Year of Award


Document Type


Degree Type

Master of Science (MS)

Degree Name

Health and Human Performance (Exercise Science Option)

Department or School/College

College of Education and Human Sciences

Committee Chair

Matt Bundle

Commitee Members

Matt Bundle, Chuck Dumke, Brent Ruby, Rich Willy


Load carriage, energy expenditure, pandolf, wildland firefighting, pack test, hiking


University of Montana

Subject Categories

Life Sciences | Medicine and Health Sciences


Despite extensive and ongoing scientific study into the metabolic requirements of load carriage, an understanding quantifying the effect of speed, load, sex and body mass has yet to come forth and the extent to which established models predict these requirements is largely untested. Specifically, because existing experimental efforts have typically focused on relatively modest walking speeds using loads representing a fixed portion of the subject’s mass, extending the available predictions to applications where individuals complete a common task carrying an identical absolute load provides estimates of unknown accuracy. PURPOSE: Here, we measured the energy use in a large subject group walking at speeds surrounding the 1.8 m s-1 necessary to successfully complete the 4.83 km USFS wildland firefighter arduous pack (20.5kg) test, and compared these results to estimates available from the prevailing models. METHODS: We measured VO2 from 61 young (age = 22.8±3.2 yrs) adults (36 males; 25 females; study range: Mb = 55.4-119.6 kg; height = 1.52-1.93 m) as they performed four, 5min trials, with a 20.5kg pack, on a level treadmill at 1.7, 1.8, 1.9 m s-1, and their individual average speed from a previously administered pack test. In addition, a subset of n=10 subjects were equipped with Douglas bags during the simulated pack test to measure steady state VO2. We used the methods of Pandolf et al. 1977 and Ludlow & Weyand 2017 to generate VO2 estimates for the individual trials we administered. RESULTS: Measured values of VO2 increased from 22.4±3.2 and 24.6±4.1 ml kg-1 min-1 at 1.7 m s-1,to 31.6±5.3 and 31.0±4.5 ml kg-1 min-1 at the fastest speed administered for males and females respectively. In contrast, the accuracy of the predictive models decreased with speed and yielded prediction errors of -12.4 and -22.9% at 1.7 m s-1 for the Pandolf and Ludlow & Weyand methods respectively, these errors were -18.0 and -32.2% at the fastest speeds administered. When evaluated at the speed subjects used in the field trial, the prediction models underestimated energy expenditure by 5.0±4.4 and 10.4±4.9 mlO2·kg-1·min-1 respectively. CONCLUSION: We conclude that existing predictive models do not retain their accuracy, and substantially underestimate measured values when applied to a group of male and female subjects undertaking relatively fast walking speeds on a flat surface with a heavy load.



© Copyright 2018 Jeffrey T. Strang, Christopher J. Alfiero, Charles Dumke, Brent Ruby, and Matthew Bundle