International Journal of Biodiversity Science, Ecosystem Services & Management

ISSN: 2151-3732 (Print) 2151-3740 (Online) Journal homepage: http://www.tandfonline.com/loi/tbsm21

Forest regrowth and cultural heritage sites in Norway and along the Norwegian St Olav pilgrim routes Esgo Kuiper & Anders Bryn To cite this article: Esgo Kuiper & Anders Bryn (2013) Forest regrowth and cultural heritage sites in Norway and along the Norwegian St Olav pilgrim routes, International Journal of Biodiversity Science, Ecosystem Services & Management, 9:1, 54-64, DOI: 10.1080/21513732.2012.711774 To link to this article: http://dx.doi.org/10.1080/21513732.2012.711774

Published online: 10 Aug 2012.

Submit your article to this journal

Article views: 121

View related articles

Citing articles: 2 View citing articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=tbsm21 Download by: [37.44.207.8]

Date: 28 January 2017, At: 00:11

International Journal of Biodiversity Science, Ecosystem Services & Management, 2013 Vol. 9, No. 1, 54–64, http://dx.doi.org/10.1080/21513732.2012.711774

Forest regrowth and cultural heritage sites in Norway and along the Norwegian St Olav pilgrim routes Esgo Kuiper* and Anders Bryn Department of Vegetation, Norwegian Forest and Landscape Institute, PO Box 115, Raveien 9, NO-1431 Aas, Norway Pilgrims travel along the main reopened St Olav pilgrim routes in Norway and visit a variety of cultural heritage types. These routes are part of a value creation programme, in which the management authorities try to increase the numbers of pilgrims. At the same time, forest regrowth is reported to reduce the landscape experience of pilgrims and to biophysically change the cultural heritage sites. However, no studies have been reported on the spatial encroachments of forests along the pilgrim routes. The purpose of this study is to analyse where forest regrowth along the main reopened pilgrim routes in Norway will appear, given the present climatic conditions, and to assess the spatial overlap of future forest regrowth with cultural heritage sites. A potential forest model and a cultural heritage sites database were combined with several baseline geographical data layers and spatially joined in geographical information systems. The results show that most of the future forest regrowth will appear in mountainous parts of the pilgrim routes, whereas many hunting sites, tradition sites and other cultural heritage sites will be overgrown by young forests. Therefore, management efforts to keep the main pilgrim routes open need to be strengthened and directed towards future risks. Keywords: archaeological sites; Askeladden; GIS modelling; potential forest regrowth; St Olav pilgrim routes; Norway

Introduction The rural landscapes of Norway have changed through time. The first half of the twentieth century marked the beginning of forest regrowth on former outfields and agricultural land, a process still ongoing today (Bryn and Daugstad 2001). Forest regrowth is a succession process following land abandonment or reduced land use (McIntosh 1999), where previously open semi-natural heaths, meadows and other habitats within the cultural landscape usually change into dense forests in Norway (Bryn 2008; Norderhaug and Johansen 2011). Forest regrowth in Norway affects a wide range of management issues. For example, forest regrowth is considered to have negative effects not only on the biodiversity of seminatural landscapes in Norway (Potthoff 2007; Fjellstad et al. 2010; Kuiper and Bryn 2011) but also on landscape characteristics that influence the landscape experience of tourists as well as pilgrims (Keller 1996; Paulsen 2005; Fyhri et al. 2009; Flø and Grimsbo 2011; Vinge and Flø 2012). Furthermore, forest regrowth has been documented to have a negative impact on the cultural heritage, for instance by biophysically altering sites through soil upheaval by roots, by hiding sites in an inaccessible and dense forest layer or through increased decay of cultural remnants in a more moist forest environment (Gaukstad 2000; Barlindhaug et al. 2007; Jantsch 2009; RA 2010; LMD 2011). Pilgrim routes were used by pilgrims for ritual journeys to Nidaros Cathedral in Trondheim, Norway (Figure 1; Halvorsen 1996). Such pilgrimages started after the fall of *Corresponding author. Email: [email protected] © 2013 Taylor & Francis

the king Olav Haraldsson during the Battle of Stiklestad in 1030 AD and the king was declared a saint on 31 August 1031 AD. The pilgrimages continued until at least 1537 AD when the Reformation marked the formal end of medieval Catholicism in Norway (Blom 1996; Kollandsrud 1997). After their reopening in 1997, the main historical routes became once again in use, increasing the numbers of pilgrims and tourists follow parts of the historical route and repeatedly visit cultural heritage sites in the proximity (Evensen 2011; RA 2011). Lately, however, public and official concern for forest regrowth along the pilgrim routes has increased (Oppland Fylkeskommune 2004; Paulsen 2005; Thorkildsen 2007; RA 2011). The concern is mainly related not only to the potentially reduced landscape experience among pilgrims but also to the above-mentioned biophysical degradation of the cultural heritage as well as the physical loss of sites (Barlindhaug et al. 2007), including the pilgrim routes themselves. The level of knowledge regarding regional and national future forest regrowth in Norway is low (Norderhaug et al. 2010). However, from the last decade, progress in spatial analyses using geographical information systems (GIS; Franklin 2009) and the availability of nationwide data sets have increased the momentum of studies that aim at predicting landscape changes, especially future forest regrowth following land-use abandonment (Bryn 2008; Bryn and Debella-Gilo 2011; Bryn et al. 2012; Hemsing and Bryn 2012). At the same time, studies that combine landscape ecology, cultural heritage management and

International Journal of Biodiversity Science, Ecosystem Services & Management 222,443

547,792

873,140

6,962,200

7,525,610

–102,905

55

Pilgrim routes

Elevation in metres High: 2465

Low: –68 50 100

6,398,790

0

200

300

400 km

Figure 1. The parts of the new pilgrim routes included in the study (black lines) and the background topography (elevation) of Norway. Note: Map information is given in Table 1.

