Full Terms & Conditions of access and use can be found at
The Journal of Human Palaeoecology
ISSN: 1461-4103 (Print) 1749-6314 (Online) Journal homepage: https://www.tandfonline.com/loi/yenv20
Archaeology and Climate Change: Evidence of a
Flash-flood During the LIA in Asturias (NW Spain)
and its Social Consequences
Jesús Fernández, Gabriel Moshenska & Eneko Iriarte
To cite this article: Jesús Fernández, Gabriel Moshenska & Eneko Iriarte (2019) Archaeology and
Climate Change: Evidence of a Flash-flood During the LIA in Asturias (NW Spain) and its Social
Consequences, Environmental Archaeology, 24:1, 38-48, DOI: 10.1080/14614103.2017.1407469
To link to this article: https://doi.org/10.1080/14614103.2017.1407469
Published online: 28 Nov 2017.
Submit your article to this journal
Article views: 103
View Crossmark data
Archaeology and Climate Change: Evidence of a Flash-flood During the LIA in
Asturias (NW Spain) and its Social Consequences
, Gabriel Moshenska
and Eneko Iriarte
UCL Institute of Archaeology, London, UK;
Laboratorio de Evolución Humana, Departamento de Historia, Geografía y Comunicación,
Universidad de Burgos, Burgos, Spain
This paper presents the results of a multidisciplinary study of the impact of climate change
during the Little Ice Age on a medieval village in Asturias, Spain. The research focused on
tracing evidence for a catastrophic flood that buried the village beneath a thick layer of
debris, including examining the remains of structures and agricultural land sealed beneath
the debris, and considering the social and economic implications of the event in the
subsequent history of the area. First, a series of test pits was excavated within the area of the
modern village to map the full extent of the damage. Following this, analysis of the
stratigraphy, architectural remains, datable artefacts and radiocarbon dating contributed
further details, while historical evidence revealed the privatisation of the agricultural land
following the catastrophe. The findings offer a snapshot of climate change and its social
contexts in a specific, under-studied area with possible implications for the study of risk
behaviour and disaster response in currently inhabited areas.
Received 17 May 2017
Accepted 14 November 2017
Climate change archeology;
Little Ice Age; flash-flood;
In this paper we present a case study of a flash-flooding
event that destroyed a medieval village in the North
West of the Iberian Peninsula, forming a large and
enduring torrential cone. Through this work we were
able to date the flooding event to close to the beginning
of the Little Ice Age, most likely between the end of the
thirteenth and the mid-fourteenth centuries. Through
further detailed analyses of the sites, stratigraphy and
relevant historical data we have begun to outline the
social consequences of the flooding and the destruction
it caused in the short, medium and long term. To this
end our archaeological data are supported by a geo-
morphological study of the basin and an analysis of
morphometric parameters to enable us to examine
the likely causes and the nature of the flash-flood and
Van de Noort’s essay on archaeological
approaches to climate change stated that ‘By offering
long-term perspectives on human interrelationships
with climate change, archaeology is well placed to
enhance an understanding of the socio-ecological
resilience of communities and their adaptive
capacity’(2011, 1046). The idea of an instrumental
archaeology at the service of studies of resilience
and endurance in the face of environmental disaster
is an inspiring one, but we would argue also for a
campaigning archaeology of climate change that
highlights the human impacts at the points where
resilience and adaptation fail.
This paper aims to contribute in a modest way to the
growing body of research into global warming that
seeks to derive humanistic and socio-politically
engaged conclusions, and drive political action. To
this end we examine events in medieval Asturias in
terms analogous to studies of climate change and its
human impacts in the contemporary world, drawing
on ideas including Naomi Klein’s notion of the
‘shock doctrine’and the operating methods of preda-
tory ‘disaster capitalism’(Klein 2007).
The Little Ice Age: Climate and
As will be made clear below, dating evidence places the
flash-flooding event described in this paper close to the
beginning of the Little Ice Age. The Little Ice Age (here-
after LIA) refers to a period between (roughly) 1300
and 1850 AD (Fagan 2000), when temperatures in
the northern hemisphere were markedly colder than
the preceding Medieval Warm period by an average
of approximately 0.3°C, and 0.8°C lower than at the
end of the twentieth century (Mann 2002). Alongside
temperature changes the LIA was characterised by,
inter alia, the growth of mountain glaciers and hydro-
logical impacts including increased rainfall (see Morel-
lón et al. 2011, for a discussion of these changes with a
specific focus on northern Spain). While it was initially
characterised as a period of consistent low temperature
the LIA is now generally understood in terms of
© Association for Environmental Archaeology 2017
CONTACT Jesús Fernández firstname.lastname@example.org
2019, VOL. 24, NO. 1, 38–48
variability and instability, with considerable regional
variation (Bradley and Jonest 1993; Pfister 1992). In
the Atlantic region in particular, changes to the Gulf
Stream at the start of the LIA contributed to irregular
patterns of rainfall both seasonally and annually,
while in the Mediterranean region and the Alps the
same period was marked by an increase in rainfall
(Benito et al. 2008).
