Content uploaded by Miguel Carrero-Pazos
Author content
All content in this area was uploaded by Miguel Carrero-Pazos on Jul 12, 2019
Content may be subject to copyright.
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
The Spatial Structure of Galician Megalithic Landscapes (NW Iberia): A Case
Study from the Monte Penide Region
Authors
Miguel Carrero-Pazos. University of Santiago de Compostela.
miguel.carrero.pazos@gmail.com
Andrew Bevan. UCL Institute of Archaeology. a.bevan@ucl.ac.uk
Mark Lake. UCL Institute of Archaeology. mark.lake@ucl.ac.uk
1. Introduction
Megalithic tombs are a common social and funerary feature of Atlantic European
landscapes during the Neolithic, involving the construction of large-scale stone
monuments, typically but not exclusively collective graves covered by a mound
(Laporte, Scarre 2011). Many explanations for the rise of megalithism and for the
meaning of particular sets of megaliths have been offered by archaeologists in the past
(Schultz Paulsson 2017), with an emphasis generally placed on their multiple rather
than singular roles in society, and their likely symbolic marking of communally-shared
resource landscapes belonging to segmentary groups (e.g. Saxe 1970; Fleming 1973;
Chapman 1981; Tilley 1994; Sherratt 1990, 1995). Such investigations therefore
highlight the importance of megalithic monuments not only as a funerary monuments
but also as visible, public landmarks for living communities (Renfrew 1976; Renfrew
1984; Delibes de Castro 1991). Their potential territorial significance has been
approached from different perspectives, including theoretical landscape archaeology
(e.g. Criado Boado, Villoch Vázquez 2000; Criado Boado 2017), historical comparanda
(e.g. Martinón-Torres 2001; Díaz Guardamino et al. 2015) and quantitative spatial
analysis (e.g. Gillings 2009; Murrieta-Flores 2012, 2014; Llobera 2015).
One of the highest concentrations of megalithic monuments along the European Atlantic
façade is Galicia (NW Iberian Peninsula), where there are more than ca.7,000
megalithic sites within an area of ca.30,000 sq.km (a density of 0.23/sq.km; Xunta de
Galicia 2013). Certain smaller geographical regions offer yet denser megalithic
landscapes, with A Costa da Morte boasting more than 500 sites in 1,740 sq.km (ca.
0.28/sq.km) or A Serra do Leboreiro’s plateau in Portugal with more than 120 sites in
1
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
75 sq.km (1.6/sq.km; see Rodríguez Casal 1997; Eguileta Franco 1999; Ferreira de
Sousa 2013; Vilas Estévez 2015; Carrero-Pazos 2018a).
The development of the Neolithic Period and the megalithic complex of northwest
Iberia probably reflects funerary ideas that originally came from the centre of Portugal
(Rodríguez Casal 1990; Prieto Martínez et al. 2012; Schultz Paulsson 2017). Although
as yet constituting only a small number of finds, and in contrast to the pattern suggested
elsewhere in Iberia, the existing archaeological evidence in the north-west suggests a
degree of continuity between the latest Epipalaeolithic hunter-gatherers and the first
food-producing communities, leading to the development a mixed economy during the
first half of the 5th millennium BC. The earliest radiocarbon date associated with
archaeological structures from this initial period is 4720-4530 BC (on an organic-rich
sediment; Ua-3267, 5780 ± 40 uncal bp), and comes from a storage pit at Monte dos
Remedios (Fábregas Valcarce et al. 2007; Prieto Martínez et al. 2012: 220). Additional
available radiocarbon dates suggest, in general dates, the appearance of simpler single-
chambered Galician megaliths over the period 4500-3500 BC, a subsequent
development of corridor dolmens (3500-2500 BC) and then a late phase of megalithism
(2500-1800 BC) characterised by small cists or burials without above-ground structures.
Despite a strong tradition of scientific study of these monuments by Galician and other
researchers (see e.g. Criado Boado 1989; Bradley 1991a, 1991b, 1998), spatial
approaches have until recently rarely extended beyond the creation of distribution maps
(Leisner 1938; Rodríguez Casal 1997), accompanied by, often sensible but informally-
expressed, comments about locational factors such as those outlined in Table 1.
[Caption] Table 1. Variables of the locational model of the megalithic culture in
Galicia as found in the literature (Carrero-Pazos 2017).
In the last few years, however, researchers have begun to adopt a more quantitative
approach to the same observations via GIS and spatial modelling (Wheatley et al. 2010),
although the emphasis has so far been on the relationship between sites and potential
travel networks across the landscape (see e.g. Llobera 2015; Rodríguez Rellán, Fábregas
Valcarce 2015; Carrero-Pazos 2018a; Carrero-Pazos, Rodríguez Casal in press). This
growing body of work has identified a series of potentially relevant environmental
2
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
variables (such as topographic prominence or natural transit routes) that exhibit
considerable statistical interdependence with one another (i.e. are highly correlated).
Given this interdependence, many of these variables could, interpretatively, be seen as
different ways of expressing the same general observation. In light of this, and after
much wider exploration of the possibilities (Carrero-Pazos 2017) we have opted to
consider just two of the simplest but most important variables – elevation and distance
to major watershed boundaries (both variables were discussed by Bradley 1991a: 78;
1991b in the case of south-west England burial mounds; by major watersheds here we
generally mean those that have maritime outlets). This simpler starting point then allows
us to conduct a more rigorous analysis and investigate both first and second-order
patterns and site sizes (for the definition of these terms, see below and Bevan et al.
2013; Baddeley et al. 2016). We conclude that these Galician monuments made quite
explicit use of the natural features of the landscape such as the ridgelines high up in
major Galician upland watersheds which were both important transit routes for travel
and major viewpoints. Moreover, we demonstrate that these monuments were organised
into spatial clusters whose internal size hierarchies exhibited non-random structure
suggesting some kind of organisational equivalence between clusters. These results can
be independently reproduced with the data and analytical workflow provided in the
accompanying repository (see Supplementary Data and Methods).
2. Study area
The study area is located in southern Galicia (NW Iberia, Figure 1) which was
identified (on the basis of previous research by one of the authors: Carrero-Pazos 2017,
2018b) as a zone of 620 sq.km with one of the most important concentrations of
megaliths in the whole Galician region. In this sub-region, 121 megalithic mounds are
known, with important concentrations at Monte Penide, Monte Vixiador and As
Pereiras. In what follows, we refer to this whole study region as ‘Monte Penide and
surroundings’, in light of the fact that Monte Penide provides the densest concentration
of sites.
[Caption] Figure 1. The study area in Galicia (NW Iberia), overlain on a shaded relief
model.
3
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
In geographical terms, this is a comparatively flat region rising from the coast to
moderate elevations (400-500 mASL). From a hydrological point of view, the
watersheds of Verdugo and Oitavén rivers stand out, as well as the Miñor watershed
which bounds the region to the south. Pico San Vicente (432 m) in Redondela, is the
highest point in the north, and has one of the largest concentrations of mounds, such as
that at Monte Penide. The main mountain range of the area is further to the East, the
elongated (N-S direction) Serra do Galiñeiro (Coto de Cales, 742 m).
