Global-Detector - GIS- and Knowledge-based Tool for a Global

Detection of the Potential for Production, Supply and Demand

Wil Hennen, Arjen Daane and Kees van Duijvendijk

Wageningen Economic Research, Wageningen University & Research, Alexanderveld 5, The Hague, The Netherlands

Keywords: Spatial Analysis, Knowledge-Base, Subjective Assessment, Face Validity, Production Potential, Location

Theory, Worldwide, R.

Abstract: Wageningen Economic Research has developed Global-Detector, a knowledge-based Geographic

Information System that aims to detect the worldwide potential for production, demand and market

strategies. At any spot in the world Global-Detector can show the values from a large amount of indicators,

such as climate, infrastructure, and land characteristics. A large set of indicators is readily available for use

without any GIS-processing, and the model builder together with the expert can instantly start building the

knowledge base for the concerning research aim. Knowledge from experts is applied to combine relevant

indicators to create new indicators. The concept of Global-Detector and 10 applications developed by this

tool are described. As a generic tool with increased flexibility, Global-Detector has many application

possibilities in a wide variety of fields.

1 INTRODUCTION

There is a growing trend towards

internationalization of the agri-food sector in order

to find new markets and reduce costs. Entrepreneurs

who want to expand their production or market

possibilities abroad are often faced with inadequate

and dispersed information about promising locations

in the world. To assemble and harmonize all the

relevant data for the best decisions may become very

costly and time-consuming, especially when data

from different areas (e.g. social and bio-physical) is

required and the search is focused on a larger group

of countries or even the whole world. Whether or

not a particular location is attractive depends on the

assessment of various factors, such as local

biophysical and climatological characteristics, local

socio-economic conditions (urbanization,

infrastructure, market access and population density)

and socio-economic conditions at country-level

(investment climate, fragility index, etc.). Local

advisors or experts are often indispensable for this

assessment, but they can only make a justifiable

valuation for their own region for which they have

access to the relevant local data. For Small and

Medium-sized Enterprises (SMEs) that are aiming to

find the best location with low costs and still

provides adequate market access, it will become

very difficult and costly to find potential areas and to

find and work with local experts from these areas to

gather relevant information. And not to mention the

difficulties to compare potential areas.

Wageningen University & Research has

developed a country-based model to map the global

potential of floriculture production (Benninga et al.,

2016). In this model the development of floriculture

production in a number of countries has been

analysed to determine the attractiveness for

floriculture production. Knowledge from experts has

been acquired and implemented as weighing factors

in the model. However, in order to analyses the

potential for regions within the country, this country-

based model does not suffice. This

led to the

development of the GIS-based

Global-Detector

model in which the attractiveness for floriculture

production can be assessed for each grid at 5’x5’

resolution (approximately 10x10 km, depending on

the latitude) in the world.

Another inspiration to develop Global-Detector was the report of

Kuhlman and Weegh (2014)

GIS = Geographic Information System

Hennen, W., Daane, A. and Duijvendijk, K.

Global-Detector - GIS- and Knowledge-based Tool for a Global Detection of the Potential for Production, Supply and Demand.

DOI: 10.5220/0006256201610168

In Proceedings of the 3rd International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2017), pages 161-168

ISBN: 978-989-758-252-3

161

2 OVERVIEW OF

GLOBAL-DETECTOR

Global-Detector is a knowledge-based Geographic

Information System developed in R that aims to

detect the worldwide potential for production,

demand and market strategies. This tool is

developed by Wageningen Economic Research. At

any spot on the world, i.e. a grid of 5’x5’, the tool

can show the values from a large amount of

indicators, such as climate, infrastructure, and land

characteristics (figure 1).

Figure 1: The concept of Global-Detector.

Relevant indicators that are available in the

library (section 2.1) are chosen by the expert. The

knowledge from one or more experts is applied to

combine relevant indicators to create a new

indicator. For each case a knowledge-based model is

developed containing a set of arithmetic rules that

prescribe how to combine the indicators chosen by

the expert (section 2.2). After the calculation of all

grids (worldwide or a specified region), the result is

presented on a map. This can be worldwide as in

figure 2, or a specific region in figure 3.

