Dependence of lake ice covers on climatic, geographic and bathymetric variables

Abstract

Information on lake ice characteristics is of interest to lake users, natural resource managers and researchers. Lake ice data for 143 North American freshwater lakes were assembled and analyzed. The principal data were observed ice-in dates, observed ice-out dates and observed ice thicknesses. Single variable linear regression and factor analysis were applied to find correlations with air temperature, lake morphometry (mean depth and surface area), latitude and topographic elevation. Strong correlations between lake ice parameters, latitude and mean air temperature were found and quantified. Lake morphometry exhibited consistent but less significant correlation to ice cover. Elevation showed varying degrees of correlation with the dependent variables. Multivariable regression analysis produced equations for ice-in date, ice-out date, ice-cover duration and maximum ice thickness. Despite low standard errors, the best regressions cannot account for the full range of annual variability in the historical records, implying the absence of influential factors from these regression models. Based on sensitivities gleaned from linear regression analysis, the effects of climate change on lake ice characteristics may also be estimated. The following are estimated shifts based on a 1 °C rise in average air temperature: ice-in date occurs ∼5 days later, ice-out date occurs ∼6 days earlier, ice cover duration (ICD) is reduced by ∼11 days, and maximum ice thickness is reduced by ∼7 cm. These values are on the order of those presented in prior studies using other methodologies and lakes.