How do centrality, population growth and urban sprawl impact farmland conversion in Norway?

Land Use Policy 59 (2016) 185–196

Contents lists available at ScienceDirect

Land Use Policy journal homepage: www.elsevier.com/locate/landusepol

How do centrality, population growth and urban sprawl impact farmland conversion in Norway? Kristine Lien Skog a,∗ , Margrete Steinnes b a b

Norwegian University of Life Sciences, Department of Landscape Architecture and Spatial Planning, Postboks 5003, 1432 Ås, Norway Statistics Norway, Department of Economics, Energy and the Environment, Postboks 1400 Rasta, 2225 Kongsvinger, Norway

a r t i c l e

i n f o

Article history: Received 4 April 2016 Received in revised form 18 August 2016 Accepted 26 August 2016 Keywords: Spatial analysis Farmland conversion Urban sprawl Centrality Food security Land use planning

a b s t r a c t A strong link has been observed between urbanization processes and conversions of farmland to built-up land. Most previous research has addressed such conversion of farmland in relation to the expansion of larger cities. However, broader analysis of both rural and central areas is needed to increase the empirical basis for these claims. Based on GIS analysis of farmland conversions, this paper aims to expand our understanding of how urban sprawl affects conversion of farmland. We find that most farmland in Norway is converted to built-up areas in relation to existing urban settlement areas. In line with population growth, urban areas in the most central municipalities experience the most significant urban sprawl and loss of farmland. Moreover, the remaining farmland is located in the same areas that already have converted the most. These areas also experience significant pressure for continued land take. Assuming current trends continue, future urban sprawl will pose great challenges for the preservation of farmland in land-use planning. New research and policies are required to handle these challenges more effectively. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction Global food production needs to double by 2050 to meet the projected demands from a rising population. Increasing crop yields is considered to be the solution to meet these demands. However, yields are no longer improving in many regions in the world. Ray et al. (2013) state that the rise of crop yields is insufficient to meet the need for increased food production. Accompanied by climate change, erosion and salinization, the world may come to face a growing agricultural crisis. Thus, we cannot take the availability of food for granted, and food supply policies to protect farmland are increasingly important. This is recognised on the international policy agenda (FAO and ITPS, 2015). The conversion of farmland to built-up land is considered to be an irreversible process (Amundson et al., 2015; Haygarth and Ritz, 2009; Jones et al., 2012; Seto et al., 2011). For historical reasons, large areas of farmland are close to urban areas, and urban sprawl is cited as the main driver of farmland conversion. However, limited empirical evidence exists to support this assumption. The aim of this paper is to increase the understanding of how urbanization processes affect the conversion of farmland to built-up land.

∗ Corresponding author. E-mail address: [email protected] (K.L. Skog). http://dx.doi.org/10.1016/j.landusepol.2016.08.035 0264-8377/© 2016 Elsevier Ltd. All rights reserved.

The methodology used to analyse spatial land use changes and the empirical results from this study can provide input for future research and land use planning policies.

1.1. Urbanization and farmland conversion People move from the countryside to the city. This continuous migration increases the proportion of people living in towns and cities. Urbanization historically has been considered a process of population concentration (Tisdale, 1942). According to FAO and ITPS (2015), more than half of the world’s population was living in urban areas in 2014. The future is increasingly urban, with approximately 66% of the world’s population likely to live in urban areas by 2050. In Norway, eighty percent of the population live in urban settlement areas (Statistics Norway), the same level as in most other European countries (Antrop, 2004), America and Oceania (FAO and ITPS, 2015). The share of the population living in urban settlement areas increases yearly, as a result of immigration and internal migration. Urbanization processes are expected to continue in all regions in the world (FAO and ITPS, 2015). Urban sprawl, which is closely connected to urbanization, can be defined as the excessive spatial growth of cities (Patel, 2014). A meta-analysis of case studies estimates that an area larger than Denmark was urbanized between 1970 and 2000 globally, and it is estimated that 1.5 million km2 is likely to be urbanized in the next

186

K.L. Skog, M. Steinnes / Land Use Policy 59 (2016) 185–196

twenty years (Seto et al., 2011). At least 275 ha of land was converted to built-up land in Europe per day between 1990 and 2000; the built-up area increased by 6% (EC, 2011). The rate of natural land converted to built-up land is also high outside Europe, especially in countries with rapidly growing economies (Gardi et al., 2015). It has been argued that urban sprawl often occurs on the most productive farmland because cities are historically built on fertile soil (Ferrara et al., 2014a; Salvati, 2013; Satterthwaite et al., 2010; Scalenghe and Marsan, 2009). In the U.S., 28% of the land that was developed for urban uses between 1982 and 2010 was classified as farmland. In the Alberta district in Canada, as much as 60% of the urban and peri-urban growth between 1988 and 2010 was located on farmland (Martellozzo et al., 2014). Urban areas are expected to experience more land use pressures (Gardi et al., 2015), which may explain the high proportion of built-up farmland in these studies. Farmland is considered to be a limited resource, and urbanization is experienced on a global scale (Amundson et al., 2015; Haygarth and Ritz, 2009; Jones et al., 2012; Seto et al., 2011). Because the global population is rising yearly, the rate of farmland conversion will probably increase rapidly. Continued urbanization is expected to have major implications for the conversion of farmland to built-up land and the future of the food supply (FAO and ITPS, 2015). While 40% of total land surface is used as cropland and/or for grazing globally (Foley et al., 2005), only 3% of the total land area in Norway is farmland. The rate of food supplied by national resources is less than 50%, and decreases yearly. Available farmland per capita in Norway is only 0.2 ha, lower than the average in the OECD countries (0,3 ha per capita). Urban settlement areas constituted only 0.7% of total land area in 2015. However, these areas increased by almost 2.300 ha yearly between 2008 and 2011, and 1.500 ha yearly between 2013 and 2015. The Norwegian Parliament has acknowledged conversion of farmland as one of the major threats to the future food supply and established a specific national target to limit farmland conversion. The Government has recommended that municipalities establish urban settlement growth borders to strengthen the prevention of farmland around urban settlements. However, we are nevertheless experiencing significant urban settlement expansion. How do these trends affect the conversion of farmland to built-up land? 1.2. Status of research There is limited empirical evidence documenting that urbanization is the principal cause of conversion of farmland to built-up land. Most previous research has only focused on case studies in areas that have experienced large-scale urbanization without also analysing non-urbanized areas for comparison (Curran-Cournane et al., 2014; Debolini et al., 2015; Martellozzo et al., 2014; Pribadi and Pauleit, 2015; Salvati, 2013; Song et al., 2015; Su et al., 2012). However, Hennig et al. (2015) also point to significant land take in rural areas, although they did not separate agricultural land take from other land take in this study. A national study in Norway from 2006 points to scattered building in the rural areas as the major cause of farmland conversions (Saglie et al., 2006). Thus, urbanization has both rural and central dimensions that need to be included (Antrop, 2004; FAO and ITPS, 2015; Hasse and Lathrop, 2003; Malik and Ali, 2015). Deng et al. (2015) also point to the need for more systematic studies in this field. To improve our understanding of how urban sprawl affects farmland, these two dimensions should be included in the GIS analysis. In addition to the rural-central dimension, the location of the conversions inside, close to or far from the urban settlement zones is relevant to mapping their contribution to urban sprawl. Burchell et al. (2005) describe urban sprawl as an unlimited outward and leapfrog expansion of low-density new development,

