Numerical Forecast Models Research Articles

MOS, Perfect Prog, and Reanalysis

Abstract Statistical postprocessing methods have been successful in correcting many defects inherent in numerical weather prediction model forecasts. Among them, model output statistics (MOS) and perfect prog have been most common, each with its own strengths and weaknesses. Here, an alternative method (called RAN) is examined that combines the two, while at the same time utilizes the information in reanalysis data. The three methods are examined from a purely formal/mathematical point of view. The results suggest that whereas MOS is expected to outperform perfect prog and RAN in terms of mean squared error, bias, and error variance, the RAN approach is expected to yield more certain and bias-free forecasts. It is suggested therefore that a real-time RAN-based postprocessor be developed for further testing.

A Bad Air Day in Houston

Abstract A case study from the Texas Air Quality Study 2000 field campaign illustrates the complex interaction of meteorological and chemical processes that produced a high-pollution event in the Houston area on 30 August 2000. High 1-h ozone concentrations of nearly 200 ppb were measured near the surface, and vertical profile data from an airborne differential-absorption lidar (DIAL) system showed that these high-ozone concentrations penetrated to heights approaching 2 km into the atmospheric boundary layer. This deep layer of pollution was transported over the surrounding countryside at night, where it then mixed out the next day to become part of the rural background levels. These background levels thus increased during the course of the multiday pollution episode. The case study illustrates many processes that numerical forecast models must faithfully represent to produce accurate quantitative predictions of peak pollutant concentrations in coastal locations such as Houston. Such accurate predictions ...

Very large inverse problems in atmosphere and ocean modelling

AbstractFor the very large nonlinear dynamical systems that arise in a wide range of physical, biological and environmental problems, the data needed to initialize a numerical forecasting model are seldom available. To generate accurate estimates of the expected states of the system, both current and future, the technique of ‘data assimilation’ is used to combine the numerical model predictions with observations of the system measured over time. Assimilation of data is an inverse problem that for very large‐scale systems is generally ill‐posed. In four‐dimensional variational assimilation schemes, the dynamical model equations provide constraints that act to spread information into data sparse regions, enabling the state of the system to be reconstructed accurately. The mechanism for this is not well understood. Singular value decomposition techniques are applied here to the observability matrix of the system in order to analyse the critical features in this process. Simplified models are used to demonstrate how information is propagated from observed regions into unobserved areas. The impact of the size of the observational noise and the temporal position of the observations is examined. The best signal‐to‐noise ratio needed to extract the most information from the observations is estimated using Tikhonov regularization theory. Copyright © 2005 John Wiley & Sons, Ltd.

A GBMR Experiment and Validation of 1DVAR-LDAS in Different Bare Soil

Land surface variables, especially the soil moisture, are central to the regulation of the land-atmosphere water, energy, and carbon interactions. The development of land hydrological parameterizations that emphasize soil moisture and temperature, together with innovative remote sensing and surface-based observations, promise a mechanism for the synthesis of regional fields of these vital hydrological variables. Data assimilation is the process of incorporating observations into a model of the land-atmosphere interaction system and filling the region of missing data with model predictions as well as providing a suite of data-constrained estimates of unobserved quantities. In order to incorporate remote sensing and surface-based observations in a numerical forecasting model that estimates land-surface parameters, we have developed a Land Data Assimilation Scheme (LDAS) that can assess the model parameters and estimates vertical profiles of soil moisture and temperature by comparing passive microwave observations through a unique minimization technique called Very Fast Simulated Annealing (VFSA). This manuscript mainly focuses on modifying LDAS and testing its capabilities in different bare soils and their surface conditions. So we conducted a field experiment and obtained well-controlled observations by employing a six channel Ground Based Microwave Radiometer (GBMR-6CH). The results of the LDAS testing were in good agreement with observational data compare to open loop simulations. Major model sensitivities to soil type, soil moisture, and surface roughness are explored in bellow section.

NNICE – a neural network aircraft icing algorithm

Although much is known about the meteorological conditions for significant aircraft icing, research studies to date have only been successful identifying conditions for general icing, i.e. clouds in the temperature range from 0 °C to about −20 °C. The aerodynamics of ice accumulation suggest three meteorological factors, cloud liquid water, droplet size, and air temperature. Only the latter is known or forecast with a significant degree of accuracy. The first two are partial functions of the atmosphere's vertical motion which is poorly known, especially in convective situations. However, favorable patterns of relative humidity and potential instability that indicate conditions for possible convection are readily discernable from sounding or numerical forecast model data. Two neural networks were taught to sort out these patterns with respect to icing intensity. Each uses a different neuron transfer function which gives each a “personality”. The “conservative” network diagnoses light icing well but has difficulty with moderate or greater icing. On the other hand, the “radical” network finds the higher intensity icing but is not as good at lower intensities. By combining the strengths of each, NNICE makes skillful icing forecasts of all intensities.

