Joint Numerical Weather Prediction Research Articles

The National Earth System Prediction Capability: Coordinating the Giant

Abstract The United States has had three operational numerical weather prediction centers since the Joint Numerical Weather Prediction Unit was closed in 1958. This led to separate paths for U.S. numerical weather prediction, research, technology, and operations, resulting in multiple community calls for better coordination. Since 2006, the three operational organizations—the U.S. Air Force, the U.S. Navy, and the National Weather Service—and, more recently, the Department of Energy, the National Aeronautics and Space Administration, the National Science Foundation, and the National Oceanic and Atmospheric Administration/Office of Oceanic and Atmospheric Research, have been working to increase coordination. This increasingly successful effort has resulted in the establishment of a National Earth System Prediction Capability (National ESPC) office with responsibility to further interagency coordination and collaboration. It has also resulted in sharing of data through an operational global ensemble, common software standards, and model components among the agencies. This article discusses the drivers, the progress, and the future of interagency collaboration.

50th Anniversary of Operational Numerical Weather Prediction

The National Centers for Environmental Prediction (NCEP), Air Force Weather Agency (AFWA), Fleet Numerical Meteorology and Oceanography Center (FNMOC), National Weather Association, and American Meteorological Society (AMS) cosponsored a “Symposium on the 50th Anniversary of Operational Numerical Weather Prediction,” on 14–17 June 2004 at the University of Maryland, College Park in College Park, Maryland. Operational numerical weather prediction (NWP) in the United States started with the Joint Numerical Weather Prediction Unit (JNWPU) on 1 July 1954, staffed by members of the U.S. Weather Bureau, the U.S. Air Force and the U.S. Navy. The origins of NCEP, AFWA, and FNMOC can all be traced to the JNWPU. The symposium celebrated the pioneering developments in NWP and the remarkable improvements in forecast skill and support of the nation's economy, well being, and national defense achieved over the last 50 years. This essay was inspired by the presentations from that symposium.

The birth of numerical weather prediction

The paper describes the major events leading gradually to operational, numerical, short-range predictions for the large-scale atmospheric flow. The theoretical foundation starting with Rossby's studies of the linearized, barotropic equation and ending a decade and a half later with the general formulation of the quasi-geostrophic, baroclinic model by Charney and Phillips is described. The problems connected with the very long waves and the inconsistences of the geostrophic approximation which were major obstacles in the first experimental forecasts are discussed. The resulting changes to divergent barotropic and baroclinic models and to the use of the balance equation are described. After the discussion of the theoretical foundation, the paper describes the major developments leading to the Meteorology Project at the Institute for Advanced Studied under the leadership of John von Neumann and Jule Charney followed by the establishment of the Joint Numerical Weather Prediction Unit in Suitland, Maryland. The inter-connected developments in Europe, taking place more-or-less at the same time, are described by concentrating on the activities in Stockholm where the barotropic model was used in many experiments leading also to operational forecasts. The further developments resulting in the use of the primitive equations and the formulation of medium-range forecasting models are not included in the paper. DOI: 10.1034/j.1600-0889.1991.t01-3-00006.x

The birth of numerical weather prediction

The paper describes the major events leading gradually to operational, numerical, short-range predictions for the large-scale atmospheric flow. The theoretical foundation starting with Rossby's studies of the linearized, barotropic equation and ending a decade and a half later with the general formulation of the quasi-geostrophic, baroclinic model by Charney and Phillips is described. The problems connected with the very long waves and the inconsistences of the geostrophic approximation which were major obstacles in the first experimental forecasts are discussed. The resulting changes to divergent barotropic and baroclinic models and to the use of the balance equation are described. After the discussion of the theoretical foundation, the paper describes the major developments leading to the Meteorology Project at the Institute for Advanced Studied under the leadership of John von Neumann and Jule Charney followed by the establishment of the Joint Numerical Weather Prediction Unit in Suitland, Maryland. The interconnected developments in Europe, taking place more-or-less at the same time, are described by concentrating on the activities in Stockholm where the barotropic model was used in many experiments leading also to operational forecasts. The further developments resulting in the use of the primitive equations and the formulation of medium-range forecasting models are not included in the paper.

Operational numerical weather prediction

A variety of analysis and forecast systems have been used routinely over the years to produce numerical weather predictions (NWP) for daily use. The Joint Numerical Weather Prediction Unit of the United States was the first operational organization established in 1955. Since then, a large number of NWP centers have evolved in the industralized nations of the world. Today their predictions serve as the primary guidance for practicing synopticians for the forecast period from about 12 hours out to 10 days.

Vertical Motion Calculations and Satellite Cloud Observations over the Western and Central United States

Abstract Vertical motions over the western and central United States during 19–21 August 1961 were computed by several methods. The resulting angles applicable, at 700 mb or 10,000 ft (600 mb for the Joint Numerical Weather Prediction vertical motion fields) were compared both subjectively and objectively to each other, to the corresponding standard type cloud and weather observations, and to approximately synoptic TIROS nephanalyses. The results of the comparisons showed only moderate agreement among the various methods and between the vertical motion analyses and the TIROS information. The component method, based on observed winds, appeared to give the best agreement with the satellite picture indications of vertical motion, though it seemed little better than some of the other methods. The use of satellite data shows definite promise as a means of evaluating computed vertical motion fields and should eventually contribute to the improvement of the procedures used to compute motions. A lag, or “trailing...