outside established urban settlement borders. However, few broad studies have analysed the location of farmland conversions in relation to existing urban settlement borders. We need both systematic digital mapping of urban settlement zones and large-scale analysis of farmland conversions and their proximity to these borders to increase this knowledge. Rural/central typology should be able to deal with rural issues at regional or local levels. Van Eupen et al. (2012) point to the need to adapt thresholds and variables to define rurality according to different contexts; one single threshold is not valid across Europe. Muilu and Rusanen (2004) also highlight the need to adapt the definitions of rurality to the needs of regional and local planning. They argue that detailed, fine-grained georeferenced data analysis is needed to identify rural diversity in sparsely populated regions. Van Eupen et al. (2012) argue that the explanatory capacity increases as scale narrows. Detailed mapping and adapted thresholds for rurality and centrality in Norway provide possibilities for efficient GIS analyses at the national level. Population development is also vital to the concept of urban sprawl; urban sprawl can be defined as an increase of scattered settlements with low population density (Prokop et al., 2011). Population density is often included when describing urban sprawl trends in research, see for instance EC (2011) and Hennig et al. (2015), but seldom analysed in relation to conversion of farmland to built-up areas. Research has most often dealt with conversion of farmland for residential purposes (see for instance Levia and Page, 2000; Polimeni, 2005). Little attention has been given to other building purposes, such as leisure building or industry. One study in Norway actually questioned the notions of residential settlement as the major cause of farmland conversions; Saglie et al. (2006) found a surprisingly high percentage of farmland converted for agriculture-related building purposes. In order to increase the empirical knowledge on how farmland conversion and urban sprawl are related, both population density and the purpose of the farmland conversions need to be established. It has been argued that the most productive farmland is situated in regions where people live, near cities or towns (Alterman, 1997; Martellozzo et al., 2014; Satterthwaite et al., 2010; Scalenghe and Marsan, 2009). However, the evidence for this assumption is limited. Few studies address the extent to which existing farmland is located in relation to urban settlement areas. Thus, further analysis of the spatial distribution of remaining farmland is required to strengthen our understanding of how future urbanization processes might impact future farmland conversion. 1.3. Objectives In order to prevent farmland conversion and manage urbanization, it is of vital importance to understand how urban sprawl impacts agricultural land take. Urban settlement areas express the development of urban expansion and provide information useful to determining urban sprawl. This paper aims to increase the empirical knowledge concerning how urban sprawl affects conversion of farmland by including urban settlement borders in both central and rural municipalities. The paper can also provide input for new methodology in GIS analyses of urban sprawl. We address the amount and location of farmland conversions and remaining farmland in relation to (1) urban settlement areas; (2) centrality of the municipality; (3) type of farmland soil; (4) purpose of farmland conversions; and (5) population development and land-use density. We define farmland conversions as agricultural land take and changes in land use from farmland to built-up land (as used by Prokop et al. (2011), Malik and Ali (2015) and FAO and ITPS (2015)). The built-up land includes sealed land, the permanent coverage

K.L. Skog, M. Steinnes / Land Use Policy 59 (2016) 185–196

187

Box 1: Categorization of the municipalities according to their centrality (Statistics Norway).

of farmland and its soil, and corresponding built-up land where agricultural production has ceased. To qualify as an urban settlement area, at least 200 persons should inhabit the area, and the distance between buildings must normally not exceed fifty meters. Municipalities normally host several urban settlement areas, but categorization of municipalities’ centrality depends on whether they are located close to an urban settlement of a certain size. For instance, the most central municipalities host or are located close to so-called regional urban centres, urban settlements with more than 50.000 inhabitants in one of the six cities in different regions of Norway (see Box 1). 2. Materials and methods 2.1. Land use change analysis Norway has detailed digital maps available, covering farmland, sealed land and urban settlement areas. The availability of these maps makes it possible to develop methodology and assess largescale location of farmland and sealed farmland in relation to urban settlement borders. The GIS analysis covers farmland conversions in Norway. The digital information covers the location and physical characteristics of agricultural soils (available at the Norwegian Institute for Bioeconomy Research (NIBIO)), construction of buildings and infrastructure, changes in urban settlement borders and land use/cover. Statistics Norway used these digital maps to assess the conversion of biological productive areas to built-up areas between 2008 and 2011 (Statistics Norway 2012). We use the mapping from 2008 as a basis for further analysis in this article. Farmland was identified by National Resource Maps. The maps serve as a register of national agricultural resources and are updated periodically by orthophoto interpretation and by fieldwork by the municipalities (Straume, 2014). Land use changes were analysed by identifying construction of new buildings in the national cadastre maps, a national property

registry that contains information and maps concerning all properties, addresses and buildings in Norway. We included elements where year of construction could be established. The basis consists of areas with buildings (in most cases properties). Roads and other infrastructure were not included in the analyses because information about the year of construction was not readily available. As of 2015, infrastructure constitutes almost 40% of the built-up area in Norway (Statistics Norway). We also used maps with borders of urban settlement areas to describe the location of the agricultural land take in relation to these settlements. Statistics Norway updates the urban settlement borders yearly. We analysed three geographical zones in particular: 1 Areas within the urban settlement boundary set by the definition applied by Statistics Norway. We could then analyse the conversion of farmland within already built-up areas and in the urban settlement expansion area. 2 Areas within a buffer zone of three kilometres outside the urban settlement boundary. We defined the buffer zone as the periurban area, describing the rural-urban transition zone between the urban area and the countryside. This allowed us to analyse urban sprawl outside of, but still relatively close to the established borders of urban settlements. 3 Areas more than three kilometres from the urban settlement boundary. These areas include rural developments outside the urban agglomeration. We dated all areas classified as built-up in 2015 using information available in the cadastre. By overlaying these built-up areas with a land resource map from 2007, we found both areas that had been farmland in 2007 but were built up in 2015 and the remaining farmland (see Step 1 in Fig. 1). The urban settlement digital maps then provided the basis for analysing the spatial distribution of converted and remaining farmland in relation to these settlements (Steps 2–4). 398 out of 428 municipalities were covered by the GIS analyses.

188

K.L. Skog, M. Steinnes / Land Use Policy 59 (2016) 185–196

Fig. 1. Illustration of the different sources and overlays for the GIS analysis.