GIS methods for Norway have appeared (Fry et al. 2004; Barlindhaug et al. 2007), but to our knowledge, no landscape ecology studies of the pilgrim routes in Norway have been carried out to date, although landscape change information has been requested (Paulsen 2005; Berg and Nesbakken 2009; RA 2011). The purpose of this study is to analyse where forest regrowth along the main reopened pilgrim routes in Norway will appear, given the present climatic conditions, and the spatial overlap of future forest regrowth with cultural heritage sites. We will also compare the distribution and abundance of cultural heritage sites along the pilgrim routes with the nationwide distribution of such sites

in Norway, as well as comment some of the implications of the findings.

Material Study area The study was carried out in Norway, covering ∼323.800 km2 and extending from 58◦ N to 71◦ N and from 5◦ E to 31◦ E. The topography rises from sea level to 2469 m above sea level (asl) and varies greatly in slope, height and geological age. Research on the cultural heritage sites was conducted for the whole of Norway,

56

E. Kuiper and A. Bryn

whereas the analyses of the pilgrim routes were restricted to the main routes in southeast and mid-Norway (Figure 1).

Cultural heritage data The cultural heritage data (Riksantikvaren 2011) implemented in this research were downloaded from Askeladden (2011), a continuously updated official digital database administrated by the Directorate of Cultural Heritage in Norway. The database was established in 2004 and provides information about the geographical location (coordinates), monument type and protection category, and in some cases the name of the cultural heritage object or site (Boaz and Steinnes 2007; Askeladden 2011). At the time of access (16 June 2011), Askeladden contained 262,100 single cultural heritage sites, but only 119,352 of these contained geometric information. Of those sites with geometric information, 72,709 were unique, single cultural heritage sites with point geometry, for example charcoal production sites, grave sites and stone-pile fields. The remaining 46,643 records were sites with polygon geometry that partly overlapped with the point-geometry sites, and were therefore excluded. Before the analyses, we also excluded all cultural heritage sites registered in the sea, because they cannot be influenced by terrestrial landscape changes. In total, we continued the study with 68,460 single cultural heritage sites with point geometry (94%). Askeladden (2011) also contains locations with several sites (polygons) and linear features (lines), but these were not included in the study.

Potential forest regrowth model The implemented potential forest regrowth model is a newly developed spatially explicit predictive GIS model (Bryn et al. 2012; Table 1). This model represents an advancement of a preliminary, low-resolution envelope model, published by Bryn and Debella-Gilo (2011). In essence, the new model identifies areas that have a climatic and edaphic potential for forest growth. Presently, the forest is absent in these identified areas due to previous land use. Areas with potential forest regrowth are

modelled by local extrapolation (within 6.25 km) of existing forests into new locations with a high prediction of forest growth based on a multivariate statistical model with a number of explanatory variables, for example temperature, precipitation, soil and vegetation. The climate variables represent the average of the last normal period (1961–1990; Bakkestuen et al. 2008). The predictive forest model is built on a maximum entropy method (Elith et al. 2010), recommended by a number of comparative studies (Ortega-Huerta and Peterson 2008; Franklin 2009; Hemsing and Bryn 2012). The implemented model has a spatial resolution of 25 × 25 m, covering mainland Norway and coastal islands. Mires, fens, boulder fields, built-up areas, existing forests and all other land-cover categories that do not support forest regrowth, as well as cultivated land, were excluded from the model by use of different map layers (Table 1; AR5, AR50 and N50). The model has been tested and has shown to be very conservative regarding the potential forest regrowth (Bryn et al. 2012). These landscapes comprise a variety of semi-natural heaths and meadows along the coast and in the boreal and subalpine vegetation zones (Bakkestuen et al. 2008). The model only predicts the potential for forest regrowth, not where the process is actually occurring today, but all locations are situated beneath the potential upper climatic forest limit, and in areas that are known to be exposed of forest regrowth (Bryn and Debella-Gilo 2011; Norderhaug and Johansen 2011; Hemsing and Bryn 2012).

The pilgrim routes The National Pilgrim Centre in Trondheim, Norway, provided a recently updated map of the pilgrim routes in Norway (Figure 1 and Table 1). The map is one of several available on the Internet (Nasjonalt Pilegrimssenter n.d. 26 January 2011) to enable pilgrims to plan their journeys. The data layer consisted of lines following the pilgrim routes and a variety of metadata. The pilgrim routes that reopened in 1997 mostly follow the original tracks dating from medieval times, but not always (Raju 2001). The new pilgrim routes have been spatially adjusted according to the landscape developments that have taken place since

Table 1. Baseline input data layer information. Data layer Forests, mires, agriculture, glaciers, waters, built-up areas, industrial areas and alpine areas Potential forest regrowth model Bare rock and boulder fields Digital elevation model (DEM) Pilgrim routes Cultural heritage sites: single archaeological sites with point geometry

Data source N50 series Bryn et al. (2012) AR5 and AR50 DTED version 2 Askeladden

Updated

Raster resolution

Data provider

2007

25 m

NMA

2011 2011 2007 2011 July 2011

25 m 25 m 25 m Lines Points

NFLI NFLI NMA NPC DCH

Notes: The N50 series and AR50 are standard topographic and land-cover maps covering entire Norway. The N50 series and AR50 is in a map scale of 1:50,000, whereas AR5 is in a map scale 1:5000. The latter cover approximately 50% of Norway, mainly the lowland parts. All layers were originally in raster format, or converted from vector to raster format, except point and lines. NMA, Norwegian Mapping Authority; NFLI, Norwegian Forest and Landscape Institute; NPC, National Pilgrim Centre; DCH, Directorate of Cultural Heritage.

