Analysing settlement dynamics using statistics based on archaeological theory

0512035 - ARÚ 2020 RIV PL eng A - Abstract
Demján, Peter
Analysing settlement dynamics using statistics based on archaeological theory.
Check Object Integrity. Book of abstracts. CAA 2019 Kraków. Kraków: Institute of Archaeology of Jagiellonian University, 2019. s. 75. ISBN 978-83-948382-7-0.
[Computer applications and quantitative methods in archaeology /47./. 23.04.2019-27.04.2019, Kraków]
R&D Projects: GA MŠMT EF16_019/0000728
Keywords : digital humanities * mathematical modeling * radiocarbon dating * settlement area theory * settlement dynamics * spatio-temporal statistics
OECD category: Archaeology
https://2019.caaconference.org/wp-content/uploads/sites/25/2019/04/CAA2019_programabstracts_v20190423.pdf

The mechanisms of settlement and land use in prehistory can be examined by testing the hypothesis that variations in spatio-temporal distribution of archaeological evidence of settlements mirror changes of the actual settlement patterns. To test this, we need a mathematical model describing the relation between the quantity of evidence and intensity of settlement activities which produced it. This model has to take into account all of the following uncertainties inherent to archaeological data: (1) The actual time and place of past events is not known and lies within the boundaries given by dating and localization of the evidence. (2) The events took place in areas which had a certain extent in time and space. (3) The observed distribution of evidence is affected by variations in feature visibility, sampling intensity and accuracy. We can test the null hypothesis that changes in spatio-temporal distribution of archaeological evidence can be explained by fluctuations stemming from these uncertainties. If we assume that an archaeologically detected site of settlement activity (as defined by the settlement area theory) represents a spatio-temporal volume, rather than a single point in time and space, we need a probabilistic method, which considers spatial and temporal dimensions equally. We propose the evidence density estimation (EDE) function, which can be applied on radiocarbon-dated evidence as well as typo-chronologically dated evidence to produce spatio-temporal distribution maps or summed distributions representing intensity of settlement activities within the examined area.
Permanent Link: http://hdl.handle.net/11104/0302362