2.2. Supplementary information from the GIS-analysis The land resource maps include registered cultivated land, surface cultivated land and grazing land. Land is categorized as farmland based on its potential to be used as such, as determined by the physical condition of agricultural soils, rather than on actual agricultural production. Fully cultivated land refers to agricultural areas that can be used for cultivating field crops or meadow and pasture renewed by ploughing. Surface cultivated land is mostly cleared and leveled in such a way that it can be mechanically harvested. Grassland is farmland suitable for pasture fields where grass species or grass-tolerant herbs cover 50% or more of the land and it is not suitable for mechanical harvesting (for more information, see Straume (2014)). The purpose of the farmland conversions in the GIS analyses is based on information from the national building cadastre map. This includes buildings and their corresponding property. The different categories used in this analysis are residential area, built-up area for farming and fishing, leisure area, business buildings, industry and mining and other purposes. As a result of limitations in maps from the transport sector, farmland conversions for transport purposes are not included.

Management information regarding the amount of farmland converted to built-up land in Norway is normally based on reporting from the municipalities. The municipalities report on what they have approved for farmland conversions, not the actual agricultural land take after the developer has realised their projects. This needs to be taken into account when using the data and discussion the actual amount of approved converted farmland. 3. Results The amount of converted and remaining farmland is important with regard to production of food and other ecosystem services. For land use planning, it is also important to understand the location of the farmland (Fig. 2). We will first present how farmland is converted in relation to urban settlement areas. Next, we analyse urban land growth and population growth in relation to farmland conversions to see how both spatial growth and population density are handled differently according to the centrality of the municipalities. We then look at the purpose of the farmland conversions to understand how different types of building establishments impact land use decisions. Finally, we describe the location of remaining farmland to understand the threat of continued urban sprawl and corresponding farmland conversions.

2.3. Compilation of land-use changes and other demographic characteristics

3.1. Conversion of farmland and its location in relation to urban settlement areas

We further analysed the GIS data of converted farmland at the municipality level. The conversion of farmland was broken down in relation to other attributes of the municipality: (1) centrality; (2) population growth in the municipalities; (3) population growth in urban areas from 2008 to 2015; (4) growth of urban land and; (5) how much farmland different municipalities have approved for conversion to built-up areas. The information is from Statistics Norway and reporting done by the municipalities. We categorized centrality of the municipalities based on their distance to urban settlement areas of certain sizes in accordance with the standard classification from Statistics Norway (see Box 1). It is also in compliance with international use of the concept “urban areas”, although the distances between buildings and minimum population required to achieve the status of urban settlement varies widely from country to country.

Eighty percent of the Norwegian population lives within urban settlement areas. Despite the low share of total land (0.7%) these areas represent, 56% of farmland conversions between 2008 and 2015 occurred inside the 2015 urban settlement borders. These conversions can be considered as densifications of existing urban areas and urban settlement expansions. The peri-urban belt constituted 10% of Norway’s total area, and 20% of the converted farmland was located here. In contrast, only 23% of the farmland conversions were dedicated for rural settlements outside the peri-urban buffer, which actually account for the majority of Norway’s total area (90%). Fully cultivated farmland is soil that can be ploughed and is capable of producing field crops. Therefore, it is an important land resource for national food supply policies (for further description, see Straume (2014)). Our results show that almost three quarters of

K.L. Skog, M. Steinnes / Land Use Policy 59 (2016) 185–196

189

1400 1200 1000 800 600 400 200 0 Within urban selements

0-3 km from urban selements > 3 km from urban selements

Fig. 2. Conversion of farmland in relation to its location inside or outside urban settlement areas 2008–2015 (hectare). N = 2143.8 ha.

the converted farmland was categorized as fully cultivated. These losses equal a loss of a production of approximately 45 thousand tons of grain per year. We also found that most of the converted farmland close to urban settlement borders has been fully cultivated. The share of conversions from fully cultivated farmland is higher within urban settlement areas (82%) than in the most rural areas more than 3 km from these urban settlements (60%).

3.2. Conversion of farmland and centrality of municipalities While urban settlement areas are registered when they have more than 200 inhabitants, municipalities are categorized in relation to their centrality – their proximity to urban settlements of certain sizes (see Box 1). Municipalities are administrative units incorporated for local self-governance, covering both urban settlement areas and less populated areas. Municipalities are considered to be most central when they are hosting or located close to one of the six biggest regional cities of Norway. The most central municipalities constitute a third of all the municipalities in Norway, but only 18% of the total land area. Despite the limited land, as much as 68.5% of the Norwegian population was living in these municipalities in 2015 (Statistics Norway). We found that central municipalities convert the most farmland to built-up area. The amount of converted farmland increases with increasing centrality of the municipality (see the red bars in Fig. 3). The most central municipalities account for more than half of the total conversions in acres. They have also converted the most farmland by far if we look at the average per municipality. Further, we find that a small number of municipalities are responsible for a relatively high share of the converted farmland. Only 5% of

the municipalities (20 out of 398) were responsible for converting one-third of the farmland. Fifteen of those 20 municipalities were among the most central municipalities. According to the law, municipalities have to approve farmland conversions before buildings are erected. We find that central municipalities had the highest share of approvals (the blue bars in Fig. 3). They accounted for 35% of all approvals. However, the quite central municipalities had the highest share of approvals on average per municipality (19%). In total, the municipalities approved almost three times as many conversions as we detected in the investigation period. The gap between approvals and detected conversions might be explained by the lack of data from the transport sector. Our analysis covered conversions of farmland to buildings, not to transport purposes. In addition, some approvals will never be realised. If we assume that the share of conversions detected in our analysis is evenly distributed among municipalities, we can see a tendency in the material. The most central municipalities converted the highest share of their approvals.

3.3. Farmland conversions, centrality and urban settlement areas We have now identified how farmland conversions correspond to urban settlement areas and centrality of municipalities. We find that conversions of farmland dominate inside and around urban settlement areas in the most central municipalities. Fig. 4 shows a clear correlation between conversion of farmland to built-up land, urban settlement area and centrality. The conversion of farmland in urban settlement areas increases with increasing centrality of the municipalities. The most central

In total

Per municipality 25

2500

20

2000

15

1500

10

1000

5

500 0

0 Central (n=146)

Quite central Less central Least central (n=58) (n=117) (n=77) Converted farmland

Central (n=146)

Quite central Less central Least central (n=58) (n=117) (n=77)

Approved conversion of farmland

Fig. 3. Conversion of farmland to built-up land and approval of farmland conversions distributed among central and rural municipalities 2008–2015 (hectares). N = 398 municipalities.(For interpretation of the references to colour in the text, the reader is referred to the web version of this article.)

