Summary
A number of local interpolation methods suitable for statistical and graphical meteorological applications, including weighted averages, least-squares polynomials and successive corrections, were compared in this study. Daily observations of temperature and geopotential height recorded in two years by 71 European stations were used as data sources; the results of the interpolations were checked against the observed values in 19 selected stations, the data of which were assumed as missing. The methods were tested in points with different conditions of data density and isotropy.
The best results were obtained with algorithms based on simple weighted averages, in one or more steps; in these methods a proper choice of the influence radii appears to be the key factor in determining the accuracy of the approximation. In particular, a successive correction (SC) method using a weighted average with a large radius as a first guess and two correction steps proved to give the highest accuracy with a reasonable computational time. The SC method was also used to generate grid-point fields of geopotential height in a six-month winter period, and the main statistical properties of this sample were compared with those of the corresponding fields analysed by the European Centre for Medium Range Weather Forecasts. The mean field, the standard deviations, and the principal components that explain the most part of the total variance were correctly reproduced in the sample generated by the SC method.
Zusammenfassung
In dieser Untersuchung werden einige lokale Interpolationsmethoden, wie gewichtete Mittel, Polynome aufgrund der kleinsten Quadrate oder eine schrittweise Korrektur, die für statistische und graphische meteorologische Anwendungen geeignet sind, miteinander verglichen. Die täglichen Beobachtungen der Temperatur und der geopotentiellen Höhe von 2 Jahren und 71 europäischen Stationen wurden als Datenmaterial herangezogen. Die Ergebnisse der Interpolationen wurden mit den Messungen von 19 ausgesuchten Stationen verglichen, deren Daten als fehlend angenommen wurden waren. Die Methoden wurden in Punkten mit verschiedener Datendichte und Datenisotropie getestet.
Die besten Ergebnisse lieferten die gewichteten Mittel - bei dieser Methode erwies sich die Wahl geeigneter Einflußradien als entscheidend für die Genauigkeit. Besonders die Methode der stufenweisen Korrektur mit einem gewichteten Mittel bei einem großen Radius als ersten und zwei Korrekturschritten erwies sich bei angemessener Rechenzen als die genaueste. Diese Methode wurde auch herangezogen, um an Gitterpunkten die geopotentiellen Höhen für eine sechsmonatige Winterperiode zu berechnen. Die statistischen Eigenschaften dieser Felder wurden mit deren des ECMWF (Europäisches Zentrum für Mittelfristvorhersagen) verglichen. Dabei wurde das mittlere Feld, die Standardabweichung und mit Einschränkungen die totale Varianz durch die oben erwähnte Methode korrekt reproduziert.
Similar content being viewed by others
References
Bergman, K. H.: Multivariate Analysis of Temperatures and Winds Using Optimum Interpolation. Mon. Weath. Rev.107, 1423–1444 (1979).
Bergthorsson, P., Döös, B. R.: Numerical Weather Map Analysis. Tellus7, 329–340 (1955).
Boone, D. R., Samuelson, G. S.: Computer Mapping of Air Quality. J. Environ. Eng. Div. Proc. ASCE103, EE6, 969–979 (1977).
Buell, C. E.: Correlation Functions for Wind and Geopotential on Isobaric Surfaces. J. Appl. Met.11, 51–59 (1972).
Buell, C. E.: The Topography of Empirical Orthogonal Functions. Preprints Fourth Conf. Probability and Statistics in Atmospheric Sciences, Tallahassee. Amer. Met. Soc., 188–193 (1975).
Buell, C. E.: On the Physical Interpretation of Empirical Orthogonal Functions. Preprints Sixth Conf. Probability and Statistics in Atmospheric Sciences, Banff, Amer. Met. Soc., 112–117 (1979).
Corby, G. A.: An Experiment in Three-Dimensional Objective Analysis. Tellus15, 432–438 (1963).
Cressman, G. P.: An Operational Objective Analysis System. Mon. Weath. Rev.87, 367–374 (1959).
Eliassen, A.: Provisional Report on Calculation of Spatial Covariance and Autocorrelation of the Pressure Field. Inst. Weather and Climate Res., Acad. Sci. Oslo, Rept. No. 5 (1954).
Endlich, R. M., Mancuso, L.: Objective Analysis of Environmental Conditions Associated with Severe Thunderstorms and Tornadoes. Mon. Weath. Rev.96, 342–350 (1968).
Gandin, L. S.: Objective Analysis of Meteorological Fields. Leningrad: Hydromet. Press 1963. (Translated from Russian by the Israelic Program for Scientific Translations, 242 pp., 1965.)
Gilchrist, B., Cressman, G. P.: An Experiment in Objective Analysis. Tellus6, 97–101 (1954).
Glahn, H. R.: Comments on: “A Comparison of Interpolation Methods for Sparse Data: Application to Wind and Concentration Fields.” J. Appl. Met.20, 88–91 (1981).
Goodin, W. R., McRae, G. J., Seinfeld, J. H.: A Comparison of Interpolation Methods for Sparse Data: Application to Wind and Concentration Fields. J. Appl. Met.18, 761–771 (1979).
Goodin, W. R., McRae, G. J., Seinfeld, J. H.: Reply. J. Appl. Met.20, 92–94 (1981).
Gustavsson, N.: A Review of Methods for Objective Analysis. Proceedings of ECMWF Seminar 1980 on Data Assimilation Methods, 15–19 September 1980, ECMWF, Reading, U.K., pp. 19–82 (1981).
Julian, P. R., Thiebaux, H. J.: On Some Properties of Correlation Functions Used in Optimum Interpolation Schemes. Mon. Weath. Rev.103, 605–613 (1975).
Karl, T. R., Koscielny, A. J., Diaz, H. F.: Potential Errors in Application of Principal Component (Eigenvector) Analysis to Geophysical Data. J. Appl. Met.21, 1183–1186 (1982).
Lorene, A.: A Global Three-Dimensional Multivariate Statistical Interpolation Scheme. Mon. Weath. Rev.109, 701–721 (1981).
Maude, A. D.: Interpolation — Mainly for Graph Plotters. The Computer16, 64–65 (1973).
McLain, D. H.: Drawing Contours from Arbitrary Data Points. Comput. J.17, 318–324 (1974).
McLain, D. H.: Two-Dimensional Interpolation From Random Data. Comput. J.19, 178–181 (1976).
Panofsky, H. A.: Objective Weather-Map Analysis. J. Met.6, 386–392 (1949).
Schlatter, T. W.: Some Experiments with a Multivariate Statistical Objective Analysis Scheme. Mon. Weath. Rev.103, 246–257 (1975).
Schlatter, T. W., Branstator, G. W., Thiel, L. G.: Testing a Global Multivariate Statistical Objective Analysis Scheme with Observed Data. Mon. Weath. Rev.104, 765–783 (1976).
Stephens, J. J., Stitt, J. M.: Optimum Influence Radii for Interpolation with the Method of Successive Corrections. Mon. Weath. Rev.98, 680–687 (1970).
Thiebaux, H. J.: Experiments with Correlation Representation for Objective Analysis. Mon. Weath. Rev.103, 617–627 (1975).
Thiebaux, H. J.: Anisotropic Correlation Functions for Objective Analysis. Mon. Weath. Rev.104, 994–1002 (1976).
Author information
Authors and Affiliations
Additional information
With 6 Figures
Rights and permissions
About this article
Cite this article
Tronci, N., Molteni, F. & Bozzini, M. A comparison of local approximation methods for the analysis of meteorological data. Arch. Met. Geoph. Biocl., Ser. B 36, 189–211 (1986). https://doi.org/10.1007/BF02278328
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF02278328