Archaeological studies of the LIA and past climate
change in general have tended to operate at regional
and larger scales: this is often for sound reasons,
including the need to gather large datasets over a
wide and varying area to carry out meaningful ana-
lyses. This has had the effect of creating a pattern
of generalised results at regional levels and above,
and has also left a relative dearth of fine-grained
studies on a small, settlement-level scale. Finer-
grained local area studies such as this can contribute
to an appreciation of regional and temporal vari-
ations in the LIA. Studies such as this that link the
climatic evidence to socio-economic processes remain
rare for the earlier stages of the LIA, when there are
far fewer historical sources than for the later periods.
It is here that archaeological evidence can be particu-
Site and Methods
The focus of this project is Villanueva, a concentrated
settlement of approximately fifty inhabitants situated
in a valley of the Cordillera Cantábrica, the mountainous
area in the centre of Asturias, north-western Spain
(Figure 1). The village is located on the fluvial terraces
of the river Trubia, a tributary of the Nalón river, and
Figure 1. Location of the study area. The bottom picture was taken from North.
ENVIRONMENTAL ARCHAEOLOGY 39
was known as S. Romano in the Middle Ages: currently
S. Romano is one of its 8 neighbourhoods, seated on
both sides of a torrential stream of the same name.
Villanueva is located on the narrow alluvial plain in
the valley bottom, 150 m above sea level. The proximity
of the river provides a range of cultivable soils. Agricul-
tural and cattle areas are distributed around the village
following the classic concentric distribution of Euro-
pean villages of medieval origin, with the orchards
dedicated to intensive agriculture closer to the village,
followed by the ‘veigas’(cereal crops areas of collective
regulation), forests and meadows in the slopes, occupy-
ing an intermediate position, and finally the uplands
dedicated to extensive livestock farming. The total
extension of the parish (San Romano) is about 6 km
The orography is very rugged which causes the
majority of this small territory present important and
large slopes. The climate is oceanic, influenced by the
sea, with cool summers, mild winters and abundant
precipitations all year rounds. During the Middle
Ages, the village was under control of the Tuñón mon-
astery. This was a major power centre at the time,
organised and built around an important pre-Roman-
esque church constructed during the 9th century (Fer-
nández Fernández 2017a).
Starting in 2009 a series of test pits were excavated as
part of a project to trace the origins and development of
the medieval village, obtaining a quantity of archaeolo-
gical data including stratigraphic and soil analyses,
radiocarbon dates, and material culture dating from pre-
history to the present. The initial findings of these test
pits forms part of an article published previously (see
Fernández Mier et al. 2014). In this paper brief mention
is made to evidence for a historic flood found in some of
the test pits, and its potentially catastrophic impacts on
the village. These flooding layers are the focus of the
current analysis, focusing on two major stratigraphic
sequence and the resulting geoarchaeological and
The topography of the torrential cone, formed during
the flood and subsequently altered by construction and
agriculture, was reconstructed using Lidar data and
GIS software. Following this, an analysis was carried
out of the morphometric parameters of the S. Romano
stream basin, together with estimates of the water vel-
ocity based on measurements of the larger stones trans-
ported by the flood employing Costa’s(1983)equations.
Radiocarbon dates were obtained for three samples
taken from the excavations: there were calibrated to an
accuracy of 2σwith 95.4% probability (Blaauw 2010)
using OxCal v4.2.2 (Bronk Ramsey 2009) with atmos-
pheric data intcal09.14C (Reimer et al. 2009).
A series of ten test pits were excavated in and around
the location of the medieval village of S. Romano
(Figure 2). Traces of flooding were found in seven of
the ten, made up of gravel and sand layers and the
remains of flood channels. The flood deposits are allu-
vial, deposited by flood related running water, and
forming an alluvial fan. These allow us to trace the
shape and extent of the torrential cone, and two test
pits in particular (TP CDR and TP MUR) offered par-
ticularly detailed insights into the nature of the flood-
ing event (Figure 2). Nowadays there is a torrent (S.