190

K.L. Skog, M. Steinnes / Land Use Policy 59 (2016) 185–196

Percentage distribuon per municipality (N=21 hectares)

Hectares distribuon per municipality (N=398 municipalies) 100 %

9,00 8,00 7,00 6,00 5,00 4,00 3,00 2,00 1,00 0,00

80 % 60 % 40 % 20 % 0% Central (n=146)

Quite central (n=77)

Less central (n=58)

> 3 km from urban selements

Least central (n=117)

Central (n=8)

0-3 km from urban selements

Quite central (n=6)

Less central (n=4)

Least central (n=3)

Within urban selements

Fig. 4. Conversion of farmland among municipalities of different centralities and their location relative to urban settlement areas 2008–2015. N = 398 municipalities, 21 ha.

municipalities convert the most farmland inside their urban settlement borders; two thirds of their conversions are converted within their urban settlement borders of 2015. Sixty percent of the Norwegian population lives in the most central urban settlement areas. In contrast, only 4% of the population lives in the least central municipalities. In these municipalities, only one quarter of the converted farmland is located inside their urban settlement areas. The least central municipalities converted more than half of their farmland far away from their urban settlement areas. 3.4. Farmland conversions and urban settlement expansion So far, we have seen how converted farmland is located inside, close to, or more than 3 km away from urban settlement borders of 2015. Conversion of farmland to built-up land in the 3 km belt outside urban settlements can be characterised as leapfrog urban expansion. In the future, these conversions can become part of an expanded urban settlement area. However, expansion of contiguous urban settlements has also taken place (see Fig. 5). We can identify them by including the expansion of urban settlement borders within the period of this GIS analysis.

The method for establishing urban settlement borders has changed during the period of this analysis. Until 2012, the property corresponding to new buildings was registered as buffer zones. From 2013, the borders were established as the actual built-up zone of the corresponding buildings. The urban settlement areas are therefore more exact, but also diminished by the introduction of the new registration method. Because of this change, we have to separate the analyses of urban settlement expansion in two periods: 2008–2011 and 2013–2015 (see Fig. 6). We found that most of the yearly converted farmland in urban settlement areas was located within the original urban settlement borders (60% in the period 2008–2011 and 69% in the period 2013–2015). However, although the extended urban settlement areas from 2008 to 2011 represent only 0.02% of the total land area, 20% of all converted farmland in the period was located here. Similarly, the urban settlement expansion zone between 2013 and 2015 represents still just 0.01% of total land area. Twenty seven percent of the converted farmland in the same period was due to urban settlement expansion. The urban settlement land expansion increased with increasing centrality of the municipalities in both periods. However, the converted farmland constitutes a minor part of the total land consumption in the expansion zone. Still, the most central municipalities had the highest share of urban settlement land consumption as converted farmland (3% in the period 2008–2011, 7% between 2013 and 2015). This leads to the highest growth in both urban settlement areas and farmland consumption in the most central municipalities. 3.5. Population growth and density

Fig. 5. Example of how farmland is converted to built-up land in urban settlement areas.

When we addressed the spatial dimension of urban sprawl both inside urban settlement areas, in the urban settlement expansion zone and in the peri-urban belt, we found that the conversion of farmland around urban settlement areas was highly influenced by urbanization trends. To determine how sprawl impacts farmland conversions, we also need to characterise the trends with regard to population development. Although the central municipalities have the most urban settlement land and farmland conversion, they use their new urban settlement area more efficiently to handle population growth. Each new inhabitant in urban settlement areas in the most central municipalities occupies less new farmland, both inside and outside

K.L. Skog, M. Steinnes / Land Use Policy 59 (2016) 185–196

0,50

2008-2011

12,00 0,90

10,08

10,00

0,40 0,30

6,00 0,20 2,01

0,10

Central (n=146)

1,95

2013-2015

7,00 6,00

0,70

5,00

8,00 0,60

6,17

0,00

0,80

6,25

191

3,38

0,50

3,77

4,00

0,40

3,00

4,00 0,30 0,20 2,00 0,10

1,19

2,00 1,00

0,00 0,00 Least Central Quite Less Quite Less central Least (n=146) central central central central (n=58) central (n=58) (n=117) (n=77) (n=77) (n=117) Yearly farmland conversions within the original urban selement borders (Hectares) Yearly farmland conversions in the urban selement expansion zone (Hectares) Yearly urban selement area growth (Hectares)

0,00

Fig. 6. Yearly urban settlement land expansion and the corresponding conversion of farmland in the periods 2008–2011 and 2013–2015 in municipalities with different centralities (Hectares per municipality). N = 398 municipalities.

urban settlement areas (Table 1). The trend is similar regarding development of new urban land: population density increases with increasing centrality of the municipalities. The least central municipalities present the opposite trend. Farmland is more often converted to built-up land outside of urban settlement areas, despite falling rural populations. 3.6. Purpose of the farmland conversions Many studies have regarded urban sprawl as only the construction of new residential buildings. This study addresses the conversions to a wider range of building purposes. Of the conversion categories assessed in this paper, we find that more than half of the converted farmland was converted to residential areas in the period of 2008–2015 (Fig. 7). Most conversions of farmland for residential purposes were located inside the urban settlement borders of 2015. Nevertheless, almost 40% of the conversions for residential purposes can be characterised as scattered building situated outside the urban settlement borders of 2015. A high share of the farmland was converted for agriculturalrelated building purposes (16%). We found that the purpose of farmland conversion inside and outside urban settlement areas differs significantly. Conversions of farmland to built-up land for agricultural building purposes also occurs most often outside urban settlement areas. In the peri-urban belt, the agricultural sector is responsible for almost 40% of the farmland converted to built-up land. Conversion for leisure building purposes increases with distance from urban settlement areas. Because of limitations in the digital maps, the transport sector was not included in the analysis. 3.7. Location of the remaining farmland We have observed that farmland is converted to built-up land primarily inside and around urban settlement areas. Given continued urbanization trends, the risk for future conversion is highly

dependent on where the remaining farmland is located. The more of the remaining farmland that is located around urban settlement areas, the higher the risk of it being built up if current trends continue. The study area had 10,317 km2 of land categorized as farmland in 2015. We see that the most central municipalities have most of the remaining farmland (Fig. 8). They have only 20% of the total land area, but almost half of the remaining farmland. Per municipality, the amount of farmland increases with increasing centrality. We also find that most of the farmland inside urban settlement areas is now built up. Only 1.3% of the remaining farmland is located here. Half of the remaining farmland is situated in the peri-urban buffer (less than 3 km from the urban settlement borders). One third of all the remaining farmland is located in the peri-urban farmland in the most central municipalities. As previously explained, farmland categorized as fully cultivated consists of agricultural areas that can be used for cultivating field crops or meadow and pasture renewed by ploughing. As is the case for farmland conversions, we find most fully cultivated farmland in the most central municipalities (see Fig. 9). More than half of the fully cultivated farmland is located here. Thirty-five percent of all farmland is located around their urban settlement areas. These areas experience the highest population and urban land growth and are most prone to risks of continued urban sprawl. The municipality of Sola on south western Norway experiences these dilemmas in their land use planning. Sola, one of the most central municipalities, approved most conversions of farmland to built-up land in the investigated period, and 45 ha of farmland was converted. They have more than 4000 ha of farmland left, with most of it categorized as fully cultivated. Ninety-five percent of the fully cultivated farmland in Sola is located inside the urban settlement borders or in the peri-urban areas around these borders. Fig. 10 illustrates the location of the remaining farmland in one of its urban settlement areas: it is surrounded by fully cultivated farmland. The case is a good demonstration of the difficulties many municipalities

Table 1 Development of population, areas and farmland in the periods 2008–2011 and 2013–2015 inside urban settlement areas (Average per municipality). N = 398 municipalities. Municipalities Centrality

