagram from Lake Estanya Gran shows that changes in human activity during the last 2000 years are reflected in lake responses. Palaeoecological data from non-pollen palynomorphs have helped to determine the causes of these environmental changes, e.g. the lake use for hemp water-retting and its limnological impacts.The main change in the non-pollen palynomorph diagram is the expansion of Desmidiaceae, indicating more acid water conditions between 1220 AD and 1760 AD. The onset of Desmidiaceae in 1220 AD may be attributed to the development of farming and probably to the construction of water channels.Desmidiaceae undergo a considerable decline after 1760 AD, coinciding with the fall in hemp production in the XVIIIth century. Consequently, the lake water recovered the oligotrophic status that had prevailed before 1220 AD.The perturbation caused by the introduction of hemp retting practices in Lake Estanya led to changes in all the proxies such as sedimentology, non-pollen palynomorphs, pollen and faunal communities.

Full text

Lake responses to historical land use changes in northern Spain: The
contribution of non-pollen palynomorphs in a multiproxy study
S. Riera
a,
⁎, J.A. López-Sáez
b,1
, R. Julià
c,2
a
Seminar of Prehistoric Study and Research, Department of Prehistory, Ancient History and Archaeology, University of Barcelona,
C/ Baldiri Reixach s/n. 08028, Barcelona, Spain
b
Institute of Earth Sciences, Jaume Almera, CSIC, C/ Lluís Solé Sabarís s/n, 08028 Barcelona, Spain
c
Laboratory of Archaeobotany. Department of Prehistory, Institute of History, CSIC, C/ Duque de Medinaceli 6, 28014 Madrid, Spain
Received 18 November 2004; accepted 20 March 2006
Available online 21 June 2006
Abstract
Environmental changes discussed in an earlier work using pollen, sedimentology, ostracods and charcoal proxies were
evaluated with new non-pollen palynomorph data. The limits of non-pollen palynomorph biozones are consistent with the main
environmental changes reported earlier in the Pre-Pyrenean Lake Estanya. The non-pollen palynomorph diagram from Lake
Estanya Gran shows that changes in human activity during the last 2000 years are reflected in lake responses. Palaeoecological data
from non-pollen palynomorphs have helped to determine the causes of these environmental changes, e.g. the lake use for hemp
water-retting and its limnological impacts.
The main change in the non-pollen palynomorph diagram is the expansion of Desmidiaceae, indicating more acid water
conditions between 1220 AD and 1760 AD. The onset of Desmidiaceae in 1220 AD may be attributed to the development of
farming and probably to the construction of water channels.
Desmidiaceae undergo a considerable decline after 1760 AD, coinciding with the fall in hemp production in the XVIIIth
century. Consequently, the lake water recovered the oligotrophic status that had prevailed before 1220 AD.
The perturbation caused by the introduction of hemp retting practices in Lake Estanya led to changes in all the proxies such as
sedimentology, non-pollen palynomorphs, pollen and faunal communities.
© 2006 Elsevier B.V. All rights reserved.
Keywords: non-pollen palynomorphs; environmental history; Late Holocene; palaeolimnology; hemp retting
1. Introduction
Hemp production was an important economic
activity in Europe in historical times and one of the
main causes of past lake perturbation. Palaeobotanical
studies have demonstrated that hemp cultivation spread
during the 1st millenium BC in Eastern and Central
Europe (van Zeist et al., 1991; Fleming and Clarke,
1998; Kroll, 2001; Bouby, 2002), especially in Roman
Review of Palaeobotany and Palynology 141 (2006) 127 –137
www.elsevier.com/locate/revpalbo
⁎Corresponding author. Tel.: +34 933333466.
E-mail addresses: rieram@ub.edu (S. Riera), alopez@ih.csic.es
(J.A. López-Sáez), rjulia@ija.csic.es (R. Julià).
1
Tel.: +34 934095410.
2
Tel.: +34 914290626.
0034-6667/$ - see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.revpalbo.2006.03.014

times (Mercuri et al., 2002). Hemp cultivation under-
went another expansion between Early Medieval times
and the XVIIth century, when the activity began to
decline (e.g. Godwin, 1967; Gaillard and Berglund,
1988; Fleming and Clarke, 1998; Mercuri et al., 2002;
Miras et al., 2003; etc.), although hemp cultivation
continued in some regions until the XIXth century
(Peglar, 1993).
It is common knowledge that the retting process in
pools brings about changes in water chemistry,
increasing acidity, eutrophication and toxicity (Sharma
and van Sumere, 1992; Cox et al., 2001). Historical
sources also record the problems of pollution due to
hemp retting (Sanz, 1995; Cox et al., 2001). Palaeoe-
cological studies based on diatoms have demonstrated
that lake eutrophication occurs during periods of hemp
retting (Bradshaw et al., 2000; Lotter, 2001). Hemp
retting induces changes in the assemblages of water
organisms, in Cannabaceae pollen percentages and in
sediment composition. Moreover, hemp retting
requires water management in the form of water
channels.
A multiproxy study carried out in the sedimento-
logical record of Lake Estanya allowed us to describe
the environmental evolution of the central Pre-
Pyrenees in the last 2000 years (Riera et al., 2004).
