Item: Multiple linear regression modeling and the correlation between surface area, snow density and burial depth of improvised snow protection
Title: Multiple linear regression modeling and the correlation between surface area, snow density and burial depth of improvised snow protection
Proceedings: Proceedings, 2012 International Snow Science Workshop, Anchorage, Alaska
Authors: Salvatore G. Candela , Eeva Latosuo M.S.
Abstract: While there is a growing body of research on snow anchors, there has never been a published study on the use of improvised objects as snow protection. In this study the potential strength of commonly utilized improvised objects (snowshoe and water bottle) were tested by exploring the relationship between three variables that can be reliably measured; snow density, burial depth of the object and effective surface area. In field tests (n= 13) of the above objects in t-slot style snow protection, observed peak forces at failure ranged from 950𝑁 to 8100𝑁, with an average of 4100𝑁. A multiple linear regression model (𝑅!= 0.64, df= 12, p= 0.02) shows that effective surface area was the most significant variable, (t= 2.32, p= 0.02) followed by snow density and burial depth. This preliminary model, which will be updated with additional data collection in a follow-up study, can be used to approximate the peak failure force for improvised snow protection and to provide a measure for judging the selection and use of improvised systems.
Keywords: improvised protection, snow density, static forces, regression modeling, peak failure, snow anchors
Digital Abstract Not Available