Search Results

You are looking at 101 - 110 of 2,092 items for :

  • Forecasting techniques x
  • Journal of the Atmospheric Sciences x
  • User-accessible content x
Clear All
Christopher M. Danforth and Eugenia Kalnay

-independent error correction did not improve the forecast skill. By adding a state-dependent empirical correction to the model, inspired by the procedure proposed by Leith, they were able to extend forecast skill up to the limits imposed by observation error. However, Leith’s technique requires the solution of a N d -dimensional linear system. As a result, before the procedure can be considered useful for operational use, a low-dimensional representation of Leith’s empirical correction operator is required

Full access
Herbert Riehl

,291-298, 1952.proaches, rather than to supplant them. We did notclaim in the 1952 paper that a final scheme for precipitation forecasting had been found, but that "theresults of the study . . . are sufficiently impressive,in our opinion, to warrant continued work with themethod." We hoped to stimulate high-level observations and further research which would deal with thefaults of the technique. Normally, a paper such as theone we wrote is accepted in this sense, rather than asan object for testing and

Full access
Robert M. White, Royce C. Derby, Duane S. Cooley, and Florence A. Seaver

contour heightsas functions of the preceding two-day sequence of contour heights at the same level. The results indicatethat the statistical procedure may provide useful forecasts for twenty-four hours over regions of the hemisphere with adequate data coverage. The utility of the system for extended-range forecasting is also discussed.1. IntroductionAll meteorological forecasting techniques imply theapplication of filters to sets of observed data. Insynoptic-subjective forecasting, the individual may

Full access
Chungu Lu and Gerald L. Browning

1. Introduction One of the major objectives of data assimilation using the four-dimensional variational/adjoint (4DVAR) technique is to generate an improved initial condition in order to obtain an improved numerical weather forecast. Conventionally, this is done by the least squares minimization of the difference (or distance) between the analyzed and observational field(s), subject to a forecast model constraint ( Lewis and Derber 1985 ; LeDimet and Talagrand 1986 ; Talagrand and Courtier

Full access
Adrian Simmons, Mariano Hortal, Graeme Kelly, Anthony McNally, Agathe Untch, and Sakari Uppala

introduction of variational techniques ( Andersson et al. 1998 ; Rabier et al. 2000 ) and direct radiance assimilation ( McNally et al. 2000 )] and in modeling [such as ECMWF’s introduction of 60-level resolution ( Untch and Simmons 1999 ) and finite-element discretization in the vertical (Untch and Hortal 2004)]. Moreover, medium-range forecasts for the Southern Hemisphere troposphere have been brought to levels of accuracy similar to those reached for the Northern Hemisphere troposphere ( Simmons and

Full access
M. A. Estoque

of the precipitation is associated with large-scaleupward motions. However, amounts are underestimated in regions where local convective showers occur,usually in the southern half of cyclones.Some of the unsatisfactory predictions are, in part,due to inaccurate 1000-mb forecasts made with thegraphical-numerical technique. This source of errorappears to have no systematic tendency to underestimate the amount of precipitation. It would beinteresting to use this approach in conjunction with1000-mb

Full access
N. A. Philips

.00011 -- 0.00580* See text for definition. B. No information from forecast A was used. C. Every 6 hr, ~i at one grid point was simply replaced by the corresponding value of ~i from the controlforecast A. No changes were made in any other variables. This is the simple replacement technique usedby Charney et al. D. Same as C, except that changes in surroundingvalues of ~, u, v were made according to Eq. (7), theTadjbakhsh method.(The availability of one observed value every 6 hrcrudely mimics the

Full access
James A. Ridout

1996 ; Lucas and Zipser 1996 ). In order to obtain insight into the potential for this type of control, in section 2 of the present study the impact on cloud growth of moisture variations of the nature of those associated with dry tongue episodes in the tropical Pacific is investigated by means of a series of quasi-cloud-resolving-scale numerical forecast experiments. These forecasts are carried out using a version of the Naval Research Laboratory's Coupled Ocean–Atmosphere Mesoscale Prediction

Full access
Robert W. Jones

large-scale atmospheric motion.Similar techniques for reliable forecasts of smallscale disturbances like the hurricane have not beenperfected for operational use. To date, several modelshave been tried experimentally for operational forecasts, by vl7. Hubert (1957), L. Hubert (1959) andVandermaii (1959) at the Joint Numerical WeatherPrediction unit in Suitland, Mar).land, and byinvestigators at Japan's numerical weather-predictioncenter, but none of these models has been highlycompetitive with

Full access
Gene E. Birchfield

,the steering-flow technique consists in removing thecirculation of the hurricane and essentially performinga trajectory calculation in the residual or steeringflow. The initial data for the two levels are determinedby taking a properly weighted average of the fourlevels of data available.From the report by Kasahara cited above, for theeight 24-hr steering-flow predictions correspondingto the eight mean-data forecasts presented here, themean magnitude of vector error E, and the meanmagnitude of the

Full access