Figure 2: Example of a presentation of the result of

Global-Detector (case ‘Potential for ornamental

horticulture (worldwide)’; green is good potential).

There might be other factors that influence the

production or demand for which there are no

indicators, e.g. cultural or demographic conditions.

Users should be aware of this shortcoming and

consider the outcome as a first step to find suitable

locations.

Figure 3: Example of a presentation of the result of

Global-Detector (case ‘Potential for Tilapia in Kenya and

surrounding countries’; dark red is good potential).

2.1 Indicators

The raw data for most indicators of Global-Detector

are downloaded open source data from the web. The

format of these freely available datasets differs:

 Gridded GIS-files with rasters in different

formats, e.g. WorldClim (2016) for global

climate, bio-physical datasets like SoilGrids

(ISRIC, 2016), or datasets with agronomic

parameters related to many crops (e.g.

FAOSTAT, 2016);

 GIS-shapefiles, e.g. a world map of rivers and

lakes;

 Tabular files with data in csv or Excel files on

country or provincial level, e.g. World

Development Indicators from World Bank

(2016).

After downloading, the data are transformed by GIS-

tools to the required 5’x5’ resolution. Tabular files

with data on country level are also gridded to the

same 5’x5’ format, in this way all grid cells of a

country have the same value. This resolution is

chosen when considering calculation time and

specificity, and many gridded sources have this

resolution. Change of resolution has either

implication for calculation time with hardly

additional information (e.g. to 1’x1’) or is not

specific enough at regional level (e.g. 30’x30’).

Not all indicators of Global-Detector are

transformed directly from open source data. Some

have been derived. For example, the indicator

‘vicinity from harbours’ has been created by using a

list of worldwide harbours and their importance, and

a specific piece of software code draws zones

around a harbour with different levels of distance. In

this way hinterlands of a harbour are incorporated

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

162

(e.g. part of Germany as hinterland for the harbours

in the Netherlands and Belgium).

To summarize, Global-Detector contains

indicators for:

 Climate, e.g. monthly temperature, solar

irradiation, precipitation;

 Infrastructure, e.g. distance to markets, vicinity

to harbours and airports;

 Land characteristics, e.g. type of soil, slope;

 Land utilisation, e.g. % cropland, area of various

crops, protected areas, N/P2O5 fertilisation;

 Population density and number of people in

radius of e.g. 20 or 250 km;

 Country level indicators retrieved from the

World Bank, FAO and other sources;

 Miscellaneous indicators, e.g. nightlight,

religion.

If an expert requires indicators that are not present in

Global-Detector but are available on the web or can

be gathered from another source, then data can be

transformed to a raster with the required template for

Global-Detector. For example, data of the World

bank and FAO have a large number of variables on

country level, each can be transformed to a raster for

Global-Detector. If an expert requires his own

indicator (e.g. export regulations for a specific

product), then he fills in a table with values for each

country. This is sequentially transformed to a

worldwide 5’x5’ indicator.

For a country, a group of countries, or a region

within a country, additional spatial data can be

brought in the model. An example is the use of an

additional map of railway infrastructure and its

influence (figure 6).

2.2 Knowledge-based Model and

Interaction with Experts

A knowledge-based model is an essential part when

Global-Detector is applied. It combines indicators

and arithmetic rules to yield the requested output,

e.g. potential for a product. From a base collection

of nearly two hundred indicators, the expert chooses

indicators that are relevant for the case of interest.

For these indicators on-the-fly maps are created by

the model builder for the specified area or for an

area that should be validated. It is indispensable that

the model builder and the expert work together in

close cooperation to develop a knowledge-based

model that contains algorithms to transform

indicator values to scores. This can be done, for

example, by developing a function that converts

ranges for the minimum, optimum and maximum

temperature to the scores of the temperature

suitability map. The expert determines parameter

values and helps the knowledge engineer (i.e. model

builder) with the construction of the arithmetic rules

for the combination of indicators. The resulting map

is shown to the expert, and after this first step, face

validity of this map can lead to the adaptation of

parameters and formulas. It is up to the expert to

gather and use additional information from literature

or gain information from other experts for the

purpose of this validation procedure. If there is a

theoretic background, the expert can use that

information when the knowledge-base is specified.