Romano stream) that incides the medieval alluvial
fan. These findings and the stratigraphic and sedimen-
tary analyses of these two test pits are described below.
This test pit was placed on the western side of the
research area, towards the centre of the torrential
Figure 2. A) Image of the village excavated and its surrounding areas with the test pits, intervention codes of two main stratigra-
phyes analyzed and reconstruction of the torrential fan. B) Enlarged image of the excavated areas. C) DEM from Lidar, topography of
the torrential fan and profile graph.
40 J. FERNÁNDEZ ET AL.
cone. It contained a stratigraphic sequence detailed in
Figure 3, beginning with a series of Roman levels (nota-
bly from the High-Imperial period) (Fernández Fer-
Above the Roman material were a number of strati-
graphic levels and structural features dating from the
ninth to eleventh centuries, which matches previous
theories on the date of origin of the medieval village.
These included a layer of blackened soils, context
010, amortising the first negative structures recorded
(S.U. 011 and 012) and related to a period in which
seem to have happened different types of agricultural
and domestic use. In this layer fauna, black pottery
and iron -mainly nails- are intermingled and
embedded with abundant carbonised vegetable matter.
Micromorphological and geochemical data from
010 confirm the presence of abundant small fragments
of domestic waste as bone, charcoal and dung dispersed
in the sediments that could indicate tilling and manur-
ing of an orchard area nearby domestic spaces or struc-
tures (Figure 4).
A detailed micromorphological and geochemical
study of the medieval archaeological units from some
test pits is ongoing. The results and discussion of
these data are out of the scope of this paper since
they are more related to the characterisation of differ-
ent medieval agricultural and paleoenvironmental pro-
cesses identified in the samples (MUR test pit,
described below, was not sampled for the micromor-
phological study since it is very similar to CDR).
Context 009 contained an abundance of charcoal
and organic matter but no archaeological material,
and it was speculated that this may have been laid
down as preparation for the structure evidenced in
contexts 008, a paved level, and 007, a floor containing
a hearth, both dated to the later medieval period
(Figure 5). This floor level varies in thickness between
5 and 10 centimetres, and contains a considerable
amount of charcoal. The hearth area is notable for
the compaction of clay burned to an orange colour.
This layer contained ceramics of a fine-grained fabric,
varying degrees of firing and combed horizontal
incisions. Faunal remains found in association with
the hearth appear to be food-related. A fragment of
charcoal from the fire was radiocarbon dated, yielding
a date between the mid-thirteenth and mid-fourteenth
centuries (Cal AD 2σ1271–1387).
The contexts above are formed by the flash-flood.
Context 006 is a one-off cutting of a channel the under-
lying contexts 007-009, and filled by contexts 004 and
005 (Figures 3 and 5). The palaeochannel is filled with
ordered gravels of various sizes in a layer more than
30 cm thick: analysis of the lithology and graded of
these gravels confirms that their origin is the nearby
stream of S. Romano rather than the River Trubia,
forty metres from the test pit. Context 004 is composed
of sub-rounded gravel in a sandy-silty matrix: context
005 is similar but less ordered. Contexts 004-006
were interpreted as representing two phases of the
flash-flood: the first erosive phase destroyed elements
of the structure and formed the palaeochannel through
the site; the second sedimentary phase saw the depo-
sition of sand and gravels. Contexts 002 and 003 over-
lying these flood layers contain early modern ceramic
remains dated to approximately the sixteenth century,
indicating the resumption of human activity in the area
following the destructive flood. Context 001 is the
This trench is located in the southern part of the
research area, closer to the River Trubia (Figures 2
Figure 3. Stratigraphy of the excavations: IT-CDR. Radiocarbon samples and dates: S1 Cal. 2 σ900–1146 AD; S2 Cal. 2 σ1271–1387
AD; S3 Cal. 2 σ1469–1635 AD.
ENVIRONMENTAL ARCHAEOLOGY 41
The excavation is on-going at the time of writing and
has not yet reached the river terrace levels, but the most
recent phase of work revealed evidence of an occupied
structure interpreted as a hut floor, with finds including
a grindstone. Postholes associated with the structure
have fills, one of which has been radiocarbon dated to
Figure 4. Micromorphological features of medieval agricultural activities. 1) Bone fragment. 2) Charcoal fragment. 3) Phosphatic
dung nodule with calcium oxalate phytoliths. 4) Goethite hypocoatings in pores indicating iron lixiviation. The random distribution
of all these features in the same layer point to anthropic manuring and tilling activity affecting surficial soils in humid environments.