International Journal of Biodiversity Science, Ecosystem Services & Management the sixteenth century (Keller 1996), but we have not found any systematized map studies that have analysed the spatial deviation between the medieval and the new pilgrim routes in Norway. Thus, the analyses reported here examined the reopened pilgrim routes from 1997.

Land cover and elevation Most land-cover types, for example mires and peat lands, existing forest, water, alpine regions, built-up areas, roads and industrial areas, were extracted from the standard topographic map series (N50; Table 1). Bare rock and boulder fields, however, were generated from other map series (AR50 and AR5; Table 1), and the redundant polygons from the standard topographic map series were excluded. A digital elevation model (DEM; Table 1) was used for the analyses of differences in altitude of the pilgrim routes regarding cultural heritage sites and forest regrowth.

Method All GIS analyses were run in ArcMap version 10.0, including the Spatial Analyst extension. The most important GIS functions used are indicated in parentheses (see also Burrough and McDonnell 1998). All data were imported with the same coordinate system (WGS84 and UTM33N) and converted to the same file types (shape) before any spatial analyses took place (function: to shape file). The shape file with lines of the pilgrim routes was cleaned for repeated lines, gaps between lines were closed and separate lines were united based on ‘area name’ (functions: erase, snapping and dissolve). This resulted in six lines, depicting the main regional routes to or from Trondheim (Figure 1). To join all map information into one new database, the different map topics were transferred onto the points with cultural heritage sites (functions: overlay, intersect and join). All cultural heritage sites within 500 m distance from the pilgrim routes were selected (function: buffer). Then, the potential forest regrowth model, the DEM and the landcover maps (Table 1) were added into the buffered zone (functions: intersect and clip), and the distance from cultural heritage sites to the closest potential forest regrowth model polygon and the pilgrim routes were calculated (functions: join and distance). The overview of the spatial distribution of cultural heritage sites (point file) was calculated with a density estimator (function: fixed kernel density estimator).

Results Potential forest regrowth along the pilgrim routes In total, potential forest regrowth will affect only small areas along the reopened pilgrim routes. Only 8% of the area within the 500 m buffer has the potential for forest regrowth. The potential forest regrowth is structurally related mainly to highland valleys and mountain

57

landscapes. More than 65% of the areas with potential forest regrowth appear at elevations higher than 500 m asl and 36% higher than 800 m asl. Today, more than 51% of the length along the pilgrim routes already leads through forested areas, whereas 20% cross-cultivated land (Figure 2). The pilgrim routes will therefore be less affected by forest regrowth than the rest of Norway; it is estimated that forest regrowth will affect 15.9% of Norway in total (Bryn et al. 2012). Cultural heritage sites along the pilgrim routes and in Norway as a whole The highest density of registered cultural heritage sites (single archaeological sites with point geometry) along the pilgrim routes is in the Oslofjord region, around Lake Mjøsa (∼100 km north of Oslo) and around Verdal (∼70 km northeast of Trondheim). The distribution of registered cultural heritage sites in Norway is highly clustered (Figure 3). The registered cultural heritage sites along the pilgrim routes differ in many aspects from those found elsewhere in Norway (Figure 4). For example, settlements and activity sites and also charcoal production sites seem to be strongly underrepresented along the pilgrim routes, whereas hunting sites and other archaeological sites are less underrepresented. Find-spot and dig sites, cooking pit sites, church sites and unknown sites are overrepresented along the pilgrim routes. The latter is registered as cultural heritage sites, but they have not been classified to a specific heritage type. Potential forest regrowth, land-cover and cultural heritage sites along the pilgrim routes The different cultural heritage sites are registered on various land-cover types along the pilgrim routes (Figures 5 and 6). Some types are closely linked to a specific landcover type, whereas others are documented in a variety of land-cover types. Since many cultural heritage types historically and functionally are linked to specific land-cover types (Fry et al. 2004), it is meaningful to address the spatial link between land-cover types and cultural heritage types. For example, almost all charcoal production sites are found within the present-day forests, whereas grave sites and stone-pile fields have been registered in all land-cover types. A number of cultural heritage sites and monument types along the pilgrim routes are closely spatially linked to the present forest distribution, for example charcoal production sites (93%), hunting sites (56%), rock carvings (54%) and iron production sites (46%). The model shows that the cultural heritage types that will be most affected by future forest regrowth include hunting sites (29%), traditional sites (25%) (known places where repeated cultural activities have taken place, for example locations for prayers or trade), other archaeological sites (15%) and grave sites (9%). Charcoal production sites, church sites

58

E. Kuiper and A. Bryn River, stream: 1.52%

Lake: 3.43%

Built-up area: 2.92%

Mire: 4.96%

Potential forest: 7.94%

Open areas: 8.02% Forest: 50.69%

Cultivated land: 20.51%

Figure 2.

