Diurnal temperature range (DTR) time series have been widely analyzed in studies of climate variability and change. At the regional scale, DTR has been related to changing levels of greenhouse gases as well as changes in land cover and land use, including urbanization and the introduction of irrigated agriculture. Meanwhile, DTR has also been related to localized influences associated with the exposure of instruments to the immediate environment. The interpretation of patterns of temporal variation in DTR therefore demands consideration of a spatial perspective.

This paper presents a comparative analysis of temporal patterns in DTR for proximate stations of the United States Historical Climatology Network located in the Ohio River Valley. Percentiles of the empirical DTR distributions are calculated by month for each year in the period of record for a station. In some cases records extend back to the late 19th century. Annual time series of the percentiles for observations taken in January, April, July, and October are plotted by station. Temporal variation in DTR percentiles is represented by scatter plots with robust smoothing curves superimposed. Comparisons are then made among proximate stations. Similar patterns among stations provide evidence of regional-scale forcings, while patterns that are dissimilar are suggestive of local-scale forcings. Station histories and metadata developed from site visits, digital orthophotography, and digital elevation models are referenced in an effort to interpret potential local-scale forcings. Results suggest that highly localized forcings can have a significant influence on the climatological record.