Figure 5. 1) Late medieval hut floor cut by flash flood channel (S.U. 008 and 009) IT-CDR. 2) NE corner, detail of the channel cut. 3) E
stratigraphic profile, detail of the channel and torrential deposit.
42 J. FERNÁNDEZ ET AL.
between the thirteenth–fourteenth centuries. Above
these structural remains there is an agricultural layer
with abundant ceramic and organic materials including
charcoal and animal bone. Like context 007 and 008 in
the previously described pit, this agricultural layer is
marked by a number of palaeochannels, and the material
culture in both marked contextsissimilar,withpottery
of equivalent dates. For this reason, it is reasonable to
interpret a contemporary late medieval date for both.
Above this layer and filling the palaeochannels there
are contexts made up of poorly ordered material includ-
ing stone, ceramic tiles, sand, pottery, bones, gravel and
sand. This thick layer appears to have been deposited
by a high-energy flow rather than by decantation, and
was itself cut by a second set of channels filled in turn
by sand, gravel and pebbles (Figure 7).
This layer, context 002, is archaeologically sterile,
and key to understand the nature and origin of the pro-
cess. It is composed with laminated sands interbedded
with finer matrices and clayey silts, with pockets of
gravels and pebbles ordered by size. Their main lithol-
ogy includes sandstone, slate and limestone of Palaeo-
zoic origin revealing a different nature than the river
Trubia barrages formed mainly by quartzites. This
lithological difference means that the origin of these
deposits is the stream of S. Romano and not the river
Trubia. Considering the structure, morphology and
lithology of the pebbels (small size, subrounded) it is
concluded that this is a torrential deposit.
These two layers (contexts 002 and 003) are inter-
preted as phases of the same flash-flood event, a first
phase (context 003) resultingfrom the massive transport
of sediment and structures destroyed by a high energy
stream in its first phase (debris flow), and a second
phase (context 002) composed of deposits of gravel,
areas and silt, accumulated in a phase of lower energy.
Context 002 and 003 are sealed by a layer of agrarian
soil rich in pottery dating (as with the previous test pit)
from the sixteenth century through to the present.
Context 003 in this trench contained a number of
very large stones, interpreted as building material
from the village as they were interspersed with broken
roof tiles (Figure 8).
Measurements of these stones were used to estimate
the speed of the water flow using the formulae devised
by Costa (1983) and the five largest stones (see Table 1).
Accordingly with Kehew, Milewski, and Soliman
(2010), there are numerous potential sources of error
Figure 7. Stratigraphy of the excavations IT-MUR, detail of the flash flood channels.
Figure 6. Stratigraphy of the excavations: IT-MUR.
ENVIRONMENTAL ARCHAEOLOGY 43
in this type of palaeohydrological analysis which could
range from 28% of average error in small drainage
basins up to 76% in large drainage basin. In conclusion
the estimates presented here would be closer to the
lower margin of error for small basins, but it is assumed
that Costa’s formula is not accurate and is based on
estimates. Nevertheless, for our research this infor-
mation is very useful combined with the rest of the
archaeological and stratigraphic data. The result was
an average speed of around 3.5 m/s. Even taking into
account this probable average error the sheer size of
the stones indicates a flow of considerable force, con-
firmed in this case by the presence of structures
destroyed, displaced and turned over.
Overall we can see four phases in these two trenches
that are indicative of a flash-flood. The first are the ero-
sion channels in the underlying contexts, the second is
the deposit containing destruction materials from the
medieval village structures, the third is a second phase
of erosion channels in this deposited layer, and the
final is a lower-energy deposit made up of sediments
from higher in the flow area, and containing no archae-
ological materials from the settlement (Table 2).
The most evidence of the flash-flood was the cre-
ation of a torrential cone. In the test pits studied
there was no evidence of alluvial cone sediments before
the event excavated, only terrace sediments, gravels and
silty clay, from the floodplain of the river Trubia. For
all these reasons it is interpreted that it was a single
event with a limited duration and chronologically
located between middle and modern ages. This cone
has been partly obscured by changes in land use over
the intervening period: the present-day neighbourhood
of S. Romano covers much of its area, and different
phases of construction and landscaping have further
affected the topography. Lidar data was used to recon-
struct the topography of the original cone, removing
the current layers of construction (Figure 2). The
resulting digital elevation model allows the proximal
and intermediate zones of the cone to be easily ident-
ified, while the distal extents have been identified in
part through the presence or absence of flood deposits
in the test pits as indicated on the map (Figure 2). In
this way the size of the cone was found to be 2.2 Ha
or approximately 5 acres). Most signifi-
cantly, the area of the cone covers more than half of
what would previously have been the most valuable
arable land in the village: the fertile, flat area in the val-
ley bottom made up of the lowest river terraces. In
mountainous areas such as this, the loss of this arable
land and the crops it contained would have had a
serious impact on the community, depending on a
number of factors including the time of year, the survi-
val of other parts of the arable land, the size of the com-
munity, its resilience and resources. The first mention
of the village name ‘Villanueva’, the new settlement
built occurs in historical documents near the end of
the fourteenth century, while the radiocarbon dates
of the flood indicate a date between the thirteenth
and fourteenth centuries. We do not have historical
documents that refer this episode, nevertheless some
information about important floods from other areas
of Asturias in the fourteen and fifteenth centuries
founts is available. Taking this information into
account, we considered the hypothesis that the event
recorded in Villanueva could be the ongoing of an
important instability climatic stage prior to that
reported by the fifteenth century documentation.