Land-cover types in a buffer of 500 m along the pilgrim routes.

and ship finds will most likely not be spatially affected by future forest regrowth along the pilgrim routes. Discussion The effect of potential forest regrowth along the pilgrim routes This study shows that the pilgrim routes will only partly be affected by forest regrowth in Norway. Approximately 8% of the area surrounding the routes (137 km2 ) will be affected, which is lower than what could be expected, given that the nationwide potential forest regrowth will affect 15.9% of Norway (Bryn et al. 2012). The affected areas are mainly found at elevations between the existing forests and the potential upper climatic forest limit (Aas and Faarlund 2000; Bryn and Debella-Gilo 2011), where the effect of previous long-term and varied land-use is still present in most regions of Norway (Bryn and Daugstad 2001). Following the results of a number of local and regional studies (Potthoff 2007; Rössler et al. 2008; Hemsing and Bryn 2012; Wehn et al. 2012), forest regrowth at these altitudes is mainly considered to be a response to reduced summer dairy farming and abandoned mining, although climate changes in the last decade might be a contributory factor. Summer dairy farming, a transhumance land-use system, dates back to the Bronze Age in Norway and is still

practised in many mountain regions, but has declined substantially since the Second World War (Bryn and Daugstad 2001). Given that more than 51% of the pilgrim routes are already within forested areas and 20% surrounded by cultivated land that is likely to be maintained in use (see discussion below), the potential forest regrowth will only influence a proportion of the open non-forested landscapes that remain today. As much as 16.3% of all the presently open non-forested areas along the pilgrim routes are within the potential forest regrowth model (Figure 2), thus indicating a perceptible potential for reduction of semi-natural mountain landscapes characterized by wide, open horizons. In addition, more forests will lead to lowered landscape diversity (Hemsing and Bryn 2011), which has been reported to be of importance for tourists in Norway (Flø and Grimsbo 2011; Vinge and Flø 2012). These changes in the historical landscapes traversed by the pilgrim routes will therefore probably affect pilgrims’ experiences of their journey, but such questions remain uncertain until landscape perception studies focusing on forest regrowth have been carried out among pilgrims (see, e.g. Fyhri et al. 2009). Cultivated land is generally far less exposed to forest regrowth than semi-natural landscapes in Norway (Bryn and Daugstad 2001), and even less so in counties where

International Journal of Biodiversity Science, Ecosystem Services & Management 252,967

578,315

903,664

7,557,810

–723,809

59

Kernel density 6,994,400

0–0.000000197 0.000000197–0.000000394 0.000000394–0.000000591 0.000000591–0.000000788 0.000000788–0.000000985 0.000000985–0.000001182 0.000001182–0.000001379 0.000001379–0.000001576 0.000001576–0.000001773

Background layers Pilgrim routes Counties

6,430,990

0

50 100

200

300

400 km

Figure 3. Kernel densities of cultural heritage sites (single archaeological sites with point geometry) in Norway and the new pilgrim routes (black lines). Note: Dark areas represent higher densities of cultural heritage sites. Map information is given in Table 1.

the new pilgrim routes cross (Fjellstad et al. 2007; Stokstad and Puschmann 2011). However, in more agricultural marginal areas, for example, the coastal counties of western or northern Norway, cultivated land is more exposed to forest regrowth (Fjellstad et al. 2010; Puschmann and Stokstad 2010). Given the relative stable area of cultivated land during the last decades within the geographical regions where the new pilgrim routes cross (Fjellstad et al. 2007; Stokstad and Puschmann 2011), the potential forest regrowth following abandoned

cultivated land was excluded from this study. In addition, there are no available spatially explicit GIS models for forest regrowth on cultivated land, although we are aware of a study working with such issues (Sang and Dramstad 2011). In our opinion, including models of abandoned cultivated land would amplify the above-mentioned effects of forest regrowth along the pilgrim routes, since some fields probably will be abandoned also in these regions (Fjellstad et al. 2010; Puschmann and Stokstad 2010).

E. Kuiper and A. Bryn 25

Percentage of total registered sites

60

20

15

10

Pilgrim route Norway

5

g

sit ta

nd

di

G ra ve

s

te

n

si

nk

d

t uc

ro

lp

s

te

io

U

g

si

si

C

H

s nt

ae

ch

r ra

e

th

s

ite

ls

ki

O

ng

oo

C

pi

ile

p e-

on

St

s

ld

ite

ts

ca

i og

ol

a

an

e

em

ttl

Se

si

iv ct

d

s

s

te

ity

r hu

r

ha

C

s

te

ch

in

t un

a co

-s

po

si

n

w no

nd Fi

s

te

sit

es

es

0

fie

ds

gs

in

rv

o R

ck

ca

ip

n

fa

tio

uc

l ra

od

on

y

ts

lit

ci

fin

Sh

en

um

on

m

ltu

pr

Ir

ld

cu

fie

ut

O

Figure 4. Cultural heritage sites (single archaeological sites with point geometry) in Norway as a whole and along the pilgrim routes. Note: Numbers are percentages of the total registered sites in Askeladden.