3 THEORETICAL

BACKGROUND AND SETTING

3.1 Model Perspective

Global-Detector can among others be used as a

suitability model aimed at assessing the potential for

different agricultural production systems, and is

intended to bridge the gap between standard

agronomic (bio-physical) crop suitability models

that provide output at grid-level on the one hand, for

example Ecocrop model (EcoCrop, 2016), GAEZ

(FAO, 2016), and on the other hand a set of global

models that focus on the agricultural sector in

general (including socio-economic and market data)

which provide output at level of administrative units

(e.g. MAGNET: see Woltjer and Kuiper, 2014).

Global-Detector uses expert knowledge to

include the assumed impact of (downscaled) socio-

economic and infrastructural variables on the

suitability of different grid cells for different

production systems. This is not restricted to crop

production systems, but can also be applied to

(mixed-) livestock, aquaculture systems, demand for

products, etc. Models that combine high-resolution

spatial data related to both environmental and socio-

economic data already exist, for example

EUClueScanner (Koomen et al., 2010; Object

Vision, 2016) and CLUMondo (Van Asselen and

Verburg, 2012). However, to our knowledge no

other model combines these different types of data at

a global scale and using expert knowledge to weigh

relative impacts, and to combine and transform data

(e.g. from temperature maps to a map showing the

deviation from optimum temperature range in

tropical regions). The EUClueScanner, however,

already uses data from other models including

IMAGE (Kram and Stehfest, 2011; IMAGE, 2016)

and MAGNET. But the prevailing difference at the

Global-Detector - GIS- and Knowledge-based Tool for a Global Detection of the Potential for Production, Supply and Demand

163

moment is the flexibility of Global-Detector that

allows the expert, within a few hours, to include a

subset of several hundreds of variables that are

relevant to make a certain area suitable for the

desired production system, based on the subjective

assessment principle (i.e. expert knowledge).

Models that provide information about the

suitability for different agricultural production

systems at the grid-level can give both policy

makers and the commercial sector a quick scan of

the expected local differences. Inclusion of climatic,

bio-physical and socio-economic variables can help

show the differences within a country (e.g.

suitability for local to local production) or assess the

impact of changes to any of these variables. Existing

models that focus on the suitability of certain

agricultural production systems generally focus on

crop production systems and the climatic and

biophysical conditions that are required for certain

crops to grow optimally.

Three main approaches to suitability analyses

exist: the limiting condition principle, the principle

of arithmetic modelling and the subjective

assessment principle (FAO, 2016a). When using the

limiting condition, the least suitable variable

determines the overall suitability. With the principle

of arithmetic modelling, variables will be assigned a

certain value, which can be operated arithmetically

to determine the final suitability. The subjective

assessment principle (‘expert model’) allows for

flexible selection and judgement over the

importance of the different variables; the variables

can be weighted and operated arithmetically to get to

a final suitability map. The latter approach requires

an expert opinion and can combine any of the

relevant aspects of the other two types of models.

Global-Detector applies the subjected assessment

principle; the expert system is the core of an

application with this model.

Sweeney et al. (2015) present and compare in a

recent review 34 journal articles and 70 web-

mapping projects that cover various aspects of food

mapping research and initiatives that have been used

to “explore complex social, economic, and

environmental components of the food system”.

Neither of these models resembles the concept of

Global-Detector, nor are we aware of another

existing comparable world-wide generic and flexible

knowledge- and GIS-based model that can be used

interactively together with an expert.

