In many urban and suburban environments there are several locations at which temperature is observed. The homogeneity of long-term records at these locations is often compromised by changes in observing locations, in instrumentation, in observing protocols, and in the surrounding site environment. Station history information (metadata) identifies these observing changes. The effects of these changes on data homogeneity are examined by analyzing the data from the network of stations in a metropolitan area. Of interest is the effect of a change on the characteristics of the frequency distribution of the data central tendencies, variability and extremes). These effects are examined both statistically and deterministically for a few case-study, urban and suburban metropolitan areas.
Knowledge about the changes in frequency distribution characteristics when observing changes occur is useful for deciding if data should be combined or adjusted. This knowledge is critical to the construction of data series that will be used in the calculation of U.S. normals.