Pollen, charcoal, ostracod and sedimentological prox-
ies provide evidence of intensive land management and
lacustrine responses to land use changes, water
management and climatic variability. In Lake Estanya,
the increase in hemp pollen coincides with changes in
other proxies, such as the disappearance of gastropod
and ostracod fauna (Riera et al., 2004). Moreover, the
presence of channels connecting the Estanya lakes
suggests water management probably related to hemp
retting and irrigation.
The Lake Estanya pollen diagram shows that
Cannabaceae values increased after 1360 AD and
reached a maximum of 25% in 1760 AD, indicating
that hemp cultivation and fiber production were
significant activities between Medieval times and the
XIXth century.
Pollen percentages of Cannabaceae allow us to
interpret the existence of hemp cultivation and hemp
retting in a water body (pools, peats and lakes).
Nevertheless, the use of pollen percentages to determine
water-retting is arguable (Whittington and Edwards,
1989). In this regard, some authors consider that hemp
pollen percentages exceeding 10–15% could be attrib-
uted to retting (e.g. Whittington and Edwards, 1989;
Peglar, 1993; Lotter, 2001) as has been confirmed by
evidence of Cannabaceae achenes in peat-ponds sedi-
ments (Bradshaw et al., 1981). By contrast, other
authors consider that these percentages must be higher
than 25% (Latalowa, 1992; Mercuri et al., 2002). Some
pollen diagrams reveal values exceeding 40–50%. In
theses cases, retting practices were confirmed (Brad-
shaw et al., 1981; Gaillard and Berglund, 1988;
Nakagawa et al., 2000; Cox et al., 2001).
Some authors suggest that other proxies could
furnish additional arguments to corroborate retting in a
water body. Lithological changes (Gaillard and Ber-
glund, 1988; Cox et al., 2001), diatoms (Bradshaw et al.,
2000; Lotter, 2001), plant fibres (Saarnisto et al., 1977)
and the presence of Potamogeton pollen (Bradshaw et
al., 1981) have been used as indicators of retting.
Our study provides new data on past land manage-
ment and on the use of Lake Estanya for retting on the
basis of non-pollen palynomorphs, which constitute a
useful tool in palaeoecological studies.
2. Description of the site
The Estanya lakes are located in the External Ranges
(reaching 900 m a.s.l.) of the Pre-Pyrenees, at the
northern boundary of the Ebro Basin (Fig. 1). This area
is made up of limestones affected by diapirs largely
composed of gypsiferous marls, dolostones, limestones,
ophite and occasional salt deposits. All the region is
deeply karstified and the Estanya lakes are dolines that
reach the water table.
The region has a Mediterranean continental climate
characterized by a long summer drought. The mean
annual rainfall is 625 mm and the mean annual
temperature is 12.2 °C (León Llamazares, 1991). The
coldest month is January with a mean temperature of
2.9 °C.
The lakes are located at the transition between the
Mediterranean (Buxo-Quercetum rotundifoliae) and the
Submediterranean (Violo-Quercetum fagineae)plant
domains (Romo, 1989; Conesa, 1991).
The present day Estanya lakes consist of three sink
holes with permanent water. Lake Estanya Gran is the
largest of these (Fig. 1). It is located at 670 m a.s.l. at
42°02′N and 0°32′E. The lake is shaped like a figure of
eight and is formed by two sink holes separated by a
ridge which is exposed in drier periods. The maximum
water depth is 22 m in the southern sink hole. The
maximum length is 850 m and its surface area is 18.8 ha
(Avila et al., 1984).
According to Avila et al. (1984), the lake is
monomictic, with a thermal stratification extending
from March to September. The water chemistry is
dominated by Ca
2+
and SO
4
2−
and the conductivity is
128 S. Riera et al. / Review of Palaeobotany and Palynology 141 (2006) 127–137

3280 μScm
−1
. The maximum content of phosphorous
is 1 μg-at PO
4
l
−1
and of nitrates 8.5 μg-at NO
3
l
−1
.
Values of alkalinity range from 2 to 3.5 meq l
−1
and pH
from 8 to 8.2. Maximum productivity of phytoplankton
occurs at the end of July (10,000 cellules ml
−1
),
coinciding with the Chlorophyceae bloom.
The lake has a negligible catchment area and is
mainly fed by underground springs. Nevertheless, the
water level in the three lakes is partially controlled by
artificial channels (Fig. 1) built with a dry-stone
technique commonly used in Medieval and Modern
times. The earliest written references to the cequia
Fig. 1. Location of Lake Estanya.
129S. Riera et al. / Review of Palaeobotany and Palynology 141 (2006) 127–137

(water channel) in the Estanya area date from the XIIth
century (Riera et al., 2004).
3. Materials and methods
The studied core EST 3.1, 157 cm long, is located at
the sill between the sink holes of Lake Estany Gran (Fig.
1). At the time of drilling (1990), the water column was
1.5 m (Riera et al., 2004). Further information on the
chronological model and other multiproxy data used in
the present discussion is available in Riera et al. (2004).