Figure 8. Context 003, destruction level, IT-MUR. Zenithal pic-
ture and drawing. Tiles are coloured in red.
Table 1. Estimation of water velocity based on the five largest
stones from context 003.
Axis of boulders in mm
V(m/s) = 0.18 DI
Where vis mean velocity and DI is b-axis
length (Costa 1983)
Velocity of water in metres
Mean is 3.5 m/s
44 J. FERNÁNDEZ ET AL.
The S. Romano Stream Basin
To understand the nature of the flash-flood it is impor-
tant to describe the geology and topography of the
basin of the S. Romano stream. The geological sub-
strate on which the stream basin sits consists of Paleo-
zoic materials ranging in age from the Ordovician to
the Carboniferous, with considerable lithological var-
iety including siliciclastic formations, and carbonated
and detrital calcareous alternations. On this substrate
different quaternary formations associated with gravity
processes can be found as scree and colluvium. The
streambed itself is extremely steep in places, with a
watershed composed of limestone resulting in escarp-
ments in some cases more than 70° and dropping
from a peak altitude of around 850 m to just 166 m
in the valley of the River Trubia, with an average gra-
dient of 27.3%. Data obtained from the analysis of
the morphometric parameters of the S. Romano
basin obtained by GIS analysis are shown in Table 3.
The stream is formed within a micro-watershed
occupying an area of 1.6 km
in the form of an
elongated oval. The highest point of the watershed
lies to the west, the Canto la Cruz, at an elevation of
850 m. Other peaks surrounding the basin include La
Rasa (799 m) El Picu Castru Mayor (665 m) and El Ser-
rón (677 m) (see Figure 9). The lowest point of the
basin is the confluence with the River Trubia men-
tioned above. The length of the main channel is
1.8 km, and there is one small sub-basin without a per-
manent watercourse. Today the main channel is an
intermittent stream that is strongest in winter and
during rainy seasons, and practically disappears during
The figures in the table above, and in particular the
steep gradient of the stream, indicate a torrential basin
with considerable potential for sediment transport. In
addition, the drainage density of 1.13 indicates a very
low hydrogeomorphological capacity in response to
extremely high precipitation contributing to the poten-
tial for violent flash-floods. Despite this, archaeological
data from the test pits indicates a long period of relative
stability, with no traces of flooding between the early
Roman period and the flash-flood in the Middle
Ages. Therefore, there is no reason to think that the
zone around the stream would have been perceived
to be a vulnerable settlement area during the
establishment of the medieval habitation area around
the ninth century. The uses of the stream basin during
this period appear to have been varied but focused
around forestry and livestock management including
communal grazing areas and private meadows in a
landscape of bocage. The climatic instability around
the onset of the LIA may have been exacerbated by
agricultural pressure and particularly the resulting
deforestation, which would have increased surface run-
off and the rapid evacuation of rainwater. This could be
considered a warning and cause for concern for the
contemporary population of the area.
