Studies of periglacial regions confirm their importance in the global carbon (C) cycle,but estimates of e.g. soil organic carbon (SOC) storage are poorly constrained and lack quantitativeestimates of errors following upscaling. In this study, a comprehensive regional SOC database from thenorthern Usa River Basin (European Russian Arctic, 55 000 km2) is used to evaluate the currentmethodology of SOC upscaling in periglacial terrain. The selection of classes for upscaling and the need forreplication in soil sampling are statistically evaluated. Upscaling using a land cover classification and a soilmap estimates SOC storage at 48.5 and 47.0 kg C m-2, respectively with 95% confidence intervals (CI)within ±8%. When corrected for spatial errors in the upscaling proxy, SOC is estimated to 46.5 kg C m-2with a 95% CI reflecting propagated variance from both natural variability and spatial errors of ±11%.Artificially decreasing the size of the database used for upscaling shows that relatively stable results can beachieved with lower replication in some upscaling classes. For future upscaling studies at large geographicscales, a priori determination of sample sizes and tests to insure unimodal and statistically independentsamples are recommended. If these prerequisites are not fulfilled, classes may be merged or subdividedprior to upscaling. Decreased spatial resolution for upscaling from 30 m to 1 km has little impact on SOCestimates in this region, but classification accuracy is dramatically reduced and land cover classes show different, sometimes non-linear, responses to scale.