The EST 3.1 core underwent a non-pollen palyno-
morph analysis, using the samples treated for pollen
counts. Pollen treatment is based on a mechanical
method (Dricot and Leroy, 1989): after dissolution in
10% HCl, the samples were dispersed in a pyrophos-
phate solution, and then sieved through 250 μm and
10 μm meshes. This procedure allowed the preservation
of some siliceous microfossils such as diatoms.
Non-pollen palynomorphs were mainly identified
according to van Geel (2001) and percentages were
calculated on the basis of the same pollen sum used for
the pollen diagram (Fig. 2)(van Geel et al., 1981).
Non-pollen palynomorphs were clustered using the
stratigraphically constrained analysis CONISS (Grimm,
1987).
4. Results
Thirty-two types of non-pollen palynomophs were
identified (Fig. 2). According to the main taxonomic
groups, these types correspond to: phanerogam remains
(2 types), algae (3 Desmidiaceae, 2 Zygnemataceae, 1
Diatom, 1 Chlorophyceae and Botryococcus sp.), 5
fungal spores, 2 Cyanobacteria, 1 Turbellarian, 1
Cladocer, 1 Acarus, 1 rotifer, 5 unidentified invertebrate,
2 testate amoebae and 1 Achritarch.
Table 1 lists the non-pollen palynomorph types and
their taxonomic identification when available.
Six non-pollen palynomorph biozones were identi-
fied by cluster analysis (Fig. 2).
4.1. Biozone EST-6 (155–118 cm depth)
This biozone is characterized by the high diversity of
types and by the greatest percentages of Botryococcus
sp. and Campylodiscus cf. clypeus in the whole
sequence. Other well-represented types are Glomus cf.
fasciculatum,Pleospora sp., Oribatei and Gloeotrichia
sp. The maximum of Botryococcus sp., the presence of a
Neorhabdocoela flat worm and a peak of Spirogyra sp.
suggest high OM deposition in a shallow environment
(Haas, 1996). This littoral environment is characterized
by warmer waters probably influenced by seasonal
drying, as suggested by the occurrence of Pleospora sp.,
Closterium idiosporum and Gloeotrichia sp. (van Geel,
1978; van Geel et al., 1981, 1983, 1989).
Two sub-biozones can be differentiated. Sub-biozone
EST-6b (155–130 cm depth) records:
(i) high diversity of types
(ii) the occurrence of Closterium idiosporum, type 90
and Sporormiella sp.
(iii) the maximum percentages of Glomus cf.
fasciculatum
Sub-biozone EST-6a (130–118 cm depth) records
high values of Spirogyra sp., Neorhabdocoela unknown
and Botryococcus sp.. The benthic species Eurycercus
cf. lamellatus, which develops in rooted vegetation
habitats (van Geel et al., 1983, 1989), is present in this
biozone 6. Ceratophyllum sp., type 176, Mougeotia sp.,
Spirogyra sp. and Botryococcus sp. suggest stagnant
waters and a low lake level (Pals et al., 1980; van Geel et
al., 1983, 1989; López Sáez et al., 1998). Low water
levels are also attested by the high values of Campy-
lodiscus cf. clypeus, a benthonic diatom living in
alkaline and saline environments (van Dam et al., 1994).
High percentages for Glomus cf. fasciculatum,
especially at the top of EST-6b, suggest erosive
phenomena (van Geel et al., 1989). Some taxa are also
indicative of grazing activities, especially the copro-
philous fungus Sporormiella sp. (van Geel, 2001; van
Geel et al., 2003) in EST-6b. The presence of the
pyrophilous fungus Coniochaeta cf. ligniaria in EST-6a
suggests wildfires (López Sáez et al., 1998).
4.2. Biozone EST-5 (118–105 cm depth)
The diversity and abundance of non-pollen palyno-
morphs decrease in this biozone. All taxa diminish but
the presence of type 119, Rivularia sp., Spirogyra sp.
and Botryococcus sp. should be pointed out.
The disappearance of Campylodiscus cf. clypeus and
the occurrence of Centropyxis ecornis, a rhizopod
growing in permanent water conditions (van Geel et
al., 1986) could indicate less saline waters and a rise in
lake level. These changes are evidenced by a decrease in
indicators of stagnant and warm waters, such as
Ceratophyllum sp., type 176, type 178, Mougeotia sp.,
Gloeotrichia sp., Spirogyra sp.,Neorhabdocoela un-
known and Botryococcus sp.
The disappearance of the Coniochaeta cf. ligniaria
(López Sáez et al., 1998), the lack of grazing indicators
130 S. Riera et al. / Review of Palaeobotany and Palynology 141 (2006) 127–137

Fig. 2. Non-pollen palynomorph diagram. Biozones were based on the stratigraphically constrained cluster analysis CONISS (Grimm, 1987).
131S. Riera et al. / Review of Palaeobotany and Palynology 141 (2006) 127–137

such as Sporormiella sp. (van Geel, 2001; van Geel et
al., 2003) and the fall in Glomus cf. fasciculatum
suggest a change from extensive grazing to arable
activities (Fig. 3A). This change favoured the expansion
of mixed forest as a result of the decline in burning
management (Riera et al., 2004).