3.2 Market and Business Perspective

According to the classical location theory,

entrepreneurs try to minimalize costs in their quest

to find appropriate locations to invest (Von Thünen,

1960). Benninga et al. (2016) give an overview of

this theory and other theories related to international

potential for production. Optimal locations have low

transport costs, proximity to raw materials and

energy, situated near markets (or harbours/airports

for export), availability of labour, etc.. Intensive

vegetable and fruit production - especially for the

fresh market (i.e. short chain production) - is best to

be situated near urban region, whereas extensive

agriculture and forestry activity is eligible to be in

the rural areas (after Von Thünen, 1960). Porter

(1990) proposes a strategic diamond model bases on

competiveness to assess the attractiveness of a

country or region, based on competiveness.

Global-Detector is able to account for factors

that are related to markets and business climate, e.g.

the ease of doing business (World Bank), the

availability and quality of infrastructure,

identification of urban and rural areas, GDP and

income inequality, and the fragility of states.

4 CASE STUDIES AND

APPLICATIONS

The availability of indicator maps and expert

knowledge is required for any application of Global-

Detector. It can be applied for various information

needs or research questions. Some of the

possibilities (with examples, see later) are:

 Detection of the potential for production, e.g.

Tilapia in ponds, ornamental horticulture, pig

production;

 The expected demand for a product, e.g.

consumption of cherry tomatoes, consumption of

milk;

 Local-for-local production, e.g. nursery (potted

plants) for the local markets, short chain fresh

vegetable production;

 Scenarios that might have an effect on the

production or demand in a region, e.g. drought,

flooding.

At a workshop a group of experts can be informed

about the circumstances in the area at issue by

showing several relevant indicator maps produced

by Global-Detector, followed by interactively

gaining region intelligence from these maps

supplemented by regional knowledge from the

participants.

Global-Detector is applied for the development

of demonstration cases and used in projects. Most

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

164

applications concern production potential. Other

types of applications, like the demand for a product,

are harder to accomplish and require additional

assumptions because of a lack of worldwide data

(e.g. difference between urban and rural

consumption).

The first case will be explained in detail whereas

the other cases will be described only briefly.

4.1 Production Potential

Potential for Ornamental Horticulture (Worldwide).

This was the research question that had led to the

development of Global-Detector and is its first

application. The steps that were taken illustrate the

way Global-Detector has been applied for this case

and to understand the development of the other

cases. A base for this case has been a country-based

model in Excel to map the global potential of

floriculture production (Benninga et al., 2016). The

steps used to apply Global-Detector for this case,

and in general for other cases, are:

 Commitment of experts for acquiring knowledge

and validation. This case consisted of two expert

groups: (1) cuttings and young plants, and (2)

flowering and bedding plants;

 Selection of indicators from Global-Detector.

These indicators are subdivided in following

groups: climate, land and soil, market &

infrastructure, and country level (e.g. fragile state

index);

 In workshops the knowledge-base of Global-

Detector is developed, i.e. creating arithmetic

formulas for combining and weighing indicators,

and setting parameters for these formulas, e.g.

the optimal temperature. The workshop is

interactive, experts can react instantaneous and

suggest adaptations of the parameters;

 Validating maps generated by the knowledge-

base. An example is the suitability of climate -

i.e. the combination of temperature, water

supply, humidity, and solar irradiation - for the

production of flowering and bedding plants

(figure 4);

 Adaptation of the set of indicators for the

knowledge-base;

 Creating the final potential map for a specific

region (e.g. figure 5) or for the whole world

(figure 2).

For the following cases these steps may differ, e.g.

bilateral sessions instead of workshops.

Figure 4: Climate suitability for flowering and beddings

plants in East Asia; dark green is good potential.

Figure 5: Potential for flowering and beddings plants in

Ethiopia; red is reasonably good potential.

Most relevant indicators for the expert model

that was developed based on expert judgement were

solar radiation, temperature and the infrastructure

indicator ‘distance to main airports’. Furthermore, a

set of indicators from World Bank (or related

sources) were applied (on country level). Results

from the 5’x5’ grids had been summarized at

province level.

Global-Detector - GIS- and Knowledge-based Tool for a Global Detection of the Potential for Production, Supply and Demand

165

Potential for Tilapia in Kenya and Surrounding

Countries.

