Probabilistic Modelling for Incorporating Uncertainty in Least Cost Path Results: a Postdictive Roman Road Case Study

Abstract The movement of past peoples in the landscape has been studied extensively through use least cost path (LCP) analysis. Although methodological issues applying LCP analysis archaeology have frequently discussed, effect DEM error on results not fully assessed. Due to this, reliability result is undermined, jeopardising how well method can confidently be used model movement. To strengthen results, this research proposes Monte Carlo simulation as a for incorporating and propagating effects results. Focusing vertical error, random fields are calculated incorporated into documented reproducible modelling process using R package leastcostpath . By graphically communicating impact probabilistic LCPs, uncertainty taken account when interpreting LCPs. applied Roman road case study, finding that incorporation identification multiple ‘least cost’ routes within landscape. Furthermore, deviation between roman suggests location was influenced by additional factors other than minimising energy expenditure. This finds derived viable graphical communication caused input data process. Therefore, it recommended LCPs become default approach contains errors.