Impact of the Flood on Buildings and the Village
Flash floods are characterised by their sudden onsets,
violent force of water, and substantial residual sediments
left in their aftermaths. They can occur in a variety of cli-
matic environments but mountainous areas are particu-
larly vulnerable. While flash floods remain rare events
they are the most lethal natural disaster in the Iberian
Peninsula, with the death of 87 people on a campsite
in the Barranco de Aras (Central Pyrenees) in 1996 a
recent example (Alcoverro, Corominas, and Gómez
1999; García-Ruiz et al. 2004). Unlike other violent cli-
matic events the unexpected nature of flash floods and
the velocity of the water makes it difficult to warn or
evacuate communities in their path. According to the
United States Army Corps of Engineers the water vel-
ocity in flash floods is also the main contributing factor
in the destruction of buildings in the path of the flood
water: in controlled experiments a 1 m depth of water
moving at 3 m/s was found to be sufficient to destroy
Table 3. Characteristics of S. Romano basin based on GIS
Basin surface 1,60 Km
Basin perimeter 6,51 Km
Mean elevation 574 m (asl)
Mean slope (%) 26,60 %
Gravelius compactness coefficient. Elongated oval basin 1,44
Length of major axis 1,80 Km
Total basin length 2,5 Km
Initial altitude 850 m (asl)
Final altitude 166 m (asl)
Elevation difference 684 m
Mean stream slope % 27,36 %
Drainage density 1,13
Table 2. Identified phases during the medieval flash-flood event registered.
Phase TP CDR TP MUR
Later context SU 003, 004. modern soils (TPQ C16) SU 001. modern soils (TPQ C16)
Phase IV Depositional. SU 002. Flood deposit (graded bedding)
Phase III Depositional. SU 005-4. Flood deposits (graded bedding) Erosive. Formation of torrential channels
Phase II Erosive phase Depositional. SU 003. Massive deposits from the medieval
village destruction. Water velocity 3.5 m/s.
Phase I Erosive. Medieval buildings destruction and displacement by flash
flood. Formation of torrential channels (SU 006).
Erosive. Medieval fields destruction by flash flood and
formation of torrential channels
Previous context SU 007-8. Medieval hut and fireplace. Chronology after s. XIII-XIV SU 004. Medieval cornfield. Chronology after s. XIII-XIV
ENVIRONMENTAL ARCHAEOLOGY 45
the walls of a typical structure (McBean et al. 1988). In
the early stages of a flash flood there is often a quantity
of mud and detritus carried along which increases the
density of the flow and its ability to transport heavy
objects, such as the stones found in the test pits (Fried-
man and Sanders 1978).
The archaeological evidence gathered in this article
indicates that the medieval village of S. Romano was
destroyed during a rapid flash flood, with water vel-
ocity around 3.5 m/s exceeding the 3 m/s figure stated
above as capable of destroying buildings (Figure 8).
The evidence of structural destruction found in the
excavations suggests an event likely to have caused
human casualties, particularly given the form of med-
ieval buildings and the greater difficulty in emergency
evacuation, as well as the limited resources for rescue
efforts in the aftermath. The flood is likely to have
destroyed the village in a period of minutes or at
most hours, covering more than half of the arable
land surrounding it in a thick layer of sterile rocky sedi-
ment. The effect on the local economy cannot be calcu-
lated, but archaeological evidence suggests that it was at
least a century before agricultural activities recom-
menced in the area affected. At around the same time
the place-name ‘Villanueva’appears in records for
the first time, and S. Romano survived in name at
least as a neighbourhood of this village. Villanueva
was principally situated on the other side of the River
Trubia from the flooded area, and farmed a different
area of arable land to the north of the old settlement.
People and Place
At the time of the flood it is likely that, in common
with similar areas across Spain and Europe, the ara-
ble land of the village of S. Romano was managed in
common by the community and managed collectively
using an ‘open field’system, although ownership of
the land lay with the Bishop of Oviedo. If similar
to the common land in the area at present, the
land as a whole would probably have been enclosed
within a boundary but within this boundary the div-
ision of land between families would have been
agreed by custom. However, the management of the
larger landscape was probably decided on a collective
level, to manage processes such as crop rotation
across the entire community. Some parts of the
land in cases of this type were not associated with
specific tenant families and were instead allocated
to different users for set periods, again by agreement.
Common lands managed as open fields were a com-
mon feature of agricultural communities in medieval
Europe: what is remarkable in the area of Asturias
around the study area is the survival of some of
these practices into the present, albeit under signifi-
cant pressure both internally and externally. One of
the aims of the fieldwork described in this paper is
to trace the origins of these patterns of land manage-
ment: while the findings remain inconclusive there is
evidence to suggest that it emerged in something
resembling its current form between the thirteenth
and fourteenth centuries.
Changes in the patterns of use and management of
common lands can be seen historically and archaeologi-
cally in the traces of past agricultural landscapes including
redundant place-names, old field boundaries, and
palaeoenvironmental evidence of different land uses
over time. Around S. Romano today the process con-
tinues, as falling population levels lead to common land
falling out of use, the breaking of long-standing associ-
ations between families and specific plots of land, and
the opportunistic enclosure of areas of common land
for construction or development by private individuals.