Roles of empirical modeling within CISM

Within the CISM project empirical models serve as baselines against which to monitor the increase in predictive skill of physics-based numerical codes as they advance. Empirical models also establish a suite of forecast models from which to evolve toward greater forecast accuracy by incorporating numerical codes as they advance enough to increase forecast skill. Establishing a suite of forecast models allows the CISM project to contribute results of use to operational space weather forecasting earlier than might otherwise be possible. Developing a suite of empirical models allows CISM to address issues of data ingestion early in the project under relatively simple conditions. Out of the data-ingestion effort has come a technique of probabilistic forecasts, which allows one to ingest information on IMF B z from solar data instead of from measurements taken only at the Lagrangian L1 point. To implement the evolution from empirical to numerical forecast models, CISM has adopted a coupled two-line system of model development, a science-model line interacting with a forecast-model line.

Assimilating GOES Brightness Temperatures. Part II: Assigning Water Vapor Wind Heights Directly from Weighting Functions

Abstract An unsolved problem with water vapor wind estimates from the upper-tropospheric 6.7-μm water vapor band on the Geostationary Operational Environmental Satellite (GOES) Imager (channel 3) is its exact placement in the vertical column. Satellite water vapor observations are known to be depth-averaged assessments of the upper-tropospheric moisture. Details about the effective averaging of upper-tropospheric observations, valid for GOES or those of other satellite platforms, are not retrieved as part of the observation. However, details about the vertical placement can be accurately estimated from forward radiative models that mimic the instrument spectral characteristics. A new method has been developed to assimilate satellite radiances or brightness temperatures directly into a numerical forecast model. A by-product of the new scheme is knowledge of the weighting functions that describe the assignment value given to each vertical layer. As a consequence, given water vapor wind data, these weighting...

Climate Prediction for Brazil's Nordeste: Performance of Empirical and Numerical Modeling Methods

Abstract Comparisons of performance of climate forecast methods require consistency in the predictand and a long common reference period. For Brazil's Nordeste, empirical methods developed at the University of Wisconsin use preseason (October–January) rainfall and January indices of the fields of meridional wind component and sea surface temperature (SST) in the tropical Atlantic and the equatorial Pacific as input to stepwise multiple regression and neural networking. These are used to predict the March–June rainfall at a network of 27 stations. An experiment at the International Research Institute for Climate Prediction, Columbia University, with a numerical model (ECHAM4.5) used global SST information through February to predict the March–June rainfall at three grid points in the Nordeste. The predictands for the empirical and numerical model forecasts are correlated at +0.96, and the period common to the independent portion of record of the empirical prediction and the numerical modeling is 1968–99. O...

Near Real‐Time Sensing of PWV from SGCAN and the Application Test in Numerical Weather Forecast

AbstractGPS data from Shanghai GPS Comprehensive Application Network (SGCAN) in summer of 2002 are analyzed. The ultra‐rapid orbits of IGS (IGU) are used directly in real‐time processing. Reliable Precipitable Water Vapor (PWV) series can be obtained with the assistance of remote stations. Considering the effect of ocean tide, the accuracy of PWV can be improved. The difference between GPS/PWV and Radiosonde observations is about 2mm. Comparison between the output of MM5 and GPS/PWV shows that GPS can be regarded as one means to verify the forecast of MM5 in near real time. Continuous GPS/PWV series in the whole Meiyu season reflect the changing processes of PWV in detail before and after the onset of Meiyu Season in Yangtze River Delta in 2002 summer. The variation of GPS/PWV shows that the onset of Meiyu season was from southwest to northeast and completed in about 7–8 hours. Rainband moved back and forth between Yangtze River Delta and Huaihe area in Meiyu season. The statistics of PWV peaks and one‐hour rain explains some short‐term forecast ability of GPS/PWV for rainstorms. The humidity condition of numerical forecast model is optimized after GPS/PWV being interpolated to the gridpoints of background field using Cressman method. The accuracy of numerical forecast is improved. One example is provided in this paper which shows the potential of GPS/PWV for improving numerical forecast.