4.3. Biozone EST-4 (105–85 cm depth)
This biozone shows a major change in the lacustrine
environment. Non-pollen palynomorphs show marked
increases in algae such as Cosmarium sp.,Euastrum
insulare var. lacustre and Tetraedron cf. T. minimum,
Campylodiscus cf. clypeus diatom and in the testate
amoeba Arcella sp. Other taxa recording slight increases
are: Pleospora sp., Closterium idiosporum,Eurycercus
cf. lamellatus and type 176.
High percentages of Campylodiscus cf. clypeus
diatom suggest more saline conditions, whereas the
increase in Eurycercus cf. lamellatus indicates the
promixity of shore vegetation. Littoral conditions are
also supported by testate amoeba Arcella sp. These data
suggest that the water level reached a new minimum and
that littoral conditions prevailed at the drilled site as
evidenced by the increases in Pleospora sp. and
Closterium idiosporum.
Very high percentages of Desmidiaceae, such as
Cosmarium sp., Euastrum insulare var. lacustre and
Tetraedro n cf. T. mi n i m u m suggest eutrophic conditions
(Bakker and van Smeerdijk, 1982) and more acid waters.
The increase in Glomus cf. fasciculatum suggests
slight soil erosion (van Geel et al., 1989), probably
because of human activities despite the absence of
fungal spores such as Sporormiella sp. and type 90 (van
Geel, 2001; van Geel et al., 2003).
4.4. Biozone EST-3 (85–50 cm depth)
This biozone is characterized by a decrease in
Desmidiaceae and Campylodiscus cf. clypeus. Diversity
and concentration are higher in sub-biozone EST-3b
than in sub-biozone EST-3a. Sub-biozone EST-3b (85–
62 cm depth) is mainly characterized by the high Arcella
sp. percentages and by increases in Closterium idios-
porum, type 90, Potamogeton sp., Centropyxis ecornis
and Botryococcus sp. Although non-pollen palyno-
morph concentration decreases in sub-biozone EST-3a
(62–50 cm depth), this sub-biozone records an increase
in Rivularia sp., Spirogyra sp. and Centropyxis ecornis.
The decline in Desmidiaceae suggests less acid waters
(López Sáez et al., 1998), whereas the disappearance of
Campylodiscus cf. clypeus indicates less saline waters,
probably because of a rise in the water level.
In the sub-biozone EST-3b, Arcella sp., type 90,
Closterium idiosporum,Neorhabdocoela unknown and
Botryococcus sp. increase, suggesting environmental
conditions closer to biozone EST-6, characterized by
more or less stagnant waters. The leaf fragments of
Potamogeton sp. appear in this sub-biozone.
In the sub-biozone EST-3a, the disappearance of
Desmidiaceae (types 332D, 332F, 332C and 371) and
Neorhabdocoela sp. indicates a change to less eutrophic
conditions and more open waters. In this regard, the
increase in Centropyxis ecornis (van Geel et al., 1986)
and the onset of the acritarch Cymatiosphaera (Pals et
al., 1980; Bakker and van Smeerdijk, 1982) corroborate
the prevalence of open waters.
This interpretation is consistent with the presence of
indicators of more mesotrophic conditions, such as type
128A and Rivularia sp. (van Geel et al., 1981, 1989;
Haas, 1996).
Table 1
List of non-pollen palynomorphs types and their taxonomic
identification in the Estanya sequence
Type Taxonomic identification Taxonomic group
137 Ceratophyllum sp. Phanerogam
241 Potamogeton sp.
60 Closterium idiosporum Algae
313 Mougeotia sp.
315 Spirogyra sp.
(Diatom) Campylodiscus cf. clypeus
766 + 901 Botryoccocus sp.
332D+ 332F Cosmarium sp.
332C Euastrum insulare var. lacustre
371 Tetraedron cf. T. minimum
207 Glomus cf. fasciculatum Fungal spores
3B Pleospora sp.
113 Sporormiella sp.
172 Coniochaeta cf. ligniaria
90
146 Gloeotrichia sp Cyanobacteria
170 Rivularia sp.
353 Neorhabdocoela unknown Faunal remains
72D Eurycercus cf. lamellatus
36 Acari, Oribatei
103 Stephanoceros eichhornii
176 Copepoda unknown
88A Invertebrate unknown
178 Invertebrate unknown
219 Invertebrate unknown
180 Unknown
352 Arcella sp. Protozoa
530 Centropyxis ecornis
116 Cymatiosphaera Incerta
119
128A
132 S. Riera et al. / Review of Palaeobotany and Palynology 141 (2006) 127–137

Fig. 3. Correlation between the seven environmental episodes established in Riera et al. (2004) and the six biozones determined by the non-pollen palynomorphs. Selected proxies from the
environmental episodes are displayed on the left (A), whereas selected non-pollen palynomorphs are shown on the right (B).
133S. Riera et al. / Review of Palaeobotany and Palynology 141 (2006) 127–137

4.5. Biozone EST-2 (50–35 cm depth)
There is an increase in the abundance and diversity of
non-pollen palynomorphs. Cosmarium sp., Euastrum
insulare var. lacustre and Tetraedron cf. T. minimum
peak again. Percentages of Neorhabdocoela oocytes,
Ceratophyllum sp. and type 176 also increase. Potamo-
geton sp. renews its presence.