The breakdown of traditional practices of common land
management and use contribute in turn to a decline in
community sense of place that further weakens the con-
nections between people and the land.
attested in Spanish history (Marcos 1999)duringthefif-
teenth-sixteenth centuries when there was a pattern of
Figure 9. S. Romano stream basin. Base map is used is a shaded relief raster derived from a DEM.
46 J. FERNÁNDEZ ET AL.
seizure of agricultural land that was deemed ‘vacant’or
unused. Therefore, natural disasters, such as the flooding
registered in Villanueva there was certainly an opportu-
In an area such as S. Romano with long-standing
customary rules regulating the agricultural spaces it is
only through the breakdown of these traditions that
the more privileged families could privatise the land.
After the natural event various acres of vacant lands
were ‘ready’to be reclaimed by these noble groups.
In the modern era the S. Romano district became
the area of the neighbourhood where the village elites
lived: this is recorded in early documents and exempli-
fied by the Muñiz-Prada family mansion which sits in
the centre of S. Romano. The Muñiz-Prada family line
died out in the 1960s but was rooted in the old feudal
families of the area: their mansion and private land
cover an area almost identical to the space covered
by the flood and the debris cone (see Figures 2 and 10).
Why, then, this concentration of social elites in an
area destroyed shortly before? It is tempting to com-
pare the strategy of land privatisation in the aftermath
of the flood to the model of ‘disaster capitalism’out-
lined by Naomi Klein in her book Shock Doctrine
(2007). Klein argues that elites exploit the aftermaths
of crises or traumatic events such as wars and natural
disasters to enact controversial laws and policies that
would normally meet strong popular resistance, but
which a traumatised, distracted or displaced popu-
lation is unable to effectively resist. Following this pat-
tern, the privatisation of land in S. Romano in the
aftermath of the flood could perhaps be described as
This paper presents evidence for a flash flood around
the fourteenth century that destroyed the village of
S. Romano and its surrounding lands, and buried its
remains beneath a layer of rocky sediment. Based on
excavations and analyses of the stratigraphy, the local
topography and hydrology we have mapped the extent
of the debris cone, the source of the floodwater and the
likely reasons for the deluge. Due to the unique nature
of the event based on excavation evidence we have pro-
posed a connection to the onset of the Little Ice Age
and the resulting rise in rainfall in the area. There
were not alluvial cone sediments before the LIA
flood, only terrace sediments, gravels and silty clay,
from the floodplain of the river Trubia. In addition,
we have sought to understand the human social and
cultural impacts of the flash flood beyond the immedi-
No potential conflict of interest was reported by the authors.
This work was supported by H2020 Marie Skłodowska-
Curie Cofund Actions [grant number ACA1 4-08].
Figure 10. Muñiz-Prada manor house and its private lands. The fan toe is approximately in the centre of the picture.
ENVIRONMENTAL ARCHAEOLOGY 47
Notes on contributors
Jesús Fernández is Honorary Ressearch Associate at UCL
Institute of Archaeology and director of La Ponte-Ecomuséu
Gabriel Moshenska is Senior Lecturer in Public Archaeology
at UCL Institute of Archaeology.
Eneko Iriarte is Geologist and Lecturer at the University of
Jesús Fernández http://orcid.org/0000-0002-2081-4277
Gabriel Moshenska http://orcid.org/0000-0003-4926-
Eneko Iriarte http://orcid.org/0000-0001-8365-5616
Alcoverro, J., J. Corominas, and M. Gómez. 1999.“The
Barranco de Arás Flood of 7 August 1996 (Biescas,
Central Pyrenees, Spain).”Engineering Geology 51: 237–
Benito, G., V. R. Thorndycraft, M. Rico, Y. Sanchez-Moya,
and A. Sopena. 2008.“Palaeoflood and Floodplain
Records from Spain: Evidence for Long-term Climate
Variability and Environmental Changes.”
Geomorphology 101: 68–77.
Blaauw, M. 2010.“Methods and Code for ‘Classical’Age-
modelling of Radiocarbon Sequences.”Quaternary
Geochronology 5 (5): 512–518.
Bradley, R. S., and P. D. Jonest. 1993.“‘Little Ice Age’
Summer Temperature Variations: Their Nature and
Relevance to Recent Global Warming Trends.”The
Holocene 3: 367–376.
Bronk Ramsey, C. 2009.“Bayesian Analysis of Radiocarbon
Dates.”Radiocarbon 51 (1): 337–360.