The megalithic monuments of these areas are well known in the literature: descriptive
studies and early excavations were carried out by G. Álvarez Limeses (1935), C. de
Mergelina (1936), P. Díaz Álvarez (1973), J. Filgueira Valverde and A. García Alén
(1977), and J. M. Hidalgo Cuñarro and F. J. Costas Goberna (1979). Later, the
excavations carried out by J. C. Abad Gallego in Cotogrande necropoleis (1990-1991;
1992-1993; 1995; 1996-1997) and Monte Penide provided very interesting case studies
of the re-utilisation of these monuments through time. More recently, intensive survey
by C. Gómez Nistal and A. Rodríguez Casal (2000), and then R. Fábregas Valcarce
(2010, orig. 2001), updated the number of known barrows in the whole area.
Nevertheless, despite a long history of research, the Galician megalithic phenomenon
has only very recently been studied from a spatial perspective (Carrero-Pazos 2017,
Figure 2).
[Caption] Figure 2. A: Current photograph from a megalithic mound in Monte Penide.
B: Field plans of the excavations carried out by C. Mergelina (1936) in some of the
monuments.
3. Data and methods
The archaeological data used in the analysis reported here is part of an archaeological
database produced by the megalithic studies group at the University of Santiago de
Compostela (GEPN-AAT). This group has catalogued 121 sites via both the
accumulation of previous fieldwork (Carrero-Pazos 2017).
Site location analysis is one of the most prolific applications of GIS in archaeology over
the last 35 years (early examples include: Kvamme 1983, 1984; Judge and Sebastian,
1988). Although these approaches have been criticised as encouraging undue
4
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
environmental determinism (Wheatley 1993, 1996, 2004; Gaffney, van Leusen 1995),
they have nevertheless remained popular, especially when allied with modern spatial
statistics which, via their ability to discriminate between first and second order effects
(Bevan et al. 2013; Baddeley et al. 2016), potentially offer a more nuanced account of
the balance between environmental and social causes of past human behaviour (see e. g.
Williams 1993; Premo 2004; Bevan, Conolly 2006; Crema et al. 2010; Hinz 2011;
Bevan 2012; Bevan et al 2013; Verhagen et al. 2013; Nakoinz, Knitter 2016). Previous
work on Galician megaliths (Carrero-Pazos 2017, 2018b) developed a predictive model
of megalithic site probability and assessed its effectiveness via a control sample,
significance tests and a gain statistic (for the latter, Kvamme 1988). The conclusions
from this wider exploratory exercise are summarised below in Table 2, which shows
which covariates appeared to be better or worse predictors of the location of megalithic
tombs in the study area.
[Caption] Table 2. Previous results from multivariate logistic regression (Carrero
Pazos, 2017: 489).
Although previous work (Carrero Pazos 2017, 2018b) identified four covariates as
important variables, an elevation cut-off is perhaps the most obvious one mentioned in
previous discussion of megalithic monuments (Bradley 1991a: 78; 1991b). As noted
above, here we deliberately retain a certain simplicity by investigating the role that just
two major variables played in this megalithic area: elevation and distance to watershed
edges (Figure 3: A, B), not least because these two are also likely to be useful proxies
for other meaningful locational priorities such as ridge-routes and high visibility
locations. We first conduct a univariate analysis of how the intensity of megalithic sites
varies with respect to these two key variables and then build a multivariate logistic
model which can account for the first-order spatial inhomogeneity in the megalithic
sites, in other words, the overall spatial trend explained by properties of the
environment. We then use this model as a control which allows us to identify the
second-order properties of the site pattern, that is, those which may be more explicable
in terms of social processes, such as what appears to be a tendency for a pre-existing
mound to encourage the construction of nearby ones.
5
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
[Caption] Figure 3. A: watershed boundaries shown on top of an elevation model (the
latter a first order covariate). B: a second covariate constructed by calculating the
distance to watershed edges (excluding coastlines and with certain watersheds merged
so that only basins with sea outlets remain). C: megalith site intensity as a function of
terrain elevation. D: megalith site intensity as a function of watershed distance (solid
lines show function estimate while grey shading is pointwise 95% confidence
envelope). E: the predicted intensity of megalithic sites in the Monte Penide as
constructed via logistic regression.
In the final part of the paper, we use the second-order results and a mark correlation
function to suggest a meaningful clustering threshold with which to identify sub-groups
of mounds using a dbscan method (Ester et al. 1996; as implemented in GRASS GIS
v.cluster). Having identified these groups, we then use a novel rank permutation method
to conclude that the tombs are distributed across them in a way that is that is broadly
(even if not perfectly) hierarchical to an extent that is unlikely to occur by chance alone.
4. First-Order Location Model
The general distribution map of megalithic monuments documents a non-random
distribution of sites over the study area (see figure 1) and we can use statistical methods
to estimate the first-order trend behind much of this variation (Baddeley et al. 2016).
First, it is useful to consider a non-parametric summary of the univariate relationships
between the dependent variable (presence/absence of megalithic sites) and each of the
two covariates (elevation and watershed edges, Figure 3C-D). These suggest that sites
are more likely to occur at elevations ranging from 400-500 m ASL and at distances
closer to watersheds than expected by chance alone. Comparison with previous work
confirms that these two variables explain most of the overall variation even when
further possible covariates are included (see Carrero-Pazos 2018a). We therefore build a
first order logistic model using both covariates and produce a prediction surface of
megalithic site intensity across the landscape. This model is then used in a second stage
to study the clustering trend of the sites by fitting known statistical models to their
spatial distribution (Figure 3E).
5. Second-Order Clustering of Mounds
6
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
We have shown that elevation and watershed distance go some way towards explaining
spatial patterning in the distribution of sites, but we can also investigate the possibility
that there are additional - perhaps more overtly social - causes. We do this by using the
logistic model of first order spatial inhomogeneity to detrend for the unevenness of
megalithic sites due to landscape preferences, which then allows us to detect whether
there is additional second-order clustering or regularity in the site point pattern across
multiple scales (Baddeley et al. 2000: 330; Palmisano 2012; Baddeley et al. 2016). The
inferential logic of this enquiry involves two main steps: first, we use a pair correlation
function – which summarises the typical point intensity found in a series of non-
cumulative buffers around each point in the dataset – to identify second order clustering
and construct a 95% critical envelope for this function via random simulations
conditioned on the first-order regression model (Baddeley et al. 2016: 225-230). We
then consider whether the residual clustering can be explained with reference to one of
the well-defined second order interaction models in the literature.
A basic pair correlation function confirms that the points are spatially clustered at
distances up to 1km (the black line remaining above the grey envelope in figure 4A-B).
More precisely, the observed function (in black) falls above the Monte Carlo critical
envelope at these short distances, before dropping back down within the expected range
after this. A second version, in which the envelope is conditioned on the first order
model (figure 4B), demonstrates that although first order trends are important, they do
not account for the overall clustering (unless a major ‘lurking’ first-order variable has
been ignored in both this and previous work, which we view as unlikely), and that the
clustering is more likely to be endogenous and second-order in character (with the
presence of one site making a nearby site more likely; see also Baddeley et al. 2016:
487). It is also interesting to note that megaliths possibly exhibit a slight dispersed
pattern at distances of ca. 2000 meters (where the black line falls just beyond or at the
lower limit of the grey envelope in figure 4A-B).
[Caption] Figure 4. Approaches to site distance and confidence. A: pair correlation
function with a 95% envelope from wholly random Poisson process. B: pair correlation
function of the observed sites with a 95% envelope conditioned on the first-order
covariates model. C: pair correlation function with a 95% envelope also conditioned on
both the first-order covariates and a second-order, area-interaction model (r=1500 m).