The increases of Eurycercus cf. lamellatus,Cerato-
phyllum sp. and Arcella sp. as well as the presence of
benthic diatom Campylodiscus cf. clypeus suggest a
slight decrease in the lake water level.
This non-pollen palynomorph composition resem-
bles that of biozone 4, suggesting the return of more
eutrophic and acid waters.
4.6. Biozone EST-1 (35–0 cm depth)
Non-pollen palynomorphs show a decrease in
diversity and abundance in the upper 35 cm of the
sequence. Increases are only recorded in type 90,
Sporormiella sp. and Spirogyra sp.
The reduction of Desmidiaceae suggests a new
change to less acid water conditions. This change
together with the low values of Botryococcus sp.
indicates a low trophic status of the lake.
The absence of thecamoebae, Eurycercus cf. lamel-
latus,Campylodiscus cf. clypeus,Closterium idios-
porum points to less saline conditions and open waters.
The increase in the coprophilous fungus Sporor-
miella sp. is evidence of grazing activities near the lake
(van Geel et al., 2003), resulting in an intensification of
soil erosion, as attested by the presence of Glomus cf.
fasciculatum (López Sáez et al., 1998).
5. Discussion
Fig. 3 displays the correlation between the seven
environmental episodes established in Riera et al. (2004)
and the six biozones of non-pollen palynomorphs.
Selected proxies from the environmental episodes are
shown in Fig. 3A(Riera et al., 2004), and selected non-
pollen palynomorph indicators of lake conditions and
land use are given in Fig. 3B.
The limits of biozones based on non-pollen palyno-
morphs (Fig. 3B) confirm the main environmental
episodes (Fig. 3A) described with pollen, sedimentolo-
gy, ostracods and charcoal (Riera et al., 2004). However,
the changes in each proxy can vary a few centimeters in
depth, suggesting that the proxies have a different
resilience and a different threshold response to environ-
mental changes.
From the bottom of the core dated from 160 AD to
1075 AD (environmental episode VII), non-pollen
palynomorphs record saline and littoral waters rich in
organic matter evidenced by Campylodiscus cf. clypeus
and Botryococcus sp. (Fig. 3B). In addition, Glomus cf.
fasciculatum indicates that the shore was close to the
drilling point. These littoral conditions are consistent
with the dominance of saline tolerant ostracod Cyprideis
torosa. High concentrations of this epiphytic ostracod
and the occurrence of gastropods and Neorhabdocoela
indicate the abundance of vegetal detritus (Fig. 3A),
which is typical of littoral areas.
The first peak of Cannabaceae pollen (5%) was
reported at 139 cm depth (Fig. 3A), suggesting the onset
of hemp cultivation circa 600–650 AD in the region
(Riera et al., 2004). This date coincides with the spread
of hemp between the Vth and IXth centuries in other
European regions (Godwin, 1967; Gaillard and Ber-
glund, 1988; Laitinen, 1996; Fleming and Clarke, 1998;
Cox et al., 2001).
Despite a lack of agrarian pollen at that time,
widespread deforestation occurred between 820 AD
and 1075 AD because of recurrent fires as evidenced by
the pyrophilous fungus Coniochaeta cf. ligniaria (Fig.
2) and the high concentration of microcharcoal particles
(Riera et al., 2004). The presence of the coprophilous
fungus Sporormiella sp. (Fig. 3B) during the period of
the highest percentages of arboreal pollen suggests
grazing activities close to the lake.
Non-pollen palynomorphs record a change between
1075 AD and 1220 AD, which correlates with the
environmental episode VI (Riera et al., 2004). Reduc-
tion in the abundance of non-pollen palynomorphs
indicates a less productive environment, whereas the
diminution of Campylodiscus cf. clypeus indicates open
water conditions and a less saline environment (Fig.
3B). The replacement of Cyprideis torosa by Candona
marchica also suggests a change towards freshwater
conditions (Fig. 3A). All these data support a rise in lake
level.
It is not easy to account for the rise in the lake level
given that medieval water channels transferring water
from one lake to another (Fig. 1) and to the irrigated
plain played an important role in lake hydrology.
Written sources indicate that water channels operated
in the region at least from 1154 AD (Riera et al., 2004).
The low values of Cannabaceae pollen suggest that
the earliest water channels were not constructed for
hemp retting.
A major change in lake conditions occurred after
1220 AD (environmental episode V), evidenced by high
abundance of Desmidiaceae, such as Cosmarium sp.
134 S. Riera et al. / Review of Palaeobotany and Palynology 141 (2006) 127–137

(Fig. 3A), indicating more acid waters and high
productivity.
Increases in Campylodiscus cf. clypeus and endo-
genic gypsum and the replacement of Candona
marchica by saline tolerant ostracods such as Herpeto-
cypris salina (Fig. 3A) suggest a decrease in the level of
the lake and more saline waters (Riera et al., 2004).
This fall in the water level has been interpreted as a
dry event corresponding to the Medieval Warm Period
(MWP) between 1220 AD and 1360 AD (Riera et al.,
2004).
