A method for monthly detection of inhomogeneities and errors in daily maximum and minimum temperatures

Two statistical tests are described that can be used to detect potential in homogeneities and errors in daily temperature observations. These tests, ba sed on neighbor comparisons, differ from existing inhomogeneity tests by ev aluating daily rather than monthly or annual observations and by focusing o n a very short record length. Standardized difference series one month in l ength are formed between a candidate station, whose daily temperature time series is being evaluated, and a number of neighboring stations. These seri es, called D-series, approximate white noise when a candidate is like its n eighbors and are other than white noise when the candidate is unlike its ne ighbors. Two white noise tests are then applied to the D-series in order to detect potential problems at the candidate station: a cross-correlation te st and a lag 1 (1-day) autocorrelation test. Examples of errors and inhomog eneities detected through the application of the two tests on observations from the National Weather Service's Cooperative Observer Network are provid ed. These tests were designed specifically to detect inhomogeneities in an operational environment, that is, while data are being routinely processed. When a potential inhomogeneity is identified, timely action can be taken a nd feedback given, if necessary, to station field managers to prevent furth er corruption of the data record. While examples are provided using observa tions from the Cooperative Observer Network, these tests may be used in any temperature observation network with sufficient station density to provide a pool of neighboring stations.