SATELLITE PICTURES AND METEOROLOGICAL ANALYSES OF A DEVELOPING LOW IN THE CENTRAL UNITED STATES1

TIROS I satellite pictures for a 3-day period about, a developing “Low system” over the central United States are shown. Cloud distribution about surface fronts and upper-air troughs was examined and compared to the “ideal” (classical) distribution. The pictures and analyses reveal that cloud patterns were significantly changed by strong subsidence and advection of dry air along a low-level wind maximum. The cloud patterns when compared to the 600-mb. vertical motions computed routinely by the Joint Numerical Weather Prediction Unit showed only partial agreement. A strong relationship between low cloud cover and surface relative humidity was found; this relationship may be useful under certain conditions in determining the stability of the air over areas where upper-air data are sparse.

FORECASTING PRECIPITATION WITH THE AID OF A HIGH-SPEED ELECTRONIC COMPUTER

A high-speed electronic computer has been programmed to prepare cloud and precipitation forecasts for the Northern Hemisphere. The basic input data are derived from the Joint Numerical Weather Prediction Unit's two-level operational model of the atmosphere. The computer predicts cloudiness, rain, snow, and amounts of precipitation. Examples of computer forecasts are given and compared with observations.

Verification of JNWP-Unit Hurricane and Typhoon Forecasts for 19591

During the 1959 season for tropical storms in the Northern Hemisphere, Joint Numerical Weather Prediction (JNWP) Unit computed operationally one or more forecast tracks for 11 hurricanes and tropical storms and 11 typhoons. The 500-mb barotropic forecast flow with the tropical vortex eliminated from the initial 500-mb analysis was employed as the steering current in obtaining these forecasts. A summary of 1959 forecasts and a table of verification of JNWP hurricane forecasts for the years 1956 through 1959 are presented. The improvement and deterioration in forecasts from one year to the next are discussed in terms of sample size, operational changes, and analysis and forecasting techniques specifically designed for forecasting trajectories of tropical cyclones.

CASE STUDIES OF NUMERICAL WIND ANALYSES

With the aid of an electronic computer, case studies of wind analyses at the 850-mb., 700-mb., 500-mb., 400-mb., 300-mb., and 200-mb. pressure levels have been made. The divergent and non-divergent wind components resulting from the u and v wind-component analyses are investigated. For the cases considered, the streamfunction fields are slightly superior to the Joint Numerical Weather Prediction operational fields, obtained initially through use of the “balance equation.” The magnitude of the horizontal wind divergence values are comparable to those obtained from the winds by previous investigators employing hand-analysis techniques. However, the divergence patterns are not sufficiently accurate for the strict requirements necessary for numerical weather forecasting.

AN ANALYSIS OF BAROTROPIC FORECAST ERRORS IN CASES OF RAPID SEA LEVEL CYCLOGENESIS1

The 24-hour 500-mb. barotropic forecasts prepared by the Joint Numerical Weather Prediction Unit (JNWPU) have been investigated in 30 cases of rapid sea level cyclogenesis. Composite error maps are presented for the region of cyclogenesis. The 500-mb. errors are found to be significantly larger when the solenoidal field at that level is strong than when it is weak.

Vertical air flow over North America

Vertical air movement at 600 mb, computed daily by the Joint Numerical Weather Prediction unit of the National Meteorological Center, and averaged by months as part of a project sponsored by the Office of Civil and Defense Mobilization, is about 0.4 cm/sec over much of North America; the standard deviation is roughly the same magnitude. Averages computed separately for the 00 UT and 12 UT observations show good agreement in the various months, and change slowly from month to month during the six months, December 1958 through May 1959, analyzed.

AN OPERATIONAL OBJECTIVE ANALYSIS SYSTEM

Abstract The system of objective weather map analysis used at the Joint Numerical Weather Prediction Unit is described. It is an integral part of the automatic data processing system, and is designed to operate with a minimum of manual supervision. The analysis method, based mainly on the method of Bergthorssen and Dooos, is essentially a method of applying corrections to a first guess field. The corrections are determined from a comparison of the data with the interpolated value of the guess field at the observation point. For the analysis of the heights of a pressure surface the reported wind is taken into account in determining the lateral gradient of the correction to be applied. A series of scans of the field is made, each scan consisting of application of corrections on a smaller lateral scale than during the previous scan. The analysis system is very flexible, and has been used to analyze many different types of variables. An example of horizontal divergence computed from a direct wind analysis is ...