7
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Having identified clustering, we then fit a known point interaction model to the
observed pattern. There are several Gibbs-type point processes that can be fitted:
examples include a so-called hard core process, in which points strictly avoid each other
up to a certain threshold, and a Strauss process in which points have constant influence
within a certain distance threshold (Nakoinz, Knitter 2016: 131). However, in our case,
the Widom-Rowlinson penetrable sphere model or area-interaction process (Baddeley,
Lieshout 1995; Baddeley et al. 2016: 519) offers a better fit and more interpretative
salience. This model generates inhibition and clustering patterns with reference to a
buffer created for all the points of the distribution, which may be interpreted as
modelling a scenario in which monuments have an area of influence within which they
either attract or repel other monuments. Figure 4C shows that, after adding such a
second-order interaction to our first-order model, the observed pair correlation function
(black line) now falls within the critical envelope of the simulation at all interaction
distances. This suggests that the jointly fitted first and second order factors are now able
to account for the observed point pattern. In other words, the distribution of megalithic
tombs can be accounted for by broad locational preferences for elevation and watershed
boundaries, as described by the regression model in tandem with local tendencies
towards mound clustering.
6. Mound Size and Shape
Given the observed second-order clustering of megalithic sites, it is interesting to
further explore the nature of this clustering and what it might imply in terms of social
organisation. If, for example, we use a mark correlation function to consider the spatial
correlation of mound volumes (referred to hereafter as ‘sizes’), there is evidence for
significant auto-correlation of these sizes, not for mounds in very close proximity but
rather for mounds spaced about 4.5km apart (Figure 5: A). Put very crudely, it would
appear that whatever process dictates tomb size repeats at approximately 4.5km
intervals, which further implies that it might be possible to detect meaningful groups by
clustering tombs using a distance to neighbour threshold of approximately half that size
(2000 m). Figure 5B shows the result of using a dbscan method (Ester et al. 1996) to
generate groups of sites using this threshold.
8
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
[Caption] Figure 5. A: results of the mark correlation function for mound
sizes/volumes. B: results of dbscan grouping analysis, showing possible groups of sites
with a territorial meaning.
The resulting 9 clustered groups of mounds make visual sense, with groups
concentrated by proximity. Thinking about possible social interpretations of these
groups an obvious further question, given the mark correlation function results, is
whether the mound sizes found within each group exhibit a non-random hierarchical
distribution, in which each group contains at least one of the larger tombs, followed by a
number of medium size tombs and finally a number of small tombs. For the sake of
clarity, in this paper ‘hierarchy’ is understood as ranking, in the sense that our argument
is about the distribution of tomb sizes characterised by their rank size. To our
knowledge there is no existing technique available to assess the hierarchical nature
distribution of tomb sizes given the uneven number of tombs with known sizes per
group, so we introduce here a novel test based on permuting the rank order sizes of the
tombs (for a complementary approach see Hennig and Lucianu 2000). The approach
(see also the Supplementary Data and Methods) operates as follows: first, rank all tombs
in descending order of size, so that the largest is ranked 1 and the smallest is ranked n,
where n is the number of tombs. Next, create a hierarchy of tomb-size ranks for each
group, so that the biggest tomb in each group is placed in hierarchical level 1, the
second biggest tomb in each group is placed in level 2, and so on. The result is a set of
tomb-size ranks for each hierarchical level, which then allows us to compare the mean
and/or sum of ranks at each hierarchical level with what we would get if we simply
allocated tombs to size-levels with no reference to their group membership. For
example, if the distribution of tombs across groups was perfectly hierarchical in the
sense suggested above and there were, say, four groups of tombs, then the tomb-size
ranks in level 1 should be 1, 2, 3 and 4, the ranks in level 2 should be 4, 5, 6 and 7, and
so on, which is exactly what we would expect if we ignored group membership. Since,
in reality, not all groups contain equal numbers of tombs, we adjust the expectation
accordingly, i.e. the hierarchical levels will not have equal numbers of members. If we
compare the observed ranks per level (i.e. those expected if a perfect size hierarchy
pertained), we see (Table 3: IdealMeans) that the ideal mean ranks increase from one
level to the next whereas that is not always true of the observed mean ranks. We also
see that the level 1 and 2 ideal mean ranks are substantially numerically smaller (i.e.
9
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
higher ranking in descending order of size) than their observed equivalents. The
conclusion that can be drawn from this is that the observed allocation of tomb sizes
across groups is not perfectly hierarchical, but it may nevertheless be closer to such an
idealised hierarchical distribution than we might expect by chance. We can examine
this possibility by conducting a Monte Carlo significance test (Hope 1968) to establish
how frequently the observed mean rank at a given level in the hierarchy is numerically
lower (i.e. higher ranking) than the equivalent mean ranks obtained in a large number of
simulations in which tomb sizes are allocated independently of group membership.
This entails repeatedly permuting (randomly shuffling the tomb sizes) and re-running
the allocation of ranks to levels by group. We conduct 999 such simulations and then in
accordance with standard practice (Manly 1991) derive p-values by comparing the
extremeness of the observed mean ranks at a given hierarchical level to the simulated
mean ranks at that level (Table 3: Pval). Note that we have not attempted to compute a
p-value for the distribution of ranks across all hierarchical levels as it is a moot point
whether the latter would constitute the testing of multiple hypotheses and so require
correction of the p-values to control the family-wise error rate (Holm 1979). The result
of this Monte Carlo simulation suggests that the observed distribution of tomb sizes in
each of the first two hierarchical levels is closer to that expected of a hierarchical
distribution (each group having a tomb ranked highly by size) than would be expected
by chance (i.e. p <= 0.05). In fact, this is true for three out of the first four hierarchical
levels. Consequently, our test supports the argument that the largest tombs are
distributed across the groups in a way that is broadly (albeit not perfectly) hierarchical,
to an extent that is unlikely to occur by chance alone.
[Caption] Table 3. Results of the permutation tests on megalithic site sizes by cluster
group.
7. Discussion
Modern spatial statistical methods allow us to move beyond a now rather stale debate
about environmental determinism, because they facilitate empirical investigation of the
interplay of different causes, as opposed to the a priori assertion of primacy according
to theoretical preference. In this case, we have been able to demonstrate that
distribution of megalithic mounds in the Monte Penide region reflects a preference for
10
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
locations with particular environmental properties, but at a more local scale the spacing
of these mounds seems to reflect some kind of social partitioning of the landscape.
The results described above allow us to conclude that megalithic sites in the Monte
Penide region concentrate at specific elevations (300-500 mASL) close to the ridgelines
that define the main watersheds draining to the sea. Once these major locational trends
are accounted for, sites still exhibit clustering within 1km of each other, probably
implying that once some megalithic landmarks were established, these encouraged the
construction of new ones nearby. These patterns elicit interesting archaeological
interpretations. First, they suggest that different groups in the regions gradually built up
mound groups on the upland side of possible local territories (see also Bakker 1976,
1991; Bradley 1991a; Criado Boado, Villoch Vázquez 2000; Murrieta-Flores 2012), on
ridge tops draining to the sea. These locations also placed such mounds in highly visible
locations on major routes of upland travel across the landscape (see also Carrero-Pazos
2018b), but we make no attempt here to tease apart which of these motivations, if any,
might have been the most important one for the mound-builders.