100%

90%

80%

70%

60%

50%

Built-up area Forest

40%

Cultivated land 30%

Open area 20%

Potential forest

10%

O

th er a

rc

H

un

tin

g

si

te s

Tr ad iti on si ha te eo s lo gi ca ls i te U s nk no w n si te R s oc k ca rv in gs G Fi ra nd v e -s s po ite O ta ut s fie nd ld di cu g ltu si Se te ra ttl s lm em on en um ts an en d ts ac tiv ity si C te oo s ki ng pi ts ite s C hu Iro r c n h pr si od te s uc tio n fa c St ilit on y C ha epi rc le oa fie lp ld ro s du ct io n si te s Sh ip fin ds

0%

Figure 5. Land-cover type (percentage) registered at the most common cultural heritage site (single archaeological sites with point geometry) types along the pilgrim routes.

A study design that generates new hypotheses During the course of this study, we found that new hypotheses were generated more often than answers provided. Let us exemplify this with church sites, one of the cultural heritage site types overrepresented along the pilgrim routes compared to the rest of Norway (Figure 4). The causes

of an overrepresentation of church sites led the research into a logical string of hypotheses generation. There can be many hypotheses for the causes behind this spatial pattern of which we will provide, in our opinion, the five most probable ways. First, pilgrims chose to travel to places where there were churches (Otera 2009), seeking places

International Journal of Biodiversity Science, Ecosystem Services & Management

61

195,949

Cultural heritage sites

Habitat type Cultivated land Forest Mire Open areas Potential forest

Elevation in metres High: 1856

6,886,510

Low: 462

0

1

2

4

6

8 km

Figure 6. Map from Dovrefjell of land-cover types and cultural heritage sites (single archaeological sites with point-geometry) along the buffered pilgrim routes. Note: Background grey shading represents topography (elevation).

in which to pray along their journey (Halvorsen 1996). Second, they chose to travel along safe routes (Keller 1996; Kollandsrud 1997), where there were many people, and thus many churches. Third, the church sites are found in densely populated areas, for which there are more records of cultural heritage sites. Fourth, the pilgrims initiated new churches along the routes. Fifth, the pilgrim routes reopened in 1997 incorporated detours to interesting church sites located outside the established medieval routes. The relative importance of the listed hypotheses is presently unknown, and there might as well be several alternative hypotheses that we have not presented. The point is that spatial analyses of cultural heritage sites, available through open access map databases such as Askeladden (2011), could introduce interesting hypotheses worth to deal with in applied science, especially in combination with landscape change studies (Fry et al. 2004; Barlindhaug et al. 2007). Spatial and thematic bias The clustered distribution of cultural heritage sites (Figure 3) raises some important questions. Does the distribution reflect a regionalized registration history or is the clustered pattern a signal of a regionalized and clustered historical land use? The density given by the Kernel estimator indicates that there is a spatial bias in the Askeladden (2011) database, originating from a regionalized registration history of cultural heritage sites. One such indicator is reflected in the highly clustered distribution

connected to the recent survey of a new military training field in Hedmark in southeast Norway. Almost 3000 cultural heritage sites were registered within a forest region of ∼225 km2 (Hedmark Fylkeskommune 2004). According to the survey report, the area was previously considered to be wilderness, with very few cultural heritage sites. However, the survey ultimately reported one of the highest densities of such sites in Norway, thus indicating that the Askeladden database is highly spatially biased towards specific regions that have been carefully inventoried. A logical consequence of such a mapping strategy is a subsequent thematically biased database (Farina 2007), given that the cultural heritage site types are not evenly distributed throughout Norway (Jacobsen and Follum 2008), as documented for church sites in the previous subchapter. The thematic bias clearly will also influence the findings of this study, so the reported spatial and thematic distribution along the pilgrim routes should be interpreted with caution. This challenge calls for an analysis of the Askeladden database (2011), with particular emphasis on spatial bias and representativeness (O’Sullivan and Unwin 2003; Meyer et al. 2007; Jantsch 2009). Askeladden (2011) is the only available database of cultural heritage sites with a national coverage of Norway, and is therefore bound to have a variety of users that will be influenced by the spatial and thematic bias. However, for studies or management of cultural heritage sites spatially restricted within the boundaries of structured and detailed surveyed regions, for example the abovementioned military training field, Askeladden will be well suited and probably without any significant spatial or

62

E. Kuiper and A. Bryn

thematic bias. The results of an ongoing monitoring and documentary survey of cultural heritage sites in 16 municipalities in Norway with the latest report period from 2010 (Sollund 2011) could, in our opinion, however, serve as a starting point for a preliminary test of the influence of spatial and thematic bias within Askeladden (2011). Another solution to overcome the challenge of bias could be to interpolate densities of cultural heritage sites between regions that have been well and structurally surveyed, and thereafter compare the expected modelled density with the one found from Askeladden. The resulting model could then serve as a null model (Gotelli 2001; Bryn et al. 2012), providing information about the expected lack of registration within regions with few cultural heritage sites. A subsequent field study from different parts of Norway could then be used to test the null model (validation) and also form the basis for a new calibrated null model (Franklin 2009). Scale and spatial precision The chosen spatial scale of analysis will influence the results of any study (Farina 2007) and also the present spatial choice of studying potential forest regrowth and cultural heritage sites within a 500 m buffer around the pilgrim routes. However, there is no specific scientific consensus for choosing the size of buffers in cultural heritage landscape studies (Kozlowski and Vass-Bowen 1997), and the buffers seems to vary in width among different studies (Fry et al. 2004; Andrews et al. 2008; Gao et al. 2009). The most important aspect is whether the chosen buffer mirrors the scientific questions of the research (Mu 2008; Bryn and Debella-Gilo 2011). As pilgrims are known to take short detours from the established routes (Berg and Nesbakken 2009), they are likely to visit cultural heritage sites that are within walking distances. The given buffer intersect of 500 m is, therefore, in our opinion, a compromise between the needed broader landscape approach and the more practical walking conditions for pilgrims. Another aspect that will influence the spatial findings within this study is the choice of data type format downloaded from the database, in this case Askeladden (2011). In this study, we only selected data with point geometry, thus excluding data with polygon or line geometry (39%). This will clearly influence the analysed densities and distribution of cultural heritage sites, because a large number of sites are missing. It will also amplify the thematic bias in our material, because, for example grave sites have more records with polygon geometry than with point geometry. Furthermore, it is not given that sites with point geometry have a more precise spatial georeference than sites with polygon or line geometry. On the contrary, the Directorate of Cultural Heritage in Norway, host for Askeladden (2011), restructured the entire database by the end of 2011, providing only polygon-geometry sites from that time. Thus, we urge to interpret the cultural heritage findings of this preliminary study with caution, having in mind the above-mentioned limited use of cultural heritage