Central (n = 146) Quite central (n = 77) Less central (n = 58) Least central (n = 117)

Inside urban settlements Average population (inhabitants)

Population density (inhabitants/km2 )

Average yearly urban population growth (percent)

Average yearly urban land growth (percent)

New urban area (m2 /new inhabitants)

Converted farmland (m2 /new inhabitants)

2015

2011

2015

2008–2011

2013–2015

2008–2011

2013–2015

2008–2011

2013–2015

2008–2011

2013–2015

23,893 11,300 5332 3611

57.7 18.0 7.2 2.7

61.4 19.2 6.9 2.85

1.8 1.2 0.9 0.7

1.6 1.0 1.7 0.9

1.0 1.0 0.7 1.1

1.4 1.2 2.9 1.4

303.2 697.0 626.7 1806.2

392.7 1099.2 980.3 2189.0

19.0 37.9 45.8 80.6

22.7 57.4 39.1 55.4

192

K.L. Skog, M. Steinnes / Land Use Policy 59 (2016) 185–196

0-3 km from urban selements (n=449,1)

Within urban selements (n=1202,8)

> 3 km from urban selements (n=491,9) 5%

6%

13 %

7%

11 %

37 %

38 % 19 % 31 % 66 %

2%

39 %

0%

20 % 6% Residenal area Built-up area for farming and fishing Other purposes

Leisure buildings Buisness buildings, industry, mining

Fig. 7. The purpose of farmland conversion in relation to its location inside or outside urban settlement areas 2008–2015 (percent). N = 2143.8 ha.

Per Municipality

In total

4000

600000

3500

500000

3000 2500

400000

2000

300000

1500

200000

1000

100000

500 0

0 Central (n=146)

Quite central (n=77)

Less central (n=58)

> 3 km from urban selements

Least central (n=117)

Central (n=146)

0-3 km from urban selements

Quite central (n=77)

Less central (n=58)

Least central (n=117)

Within urban selements

Fig. 8. Remaining farmland distributed between municipalities of different centralities and their corresponding growth in population and urban settlement areas 2015 (hectares). N = 398 municipalities.

In total

Per municipality 4000

600000

3500

500000

3000 400000

2500 2000

300000

1500

200000

1000 100000

500 0

0 Central (n=146)

Quite central (n=77)

Pasture

Less central (n=58)

Least central (n=117) Surface culvated

Central (n=146)

Quite central (n=77)

Less central (n=58)

Least central (n=117)

Fully culvated

Fig. 9. Type of remaining farmland distributed between municipalities of different centralities and their corresponding growth in population and urban settlement areas in 2015 (hectares). N = 398 municipalities.

K.L. Skog, M. Steinnes / Land Use Policy 59 (2016) 185–196

Fig. 10. Example of urban settlement borders (red) and land cover in the central Sola municipality. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

experience in daily land-use planning to tackle population growth: to avoid conversion of farmland, scattered building and continued urban expansion. There are few alternatives left. 4. Discussion 4.1. Is the Norwegian case relevant to others? to what extent farmland conversion can be identified depends on methodology and on the quality of digital maps (Thompson and Prokopy, 2009; Tulloch et al., 2003). For instance, FAO and ITPS (2015) mainly cover the actual sealing of soil, while in this study, we include the corresponding land use changes. The availability of detailed land use maps in Norway made it possible to gather detailed data on land use changes. No areas in Norway qualify as predominantly urban based on the OECD scale. Therefore, in order to take into account the regional context, Norway normally uses a finer scale in the definition of centrality, as recommended by Muilu and Rusanen (2004) and Van Eupen et al. (2012). Despite different thresholds for rurality and urban settlement areas, we argue that our results from Norway is relevant for comparison with other countries. First, despite the different scales, the spatial distinction between rural and central areas in this paper is, in general terms, in accordance with the OECD definitions. The most central municipalities in Norway are located around the regional urban settlements also registered as central in the OECD mapping (see the mapping in Van Eupen et al. (2012)). Second, the amount of converted farmland is not the main component, but how urban sprawl and centrality impact farmland conversions. Urbanization trends are familiar across Europe and on other continents. People move towards larger urban settlement areas either as immigrants or from more rural areas. In Norway, as in all regions of the world (FAO and ITPS, 2015), such migration leads to urban land take. Therefore, the land use change patterns are far more interesting than the amount of converted farmland itself. Our results provide an empirical basis for claims regarding how urban sprawl and centralization are related to farmland conversion adequate for comparisons with other research. 4.2. Centrality, urban settlement areas and farmland conversions The distinction between the different levels of centrality in municipalities depends on their distance to an urban settlement area of a certain size, from larger regional cities to smaller villages. The smaller the distance to larger urban settlement areas, the

193

more central the municipality. In the context of our study, centrality proved to be a useful concept that increase the understanding of how urbanization affects farmland conversion. Urban areas are expected to have a higher land-take pressure (Gardi et al., 2015). Our study shows a clear correlation between municipal centrality and conversion of agricultural land. The most central municipalities occupy only 19% of total land, but have converted the most farmland by far. They host 68% of the total population and have experienced the most population growth. These patterns might be found elsewhere, even when centrality defined differently. For instance, Hasse and Lathrop (2003) identified the hotspots of total farmland losses in the most urban areas in New Jersey. Tóth (2012) found a similar trend in the European Union – farmland conversion is increasing with population growth, and countries with more developed economies convert more farmland. Nevertheless, urbanization also affects less central municipalities, but on a smaller scale. The use of urban settlement borders and buffer zones around the urban settlements made it possible to detect outward and leapfrog expansion of new development as a way of explaining urban sprawl. Our methodology demonstrates new possibilities for GIS analysis in order to understand the spatial dimension of urban sprawl. We found that most farmland conversions took place inside the 2015 urban settlement areas; 56% of the farmland was converted in less than one per cent of the total land area hosting eighty per cent of the total population. However, scattered building less than 3 km from urban settlements areas accounted for 20%; this development can be characterised as leapfrog expansion. These patterns are most significant in the most central municipalities, which converted a higher share of farmland both within their urban settlement areas and in their peri-urban areas. The most central municipalities has also expanded their urban settlement area most in the investigation period. These findings support other research stating that urban sprawl has serious consequences for conversion of farmland (FAO and ITPS, 2015; Martellozzo et al., 2014; Pribadi and Pauleit, 2015; Seto et al., 2011). Burchell et al. (2005) also consider urban sprawl to include lowdensity new developments. The dimensions of population density and spatial urban growth add new aspects to our study. With regard to density, the most central municipalities used the newly developed land most efficiently; both existing and new urban land were more densely populated. Each new citizen occupy less farmland and new urban land in the most central municipalities, compared to the less central municipalities. This corresponds with findings by Hasse and Lathrop (2003); the highest consumption of farmland per capita was found in the rural areas in New Jersey. The capital city of Norway, Oslo, demonstrates as an example of increased urban population density (Næss et al., 2011). Fulton et al. (2001) state that an area experiences urban sprawl if land is sealed at a faster rate than the growth of the population. Considered as percent growth, this only happened in the least central municipalities. However, the large number of new inhabitants in the central municipalities creates the need for more new urban settlement areas. Compact housing policies are not sufficient to avoid high rates of farmland conversion. Gardi et al. (2015) state that the large-scale urban expansion in European cities is becoming more similar to the dispersed urban areas in the U.S. The methods selected for GIS analysis highly influence conclusions regarding how urban sprawl and farmland conversions are related. Saglie et al. (2006) analysed conversion of farmland to built-up land in Norway based on similar digital maps, but came to the opposite conclusion. They concluded that two-thirds of farmland conversions was caused by scattered building, and only one-third was the result of urban expansion. Different research methods influenced our conclusions. Saglie et al. (2006) did not

194

K.L. Skog, M. Steinnes / Land Use Policy 59 (2016) 185–196

include farmland conversions inside original urban settlement borders nor the peri-urban buffers around the urban settlement areas in their study. This demonstrates how different methodologies can lead to different conclusions.