A STUDY OF ENERGY CONVERSION AND MERIDIONAL CIRCULATION FOR THE LARGE-SCALE MOTION IN THE ATMOSPHERE1

Energy conversion between potential and kinetic energy is considered. Section 2 contains the derivations which are necessary to compute the energy conversion for a large region on the basis of vertical velocities and relative topography as obtained from a two-parameter model presently used by the Joint Numerical Weather Prediction Unit. The energy conversion is divided into three parts: (1) energy conversion due to a mean vertical velocity over the region, (2) energy conversion in meridional planes, and (3) energy conversion in the zonal planes. Section 3 contains a discussion of the results obtained for the months January and April 1959. The energy conversion is positive for each day in both months, but the conversion in the meridional planes has a different sign in the two months, being positive in January and negative in April. The pattern of the mean meridional circulation is discussed and the frictional dissipation estimated. Section 4 describes a procedure for an evaluation of the energy conversion for the different wave numbers, and discusses the results for the same two months. The same section contains a comparison with results obtained from a linear, adiabatic theory. Section 5 contains a discussion of the modifications to the results in section 4 caused by the diabatic heating of the atmosphere It is made plausible that the maximum conversion found for the small wave numbers by an adiabatic computation is greatly altered due to the effects of the heating.

THE ERROR IN NUMERICAL FORECASTS DUE TO RETROGRESSION OF ULTRA-LONG WAVES

Abstract Error charts for the numerical barotropic forecasts prepared at the Joint Numerical Weather Prediction Unit since October 1957 have revealed retrogressive patterns of very long wavelength. These errors are shown to be due to changes in the large-scale components predicted by the numerical model. These components are actually quasi-stationary in the atmosphere. Forecasts prepared with an approximation to these components held unchanged show significantly increased accuracy. Finally, some of the difficulties in developing a more acceptable physical approach to this problem are outlined.

A COMPARISON OF JNWP TRAJECTORY FORECASTS WITH TRANSOSONDE FLIGHTS

Abstract Trajectories at 300 mb. numerically computed from the Joint Numerical Weather Prediction Unit's operational forecasts are compared with transosonde balloon tracks. The method developed by Hubert is shown to provide trajectories of considerable accuracy.

NUMERICAL METHODS IN WEATHER PREDICTION: I. THE BALANCE EQUATION*

Abstract Two methods of solving the balance equation are outlined. Both methods have been used successfully on a daily operational basis at the Joint Numerical Weather Prediction Unit for a period of more than a year. Solutions were on the operational grid of 30 × 34 points spaced at 381-km. intervals.

The relation between standard deviation of contour heights and standard vector deviation of wind

Quarterly Journal of the Royal Meteorological SocietyVolume 83, Issue 358 p. 553-554 Correspondence The relation between standard deviation of contour heights and standard vector deviation of wind P. D. Thompson, P. D. Thompson Development Section, Joint Numerical Weather Prediction Unit, U.S. Weather Bureau, WashingtonSearch for more papers by this authorA. F. Jenkinson, A. F. Jenkinson The Waite Institute, Adelaide, S. AustraliaSearch for more papers by this author P. D. Thompson, P. D. Thompson Development Section, Joint Numerical Weather Prediction Unit, U.S. Weather Bureau, WashingtonSearch for more papers by this authorA. F. Jenkinson, A. F. Jenkinson The Waite Institute, Adelaide, S. AustraliaSearch for more papers by this author First published: October 1957 https://doi.org/10.1002/qj.49708335815Citations: 3AboutPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat No abstract is available for this article.Citing Literature Volume83, Issue358October 1957Pages 553-554 RelatedInformation

A STUDY OF NUMERICAL FORECASTING ERRORS

Studies of numerical forecasting errors have revealed that the principal difficulties encountered in models currently operational at the Joint Numerical Weather Prediction Unit arise from errors in the 500-mb. forecasts. These errors are common to both barotropic and baroclinic models. The errors in the 1000–500-mb. thickness forecasts are considerably smaller. Results show that systematic errors are introduced (1) by the use of the geostrophic approximation and (2) by the approximations used on the boundaries. The cases presented in this paper show that the first type of error is virtually nonexistent in non-geostrophic, barotropic forecasts. Boundary errors are particularly serious in cases where the boundaries are meteorologically active, suggesting the future use of hemispheric forecast grids. The elimination of these two types of errors from all forecasts appears to be essential before the smaller errors, such as truncation errors and errors due to baroclinic development, non-adiabatic effects, etc., can be isolated and studied satisfactorily.

A PRELIMINARY REPORT ON NUMERICAL SEA CONDITION FORECASTS

Since July 1956 the Joint Numerical Weather Prediction Unit at Suitland, Md., has been making machine forecasts of sea conditions on an operational basis. These prognoses are based on the 1,000-mb. wind forecasts derived from the two-level, thermotropic model currently in use at JNWP. Two different sea-forecast models have been tested to date. The first utilized only the forecast winds at the end of the forecast period and therefore yielded “fully developed waves.” A model incorporating “duration” in a crude manner is now in daily operation. This paper describes both methods, compares the numerical results with observed conditions, and outlines future plans.