sites from the database, as well as the inherited spatial and thematic bias. Directing management plans towards future risks of forest regrowth Forest regrowth in Norway is a documented phenomenon appearing in many regions and is mainly regarded as having negative influences on a variety of value fields, such as tourism, cultural heritage and biodiversity (Barlindhaug et al. 2007; Bryn and Debella-Gilo 2011; Flø and Grimsbo 2011; Kuiper and Bryn 2011; Norderhaug and Johansen 2011). Therefore, the findings of this study clearly point to the future need of landscape management along parts of the pilgrim routes in Norway. It will most likely be necessary to take practical management steps in order to decelerate the progressing forest regrowth along specific parts of the routes. To enable a landscape management plan directed towards future risks of forest regrowth (Barlindhaug et al. 2007), we have shown that the spatial analysis approach, involving a predictive potential forest regrowth model and a database of cultural heritage sites (Askeladden 2011; Bryn et al. 2012), could provide important aspects of the information needed. By introducing this approach, we enable a preliminary method for risk assessment regarding cultural heritage sites and future forest regrowth along the pilgrim routes. If the landscape management authorities would like to plan for future management of cultural heritage sites before the negative changes of forest regrowth have appeared, it is our belief that the introduced methods could be recommended. Acknowledgements This article has been financially supported by the Research Council of Norway (project no. 189977). The authors are grateful to Eiler Munksgaard (Pilegrimssenter), Evy Berg, Joel Boaz and Sverre Steinnes (Riksantikvaren) and Catriona Turner. The authors express their sincere gratitude to two anonymous reviewers.

References Aas B, Faarlund T. 2000. Forest limits and the subalpine birch belt in north Europe with a focus on Norway. AmS-Varia. 37:103–147. Andrews BN, LaBelle JM, Seebach JD. 2008. Spatial variability in the Folsom archaeological record: a multi-scalar approach. Am Antiq. 73(3):464–490. Askeladden [Internet]. 2011. Riksantikvaren; [cited 2011 Jun 16]. Available from: https://askeladden.ra.no/Askeladden/Pages/ LoginPage.aspx?ReturnUrl=%2faskeladden. Bakkestuen V, Erikstad L, Halvorsen R. 2008. Step-less models for regional environmental variation in Norway. J Biogeogr. 35(10):1906–1922. Barlindhaug S, Holm-Olsen IM, Tømmervik H. 2007. Monitoring archaeological sites in a changing landscape: using multitemporal satellite remote sensing as an “early warning” method for detecting regrowth processes. Archaeo Prosp. 14(4):231–244. Berg SK, Nesbakken A. 2009. I pilegrimenes fotefar: pilegrimsleden som verdiskapningsprosjekt [In the footsteps of