4.3. Purpose of the farmland conversions The purpose of farmland conversions is seldom analysed, even though it adds useful information for land use planners. When including construction of all new buildings and their corresponding built-up area between 2008 and 2015, we find that farmland is most often converted for residential building purposes. By including peri-urban and more peripheral areas in the analyses, a more nuanced picture appears. Agricultural development is the second most common building purpose, and accounts for half of the conversions in the peri-urban belts. This corresponds with findings by Saglie et al. (2006). However, agricultural building on farmland has natural explanations. For many landowners, few alternatives to farmland are provided as possible locations for new buildings. In addition, farmers have received agricultural support for developing new businesses on their farms. Substantial subsidies are provided to support investment in new and more efficient production buildings. In addition, the legislation that licenses these conversions states that farmland conversions are accepted when they strengthen the agricultural activity. This might explain the high level of agricultural building on farmland. It is partly supported by the agricultural policy itself. The accessibility of places in relation to transportation routes contributes to land use changes and affects the landscape, allowing urban sprawl and development of cities. According to reporting from the municipalities, the transport sector causes substantial direct land take from farmland. In addition, transport might lead to new conversions of farmland because infrastructure creates new possibilities for land development (Antrop, 2004). Unfortunately, these conversions are not covered in this study. Availability of reliable detailed maps showing when new infrastructure is established will provide great opportunities for future analyses of how this development affects conversion of farmland.

4.4. A risk of misleading reporting Municipalities have to approve conversions of farmland to building purposes before the farmland is built on, and the Norwegian Parliament bases their policies on this reporting. Permissions indicate the municipalities’ willingness to convert farmland, while actual converted farmland indicates the market potential to realise them. The rate of permissions is much higher than the rate of actual conversions during the investigation period. This gap is partly explained by lack of data on conversion to transport purposes in our analysis. Secondly, some approvals will most likely never be realised. However, the gap was greatest in the least central municipalities. Therefore, permissions will most likely be realised in more central municipalities. Current national reporting of farmland conversions does not cover these differences. The high share of conversions to agricultural purposes is not in accordance with relative share in the reporting from the municipalities. Thus, our findings might indicate the consequences of not including these approvals sufficiently in the reporting from the municipalities. Or, the permissions given to the agricultural building purposes are more often realised. However, the GIS analysis can contribute to more accurate understanding of farmland conversion trends and guide the Parliament in creating more effective policy tools.

4.5. Remaining farmland: location and impact for food supply By studying the spatial distribution of remaining farmland, we better understand why the most central municipalities have converted the most: most of the remaining farmland is located here. Little farmland remains inside their urban settlement areas. Most of the remaining farmland is located immediately outside the urban settlement borders. These municipalities are also those that experience the highest population growth and have fewer options for future land take. Continued population growth and urban sprawl indicate that farmland losses will continue on a large scale. Given the limited potential for food production and farmland expansion in Norway, continued urban expansion will probably affect our future self-supply of food. Our findings demonstrates the extent of global concern; continued urbanization leads to conversion of farmland to built-up land that might affect food security on a global scale (FAO and ITPS, 2015). The same challenges are identified in the U.S. and Canada (Francis et al., 2012) and in other parts of Europe (Gardi et al., 2015). Agricultural areas categorized as fully cultivated is considered as the most important farmland for food security policies in Norway. Central municipalities provide half of the remaining fully cultivated farmland in Norway. More than one-third of the fully cultivated farmland is located in urban and peri-urban settlement areas in the most central municipalities. Tóth (2012) found a clear relationship between productivity of farmland and population development in the European Union; conversion of high quality farmland increased with increasing population growth rates. These findings also strengthen the general understanding that farmland is most productive close to urban areas, as expressed by Alterman (1997), Olson and Lyson (1999), Scalenghe and Marsan (2009), Satterthwaite et al. (2010) and FAO and ITPS (2015).

4.6. Implications for future land use policies In the context of climate and land-use changes, availability of food cannot be taken for granted. Food production will most likely not meet the demand for a rising global population (Ray et al., 2013). Urbanization is considered to be a threat for future food supply in all regions of the world (FAO and ITPS, 2015). Researchers have found that urban agriculture has low potential to increase global food security in both high-income and developing countries (Badami and Ramankutty, 2015; Poulsen et al., 2015; Warren et al., 2015; Zezza and Tasciotti, 2010). Thus, effective food production requires protection of farmland. Blum (2013) and Nortcliff (2013) predict that conversion of farmland to built-up land will increase globally due to continuing migration of rural dwellers to urban areas. Countries such as Norway should also contribute to global food security by protecting our limited farmland resources (Vinge, 2015). The Norwegian Parliament also acknowledged the importance of preserving farmland when they strengthened their policies to avoid farmland conversions. Urbanization is a global trend with both social, cultural and economic drivers, and policies suggested by FAO and ITPS (2015) to reverse these trends seem difficult to implement. This study shows how urban sprawl, population growth and centrality impact farmland conversions in Norway; despite established land use targets to combat urban sprawl and protect farmland, conversions occur more often where the development pressure is highest. Given the urban expansion that have taken place in Norway, policies seems not sufficiently effective to minimise farmland conversions at national scale. It might be time to think differently also in land use policies. Barthel and Isendahl (2013) argue that integrating farmland as a part of larger urban landscapes is a step towards a

