Journal of the Atmospheric Sciences · 1972 · 19 citations · 0 references
Natural SciencesEarth SciencesMeteorologyAtmospheric DynamicsNumerical Weather PredictionWeather ForecastingData AssimilationNumerical AnalysisNumerical Weather Prediction Data AssimilationMeteorological Data AnalysisOrthogonal Polynomial ApproximationOrthogonal Polynomial BasesOrthogonal Polynomial ExpansionsOrthogonal PolynomialsPolynomial Basis FunctionsWindWind Observation Data Analysis
With the aid of orthogonal polynomials, generated by means of the Gram-Schmidt process, it is possible to fit polynomial functions of the form z=f (x, y) or z=f (x, y, p) to large bodies of data irregularly distributed in two or three dimensions. The results of some experiments with radiosonde pressure heights and wind data are shown. With adequate computing power the technique, which extends naturally to four dimensions, will afford an alternative to current, mainly grid point, techniques of analysis.