Performance of the surface observation network during MAP

The comprehensive data collection effort in the framework of the Mesoscale Alpine Programme (MAP) offers the unprecedented possibility to carry out a careful evaluation of the performance of one of the densest surface observational networks of the world. Quality control of meteorological data today is usually seen as an integrated step of prognostic model initialization. If mesoscale models should be validated independently, however, a model independent quality checking procedure becomes important, especially when operating over complex terrain. For that reason DAQUAMAP, a project sponsored by the EUMETNET programme MAP-NWS, was conducted to make a high quality data set available to the scientific community testing high resolution numerical weather forecast models as well as performing diagnostic studies. The applied method of quality control consists of an automatic spatial consistency check of primary atmospheric variables. It allows to recognize gross errors and biases of individual station data and to derive station characteristics as well. The latter is especially important when validating model results with single station data. Due to a separate treatment of GTS and combined GTS and non-GTS station data a distinction of the performance of both networks could be achieved. Also a comparison between a similar exercise done with the ALPEX data set in 1982 and the MAP data set has been carried out. This allows to assess the effect of automatization in the meteorological observation networks which has taken place during the last 20 years.

Accuracy analysis of the Grey Markov forecasting model

This work reorganized a new Grey Markov numerical forecasting model to as an application model. 30-day commercial papers primary market interest rates were used as an analytical example to examine the state number, state interval range and average accuracy of the Grey forecasting model, and how they influence the accuracy of the Grey Markov forecasting model. The analytical results reveal improving the average accuracy of the Grey forecasting model, also increases the accuracy of the Grey Markov forecasting model.

Short-Range (0–48 h) Numerical Prediction of Convective Occurrence, Mode, and Location

Abstract A verification of high-resolution (6-km grid spacing) short-range (0–48 h) numerical model forecasts of warm-season convective occurrence, mode, and location was conducted over the Lake Michigan region. All available days from 5 April through 20 September 1999 were evaluated using 0.5° base reflectivity and accumulated precipitation products from the national radar network and the day-1 (0–24 h) and day-2 (24–48 h) forecasts from a quasi-operational version of the fifth-generation Pennsylvania State University–National Center for Atmospheric Research Mesoscale Model (MM5). Contingency measures show forecast skill for convective occurrence is high, with day-1 (day 2) equitable threat scores and Kuipers skill scores (KSS) of 0.69 (0.60) and 0.84 (0.75), respectively. Forecast skill in predicting convective mode (defined as linear, multicellular, or isolated) is also high, with KSS of 0.91 (0.86) for day 1 (day 2). Median timing errors for convective initiation/dissipation were within 2.5 h for all ...

Analysis and Prediction of Cirrus-Top Altitude and Ice Water Path in a Mesoscale Area

Abstract Vertical distributions of clouds have been a focus of many studies, motivated by their importance in radiative transfer processes in climate models. This study examines the horizontal distribution of cirrus clouds by means of satellite imagery analyses and numerical weather prediction model forecasts. A ground-truth dataset based on two aircraft mission periods flying particle probes through cirrus over a ground-based cloud radar is developed. Particle probe measurements in the cirrus clouds are used to compute ice water content and radar reflectivity averages in short time periods (25–30 s). Relationships for ice water content as a function of reflectivity are developed for 6-K ambient temperature categories. These relationships are applied to the radar-measured short-term-averaged reflectivities to compute vertical profiles of ice water content, which are vertically integrated over the depth of the observed cirrus clouds to form ice water path estimates. These and cloud-top height are compared ...

An Objective Climatology, Classification Scheme, and Assessment of Sensible Weather Impacts for Appalachian Cold-Air Damming

Abstract The geostrophic adjustment process for a rotating, stratified atmosphere in the presence of an orographic barrier may be manifest as a phenomenon known as “cold-air damming” (CAD). The degree of blocking by an orographic barrier, and therefore CAD intensity, is related to the static stability of the upstream air mass. When precipitation falls into dry near-surface air, differential evaporational cooling can increase static stability, and strengthen or initiate CAD. The sheltering effect of clouds can also maintain surface-based stability. Therefore, the ability of numerical forecast models to accurately predict CAD requires adequate representation of cloud and precipitation processes. Operational forecasters in the Appalachian damming region have previously developed a subjective classification scheme that distinguishes those CAD events that are heavily influenced by diabatic processes from those that are dominated by synoptic-scale forcing. In this study the subjective scheme is formalized in or...

The 3-hour-interval prediction of ground-level temperature in South Korea using dynamic linear models

The 3-hour-interval prediction of ground-level temperature from +00 h out to +45 h in South Korea (38 stations) is performed using the DLM (dynamic linear model) in order to eliminate the systematic error of numerical model forecasts. Numerical model forecasts and observations are used as input values of the DLM. According to the comparison of the DLM forecasts to the KFM (Kalman filter model) forecasts with RMSE and bias, the DLM is useful to improve the accuracy of prediction.