K.L. Skog, M. Steinnes / Land Use Policy 59 (2016) 185–196

more sustainable future. Lovell et al. (2010) provide a framework for evaluating and designing urban agricultural landscapes. Urban agriculture parks are also recommended by Paül and McKenzie (2013). However, they find that passive protection land use measures have not been sufficient to avoid farmland conversion in peri-urban areas around Barcelona. Different stakeholders have successfully collaborated to achieve a stronger protection of surrounded farmland. Colding and Barthel (2013) suggest that these areas could be managed as green commons to promote collective ownership of future land resources. Given the liberal planning regime and local land use decision making in Norway (Næss et al., 2011; Falleth et al., 2010), integrated agricultural and land use policies might be useful to highlight production of common goods provided by farmland and limit farmland conversion. However, more research is required to identify best practice in land use policy to protect farmland and secure future supply of food. 5. Conclusions This paper increases our understanding of how population growth, urban sprawl and centrality affects farmland conversion spatially. Conversion of farmland to built-up land is seldom analysed in detail on such a large scale. The existence of digital maps makes such analyses possible in Norway. The relationship between centrality, spatial increase in urban settlement areas and farmland conversion highlights how agricultural land is negatively affected by population growth and urbanization processes. The fact that this conversion is most often located near cities demonstrates urbanization’s function as a driver for the loss of farmland. Independent from the scale of centrality and urban settlement areas, similar patterns can be expected to be found outside Norway. There is a tremendous opportunity to integrate soil and land use, given the new possibilities in data generation (Haygarth and Ritz, 2009). We do so by including urban settlement borders, physical production capacities of agricultural soil, centrality, purpose of farmland conversions, population changes, and urban settlement land growth in this study. We demonstrate the usefulness of detailed digital maps and definitions of rurality, as well as new methodological possibilities for future research when detailed digital maps and statistics are available. High resolution maps are needed to obtain reliable data for good decision-making for future land use (Ferrara et al., 2014b). Data is still limited on national, regional and global scales. Better maps of farmland and built-up land are vital to understanding land use changes and to formulating adequate responses (FAO and ITPS, 2015). More research is necessary to increase the empirical understanding of how urban urbanization impacts land use decision making and farmland losses in rural and central areas. Future research should also address how new transport infrastructure and corresponding fragmentation of farmland affect farmland conversion. In addition, it is important to study how land use policies can strengthen protection of farmland and the corresponding supply of food. Land use planning and policies must demonstrate how we can change the consumption of farmland by containing urban growth. Acknowledgments This study has been conducted as a part of the food program commissioned and financed by the Norwegian University of Life Sciences (NMBU). We would like to thank Eirin Hongslo at NMBU and Wenche Dramstad from the Norwegian Institute of Bioeconomy Research (NIBIO) for useful comments and feedback, Gunnar Tenge at NMBU for technical advice, colleagues from the Institute of Landscape Architecture and Spatial Planning for support and constructive advice and Statistics Norway for a great cooperation.

195

References Alterman, R., 1997. The challenge of farmland preservation: lessons from a six-nation comparison. J. Am. Plann. Assoc. 63, 220–243. Amundson, R., Berhe, A.A., Hopmans, J.W., Olson, C., Sztein, A.E., Sparks, D.L., 2015. Soil and human security in the 21st century. Science 348. Antrop, M., 2004. Landscape change and the urbanization process in Europe. Landsc. Urban Plann. 67, 9–26. Badami, M.G., Ramankutty, N., 2015. Urban agriculture and food security: a critique based on an assessment of urban land constraints. Glob. Food Secur. 4, 8–15. Barthel, S., Isendahl, C., 2013. Urban gardens, agriculture, and water management: sources of resilience for long-term food security in cities. Ecol. Econ. 86, 224–234. Blum, W.E.H., 2013. Soil and land resources for agricultural production: general trends and future scenarios-a worldwide perspective. Int. Soil Water Conserv. Res. 1, 1–14. Burchell, R., Downs, A., McCann, B., Mukherji, S., 2005. Sprawl Costs: Economic Impacts of Unchecked Development. Island Press. Colding, J., Barthel, S., 2013. The potential of ‘Urban Green Commons’ in the resilience building of cities. Ecol. Econ. 86, 156–166. Curran-Cournane, F., Vaughan, M., Memon, A., Fredrickson, C., 2014. Trade-offs between high class land and development: recent and future pressures on Auckland’s valuable soil resources. Land Use Policy 39, 146–154. Debolini, M., Valette, E., Franc¸ois, M., Chéry, J.-P., 2015. Mapping land use competition in the rural-urban fringe and future perspectives on land policies: a case study of Meknc¸s (Morocco). Land Use Policy 47, 373–381. Deng, X., Huang, J., Rozelle, S., Zhang, J., Li, Z., 2015. Impact of urbanization on cultivated land changes in China. Land Use Policy 45, 1–7. EC, 2011. Overview of best practices for limiting soil sealing or mitigating its effects in EU-27. In: Prokop, G., Jobstmann, H., Schönbauer, A. (Eds.), Final Report Contracted by the European Commission, 23. DG Environment, Brussels. Falleth, E.I., Hanssen, G.S., Saglie, I.L., 2010. Challenges to democracy in marked-oriented urban planning in Norway. Eur. Plann. Stud. 18, 737–753. FAO, ITPS, 2015. Status of the World’s Soil Resources (SWSR) –Main Report. Food and Agriculture Organization of the United Nations and Intergovernmental Technical Panel on Soils, Rome, Italy650. Ferrara, A., Salvati, L., Sabbi, A., Colantoni, A., 2014a. Soil resources, land cover changes and rural areas: towards a spatial mismatch? Sci. Total Environ. 478, 116–122. Ferrara, C., Salvati, L., Tombolini, I., 2014b. An integrated evaluation of soil resource depletion from diachronic settlement maps and soil cartography in peri-urban Rome Italy. Geoderma 232–234, 394–405. Foley, J.A., DeFries, R., Asner, G.P., Barford, C., Bonan, G., Carpenter, S.R., Chapin, F.S., Coe, M.T., Daily, G.C., Gibbs, H.K., Helkowski, J.H., Holloway, T., Howard, E.A., Kucharik, C.J., Monfreda, C., Patz, J.A., Prentice, I.C., Ramankutty, N., Snyder, P.K., 2005. Global consequences of land use. Science 309, 570–574. Francis, C.A., Hansen, T.E., Fox, A.A., Hesje, P.J., Nelson, H.E., Lawseth, A.E., English, A., 2012. Farmland conversion to non-agricultural uses in the US and Canada: current impacts and concerns for the future. Int. J. Agric. Sustain. 10, 8–24. Fulton, W.B., Pendall, R., Nguyen, M., Harrison, A., 2001. Who Sprawls Most?: How Growth Patterns Differ Across the US. Brookings Institution, Center on Urban and Metropolitan Policy Washington, DC. Gardi, C., Panagos, P., Van Liedekerke, M., Bosco, C., De Brogniez, D., 2015. Land take and food security: assessment of land take on the agricultural production in Europe. J. Environ. Plann. Manage. 58, 898–912. Hasse, J.E., Lathrop, R.G., 2003. Land resource impact indicators of urban sprawl. Appl. Geogr. 23, 159–175. Haygarth, P.M., Ritz, K., 2009. The future of soils and land use in the UK: soil systems for the provision of land-based ecosystem services. Land Use Policy 26 (Suppl. 1), S187–S197. Hennig, E.I., Schwick, C., Soukup, T., Orlitová, E., Kienast, F., Jaeger, J.A.G., 2015. Multi-scale analysis of urban sprawl in Europe: towards a european de-sprawling strategy. Land Use Policy 49, 483–498. Jones, A., Panagos, P., Barcelo, S., Bouraoui, F., Bosco, C., Dewitte, O., Gardi, C., Erhard, M., Hervás, J., Hiederer, R., Jeffery, S., Lükewille, A., Marmo, L., Montanarella, L., Olazábal, C., Petersen, J.E., Penizek, V., Strassburger, T., Tóth, G., Van Den Eeckhaut, M., Van Liedekerke, M., Verheijen, F., Viestova, E., Yigini, Y., 2012. The state of soil in Europe-a contribution of the JRC to the European Environment Agency’s Environment State and Outlook Report–SOER 2010, Reference report, European Commission, Joint Research Centre, Institute for Environment and Sustainability, 76 pp, https://ec.europa.eu/jrc/en/ publication/reference-reports/state-soil-europe-contribution-jrc-europeanenvironment-agency-s-environment-state-and-outlook. Levia, J.D.F., Page, D.R., 2000. The use of cluster analysis in distinguishing farmland prone to residential development: a case study of Sterling, Massachusetts. Environ. Manag. 25, 541–548. Lovell, S.T., Nathan, C.A., Olson, M.B., Mendez, V.E., Kominami, H.C., Erickson, D.L., Morris, K.S., Morris, W.B., 2010. Integrating agroecology and landscape multifunctionality in Vermont: an evolving framework to evaluate the design of agroecosystems. Agric. Syst. 103, 327–341. Malik, R., Ali, M., 2015. The impact of urbanization on agriculture sector: a case study of Peshawar, Pakistan. J. Res. Dev. Manag. 8, 79–85. Martellozzo, F., Ramankutty, N., Hall, R.J., Price, D.T., Purdy, B., Friedl, M.A., 2014. Urbanization and the loss of prime farmland: a case study in the calgary-edmonton corridor of Alberta. Reg. Environ. Change 15, 881–893.