A decline in the water level also occurrred between
820 AD and 1075 AD (Riera et al., 2004), but non-
pollen palynomorphs show different trophic status
conditions after 1220 AD. This change in trophic status
could be attributed to forest clearances and to the
expansion of farming activities, mainly olive cultivation
(Riera et al., 2004). The main difference between the
two episodes of low water level lies in the land use, i.e.
the expansion of olive cultivation after 1220 AD.
From 1360 AD until 1580 AD (environmental
episode IV), non-pollen palynomorphs recorded more
open water conditions due to the progressive increase in
water level. Between 1500 AD and 1580 AD, the low
presence of non-pollen palynomorphs (Fig. 2) suggests
a maximum water level which is in accordance with lake
water level reconstruction (Fig. 3A). The prevalence of
ostracod Candona marchica also indicates freshwater
conditions (Fig. 3A). Other proxies such as sedimen-
tology and gastropods confirm a deeper water column
(Riera et al., 2004).
One of the main features of this episode is the
expansion of cereal, olive and hemp cultivations (Fig. 3A).
Cannabaceae pollen attains values of 12%, confirm-
ing the spread of hemp cultivation near the lake. The
first references to hemp cultivation in written sources
date from the XIVth century (de Asso, 1798), which is
coeval with the medieval expansion of hemp in Europe
(e.g. Peglar, 1993; Fleming and Clarke, 1998; Mercuri et
al., 2002).
However, although these percentages of Cannabaceae
confirm hemp cultivation, some authors suggest that they
may not necessarily indicate retting in the lake.
Nevertheless, in the Estanya core, other proxies such as
the increase in Potamogeton sp. (Fig. 3A) suggest the
practice of retting (Bradshaw et al., 1981)althoughno
Potamogeton pollen has been found in the Estanya
sequence (Riera et al., 2004). The coincidence of
Potamogeton sp.remains with high percentages of
Cannabaceae can be attributed to human influence. In
this regard, the activities related to hemp processing
include cleaning and digging of the water channels. This
management could result in plant remains being dis-
charged into the lake, although there is no evidence of this.
Moreover, the disappearance of the gastropod fauna circa
1400 AD could be attributed to lake water toxicity.
A new change in the non-pollen palynomorph
diagram occurs between 1580 AD and 1760 AD
(environmental episode III). There is a change from
open water to more littoral environments, resulting
in conditions similar to those of episode V (1220–
1360 AD). This low water level is also deduced from
increases in Campylodiscus cf. clypeus (Fig. 3B), in the
saline tolerant ostracod Cyprideis torosa and in the
endogenic gypsum (Fig. 3A).
This episode corresponds to the maximum agrarian
expansion during the XVIIth and XVIIIth centuries. The
maximum hemp production in Spain occurred at the end
of the XVIIIth century because of the high demand from
the Spanish navy (Sanz, 1995). Hemp pollen percen-
tages reach 25%, a value that has been considered by
some authors as an indicator for hemp retting (Latalowa,
1992; Peglar, 1993; Mercuri et al., 2002). The expansion
of Potamogeton sp. remains corroborates the retting
practices (Bradshaw et al., 1981). Moreover, the high
percentage of detrital minerals (Fig. 3a) coinciding with
the Cannabaceae pollen peak indicates shore perturba-
tion due to the use of rocks to weigh down the bundles
of hemp. The ostracod population disappeared circa
1650 AD because of high water toxicity when the use of
retting was at its height.
Desmidiaceae such as Cosmarium sp. underwent a
sharp decline after 1760 AD (environmental episode II)
and the water composition recovered its chemical
properties that are characteristic of karstic oligotrophic
lakes.
A new lake status was attained after the decrease of
hemp production and retting practices (environmental
episodes I and II). Cannabaceae pollen diminishes to
less than 10% after 1760 AD as a result of the Spanish
hemp crisis (Sanz, 1995). However, hemp cultivation
still covered 53 ha around the lake in 1846 AD (Madoz,
1846) and the activity persisted into the first half of the
XXth century (Violant-Simorra, 1934). The XXth
century is characterized by depopulation and decline
in crop cultivation. In parallel, grazing activities
expanded as evidenced by an increase in Sporormiella
sp. (Fig. 3B).
6. Conclusions
Changes in hydrology and land use are recorded by
the non-pollen palynomorph assemblages. The main
environmental episodes based on pollen, sedimentology,
135S. Riera et al. / Review of Palaeobotany and Palynology 141 (2006) 127–137

ostracods and charcoal (Riera et al., 2004)were
confirmed by the limits of the non-pollen palynomorph
biozones. Palaeoecological data from non-pollen paly-
nomorphs have helped to determine the causes of
environmental change, e.g. the lake use for retting and
its limnological impacts.
The main change in the non-pollen palynomorph
diagram is highlighted by Desmidiaceae expansion, indicat-
ing more acid water conditions between 1220 AD and 1760
AD. The onset of Desmidiaceae in 1220 AD could be
ascribed to three factors: the expansion of farming, the low
water levels during the MWP (Riera et al., 2004)andthe
management of the lake water using water channels.