Relationship of the Arctic and Antarctic Oscillations to the Outgoing Longwave Radiation

Abstract While several mechanisms have been suggested to account for the association of the Arctic and Antarctic Oscillations (AO/AAO) with atmospheric parameters, this paper focuses on the relationship with the atmospheric outgoing longwave radiation (OLR). The main objective of this paper is to demonstrate through AO/AAO composite analysis that the NCEP–NCAR reanalysis OLR agrees with the independent observations of the NASA Earth Radiation Budget Experiment (ERBE) broadband satellite instruments both in zonal averages and in geographically mapped space, and to verify AO/AAO characterized general circulations derived from models and analyses. The results indicate several pronounced areas of storminess that are AO/AAO dependent. One is the well-known variation over the North Atlantic Ocean toward Europe. Also, several major areas are indicated in the tropical region—one in the Indian Ocean and the others in the west and central Pacific Ocean. In addition to demonstrating that the signals are statistically significant, also tested is the relationship of the features to other well-known tropical forcing mechanisms: the Madden–Julian oscillation (MJO) and the El Niño–La Niña sea surface temperature variations. It is shown that the tropical features do, in fact, have a strong relationship to the MJO, which indicates an additional tropical–extratropical interaction. With regard to the sea surface temperature, no correlation associated with the AO/AAO variability is seen. These associations with the cloudiness and atmospheric radiation budget suggest that if there is to be improvement of numerical model forecasts to an extended time period that numerical model radiation physics will have to be taken into consideration and improved.

Un outil global de prévision des ventes adapté à la distribution textile

Competition and globalisation imply a very accurate production and sourcing management of the Textile-Apparel-Distribution network actors. Such an organization requires a sales forecasting system answering to versatile textile market and distributor needs. Actually, the numerous forecasting models are generally unsuited to the textile context. We propose a forecasting system, composed of several models which perform forecasts on various horizons and at different sales aggregation levels. This system is based on techniques resulting from soft computing such fuzzy logic, neural networks or the evolutionary procedures, authorizing the processing of uncertain data. Performances of our models are then evaluated and analysed on real data from a important French textile distributor.

Evaluation of PLANT-Plus, a decision support system for control of early blight on potatoes in South Africa

Early blight is a major foliar disease of potatoes in most potato growing regions of the world. Numerous forecasting models have been developed and evaluated for the control of early blight on potatoes; however, prior to the development of the PLANT-Plus early blight forecaster none were applicable to South African conditions and cultivars. Field trials, conducted in spring and fall at one location and in fall at another, were used to evaluate the model. These trials compared the efficacy of fungicide programmes applied according to PLANT-Plus with a conventional calendar-based spray programme for control of early blight in South Africa. The PLANT-Plus early blight decision support system reduced the number of fungicide applications without negatively affecting yield or returns. Area under the disease progress curve-values were not significantly different for the PLANT-Plus and conventional treatments, but both were significantly lower than those of the unsprayed control. The PLANT-Plus early blight forecasting model holds great potential for use in the South African potato industry, but must be incorporated into the existing PLANT-Plus late blight forecaster in order to be more cost-effective and have multipurpose applicability.

Phase-Correcting Data Assimilation and Application to Storm-Scale Numerical Weather Prediction. Part II: Application to a Severe Storm Outbreak

Abstract A scheme to correct phase errors in numerical model forecasts using Doppler radar, radiosonde, profiler, and surface data is demonstrated to improve forecasts in a complex severe thunderstorm situation. The technique is designed to directly address forecast phase errors or initial position errors as part of a data assimilation strategy. In the demonstration the phase error correction is applied near the time of initial cell development and the forecast results are compared to the uncorrected forecast and forecasts made using an analysis created at the time of the observations. Forecasts are verified qualitatively for the position of thunderstorm cells and quantitatively for accumulated precipitation. It is shown that the scheme can successfully correct errors in thunderstorm locations and it has a positive influence on the subsequent forecast. The advantage of the phase correction over the control lasts for about 3 h despite storm dissipation and regeneration, and interactions among multiple storms.

Parameterizing Ice Cloud Inhomogeneity and the Overlap of Inhomogeneities Using Cloud Radar Data

Abstract Cloud variability on scales smaller than the gridbox size of numerical forecast and climate models is believed to be important in determining the radiative effects of clouds, and increasingly attempts are being made to parameterize these fluctuations in the radiation schemes of current models. In order to calculate the radiative effects of an inhomogeneous cloud, a model needs to know not only the degree of variability within a gridbox but also the degree to which the inhomogeneities in vertically adjacent levels are overlapped. In this paper these two parameters are derived for ice clouds from an 18-month midlatitude 94-GHz cloud radar dataset and parameterized in terms of horizontal gridbox size (d), the vertical shear of the horizontal wind (s), and the vertical position in the cloud. The vertical decorrelation length Δz0 (i.e., the depth over which the correlation coefficient of either ice water content or optical extinction coefficient in separate vertical levels falls to e−1) is found to be...