To proof the applicability of Global-Detector was

the development of a map of Kenya and

surroundings to indicate the best pond locations for

Tilapia. In three sessions with an expert, who has

also regional knowledge, a knowledge-base was

developed. Results depicted on Google maps were

validated by this expert (figure 3). After a positive

validation, the model can have opportunities for a

worldwide applicability - presumably with some

modifications for non-tropic regions. The Tilapia

case was valuable since it gained experience in

interacting with an expert and the way validation

was done.

Potential for Avocado Production in Ethiopia.

For a project initiated by the Dutch organization

‘GroentenFruit Huis’, Global-Detector was applied

to assess the best areas for avocado production in

Ethiopia when accounting for the infrastructure.

New maps (i.e. indicators) had to be developed to

indicate the location of (potential) railways and

stations (figure 6). This has shown the flexibility of

Global-Detector to make and use additional specific

indicators. Soil characteristics from SoilGrids, like

pH, organic content, clay and drainage, had also

been used as additional indicators.

Figure 6: Scope of railways and stations in Ethiopia

(additional Global-Detector indicator for one country).

Potential for Tomato Production (Worldwide).

The Dutch company Prominent is interested in the

best locations for tomato production worldwide.

Entrepreneurs can be aided in their quest for

locations by means of the results of the model. In a

workshop indicators were identified and knowledge

was acquired on how to combine indicators. For

example, the combination of the maximum and the

minimum temperature in the hottest 3 months

accounting for optimal temperature combined with

enough difference between minimum and maximum

temperature.

Potential for Pig Meat Production in South-East

Asia.

Together with an expert on pig meat production, a

knowledge base was developed to assess the

potential for pig meat production in South-East Asia.

Besides indicators on 5’x5’ and country level, an

additional level was applied. Province level

indicators maps for the main religions had to be

developed (from tabular base data from Johnson and

Grim, 2008) and applied. The value of this case for

Global-Detector is the flexibility to incorporate also

a province level for regional information.

4.2 Demand (Consumption)

Demand of Cherry Tomatoes in Africa and Europe.

This application was developed to show that Global-

Detector is also suitable for the demand, so not just

for production (supply). For Africa and Europe the

indicators and a knowledge base was used to

estimate the amount of expected cherry tomato

consumption. FAO data on imports of (luxury)

vegetables had been transformed to an indicator;

seven products were weighted and combined and

divided by the population to yield the consumption

by import per person. The estimation of the total

consumption per grid cell is done by a combination

of expected % of consumers, the availability (based

on infrastructure) and the expected consumption per

capita.

4.3 Classification for Decision Making

Agro-ecological Potential and Climate Vulnerability

in Mali and Burkina Faso.

Global-Detector was applied to map the agro-

ecological potential and climate vulnerability of

Mali and Burkina Faso as a case study. This pilot

project aimed to test the possibilities for enabling

strategic discussion on future climate smart Food

and Nutrition Security interventions in Sub Saharan

Africa, by identifying areas with high/low food

system dynamics and high/low climate dynamics.

These classifications have been presented on a map

and discussed with experts in a workshop.

Classification of Supply and Demand of Onions.

For a project initiated by the Dutch organization

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

166

‘GroentenFruit Huis’ Global-Detector was applied to

downscale information from country level (e.g. data

from World Bank and FAOSTAT) to grid level. The

result is a map with grids classified as ‘import’,

‘export’ and ‘local’. The distinction between urban

and rural areas and the balance between supply and

demand led to these classifications.

4.4 Global-Detector to Analyse

Metropolitan Assignments

Detection of Metropolitan Land use Options

(Metropolitan Solutions).

In the project Metropolitan Solutions one of the aims

is to investigate the options for land use in

metropolitan regions, accounting for the demand

(consumption) and supply (production) of food and

recreation. In an expert model, indicators of Global-

Detectors were used to assign regions for urban

recreation, rural recreation, short chain fresh food

production, intensive meat production, vegetables

and fruit production, arable production and grazing

areas. Effects of options - like urban population

growth and flood by sea - on the supply and demand

of food and recreation had been calculated. For the

demand of food so-called ‘Food Metres’ were used,

indicating the number of required hectares per 1000

people (Wascher et al., 2015). Currently another

model is being developed to specify ‘Food Metres’

or the ‘consumptive food print’ for all countries in

the world, and thus having a country specific ‘Food

Metres’.

