Open AccessTesting least cost path (LCP) models for travel time and kilocalorie expenditure: Implications for landscape genomics
Author(s) -
Kyle Gowen,
Timothy S. de Smet
Publication year - 2020
Publication title -
plos one
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.99
H-Index - 332
ISSN - 1932-6203
DOI - 10.1371/journal.pone.0239387
Subject(s) - terrain , sample (material) , statistics , computer science , environmental science , ecology , physical geography , econometrics , mathematics , geography , biology , physics , thermodynamics
Least Cost Path (LCP) analysis allows a user to define a cost parameter through which cost of movement can be assessed using Geographical Information Systems (GIS). These analyses are commonly used to construct theoretical movement through a landscape, which has been useful for creating hypotheses concerning prehistoric archaeology and landscape genomics. However, LCP analysis is commonly employed without testing the generated LCP(s), complicating its usefulness as a methodological tool. This paper proposes a model for analyzing movement in ArcGIS by using topography data to calculate slope. This slope data is then then used to calculate LCPs based on travel time and kilocalorie expenditure. LCPs were constructed in the Nature Preserve at Binghamton University, a 182-acre area that consists of wetland and mountainous terrain, and a Fitbit® Surge activity monitor was used to test the accuracy of the model’s predictions. Paired sample t-tests show a lack of significant difference between calculated and walked time in our analysis (p = .953), suggesting that our model can estimate travel time between two points based solely on slope of the region. Paired sample t-tests also show a lack of significant difference between calculated and observed kilocalorie expenditure (p = .930), suggesting that our model is also capable of estimating kilocalorie expenditure associated with movement between two points. Finally, paired sample t-tests confirm that straight line distances do not reflect real movement through a terrain (p = .009), highlighting the need for alternate measures of movement when analyzing the effects of local landscape on movement. Our current model shows strength in its estimations of travel time and kilocalorie expenditure based on topography of a region–future iterations of the model need to establish the statistical similarity between our model’s estimations and recorded values for walking time and kilocalorie expenditure.