Beyond the broader locational preferences, the distribution of mound volumes/sizes in
the nine identified clusters exhibit a non-random hierarchical pattern, with a larger
mound per group and then smaller ones around that, with what appears to be a
preference for the large example to be at or near the meeting point of several watersheds
(and upland ridge-routes). At risk of straying too far beyond the evidence, this
observation suggests that the Monte Penide region was probably occupied by several
groups of roughly equivalent status. Furthermore, without wishing to delineate strict
territories without better dating and wider evidence, it is still a useful thought
experiment to imagine rough zones of activity that often gave each of these identified
groups a strip of land from mountain to sea of ca.70 sq.km, and larger still if such areas
also extended over into the inland valleys.
Such conclusions fairly match with the spatial structure of megalithic landscapes in
other Iberian and European regions, such as for example in Falbygden, Sweden
(Sjögren 2010) where mounds tend to be found in marginal zones and in clusters, with a
large tomb in the centre of the area. In that case, the interpretation was that such
patterning related to social factors, such as the organisation of settlement or social
11
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
groups, with mounds acting as territorial markers. Similarly, the role of megalithic
architectures as landmarks, waypoints or territorial markers that have helped to structure
movement along emerging path networks has been very well analysed in south-Iberian
megaliths (Wheatley et al. 2010). On the other hand, linear arrangements of barrows in
the Netherlands and Ukraine have been interpreted as being the result of processes of
extension through time, probably representing lineages or family necropolis (a trend
especially visible in the Bronze Age, e.g. Bourgeois 2013; Makarowicz et al. 2018).
Furthermore, differences in tomb orientation may in certain circumstances indicate
specific cultural or funerary trends (Hoskin 2001; Silva 2010; Prendergast 2016), but
this is still under initial investigation in Galicia (see e.g. Corujo-Tilve, Domínguez
Márquez 2014; González García et al. 2017; 2018).
Future work in Galicia would clearly benefit from better radiocarbon dating of mound
use-lives and mound groups, and some efforts in this direction are already underway in
the Barbanza peninsula, west of Galicia (Rodríguez Rellán, Fábregas Valcarce in press)
in order to elucidate a better ‘biography’ of megalithic sequences (Rodríguez Casal
1989; Alonso Mathías, Bello Diéguez 1995; Mañana Borrazás 2005; Bóveda Fernández,
Vilaseco Vázquez 2015). Likewise, there is much still to infer about what the possible
territories really mean in terms of anticipated community sizes and social structure,
something more visible in later periods such as the Bronze Age. However the above
exploration already suggests new theory and method that might be applied to a much
wider set of Galician and European megalithic evidence.
Acknowledgements
The archaeological dataset used in this work was kindly provided by Prof. Antón A.
Rodríguez Casal from the University of Santiago de Compostela, as a part of the project
“Archaeology and Ecology of the Megalithic Complex in the South of Galicia” (1998-
2000), funded by the regional government of Galicia (Xunta de Galicia). The digital
elevation model (LiDAR based) was obtained from the Spanish National Cartographic
Service (http://centrodedescargas.cnig.es/CentroDescargas/index.jsp), derived from
DTM25 CC BY 4.0 https://www.ign.es/. We used GRASS GIS 7.4 (GRASS
Development Team 2015) and the R statistical environment for all the analysis (R
Development Core Team 2008, especially the spatstat package, Baddeley et al. 2016). A
first draft version of this paper (poster) was presented in the International Colloquium
12
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
on Digital Archaeology: Quantitative approaches, spatial statistics and socioecological
modelling, University of Bern, Oeschger Centre for Climate Change Research, Bern
(04/02/2019-06/02/2019). Our thanks also to the reviewers and the editor for their
helpful and useful comments.
Supplementary Data and Methods
To enable re-use of our materials and improve reproducibility and transparency, we
include the raw data and R code used for all the analysis and visualisations contained in
this paper in our supplemental online material at
(http://dx.doi.org/10.17632/3sb4hwrrw9.1). The data used in this paper is released
under the CC-BY4.0 license and our code with the MIT license to encourage maximum
reuse.
References
Abad Gallego, J. C. (1990-1991). La excavación de la mámoa nº2 de Cotogrande
(Cabral-Vigo). Castrelos 3-4: 29-43.
Abad Gallego, J. C. (1992-1993). Balance de las actuaciones arqueológicas llevadas a
cabo en la necrópolis megalítica de Cotogrande (Cabral). Castrelos 5-6: 7-28.
Abad Gallego, J. C. (1995). Un ejemplo de readaptaciones constructivas en un
enterramiento tumular: Cotogrande nº5. Minius IV: 13-30.
Abad Gallego, J. C. (1996-1997). Materiales líticos procedentes de la necrópolis
megalítica de Monte Penide (Redondela). Castrelos 9-10: 11-16.
Alonso Mathías, F., Bello Diéguez, J. (1995). Aportaciones del monumento de Dombate
al megalitismo noroccidental: dataciones de Carbono 14 y su contexto arqueológico.
Trabalhos de Antropologia e Etnologia 35 (3): 153-181.
Álvarez Limeses, G. (1935). Geografía del Reino de Galicia. Tomo de la Provincia de
Pontevedra. Barcelona: Casa Editorial Alberto Martin.
Baddeley, A., Van Lieshout, M. N. M. (1995). Area-Interaction point processes. Annals
of the Institute of Statistical Mathematics, 47: 601–619.
13
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Baddeley, A., Moller, J., Waagepetersen, R. (2000). Non-and semi-parametric
estimation of interaction in inhomogeneous point patterns. Statistica Neerlandica 54:
329-350.
Baddeley, A., Rubak, E., Turner, R. (2016). Spatial Point Patterns: Methodology and
Applications with R. London: CRC Press.
Baddeley, A., Turner, R. (2005). Modelling Spatial Point Patterns in R. In Baddeley, A.,
Gregori, P., Mateu, J., Stoica, R., Stoyan, D. Case Studies in Spatial Point Process
Modeling (pp. 23-74). New York: Springer.
Bakker, J. A. (1976). On the Possibility of Reconstructing Roads from the TRB Period.
Berichten van de Rijksdienst voor het Oudheidkundig Bodemonderzoek 26: 63-91.
Bakker, J. A. (1991). Prehistoric Long-Distance Roads in North-West Europe. In
Lichardus, J. (Ed.). Die Kupferzeit als historische Epoche. Symposium Saarbrücken und
Otzenhausen 6-13, November 1988. Bonn: Dr Rudolf Habelt GmbH: 505-528.
Bello Diéguez, J. M. (1995). Arquitectura, arte parietal y manifestaciones escultóricas
en el Megalitismo noroccidental. In Losada, F., Castro Pérez, L. (Eds), Arqueoloxía e
arte na Galicia prehistórica e romana (pp. 29-98). Coruña: Museo Arqueolóxico e
Histórico de A Coruña.
Bello Diéguez, J. M.ª, Criado Boado, F., Vázquez Varela, J. M. (1982a). Aproximación
a un modelo económico-social del megalitismo del noroeste peninsular. Brigantium, 3:
33-39.
Bello Diéguez, J. M.ª, Criado Boado, F., Vázquez Varela, J. M. (1982b). Sobre la
cultura megalítica y los caminos antiguos en Galicia. El Museo de Pontevedra, 26: 143-
164.
Bello Diéguez, J. M.ª, Criado Boado, F., Vázquez Varela, J. M. (1982c). Megalitismo y
medio físico en el Noroeste de la Península Ibérica: Estado de la cuestión y
perspectivas. Zephyrus, XXXIV-XXXV: 109-118.
Bello Diéguez, J. M.ª, Criado Boado, F., Vázquez Varela, J. M. (1984). Medio físico y
sociedades megalíticas: aproximación a los problemas constructivos de los megalitos en
el NW peninsular. Gallaecia, 7-8: 31-57.