International Journal of Biodiversity Science, Ecosystem Services & Management pilgrims: the pilgrim routes in the Value Creation Project]. Oslo (Norway): Norsk Institutt for Kulturminneforskning. NIKU Rapport No.: 33. Blom GA. 1996. Nidaros – vandringens mål [Nidaros – migration targets]. In: Kollandsrud M, editor. Natur, kultur og tro i middelalderen [Nature, culture and faith in the Middle Ages]. Oslo (Norway): Riksantikvaren. p. 72–79. Boaz J, Steinnes S. 2007. Askeladden – ny reiskap for gammalt tema [Askeladden – new tool for old theme]. Oslo (Norway): Riksantikvaren. p. 4–5. Årringer September 2007. Bryn A. 2008. Recent forest limit changes in south-east Norway: effects of climate change or regrowth after abandoned utilisation? Norsk Geografisk Tidsskrift–Nor. J Geogr. 62(4):251–270. Bryn A, Daugstad K. 2001. Summer farming in the subalpine birch forest. In: Wielgolaski FE, editor. Nordic mountain birch ecosystem. UNESCO Man and Biosphere Series 27. New York (NY): Parthenon. p. 307–315. Bryn A, Debella-Gilo M. 2011. GIS-based prognosis of potential forest regeneration affecting tourism destinations and cultural landscapes in south Norway. Scand J Hosp Tour. 11(2):166–189. Bryn A, Dourojeanni P, Hemsing LØ, O’Donnel S. 2012. A high resolution GIS null-model of potential forest expansion following land use changes in Norway. Scand. J For Res. 28. doi:10.1080/02827581.2012.689005. Burrough PA, McDonnell RA. 1998. Principles of geographical information systems. Oxford (UK): Oxford University Press. Elith J, Phillips SJ, Hastie T, Dudik M, Chee YE, Yates CJ. 2010. A statistical explanation of MaxEnt for ecologists. Diversity Distrib. 17(1):43–57. Evensen HB. 2011 Jun 30. Grenseløse pilegrimer [Pilgrims without boundaries]. Hamar Dagblad:12. Farina A. 2007. Principles and methods in landscape ecology: towards a science of landscape. Dordrecht (The Netherlands): Springer. Fjellstad W, Dramstad W, Huso B. 2007. 3Q – Jordbrukets kulturlandskap – Status og utviklingstrekk. Fylker: Hedmark og Oppland [3Q – The agricultural landscapes – Status and trends. Counties: Hedmark and Oppland]. Ås (Norway): Norsk Institutt for Skog og Landskap. Dokument 4. Fjellstad W, Norderhaug A, Ødegaard F. 2010. Agricultural habitats. In: Kålås JA, Henriksen S, Skjelseth S, Viken Å, editors. Environmental conditions and impacts for red list species. Trondheim (Norway): Norwegian Biodiversity Information Centre. p. 39–50. Flø BE, Grimsbo A-MH. 2011 Feb 2. I opphavet var skogen [In the beginning there was forest]. Chronicle Nationen. 28. Franklin J. 2009. Mapping species distributions: spatial inference and prediction. Cambridge Series in Ecology, Biodiversity and Conservation. Cambridge (UK): Cambridge University Press. Fry GLA, Skar B, Jerpåsen G, Bakkestuen V, Erikstad L. 2004. Locating archaeological sites in the landscape: a hierarchical approach based on landscape indicators. Landsc Urban Plan. 67(1–4):97–107. Fyhri A, Jacobsen JKS, Tømmervik H. 2009. Tourists’ landscape perceptions and preferences in a Scandinavian coastal region. Landsc Urban Plan. 94(4):202–211. Gao C, Wang X, Jiang T, Jin G. 2009. Spatial distribution of archaeological sites in lakeshore of Chaohu Lake in China based on GIS. Chin Geogra Sci. 19(4):333–340. Gaukstad E. 2000. Jordbrukslandskapets kulturverdier – utfordring i et tverrsektorielt samarbeid [Cultural values within the agricultural landscape – challenges in a cross-sectorial cooperation]. Copenhagen (Denmark): Nordisk Ministerråd. Report Nord 2000:18. Gotelli NJ. 2001. Research frontiers in null model analysis. Glob Ecol Biogeogr. 10(4):337–343.

63

Halvorsen PB. 1996. Pilegrimsvandringene i middelalderen – den historiske og religiøse bakgrunnen [Pilgrimage in the Middle Ages – the historical and religious background]. In: Kollandsrud M, editor. Natur, kultur og tro i middelalderen [Nature, culture and faith in the Middle Ages]. Oslo (Norway): Riksantikvaren. p. 9–22. Hedmark Fylkeskommune. 2004. Kulturminner for Hedmarks framtid: Fylkesdelsplan for vern og bruk av kulturminner fra 2005 [The future of cultural heritage in Hedmark: county plan for the conservation and use of cultural heritage since 2005]. Hamar (Norway): Hedmark Fylkeskommune. Hemsing LØ, Bryn A. 2011. Changes in landscape- and vegetation heterogeneity of rural Norway. In: Nesheim L. editor. Biodiversity in agriculture – lessons learned and future directions. Stockholm (Sweden): Nordic Association for Agricultural Research. p. 33–38. NJF Report No.: 7(1). Hemsing LØ, Bryn A. 2012. Three methods for modelling potential natural vegetation (PNV) compared: a methodological case study from south-central Norway. Norsk Geografisk Tidsskrift–Nor J Geogr. 66(1):11–29. Jacobsen H, Follum J-R. 2008. Kulturminner i Norge: Spor etter mennesker gjennom 10,000 år [Cultural heritage in Norway: traces of people through 10,000 years]. Oslo (Norway): Tun Forlag. Jantsch AK. 2009. Kvalitetsheving av Askeladden [Quality improvements of Askeladden]. Tønsberg (Norway): Vestfold Fylkeskommune. Keller C. 1996. Pilegrimens landskap i middelalderen og i dag [Pilgrim landscape in the Middle Ages and today]. In: Kollandsrud M, editor. Natur, kultur og tro i middelalderen [Nature, culture and faith in the Middle Ages]. Oslo (Norway): Riksantikvaren. p. 35–48. Kollandsrud M. 1997. Pilegrimsleden til Nidaros [The pilgrim’s route to Nidaros]. Oslo (Norway): Norsk Forlag. Kozlowski J, Vass-Bowen N. 1997. Buffering external threats to heritage conservation areas: a planner’s perspective. Landsc Urban Plan. 37(3–4):245–267. Kuiper E, Bryn A. 2011. Red-listed plants in semi-natural landscapes. In: Nesheim L, editor. Biodiversity in agriculture – lessons learned and future directions. Stockholm (Sweden): Nordic Association for Agricultural Research. p. 68–74. NJF Report No.: 7(1). LMD. 2011. Landbruks- og matpolitikken. Velkommen til bords [Agriculture and food policy. Welcome to the table]. Oslo (Norway): Det kongelige Landbruks- og Matdepartement. Melding til Stortinget nummer 9. McIntosh R. 1999. The succession of succession: a lexical chronology. Bull Ecol Soc Am. 80(4):256–265. Meyer E, Grussenmeyer P, Perrin J-P, Durand A, Drap P. 2007. A web information system for the management and the dissemination of Cultural Heritage data. J Cult Herit. 8(4):396–411. Mu L. 2008. A shape-base d buffering method. Envir Plan B. 35(3):399–412. Nasjonalt Pilegrimssenter [Internet]. n.d. Pilegrimsleden til Nidaros. [cited 2011 Jul 5]. Available from: http://www. pilgrim.info. Norderhaug A, Bele B, Bratli H, Stabbetorp O. 2010. Åpent lavland [Open lowland]. In: Nybø S, editor. Naturindeks for Norge 2010 [Nature index for Norway 2010]. Trondheim (Norway): Direktoratet for Naturforvaltning. p. 70–78. Norderhaug A, Johansen L. 2011. Kulturmark og boreal hei [Semi-natural fields and boreal heath]. In: Lindgaard A, Henriksen S, editors. Norsk rødliste for naturtyper 2011 [Norwegian red-list for nature types 2011]. Trondheim (Norway): Norwegian Biodiversity Information Centre. p. 81–85. DN-Utredning 3. Oppland Fylkeskommune. 2004. Pilegrimsleder i Norge [Pilgrims’ routes in Norway]. Lillehammer (Norway): Oppland Fylkeskommune. Statusrapport No.: 2004.