Desmidiaceae drastically diminished, coinciding with the
retraction of hemp production after 1760 AD, when water
recovered its former oligotrophic status.
High Cannabaceae pollen percentages between 1580 AD
and 1760 AD strongly suggest that retting occurred. This
conclusion is corroborated by changes in other proxies.
Potamogeton sp. underwent an expansion as reported by
Bradshaw et al. (1981). Moreover, the disappearance of
gastropods circa 1400 AD and ostracods circa 1650 AD
could be attributed to water toxicity caused by retting.
Furthermore, the detrital minerals also constitute a reliable
indicator of widespread retting.
Acknowledgements
This work was supported by the DGICYT (Spanish
Government) INVACAS-Project REN 2002-04592. The
authors are indebted to the useful comments and
suggestions of two anonymous reviewers.
References
Avila, A., Burrel, J.L., Domingo, A., Fernández, E., Godall, J., Llopart,
J.M., 1984. Limnología del Lago Grande de Estanya (Huesca).
Oecol. Aquat. 7, 3–24.
Bakker, M., van Smeerdijk, D.G., 1982. A palaeoecological study of a
Late Holocene section from “Het Ilperveld”, western Netherlands.
Rev. Palaeobot. Palynol. 36, 95–163.
Bouby, L., 2002. Le chanvre (Cannabis sativa L.): une plante cultivée
à la fin de l'âge du Fer en France du Sud-Ouest? C.R. Palevol. 1,
89–95.
Bradshaw, R.H.W., Coxon, P., Greig, J.R.A., Hall, A.R., 1981. New
fossil evidence for the past cultivation and processing of hemp
Cannabis sativa L. in eastern England. New Phytol. 89, 503–510.
Bradshaw, E.G., Rasmussen, P., Anderson, J., 2000. Long-term lake
responses to changing landscape: evidence from Dallund Sø,
Funen, Denmark. Department of Quaternary Geology, Lund
University LUNDQUA Report 37, Lund. http://www.sdu.dk/
Hum/ForandLand/histforsk/pollen/LAKERES.PDF.
Conesa, J.A., 1991. Flora i vegetació de les Serres Marginals Pre-
pirinenques compreses entre els rius Segre i Noguera Ribagorçana.
Ph. D. Thesis, University of Barcelona, Spain.
Cox, M., Chandler, J., Cox, Ch., 2001. The archaeological significance
of patterns of anomalous vegetation on a raised mire in the Solway
estuary and the processes involved in their formation. J. Archaeol.
Sci. 28, 1–18.
de Asso, I., 1798. Historia de la Economía Política de Aragón. CSIC-
Estación de Estudios Pirenaicos, 1947. Zaragoza.
Dricot, E., Leroy, S., 1989. Peptization and sieving for palynological
purposes. Geobound 2, 114–126.
Fleming, M.P., Clarke, R.C., 1998. Physical evidence for the antiquity
of Cannabis sativa L. J. Int. Hemp Association 5, 80–92.
Gaillard, M.-J., Berglund, B.E., 1988. Land-use history during the last
2700 years in the area of Bjäresjö, Southern Sweden. In: Birks, H.
H., Birks, H.J.B., Kaland, P.E., Moe, D. (Eds.), The Cultural
Landscape. Past, Present and Future. Cambridge University Press,
Cambridge, pp. 409–428.
Godwin, H., 1967. Pollen-analytic evidence for the cultivation of
Cannabis in England. Rev. Palaeobot. Palynol. 4, 71–80.
Grimm, E.C., 1987. CONISS: a Fortan 77 program for stratigraphi-
cally constrained cluster analysis by the method of incremental
sum of squares. Geosci. Computer 13, 13–35.
Haas, J.N., 1996. Neorhabdocoela oocytes—palaeoecological
indicators found in pollen preparations from Holocene
freshwater lake sediments. Rev. Palaeobot. Palynol. 91,
371–382.
Kroll, H., 2001. Literature on archaeological remains of cultivated
plants (1999/2000). Veget. Hist. Archaeobot. 10, 31–68.
Laitinen, E., 1996. History of hemp in Finland. J. Int. Hemp
Association 3 (1) http://www.calyx.net/~olsen/HEMP/IHA/
iha03115.html.
Latalowa, M., 1992. Man and vegetation in the pollen diagrams from
Wolin Island (NW Poland). Acta Palaeobot. 32, 123–249.
León Llamazares, A., 1991. Caracterización Agroclimática de la
Provincia de Huesca. MAPA, Madrid.
López Sáez, J.A., van Geel, B., Farbos-Texier, S., Diot, M.F., 1998.
Remarques paléoécologiques à propos de quelques palynomorphes
non-polliniques provenant de sédiments quaternaires en France.
Rev. Paléobiol. 17 (2), 445–459.
Lotter, A.F., 2001. The palaeolimnology of Soppensee (central
Switzerland), as evidenced by diatoms, pollen, and fossil pigment
analyses. J. Paleolimnol. 25, 65–79.
Madoz, P., 1846. Diccionario Geográfico-Estadístico-Histórico de
España y sus Posesiones de Ultramar. Est. Tip. P. Madoz y L.