14
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Bello Diéguez, J. M.ª, Criado Boado, F., Vázquez Varela, J. M. (1987). La cultura
megalítica de la provincia de la Coruña y sus relaciones con el marco natural:
implicaciones socio-económicas. A Coruña: Excma. Diputación provincial de A
Coruña.
Bevan, A., Conolly, J. (2006). Multiscalar approaches to settlement pattern analysis. In
G. Lock, Molyneaux, B. (Eds.), Confronting scale in archaeology: issues of theory and
practice (pp. 217-234). New York: Springer.
Bevan, A. (2012). Spatial methods for analysing large-scale artefact inventories.
Antiquity 86 (443): 493-506.
Bevan, A., Crema, E., Li, X. and A. Palmisano (2013). Intensities, interactions and
uncertainties: some new approaches to archaeological distributions. In: Bevan,
A. and M. Lake (eds.) Computational Approaches to Archaeological Spaces (pp. 27-
52). Walnut Creek: Left Coast Press.
Bóveda Fernández, M.ª J., Vilaseco Vázquez, X. I. (2015). La cámara megalítica de
Chousa Nova 1 (Silleda, Pontevedra). ¿Rotura intencional o colapso? In Gonçalves, V.
S., Diniz, M., Sousa, A. C. (Eds.), Actas do 5º Congreso do Neolítico Peninsular.
Faculdade de Letras da Universidade de Lisboa, 7-9 abril 2011 (pp. 564-570). Lisboa:
UNIARQ WAPS.
Bourgeois, Q. (2013). Monuments on the Horizon. The formation of the barrow
landscape throughout the 3rd and 2nd millennium BC. Leiden: Sidestone Press.
Bradley, R. (1991a). Rock art and the perception of landscape. Cambridge
Archaeological Journal, 1: 77-101.
Bradley, R. (1991b). Monuments and places. In Garwood, P., Jennings, D., Skeates, R.,
Thoms, J. Sacred and Profane. Oxford: Oxford University Committee for Archaeology.
Bradley, R. (1998). The significance of monuments. On the shaping of human
experience in Neolithic and bronze age Europe. London: Routledge.
Carrero-Pazos, M. (2015). Sobre geografía y espacio funerario megalítico en Galicia.
Aproximación a su interpretación. In E. Ramil Rego (Ed.), 3 Congreso Internacional de
Arqueoloxía de Vilalba (Vilalba, 2015) (Férvedes 8) (pp. 153-161). Vilalba: Museo de
Prehistoria e Arqueoloxía de Vilalba.
15
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Carrero-Pazos, M. (2017). The Megalithic Phenomenon in Galicia. Contributions from
Geographic Information Systems and Spatial Statistics to the Study of Site Location
Patterns. PhD Research. University of Santiago de Compostela. Santiago de
Compostela. http://hdl.handle.net/10347/15823
Carrero-Pazos, M. (2018a). Beyond the scale. Building formal approaches for the study
of spatial patterns in Galician moundscapes (NW Iberian Peninsula). Journal of
Archaeological Science: Reports, 19: 538-551.
Carrero-Pazos, M. (2018b). Modelando dinámicas de movilidad y visibilidad en los
paisajes megalíticos gallegos. El caso del Monte de Santa Mariña y su entorno
(Comarca de Sarria, Lugo). Trabajos de Prehistoria. 75 (2): 287-306. DOI:
https://doi.org/10.3989/tp.2018.12216
Carrero-Pazos, M. (2019). Density, intensity and clustering patterns in the spatial
distribution of Galician megaliths (NW Iberian Peninsula). Archaeol Anthropol Sci. 11,
5: 2097-2108. https://doi.org/10.1007/s12520-018-0662-2
Carrero-Pazos, M., Rodríguez Casal, A. A. (in press). General and local spatial trends in
Galician megalithic landscapes. In Hinz, M., Müller, J. Megaliths, societies and
landscapes. Early monumentality and social differentiation in Neolithic Europe. Bonn:
Habelt.
Castillo López, Á. d. (1927). Un depósito de hachas del bronce de talón. Boletín de la
Real academia Galega, 17 (194): 33-42.
Chapman, R. (1981). Archaeological theory and communal burial in prehistoric Europe.
In I. Hodder, Isaac, G., Hammond, N. (Eds.), Pattern to process. Studies in honour of
David Clarke (pp. 387-412). Cambridge: Cambridge University Press.
Corujo-Tilve, J. P., Domínguez-Márquez, M. T. (2014). “Islas de luz”, el factor de
localización de las necrópolis tumulares en el Suroeste de Galicia. Gallaecia 33, 59-96.
Crema, E., Bevan, A., Lake, M. (2010). A probabilistic framework for assessing spatio-
temporal point patterns in the archaeological record. Journal of Archaeological Science
37(5): 1118-1130.
16
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Criado Boado, F. (1988). Contribución al estudio de las relaciones entre las
comunidades megalíticas del noroeste peninsular y su medio natural: implicaciones
socio-económicas. Santiago de Compostela. Universidade de Santiago de Compostela.
Criado Boado, F. (1989). Megalitos, Espacio, Pensamiento. Trabajos de Prehistoria, 46:
75-98.
Criado Boado, F. (2017). A paisaxe como enxeñería inversa da materia e a memoria. In
VVAA. Paisaxe e Patrimonio - Un percorrido polo territorio a través do Arquivo de
Galicia. Santiago de Compostela: Xunta de Galicia. Consellería de Cultura, Educación e
Ordenación Universitaria, pp. 149-157.
Criado Boado, F., Grajal Blanco, M. (1981). Relación entre la distribución de mámoas y
el medio físico de la zona de Sobrado-Curtis. Brigantium, 2, 7-26.
Criado Boado, F., Vaquero Lastres, J. (1991). El fenómeno megalítico y tumular:
formas diversas de pasado monumental. In F. Criado Boado (Ed.), Arqueología del
paisaje. El área Bocelo-Furelos entre los tiempos paleolíticos y medievales (Vol.
Arqueoloxía/Investigación: 6, pp. 129-146). A Coruña: Xunta de Galicia.
Criado Boado, F., Villoch Vázquez, V. (2000). Monumentalizing landscape: from
present perception to the past meaning of Galician Megalithism (north-west Iberian
Peninsula). European Journal of Archaeology, 3 (2): 188-216.
Delibes de Castro, G. (1991). Megalitos ¿Todavía una civilización de muertos?
Arqrítica. Crítica de Arqueología Española, 2 (2): 9-10.
Díaz Álvarez, P. (1973). Noticias de prehistoria en torno a Vigo y su ría. Vigo: Caja de
Ahorros Municipal de Vigo.
Díaz Guardamino, M., García Sanjuán, L., Wheatey, D. (2015). The Lives of Prehistoric
Monuments in Iron Age, Roman, and Medieval Europe. Oxford: Oxford University
Press.
Díaz Sanjurjo, M. (1904). Los caminos antiguos y el itinerario nº 18 de Antonino en la
provincia de Orense. Boletín de la Comisión de Monumentos Históricos y Artísticos de
Orense II: 37-43.
Eguileta Franco, J. Mª. (1999). A Baixa Limia na Prehistoria Recente. Arqueoloxía
dunha Paisaxe na Galicia interior. Ourense: Excma. Diputación Provincial de Ourense.
