Voorspellen van het Verleden: GIS en Predictive Modelling'

1
Voorspellen van het Verleden GIS en Predictive
Modelling.

• Hans Kamermans
• Universiteit Leiden
• 13-04-2005

2
Predictive Modelling

• Predictive Modelling is a technique to predict,
  at a minimum, the location of archaeological
  sites or materials in a region, based either on
  the observed pattern in a sample or on
  assumptions about human behaviour.
• (Kohler Parker 1986 400)

3
Reasons to apply predictive modelling in
archaeology

• To gain insight into former human behaviour in
  the landscape an academic research application.
• To predict archaeological site location to guide
  future developments in the modern landscape an
  archaeological heritage management application.

4
Valletta (Malta)
5
  European Treaty Series - No. 143           EUROP
EAN CONVENTION ON THE PROTECTION OF THE
ARCHAELOGICAL HERITAGE (REVISED)   Valletta,
16.I.1992

6

• Article 1
• The aim of this (revised) Convention is to
  protect the archaeological heritage as a source
  of the European collective memory and as an
  instrument for historical and scientific study.

7

• Article 5
• Each Party undertakes
• to seek to reconcile and combine the respective
  requirements of archaeology and development plans
  by ensuring that archaeologists participate

8
Article 5

• in planning policies designed to ensure
  well-balanced strategies for the protection,
  conservation and enhancement of sites of
  archaeological interest
• in the various stages of development schemes

9
Indicative Map of Archaeological Values of the
Netherlands (IKAW)
10
Our goals

• A thorough analysis of the various models and
  methods used in current predictive modelling
  practice,
• The exploration of possibilities for
  methodological improvement, and
• The formulation of recommendations for the
  Netherlands Handbook of Archaeological Quality
  Norms.

11
(No Transcript)
12
"When you can measure what you are speaking about
and express it in numbers, you know something
about it." Lecture to the Institution of Civil
Engineers, 3 May 1883 Lord Kelvin (William
Thomson)
13
(No Transcript)
14
IKAW

• Correlatie archeologische vindplaatsen met
  variabelen van het natuurlijke landschap
• Aangevuld met expert opinion
• Geeft kaart met indicatie waar archeologie in de
  bodem zit

15
IKAW

• Correlatie archeologische vindplaatsen met
  variabelen van het natuurlijke landschap
• Kritiek t.a.v. de gebruikte variabelen
• Hoe heeft het natuurlijke landschap de locatie
  keuze van mensen in het verleden beïnvloed?
• Zijn er nog andere factoren te bedenken?
  Culturele misschien?
• En zou die factoren niet verschillen per
  archeologische periode? (Neolithicum versus
  Middeleeuwen)

16
IKAW

• Correlatie archeologische vindplaatsen met
  variabelen van het natuurlijke landschap
• Kritiek t.a.v. de techniek
• Is de correlatie getest? Vergeleken met een
  willekeurige verspreiding?
• Is er gekeken of er een causaal verband is?

17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
IKAW

• Correlatie archeologische vindplaatsen met
  variabelen van het natuurlijke landschap
• Aangevuld met expert opinion
• Kritiek
• Wat is expert opinion? Intuïtie?

21
IKAW

• Correlatie archeologische vindplaatsen met
  variabelen van het natuurlijke landschap
• Aangevuld met expert opinion
• Geeft kaart met indicatie waar archeologie in de
  bodem zit
• Kritiek
• Klopt het een beetje?

22
Perspectives

• Archaeological heritage management perspective
  predictive modelling is a cost-effective
  alternative to large-scale archaeological
  prospection
• Academic research perspective predictive
  modelling should not be used for archaeological
  heritage management

23
Brandt et al. (1992 276)

• Given the numerous difficulties and data
  limitations that confront archaeological model
  development in the Netherlands, together with
  many simplifications and assumptions that one
  must make, it is imperative that sufficient tests
  be performed before any serious consideration of
  a model is undertaken.

24
David Wheatley (2003)

• However, it has been argued above that
  correlative predictive modelling does not
  actually work very well and, more significantly,
  will lead to an increasingly unrepresentative
  archaeological record. If resource management
  requires a methodology that does work and will
  lead to a more representative record, then it
  follows from this that archaeology would be
  better served by a focus on well-designed and
  properly implemented sampling strategies, rather
  than correlative predictive models.

25
Find more sitesin high probability areas
Biased database
Look more carefully in high probability areas
Biased model
Wheatley 2003
26
Thomas Whitley (2004)

• In many cases it is too costly or even
  impossible to do an correlative predictive model
  and ultimately the resulting model does not
  provide better insight into site placement
  processes than intuition.

27
Conclusie

• The models should not reach land managing
  officials and certainly not the planners. Their
  only role should be in an initial phase, to aid
  archaeologists to stratify an area in order to
  plan various forms of archaeological prospection
  on the basis of a good sampling design.

28
Malta Magazine April 2005