Sagasti, Madrid.
Mercuri, A.M., Accorsi, C.A., Bandini, M., 2002. The long history of
Cannabis and its cultivation by the Romans in central Italy, shown
by pollen records from Lago Albano and Lago di Nemi. Veget.
Hist. Archaeobot. 11, 263–276.
Miras, Y., Guenet, P., Surmely, F., Michelin, Y., Tible, R., Walter-
Simonnet, A.-V., Richard, H., 2003. Histoire de l'environnement et
des dynamiques agro-pastorales dans le Massif du Cantal: la
tourbière de Roussy (Saint-Projet-de-Salers, France). Quaternaire
14, 265–278.
Nakagawa, T., Beaulieu, J.L., Kitagawa, H., 2000. Pollen-derived
history of timber exploitation from the Roman period onwards in
the Romanche valley, central French Alps. Veget. Hist. Archae-
obot. 9, 85–89.
Pals, J.P., van Geel, B., Delfos, A., 1980. Palaeoecological studies
in the Klokkeweel bog near Hoogkarspel (Prov. of Noord-
Holland). Rev. Palaeobot. Palynol. 30, 371–418.
Peglar, S.M., 1993. The development of the cultural landscape around
Diss Mere, Norfolk, UK during the past 7000 years. Rev.
Palaeobot. Palynol. 76, 1–43.
136 S. Riera et al. / Review of Palaeobotany and Palynology 141 (2006) 127–137

Riera, S., Wansard, G., Julià, R., 2004. 2000-year environmental
history of a karstic lake in the Mediterranean Pre-Pyrenees: the
Estanya lakes (Spain). Catena 55, 293–324.
Romo, A.M., 1989. Flora i vegetació del Montsec (Pre-Pirineus
Catalans). Arxius de la Secció de Ciències, vol. 90. Institut
d'Estudis Catalans, Barcelona.
Saarnisto, M., Huttunen, P., Tolonen, K., 1977. Annual lamination of
sediments in Lake Lovojärvi, southern Finland, during the past
600 years. Ann. Bot. Fenn. 14, 35–45.
Sanz, V., 1995. D'artesans a Proletaris : La Manufactura del Cànem a
Castelló, 1732–1843. Servei de Publicacions, Diputació de
Castelló, Castelló.
Sharma, H.S.S., van Sumere, C.F., 1992. The Biology and Processing
of Flax. M. Publications, Belfast, p. 576.
van Dam, H., Mertens, A., Sinkeldam, J., 1994. A coded checklist and
ecological indicator values of freshwater diatoms from the
Netherlands. Neth. J. Aquat. Ecol. 28, 117–133.
van Geel, B., 1978. A palaeoecological study of Holocene peat bog
sections in Germany and the Netherlands. Rev. Palaeobot. Palynol.
25, 1–120.
van Geel, B., 2001. Non-pollen palynomorphs. In: Smol, J.P., Birks, H.
J.B., Last, W.M. (Eds.), Tracking Environmental Change Using
Lake Sediments: Terrestrial, Algal and Silicaceous Indicators, vol.
3. Kluwer Academic Publishers, Dordrecht, pp. 99–119.
van Geel, B., Bohncke, S.J.P., Dee, H., 1981. A palaeoecological study
of an Upper Late Glacial and Holocene sequence from ‘De
Borchert’, The Netherlands. Rev. Palaeobot. Palynol. 31,
367–448.
van Geel, B., Hallewas, D.P., Pals, J.P., 1983. A Late Holocene deposit
under the Westfriese Zeedijk near Enkhuizen (Prov. of Noord-
Holland, The Netherlands): palaeoecological and archaeological
aspects. Rev. Palaeobot. Palynol. 38, 269–335.
van Geel, B., Bos, J.M., Pals, J.P., 1986. Archaeological and
palaeoecological aspects of a medieval house terp in a reclaimed
raiseg bog area in North Holland. Ber. Rijksd. Oudheidkd.
Bodemonderz. 33, 419–444.
van Geel, B., Coope, G.R., van der Hammen, T., 1989. Palaeoecology
and stratigraphy of the Lateglacial type section at Usselo (The
Netherlands). Rev. Palaeobot. Palynol. 60, 25–129.
van Geel, B., Buurman, J., Brinkkemper, O., Schelvis, J., Aptroot, A.,
van Reenen, G., Hakbijl, T., 2003. Environmental reconstruction
of a Roman period settlement site in Uitgeest (The Netherlands),
with special reference to coprophilous fungi. J. Archaeol. Sci. 30
(7), 873–883.
van Zeist, W., Wasylikowa, K., Behre, K.E., 1991. Progress in Old
World Palaeoethnobotany. Balkema, Rotterdam.
Violant-Simorra, R., 1934. Elaboració del Cànem i de la llana al
Pallars. Indústries Casolanes, Barcelona.
Whittington, G., Edwards, K.J., 1989. Problems in the interpretation of
Cannabaceae pollen in the stratigraphic record. Pollen Spores 31
(1,2), 79–96.
137S. Riera et al. / Review of Palaeobotany and Palynology 141 (2006) 127–137