64

E. Kuiper and A. Bryn

Ortega-Huerta MA, Peterson AT. 2008. Modelling ecological niches and predicting geographic distributions: a test of six presence-only methods. Rev Mex Biodiv. 79(1):205–216. O’Sullivan D, Unwin DJ. 2003. Geographic information analysis. Hoboken (NJ): John Wiley & Sons. Otera FA. 2009. Santiago and the roads to Santiago: a cultural landscape, a culture of landscape. Boletin de la Asociacion de Geografos Espanoles [Internet]. 51:203–218. [cited 2012 Aug 1]. Available from: http://age.ieg.csic.es/boletin/51/24ALONSO.pdf. Paulsen IS. 2005. Pilegrimsleden fra Oslo til Trondheim. Et møte med vandrere, kulturminner, steder og landskap [The pilgrim route from Oslo to Trondheim]. A meeting with walkers, cultural heritage, places and landscapes] [master’s thesis in Archaeology]. Tromsø (Norway): University of Tromsø. Potthoff K. 2007. Persistence of alpine grass-dominated vegetation on abandoned mountain summer farms in Western Norway. Norsk Geografisk Tidsskrift–Nor J Geogr. 61(4):192–206. Puschmann O, Stokstad G. 2010. Status og utvikling i jordbrukets kulturlandskap i Nordland, Troms og Finnmark [Status and development of agricultural landscapes in Nordland, Troms and Finnmark]. Ås (Norway): Norsk Institutt for Skog og Landskap. Rapport No.: 6. RA. 2010. Informasjonsark om vegetasjon [Information sheet about vegetation]. Oslo (Norway): Riksantikvaren.Ark 7.3.5. RA. 2011. Kulturarv som varer. Verdiskapningsprogrammet på kulturminneområdet 2006–2010 [Heritage lasting. Value creation programme on the cultural area 2006–2010]. Oslo (Norway): Riksantikvaren. Raju A. 2001. Pilgrim road to Nidaros. St Olav’s way: Oslo to Trondheim. Cumbria (UK): Cicerone. Riksantikvaren. 2011. Kulturminner [Cultural heritage]. Oslo (Norway): Riksantikvaren.

Rössler O, Bräuning A, Löffler J. 2008. Dynamics and driving forces of treeline fluctuation and regeneration in central Norway during the past decades. Erdkunde. 62(2): 117–128. Sang N, Dramstad W. 2011. Mapping estimates of future rates of farm abandonment based on historic subsidy information. Oral presentation at the conference “People and nature in mountains”; 2011 Sep 23; Trondheim (Norway): Norwegian University of Science and Technology. Sollund M-LB. 2011. Fortidens minner i Trondheim kommune, Sør-Trøndelag 2010 [Cultural heritage in Trondheim Municipality, Sør-Trøndelag 2010]. Oslo (Norway): Norsk institutt for kulturminneforskning. NIKU Tema 36. Stokstad G, Puschmann O. 2011. Status og utvikling i jordbrukets kulturlandskap. Hordaland, Sogn og Fjordane, Møre og Romsdal, Sør-Trøndelag og Nord-Trøndelag [Status and development of the agricultural landscape. Hordaland, Sogn og Fjordane, Møre og Romsdal, Sør-Trøndelag and NordTrøndelag]. Ås (Norway): Norsk institutt for skog og landskap. Rapport No.: 13. Thorkildsen I. 2007. Pilegrim: Å vandre i fremmed land [Pilgrimage: Walking in a foreign country]. Ildsjelen. 4: 34–37. Vinge H, Flø BE. 2012. På gjengrodde stier. Turisters forståelser av et landskap i endring [On overgrown paths. Tourists’ perceptions of a landscape in transition]. In: Rønningen M, Slåtten T, editors. Innovasjon og næringsutvikling i en reiselivskontekst [Innovation and economic development in a tourism context]. Bergen (Norway): Fagbokforlaget. p. 279–296. Wehn S, Olsson GA, Hanssen SK. 2012. Forest line changes after 1960 in a Norwegian mountain region: implications for the future? Norsk Geografisk Tidsskrift–Nor. J Geogr. 66(1): 2–10.