Classification of Cities Worldwide.

Another application within the project Metropolitan

Solutions is the classification of nearly 850 of the

worlds’ largest cities by making use of Global-

Detector. Each city has to be characterized by a few

dozen aspects like the risk of flooding, the degree of

urbanisation, infrastructure, available urban

recreation and urban agriculture, possibilities to

expand, etc.. By means of cluster analyses or with

software programmes that search for comparable

cities, e.g. the FaceIT tool based on genetic

algorithms (Hennen, 2009), groups of cities that

share common characteristics can be identified.

5 DISCUSSION AND

CONCLUSION

The wide variety of applications in the previous

chapter shows that Global-Detector is a generic tool

for knowledge-based spatial analysis. Since the large

set of indicators is readily available for use without

any GIS-processing, the model builder together with

the expert can start building the knowledge base for

the concerning research aim instantaneously. In this

way the result is reached considerable faster and

more efficient compared to a customary approach

where often tedious and time-consuming data

tracing, collection and GIS-processing have to take

place before the actual spatial analyses can start.

Due to the readily availability of the indicators

(maps) in Global-Detector, the spatial analyst (i.e.

knowledge engineer together with expert) can

instantly work on the knowledge level to create an

application.

The knowledge base is imperative in any

application built with Global-Detector, this makes

the expert an unavoidable agent. After all, spatial

analysis is a knowledge driven process when the

algorithms and scientific methods are not explicitly

available, and merely exist as ‘tacit knowledge’ in

the head of an expert. Experts are also crucial to

validate the outcome and to communicate its worth

and shortcomings.

An application developed with Global-Detector

can only make use of the available indicators.

Despite a considerable amount of indicators there

might be factors for which no indicators exist in the

stack of Global-Detector. For example cultural or

region-specific aspects, or regional knowledge that a

river carries no water during some months in the

year. Results should therefore be considered as a

‘quick scan’ or as a first step that has to be followed

by more detailed analyses with additional (regional)

data or models. Global-Detector’s resolution of

5’x5’ might also be too coarse grained when small

regions are subject of an investigation. Great care is

needed when the individual grids contain a large

variety of landscape elements.

Global-Detector is valuable when applied in

conjunction with existing models, e.g. with the

MAGNET model (Woltjer and Kuiper, 2014). In this

way Global-Detector can benefit from additional

specialized data, and an existing model can use

Global-Detector‘s downscaling possibilities, and so

increase the value of both models (Bartelings and

Hennen, 2016). Promising efforts in this direction

have been made.

The initial goal of Global-Detector and the

reason why it has been developed, - i.e.

entrepreneurs want to be aided in their search for

promising locations in the world to expand their

production or market possibilities abroad - is

attained since locations can be spotted, analysed and

Global-Detector - GIS- and Knowledge-based Tool for a Global Detection of the Potential for Production, Supply and Demand

167

compared with the model. Examples are the

potential for ornamental horticulture and tomatoes.

The horticultural sector has pronounced their interest

to use it for other products. As a generic tool with

increased flexibility, Global-Detector will have

many application possibilities in a wide variety of

projects.

REFERENCES

Bartelings, H., Hennen, W., 2016. MAGNET and Global-

Detector. LEI Wageningen UR, internal project

document.

Benninga, J., Hennen, W., Dijkxhoorn, Y., Galen, M. van,

2016. Internationaal Productie Potentie Model

Sierteelt, modelbeschrijving. LEI Wageningen UR,

report in press.

EcoCrop, 2016. Expert cross checking and validation of

crop niche predictions. URL:

http://gisweb.ciat.cgiar.org/ClimateChange/EcoCropF

B/ (Retrieved: 11/16/2016).

EarthStat, 2016. EarthStat data download. URL:

http://www.earthstat.org/data-download/ (Retrieved:

11/21/2016).