17
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Ester, M., H.-P. Kriegel, J. Sander, and X. Xu (1996). “A Density-Based Algorithm for
Discovering Clusters in Large Spatial Databases with Noise”. In: Proceedings of 2nd
International Conference on Knowledge Discovery and Data Mining (KDD-96). Ed. By
Simoudis, E., J. Han, and U.M. Fayyad. Portland: AAAI Press, pp. 226–231.
Fábregas Valcarce, R. (2010). Os petróglifos e o seu contexto. Vigo: Instituto de
Estudios Vigueses.
Fábregas Valcarce, R., Bonilla Rodríguez, A., César Vila, M. (2007). Monte dos
Remedios (Moaña, Pontevedra): un asentamiento de la Prehistoria reciente. Santiago de
Compostela: Tórculo.
Ferreira de Sousa, M. G. (2013). O fenómeno tumular e megalítico na região galaico-
portuguesa do Minho. Santiago de Compostela: Tese doutoral. Universidade de
Santiago de Compostela.
Filgueira Valverde, J., García Alén, A. (1977). Inventario de monumentos megalíticos
de la provincia de Pontevedra. El Museo de Pontevedra 31: 43-130.
Filgueiras Rey, A. I., Rodríguez Fernández, T. (1994). Túmulos y petroglifos. La
construcción de un espacio funerario. Aproximación a sus implicaciones simbólicas.
Estudio en la Galicia centro-oriental: Samos y Sarria. Espacio, Tiempo y Forma, Serie I,
Prehistoria y Arqueología, 7: 211-253.
Fleming, A. (1973). Tombs for the living. Man, 8 (2): 177-193.
Gaffney, V., Van Leusen, P. M. (1995). Postscript: GIS, environmental determinism
and archaeology. A parallel text. En G. Lock, Stancic, Z. (Eds.), Archaeology and
Geographical Information System: A European Perspective (pp. 367-382). London:
Taylor and Francis.
Gillings, M. (2009). Visual Affordance, Landscape, and the Megaliths of Alderney.
Oxford Journal of Archaeology, 28: 335-356.
Gómez Nistal, C., Rodríguez Casal, A. A. (2000). El fenómeno tumular y megalítico en
la Galicia suroccidental: aspectos historiográficos y estado actual de la investigación. In
V. O. Jorge (Ed.), Congresso de Arqueologia Peninsular (3º. 2000, Vila Real)-
Neolitizaçao e Megalitismo na Peninsula Iberica (pp. 321-336). Porto: ADECAP.
18
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Gómez Vila, J. (2005). Caminos y túmulos: aproximación al estudio de los caminos
megalíticos en el noroeste peninsular. En P. Arias Cabal, Ontañón Peredo, R., García
Moncó, C. (Eds.), Actas del III Congreso del Neolítico en la Península Ibérica
(Santander, 2003) (pp. 405-412). Santander: Universidad de Cantabria.
González García, C., Criado-Boado, F., Vilas Estévez, B. (2017). Megalithic skyscapes
in Galicia. Culture and Cosmos 21: 87-103.
González García, C., Vilas Estévez, B., López Romero, E., Mañana Borrazás, P. (2019).
Domesticating Light and Shadows in the Neolithic: The Dombate Passage Grave (A
Coruña, Spain). Cambridge Archaeological Journal 29 (2): 327-343.
GRASS Development Team (2015). Geographic Resources Analysis Support System
(GRASS) Software, Version 7.0.2. Open Source Geospatial Foundation. Electronic
document: http://grass.osgeo.orgHidalgo Cuñarro, J. M., Costas Goberna, F. J. (1979).
El Gran conjunto megalítico de Monte Penide. Redondela-Pontevedra. El Museo de
Pontevedra, 34: 1-93.
Hennig, v. H., Lucianu, C. (2000). Zipf, Christaller, Gräberfelder. Sind latent
besieldlungsstrukturen der hallstattzeit aus der verteilung der nekropolen ersichtlich?
Archäologisches Korrespondenblatt 4: 527-548.
Hinz, M. (2011). Who for Whom? Ritual Architecture in the Light of the related
Population. In Furholt, M., Müller, J., Lüth, F. (Eds.), Megaliths and identities: early
monuments in Neolithic societies from the Atlantic to the Baltic; 3rd European
Megalithic Studies Group meeting 13th - 15th of May 2010 at Kiel University (pp. 129-
142). Habelt: Bonn.
Holm, S. (1979). A simple sequentially rejective multiple test procedure. Scandinavian
Journal of Statistics 6 (2): 65-70.
Hope, A.C.A. (1968). A simplified Monte Carlo significance test proedure. Journal of
the Royal Statistical Society B30: 582-598.
Hoskin, M. (2001). Tombs, temples and their orientations. Ocarina Books: Bognor
Regis.
19
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Judge, J. W., Sebastian, L. (Eds) (1988). Quantifying the present and predicting the
past: theory, method and application of archaeological predictive modelling. Denver:
U.S. Department of the Interior, Bureau of Land Management.
Kvamme, K.L. (1983). A manual for predictive site location models: examples from the
Grand Junction District, Colorado. Grand Junction: District Bureau of Land
Management.
Kvamme, K.L. (1984). Models of prehistoric site location near Pinyon Canyon,
Colorado. In: Condie, C. J. (Ed), Papers of the Philmont conference on the archaeology
of northeastern New Mexico (pp. 349-370). New Mexico: Archaeological Council,
Albuquerque.
Kvamme, K.L. (1988). Development and testing of quantitative models, In W.J. Judge
and L. Sebastian (eds.) Quantifying the Present and Predicting the Past: Theory,
Method, and Application of Archaeological Predictive Modeling: 325-428. Denver:
U.S. Department of the Interior, Bureau of Land Management.
Laporte, L., Scarre, C. (2011). The Megalithic Architectures of Europe. Oxford:
Oxbow Books.
Leisner, G. (1938). Verbreitung und Typologie der Galizisch-Nordportugiesischen
Megalithgräber. München: J. B. Lindin.
López Cuevillas, F. (1925). As mámoas do Concello de Lobeira. Boletín de la Real
Academia Gallega, XV: 25-39.
López Cuevillas, F. (1959). La época megalítica en el Noroeste de la Península.
Caesaraugusta, 13-14: 21-77.
Llobera, M. (2015). Working the Digital: Some thoughts from Landscape Archaeology.
In R. Chapman, Wylie, A. (Eds.), Material Evidence: Learning from Archaeological
Practice (pp. 173-188). Abingdon: Routledge.
Maciñeira, F. (1943-1944). Túmulos prehistóricos. Inventario descriptivo de los 286
túmulos prehistóricos hasta ahora descubiertos en la avanzada comarca del cabo
Ortegal. Boletín de la Real Academia Gallega, 23-24: 15-34.
Makarowicz, P., Goslar, T., Niebieszczańskia, J., Cwalińskia, M., Kochkind, I. T.,
Romaniszyna, J., Lysenkoe, S. D., Ważnyf, T. (2018). Middle Bronze Age societies and
20
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
barrow line chronology. A case study from the Bukivna ‘necropolis’, Upper Dniester
Basin, Ukraine. Journal of Archaeological Science, 95: 40-51.
Mañana Borrazás, P. (2005). Túmulo 5 de Forno dos Mouros (Ortigueira, A Coruña).
Primeiros resultados. Cuadernos de Estudios Gallegos LII, 118: 39-79.
Manly, B.F.J. (1991). Randomization and Monte Carlo Methods in Biology. London:
Chapman and Hall.
