Accession Number : ADA258569


Title :   The Development of the Navy's Multivariate Optimum Interpolation Analysis System.


Descriptive Note : Final rept.,


Corporate Author : NAVAL OCEANOGRAPHIC AND ATMOSPHERIC RESEARCH LAB MONTEREY CA


Personal Author(s) : Barker, E H


Full Text : https://apps.dtic.mil/dtic/tr/fulltext/u2/a258569.pdf


Report Date : Jul 1992


Pagination or Media Count : 41


Abstract : The equations for the Navy's multivariate optimum interpolation (MVOI) analysis system for atmospheric analysis are presented in complete form. Included are the derivations for horizontal and vertical covariances that retain the geostrophic constraint a various kinds of observations such as pressure thickness, pressure height, and winds. Methods that scale the observation values to the resulting nondimensional values so that they are consistent with the constraints are presented. The derivations of the analysis equations are provided using optimal theory to minimize the analysis error variance. The derivation of the internal data checking procedure is given. The method compares suspicious values with neighboring values and removes rejected values in a computationally consistent manner. Validation experiments that show the exactness of the constraints, expected analysis error, and the method for analysis of wind around the poles are presented. These experiments have known results, and therefore are useful in catching design and program errors. A brief description is provided of the observation selection procedures and internal quality control, and some recommendations for improvement are given.


Descriptors :   *MULTIVARIATE ANALYSIS , *MATHEMATICAL PREDICTION , *INTERPOLATION , *ATMOSPHERICS , CONTROL , THICKNESS , THEORY , SCALE , PRECISION , QUALITY CONTROL , HEIGHT , SELECTION , EQUATIONS , COVARIANCE , VALUE , PRESSURE , ERRORS , INTERNAL , QUALITY , WIND , OBSERVATION , VALIDATION


Subject Categories : Atmospheric Physics
      Numerical Mathematics


Distribution Statement : APPROVED FOR PUBLIC RELEASE