FAO, 2016. GAEZ - Global Agro-Ecological Zones. URL:

http://www.fao.org/nr/gaez/en/ (Retrieved:

11/16/2016).

FAO, 2016a. Land evaluation for development. URL:

https://www.mpl.ird.fr/crea/taller-

colombia/FAO/AGLL/pdfdocs/landev.pdf (Retrieved:

11/16/2016).

FAOSTAT, 2016. FAOSTAT Statistics division. URL:

http://faostat3.fao.org/home/E (Retrieved:

11/21/2016).

Hennen, W.H.G.J., 2009. Comparing companies and

strategies: a Genetic Algorithms approach. JIAC2009.

July 6-8 2009, Wageningen.

IMAGE, 2106. IMAGE : Integrated Model to Assess the

Global Environment. URL:

http://themasites.pbl.nl/models/image/index.php/IMA

GE_framework (Retrieved: 11/16/2016).

ISRIC, 2016. Soilgrids. ISRIC World Soil Information.

URL: http://www.isric.org/content/soilgrids

(Retrieved: 11/19/2016).

Johnson, T.M., Grim, B.J., 2008. Eds. ‘World Religion

Database’, Leiden/Boston, Brill, 2008. See

http://www.worldreligiondatabase.org .

Koomen, E., Diogo, V., Hilferink, M., Beek, M.C.J. van

der, 2010. EU-ClueScanner100m; model description

and validation results. (Internal report). Amsterdam:

VU Amsterdam.

Kram, T., Stehfest, E., 2011. Integrated Modeling of

Global Environmental Change. In: Land use, climate

change and biodiversity modeling: perspectives and

applications. Trisurat, Y., Shrestha, R.P. and

Alkemade, R. (Eds.), pp.104-118.

Kuhlman, T., Weegh, J. op de, 2014. Meloenen in Afrika;

Of: hoe men met behulp van GIS-data potentiële

teeltgebieden voor een gewas kan bepalen. LEI,

Wageningen UR.

Object Vision, 2016. EUClueScanner. URL:

http://www.objectvision.nl/gallery/products/eucluesca

nner (Retrieved: 11/16/2016).

Porter, M.E., 1990. The competitive advantage of nations.

New York: Free Press.

Sweeney, G., Hand, M., Kaiser, M. , Clark, J.K., Rogers,

C., Spees, C., 2015. The State of Food Mapping

Academic Literature Since 2008 and Review of Online

GIS-based Food Mapping Resources. Journal of

Planning Literature, 31 (2), 123-219.

Van Asselen, S., Verburg, P.H., 2012. A land system

representation for global assessments and land-use

modeling. Global Change Biology, 18, 3125-3148.

http://dx.doi.org/10.1111/j.1365-2486.2012.02759.x

Von Thünen, J.H., 1930. Der isolierte Staat in Beziehung

auf Landwirthschaft und Nationalokonomie. 3 vols.

Jena (Germany): Fischer, 1930.

Woltjer, G.B., Kuiper, M.H., 2014. The MAGNET Model:

Module description. Wageningen, LEI Wageningen

UR (University & Research centre), LEI Report 14-

057. 146 pp.; 18 fig.; 45 tab.; 39 ref.

Wascher, D.M., Piorr, A., Pintar, M., Kneafsey, M., Sali,

G., Corsi, S.L., Mbatia, T., Jeurissen, L.J.J.,

Arciniegas, G., Glavan, M., Doernberg, A., Zasada, I.,

Groot, J., Schmutz, U., Bos, E., Venn, L., Monaco, F.,

Simiyu, R., Owour, S., Asselt, E.D. van, Fels, H.J. van

der, Eupen, M. van, 2015. FOODMETRES –

Metropolitan food planning connecting the local with

the global. Urban Agriculture Magazine (2015)29. -

ISSN 1571-6244 - p. 41 - 44.

World Bank, 2016. World Bank : World Development

Indicators. http://data.worldbank.org/products/wdi

Worldclim, 2016. WorldClim - Global Climate Data.

URL: http://www.worldclim.org/ (Retrieved:

11/19/2016).

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

168