Martinón-Torres, M. (2001). Os monumentos megalíticos despois do Megalitismo.
Arqueoloxía e Historia dos megalitos a través das fontes escritas (S. VI-S. XIX). Valga:
Concello de Valga.
Méndez Fernández, F. (1998). Definición y análisis de poblados de la Edad del Bronce
en Galicia. In R. Fábregas Valcarce (Ed.), A Idade do Bronce en Galicia: novas
perspectivas (Vol. 77, pp. 153-190). A Coruña: Cadernos do Seminario de Sargadelos.
Mergelina, C.D. (1936). Notas sobre arqueología gallega. Las mámoas del Chan da
Pedra do Couto y de San Colmado. Boletín del Seminario de Estudios de Arte y
Arqueología de Valladolid, 10: 1-33.
Murrieta-Flores, P. (2012). Understanding human movement through spatial
technologies: the role of natural areas of transit in the Late Prehistory of South-western
Iberia. Trabajos de Prehistoria, 69 (1): 103-122.
Murrieta-Flores, P. (2014). Developing computational approaches for the study of
movement: assessing the role of visibility and landscape markers in terrestrial
navigation during Iberian Late Prehistory. In S. Polla and P. Verhagen (Eds.),
Computational Approaches to the Study of Movement in Archaeology Theory, Practice
and Interpretation of Factors and Effects of Long Term Landscape Formation and
Transformation (pp. 99–132). Berlin: De Gruyter.
Nakoinz, O., Knitter, D. (2016). Modelling Human Behaviour in Landscapes. Basic
Concepts and Modelling Elements. Springer: Switzerland.
Palmisano, A. (2012). Zooming patterns among the scales: a statistic technique to detect
spatial patterns among settlements. In G. Earl, Sly, T., Chrisanty, A., Murrieta Flores,
P., Papadopoulos, C., Romanowska, I., Wheatley, D. (Eds.), Archaeology in the Digital
Era (pp. 348-356). Southampton: CAA Proceedings.
21
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Premo, L. (2004). Local spatial autocorrelation statistics quantify multi-scale patterns in
distributional data: an example from the Maya Lowlands. Journal of Archaeological
Science 31: 855-866.
Prieto Martínez, P., Mañana Borrazás, P., Costas Casais, M., Criado Boado, F., López
Sáez, J. A., Carrión Marco, Y., Martínez Cortizas, A. (2012). Galicia. In M. A. Rojo
Guerra, Garrido Pena, R., García Martínez de Lagrán, Í. (Eds.), El Neolítico en la
Península Ibérica y su contexto europeo (pp. 215-253). Madrid: Cátedra.
R Development Core Team (2008). R: A language and environment for statistical
computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-
07-0, URL http://www.R-project.org.
Renfrew, C. (1976). Megaliths, Territories and Populations. In S.J. De Laet (Ed.),
Acculturation and Continuity in Atlantic Europe, Mainly During the Neolithic period
and the Bronze Age (pp. 198-220). Brugge: De Tempel.
Renfrew, C. (1984). Arqueología social de los monumentos megalíticos. Investigación y
Ciencia, 88: 70-79.
Rodríguez Casal, A. (1989). La necrópolis megalítica de Parxubeira. A Coruña:
Monografías urxentes do Museu de San Antón, nº 4. Museu Arqueolóxico e histórico de
San Antón.
Rodríguez Casal, A. A. (1997). O Neolítico atlántico e as orixes do Megalitismo: actas
do Coloquio Internacional (Santiago de Compostela, abril de 1996). Santiago de
Compostela: Universidade de Santiago de Compostela.
Rodríguez Rellán, C., Fábregas Valcarce, R. (2015). Arte rupestre galaica: unha achega
dende a estatística espacial e os SIX. In A. A. Rodríguez Casal, Blanco Chao, R.,
Sistemas de Información Geográfica, gestión del territorio y conocimiento histórico
(Sémata, Ciencias Sociais e Humanidades, 27) (pp. 323-348). Santiago de Compostela:
Universidade de Santiago de Compostela.
Rodríguez Rellán, C., Fábregas Valcarce, R. (in press). Monuments on the move.
Assessing megalith’s interaction with NW Iberian landscapes. In Hinz, M., Müller, J.
Megaliths, societies and landscapes. Early monumentality and social differentiation in
Neolithic Europe. Bonn: Habelt.
22
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Santos Estévez, M. (2008). Petroglifos y paisaje social en la prehistoria reciente del
noroeste de la Península Ibérica (TAPA, Vol. 38). Santiago de Compostela: Instituto de
Estudos Galego Padre Sarmiento. CSIC.
Santos Estévez, M., Parcero Oubiña, C., Criado Boado, F. (1997). De la arqueología
simbólica del paisaje a la arqueología de los paisajes sagrados. Trabajos de Prehistoria,
54 (2): 61-80.
Saxe, A. A. (1970). Social dimensions of mortuary practices. Michigan: University of
Michigan.
Schultz Paulsson, B. (2017). Time and Stone: The Emergence and Development of
Megaliths and Megalithic Societies in Europe. Oxford: Archaeopress.
Sherratt, A. (1990). The genesis of megaliths: monumentality, ethnicity and social
complexity in Neolithic North-West Europe. World Archaeology, Soils and Early
Agriculture, 22 (2): 147-167.
Sherratt, A. (1995). Instruments of Conversion? The Role of Megaliths in the
Mesolithic/Neolithic Transition in Northwest Europe. Oxford Journal of Archaeology,
14: 245-260.
Silva, F. (2010). Cosmology and the Neolithic: A New Survey of Neolithic
Dolmens in Central Portugal. Journal of Cosmology 9, 3094-3106.
Sjögren, K. G. (2010). Megaliths, landscapes and identities: the case of Falbygden,
Sweden. In Furholt, M., Lüth, F., Müller, J. (Eds.). Megaliths and identities. Early
Monuments and Neolithic Societies from the Atlantic to the Baltic (pp. 155-166).
Habelt, Bonn.
Tilley, C. (1994). A phenomenology of Landscape. Paths, Places and Monuments.
Oxford: Berg.
Vaquero Lastres, J. (1990). Ríos y tumbas. Sobre el emplazamiento de túmulos en el
NW peninsular. Trabalhos de Antropologia e Etnologia, XXX (1-4): 151-173.
Vaquero Lastres, J. (1991-1992). Del análisis del emplazamiento al estudio de la
distribución de túmulos en el NW. Brigantium, 7: 151-176.
23
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Vaquero Lastres, J. (1993-1994). Galiñeiro, paso de lobos, novios y héroes. Sobre el
emplazamiento de los túmulos del Noroeste. Cuadernos de Estudios Gallegos, XLI
(106): 11-39.
Verhagen, P., Nuninger, L., Tourneux, F.P., Bertoncello, F., Jeneson, K. (2013).
Introducing the human factor in predictive modelling: a work in progress. In: G. Earl, T.
Sly, A. Chrysanthi, P. Murrieta-Flores, C. Papadopoulos, I. Romanowka, D. Wheatley
(eds). Archaeology in the digital era. Papers from the 40th annual conference of
computer applications and quantitative methods in archaeology (CAA), Southampton,
26–29 March 2012. Amsterdam University Press, Amsterdam: 379-388.
Vilas Estévez, B. (2015). Estudio de las orientaciones y emplazamientos de los túmulos
de la necrópolis de la Serra do Leboreiro en base a la Arqueología del Paisaje y la
Arqueoastronomía. Santiago de Compostela: Universidade de Santiago de Compostela.
