Surface sediment diatom assemblages from 51 Alaskan lakes, distributed along a north-south transect, were enumerated in order to develop transfer functions that could be used to make inferences of past climatic and environmental change. Environmental variables that were found to be the strongest predictors of the diatom assemblages, identified through Canonical Correspondence Analyses, were ionic and nutrient concentrations, as well as lake depth. A number of weighted-averaging (WA) regression-calibration techniques were employed to develop transfer functions for lake-water conductivity, total phosphorus concentration, and lake depth, but simple WA with classical deshrinking produced the most robust models for all variables. The strength of these models (r2boot ranged between 0.52 and 0.53) is comparable to those generated from other northern calibration studies. Application of our models to fossil diatom assemblages could provide information on the magnitude of past environmental change, and may serve as a basis for assessing anthropogenically induced impacts.