196

K.L. Skog, M. Steinnes / Land Use Policy 59 (2016) 185–196

Muilu, T., Rusanen, J., 2004. Changes in population and industries in the rural areas of Finland: from analysis of administrative regions to GIS based approach. Chang. Funct. Rural Areas Baltic Sea Reg., 109–117. Næss, P., Næss, T., Strand, A., 2011. Oslo’s farewell to urban sprawl. Eur. Plann. Stud. 19, 113–139. Nortcliff, S., 2013. Soils and food security. Niger. J. Technol. Res. 7. Olson, R.K., Lyson, T.A., 1999. Under the Blade: The Conversion of Agricultural Landscapes. Paül, V., McKenzie, F.H., 2013. Peri-urban farmland conservation and development of alternative food networks: insights from a case-study area in metropolitan Barcelona (Catalonia Spain). Land Use Policy 30, 94–105. Patel, P., 2014. Cities without sprawl-a utopian concept. Int. J. Res. 1, 669–679. Polimeni, J., 2005. Simulating agricultural conversion to residential use in the Hudson River Valley: scenario analyses and case studies. Agric. Hum. Values 22, 377–393. Poulsen, M.N., McNab, P.R., Clayton, M.L., Neff, R.A., 2015. A systematic review of urban agriculture and food security impacts in low-income countries. Food Policy 55, 131–146. Pribadi, D.O., Pauleit, S., 2015. The dynamics of peri-urban agriculture during rapid urbanization of Jabodetabek metropolitan area. Land Use Policy 48, 13–24. Prokop, G., Jobstmann, H., Schönbauer, A., 2011. Overview of best practices for limiting soil sealing or mitigating its effects in EU-27. Eur. Commun. 227. Ray, D.K., Mueller, N.D., West, P.C., Foley, J.A., 2013. Yield trends are insufficient to double global crop production by 2050. PLoS One 8, e66428. Saglie, I.-L., Falleth, E.I., Bloch, V.V.H., Bye, A.S., Steinnes, M., 2006. Scattered building and protection of farmland. The scale and the causes of scattered building on agricultural land. NIBR-report 2006, 6, 147 pp, http://www.hioa. no/Om-HiOA/Senter-for-velferds-og-arbeidslivsforskning/NIBR/Publikasjoner/ Publikasjoner-norsk/Spredt-utbygging-og-jordvern. Salvati, L., 2013. Monitoring high-quality soil consumption driven by urban pressure in a growing city (Rome, Italy). Cities 31, 349–356. Satterthwaite, D., McGranahan, G., Tacoli, C., 2010. Urbanization and its implications for food and farming. Philos. Trans. Royal Soc. London B: Biol. Sci. 365, 2809–2820. Scalenghe, R., Marsan, F.A., 2009. The anthropogenic sealing of soils in urban areas. Landsc. Urban Plann. 90, 1–10. Seto, K.C., Fragkias, M., Güneralp, B., Reilly, M.K., 2011. A meta-analysis of global urban land expansion. PLoS One 6, e23777. Song, W., Pijanowski, B.C., Tayyebi, A., 2015. Urban expansion and its consumption of high-quality farmland in Beijing, China. Ecol. Indic. 54, 60–70.

Straume, K., 2014. The social construction of a land cover map and its implications for geographical information systems (GIS) as a management tool. Land Use Policy 39, 44–53. Su, S., Xiao, R., Zhang, Y., 2012. Multi-scale analysis of spatially varying relationships between agricultural landscape patterns and urbanization using geographically weighted regression. Appl. Geogr. 32, 360–375. Tóth, G., 2012. Impact of land-take on the land resource base for crop production in the European Union. Sci. Total Environ. 435–436, 202–214. Thompson, A.W., Prokopy, L.S., 2009. Tracking urban sprawl: using spatial data to inform farmland preservation policy. Land Use Policy 26, 194–202. Tisdale, H., 1942. The process of urbanization. Social Forces 20, 311–316. Tulloch, D.L., Myers, J.R., Hasse, J.E., Parks, P.J., Lathrop, R.G., 2003. Integrating GIS into farmland preservation policy and decision making. Landsc. Urban Plann. 63, 33–48. Van Eupen, M., Metzger, M.J., Pérez-Soba, M., Verburg, P.H., van Doorn, A., Bunce, R.G.H., 2012. A rural typology for strategic European policies. Land Use Policy 29, 473–482. Vinge, H., 2015. Food Security, Food Sovereignty, and the Nation-State. Food Sovereignty in International Context: Discourse, Politics and Practice of Place., pp. 87. Warren, E., Hawkesworth, S., Knai, C., 2015. Investigating the association between urban agriculture and food security, dietary diversity, and nutritional status: a systematic literature review. Food Policy 53, 54–66. Zezza, A., Tasciotti, L., 2010. Urban agriculture, poverty, and food security: empirical evidence from a sample of developing countries. Food policy 35, 265–273.

Internet http://www.oecd.org/std/OECD-Statistics-NewsletterSept2013.pdf. http://www.ssb.no/en/natur-og-miljo/statistikker/arealstat/ aar/2015-09-11. http://www.ssb.no/en/natur-og-miljo/statistikker/arealstat/ aar/2015-09-11?fane=om content. http://www.ssb.no/en/natur-og-milj%C3%B8/statistikker/ arealbruk.