Villoch Vázquez, V. (1995). Monumentos y petroglifos: la construcción del espacio en
las sociedades constructoras de túmulos del noroeste peninsular. Trabajos de
Prehistoria, 52 (1): 39-55.
Villoch Vázquez, V. (2000). La configuración del espacio entre las sociedades
constructoras de túmulos en Galicia: estudios de emplazamiento tumular. Santiago de
Compostela: Universidade de Santiago de Compostela.
Wheatley, D. (1993). Going over old ground: GIS, archaeological theory and the act of
perception. In: Andresen, J., Madsen, T., Scollar, I. (Eds), Computing the past.
Computer applications and quantitative methods in archaeology. CAA92 (pp- 133-138).
Aarhus University Press, Aarhus.
Wheatley, D. (1996). Between the lines: the role of GIS-based predictive modelling in
the interpretation of extensive survey data. Analecta Praehistorica Leidensia 28: 275-
292.
Wheatley, D. (2004). Making space for an archaeology of place. Internet Archaeology
15. 10.11141/ia.15.10
Wheatley, D. W., García Sanjuán, L., Murrieta Flores, P., Márquez Pérez, J. (2010).
The megalithic phenomenon in Southern Spain: Approaching the landscape dimension.
Oxford Journal of Archaeology 29 (4): 387-405.
24
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Williams, J. T. (1993). Spatial autocorrelation and the Classic Maya collapse: one
technique, one conclusion. Journal of Archaeological Science 20: 705-709.
25
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Table 1. Variables of the locational model of the megalithic culture in Galicia as found
in the literature (Carrero-Pazos 2017).
Variable Reference
Geology
(granite areas) Leisner 1938; Bello Diéguez et al. 1982a, 1987
Edaphology
(sites close to tillage
areas)
Criado Boado, Grajal Blanco 1981
Altitude
(high elevation areas) López Cuevillas 1959; Criado Boado 1988
Topographic prominence
and visual impact Criado Boado 1988
Relation with transit
network
Díaz Sanjurjo 1904; Castillo López 1927; López
Cuevillas 1925; Maciñeira 1943-1944; Bello Diéguez et
al. 1982b, 1982c; Criado Boado, Vaquero Lastres 1991;
Vaquero Lastres 1991-1992, 1993-1994; Eguileta
Franco 1999; Gómez Vila 2005
Visual impact and water
areas
Vaquero Lastres 1990; Méndez Fernández 1998; Villoch
Vázquez 2000; Santos Estévez 2008
Relation with other natural
features and
archaeological remains
(petroglyphs)
Filgueiras Rey, Rodríguez Fernández 1994; Villoch
Vázquez 1995, 2000; Santos Estévez et al., 1997
26
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Table 2. Previous results from multivariate logistic regression (Carrero Pazos, 2017:
489).
A: Univariate logistic regression
Dependent variable: presence of sites
Covariable p value Predictive value
Altitude 0 Good
Geology 0.060 Bad
Slope 0.271 Bad
Landforms 0.010 Good
Least cost path density 3.597e-10 Good
Cost of passage from potential routes 0.624 Bad
Potential river network 0.001 Good
Cost of passage from highly accumulated water
zones 6.873e-08 Good
Visual prominence of landscape 0.610 Bad
Topographic prominence (100 m) 0.160 Bad
Topographic prominence (1000 m) 2.166e-10 Good
Wetlands 0.069 Bad
B: Multivariate logistic regression
Coeficients Estimate Std.Error Z
value Pr (>|z|)
(Intercept)
-3.420 0.646 -5.287
1.24e-
07
Altitude
0.013 0.002 5.853
4.83e-
09
Least cost path density 0.065 0.017 3.806 0.000
Potential river network -0.045 0.021 -2.100 0.035
Cost of passage from water zones -0.001 0.00 -1.752 0.079
Eliminated covariates: Topographic prominence index (1000 m)
27
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Table 3. Results of the permutation tests on megalithic site sizes by cluster group.
Size-
Level RankSums RankMeans IdealSums IdealMeans SimRank MinSimRank Pval
1 203.5 22.61111 45 5 5 18.5 0.005
2 239.5 29.9375 108 13.5 25 22.5 0.025
3 238 34 147 21 86.5 14.428571 0.0865
4 121 24.2 135 27 25 7.8 0.025
5 112 28 126 31.5 68 10 0.068
6 128.5 32.125 142.5 35.625 148 10.125 0.148
7 142 35.5 157.5 39.375 190.5 8.75 0.1905
8 165 41.25 174 43.5 331.5 6.5 0.3315
9 179.5 44.875 190 47.5 420 13 0.42
10 109 36.33333 153 51 239 5.666667 0.239
11 125.5 41.83333 162 54 376.5 4 0.3765
12 147 49 171 57 532.5 7 0.5325
13 165 55 180 60 693 9 0.693
14 177 59 189 63 759.5 4.666667 0.7595
15 187.5 62.5 198 66 856 4.666667 0.856
16 196.5 65.5 207 69 860.5 5.666667 0.8605
17 212 70.66667 216 72 928 7 0.928
18 225 75 225 75 958 6.666667 0.958
19 143 71.5 155 77.5 883 2.5 0.883
20 150 75 159 79.5 914.5 2.5 0.9145
21 152 76 163 81.5 927.5 2 0.9275
22 154 77 167 83.5 923.5 4 0.9235
23 65 65 85.5 85.5 680 1 0.68
24 68 68 85.5 85.5 692.5 1 0.6925
25 70 70 87 87 748 1 0.748
26 72 72 88 88 762 1 0.762
27 73 73 90 90 759 1 0.759
28 83 83 90 90 893 1 0.893
29 87 87 90 90 915.5 1 0.9155
30 88 88 92.5 92.5 928 1 0.928
31 92.5 92.5 92.5 92.5 984 1 0.984
32 94 94 94 94 996.5 1 0.9965
28
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Figure 1. The study area in Galicia (NW Iberia), overlain on a shaded relief model.
29
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Figure 2. A: Current photograph from a megalithic mound in Monte Penide. B: Field
plans of the excavations carried out by C. Mergelina (1936) in some of the monuments.
30
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Figure 3. A: Watershed boundaries shown on top of an elevation model (the latter a
first order covariate). B: A second covariate constructed by calculating the distance to
watershed edges (excluding coastlines and with certain watersheds merged so that only
basins with sea outlets remain). C: Megalith site intensity as a function of terrain
elevation. D: Megalith site intensity as a function of watershed distance (solid lines
show function estimate while grey shading is pointwise 95% confidence envelope). E:
The predicted log intensity of megalithic sites in the Monte Penide as constructed via
logistic regression.
31
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
32
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Figure 4. Approaches to site distance and confidence. A: Pair correlation function with
a 95% envelope from wholly random Poisson process. B: Pair correlation function of
the observed sites with a 95% envelope conditioned on the first-order covariates model.
C: Pair correlation function with a 95% envelope also conditioned on both the first-
order covariates and a second-order, area-interaction model (r=1500 m).
33
Final version: Carrero-Pazos, M., Bevan, A., Lake, M. 2019. The spatial
structure of Galician megalithic landscapes (NW iberia): A case study from
the Monte Penide region. Journal of Archaeological Science 108.
https://doi.org/10.1016/j.jas.2019.05.004
Figure 5. A: Results of the mark correlation function for mound sizes/volumes. B:
Results of dbscan grouping analysis, showing possible groups of sites with a territorial
meaning.
34