Costa, J. E. 1983.“Paleohydraulic Reconstruction of Flash-
flood Peaks from Boulder Deposits in the Colorado
Front Range.”Geological Society of America Bulletin 94
Fagan, B. 2000.The Little Ice Age. How Climate Made
History 1300-1850. New York: Basic Books.
Fernández Fernández, J. 2014a.Estudios Multiescalares Sobre
la Alta Edad Media en el Valle del Trubia. Servicio de pub-
licaciones de la Universidad de Oviedo.
Fernández Fernández, J. 2014b.“Aproximación arqueológica
a la génesis y evolución del poblamiento medieval en un
territorio de media montaña (bajo Valle del Trubia, con-
cejos de Proaza, Santo Adriano y Oviedo).”In
Excavaciones arqueológicas en Asturias, vol. 7, 331–342.
Oviedo: Consejería de Cultura del Principado de Asturias.
Fernández Fernández, J. 2017a.“Reyes, obispos y campesi-
nos: territorio y poblamiento durante la Alta Edad
Media en el Valle del Trubia, Asturias (siglos VIII-XII).”
Studia Historica. Historia Medieval 35 (1): 13–47.
Fernández Fernández, J. 2017b.“Arqueología de una aldea
medieval y su espacio agrario: Villanueva de Santo
Adriano (Asturias, noroeste de la península Ibérica).”
Historia Agraria 72: 69–106.
Fernández Mier, M., J. Fernández Fernández, P. Alonso
González, J. A. López Sáez, S. Pérez Díaz, and B.
Hernández Beloqui. 2014.“The Investigation of
Currently Inhabited Villages of Medieval Origin:
Agrarian Archaeology in Asturias (Spain).”Quaternary
International 346: 41–55.
Friedman, G. M., and J. E. Sanders. 1978.Principles of
Sedimentology. New York: John Wiley & Sons, 792 pp.
García-Ruiz, J. M., S. White, C. Martí-Bono, B. Valero, M. P.
Errea, and A. Gómez Villar. 2004.“La avenida del barranco
de Arás y los riesgos hidrológicos en el Pirineo central
español.”In Geografía física de Aragón. Aspectos generales
y temáticos, edited by J. L. Peña, L. A. Longares, and M.
Sánchez, 131–140. Zaragoza: Universidad de Zaragoza e
Institución Fernando el Católico.
Kehew, A. E., A. Milewski, and F. Soliman. 2010.
“Reconstructing an Extreme Flood from Boulder
Transport and Rainfall–Runoff Modelling: Wadi Isla,
South Sinai, Egypt.”Global and Planetary Change 70
Klein, N. 2007.The Shock Doctrine: The Rise of Disaster
Capitalism. Toronto: A.A. Knopf Canada.
Mann, Michael E. 2002.“Little Ice Age.”Encyclopedia of
Global Environmental Change 1: 504–509.
Marcos, A. 1999.“Estructuras de la propiedad en la época
moderna: Evolución y variantes peninsulares.”In
Historia de la propiedad en España, siglos XV-XX:
Encuentro interdisciplinar, Salamanca, 3-6 de junio de
1998, Coords. R. Robledo, E. Torijano, S. de Dios, and J.
Infante, 113–162. Salamanca: Centro de Estudios
McBean, E. A., J. Gorrie, M. Fortin, J. Ding, and R. Monlton.
1988.“Adjustment Factors for Flood Damage Curves.”
Journal of Water Resources Planning and Management
114 (6): 635–646.
Morellón, M., B. Valero-Garcés, P. González-Sampériz, T.
Vegas-Vilarrúbia, E. Rubio, M. Rieradevall, A. Delgado-
Huertas, et al. 2011.“Climate Changes and Human
Activities Recorded in the Sediments of Lake Estanya
(NE Spain) during the Medieval Warm Period and Little
Ice Age.”Journal of Paleolimnology 46: 423–452.
Pfister, C. 1992.“Monthly Temperature and Precipitation in
Central Europe 1525–1979: Quantifying Documentary
Evidence on Weather and Its Effects.”In Climate Since
A.D. 1500, edited by R. S. Bradley, and P. D. Jones, 118–
142. New York: Routledge.
Reimer, P. J., M. G. L. Baillie, E. Bard, A. Bayliss, J. W.
Beck, P. G. Blackwell, C. Bronk Ramsey, et al. 2009.
“IntCal09 and Marine09 Radiocarbon Age Calibration
Curves, 0–50,000 Years cal BP.”Radiocarbon 51 (4):
Van de Noort, R. 2011.“Conceptualising Climate Change
Archaeology.”Antiquity 85 (329): 1039–1048.
48 J. FERNÁNDEZ ET AL.