Retrospective metropolitan population statistics have long been used to compute adjustments to urban-induced temperature biases at U.S. Historical Climatology Network (U.S. HCN) stations. These values were compared with gridded population statistics surrounding each U.S. HCN station. The latter data set was created by coupling county-scale retrospective Census Bureau population statistics with 1990 1 km2 gridded Census Bureau population density values. Differences between the data appear to be related to population values based on political jurisdictions in the former data set.