Abstract

In a strongly coupled data assimilation (DA), a cross-fluid covariance is specified that allows measurements from a coupled fluid (e.g., atmosphere) to directly impact analysis increments in a target fluid (e.g., ocean). The exhaustive solution to this coupled DA problem calls for a covariance where all available measurements can influence all grid points in all fluids. Solution of such a large algebraic problem is computationally expensive, often calls for a substantial rewrite of existing fluid-specific DA systems, and, as shown in this paper, can be avoided.

The proposed interface solver assumes that covariances between coupled measurements and target fluid are often close to null (e.g., between stratospheric observations and the deep ocean within a 6-h forecast cycle). In the interface solver, two separate DA solvers are run in parallel: one that produces an analysis solution in the atmosphere, and one in the ocean. Each system uses a coupled observation vector where in addition to resident measurements in the target fluid it also includes nonresident measurements in the coupled fluid that are likely to have significant influence on the analysis in the target fluid (interface measurements). An ensemble-based method is employed and a localization function for coupled ensembles is proposed. Using a coupled model for the Mediterranean Sea (in a twin setting), it is demonstrated that (i) the solution of the interface solver converges to the exhaustive solution and (ii) that in presence of poorly known error covariances, the interface solver can be configured to produce a more accurate solution than an exhaustive solver.

Abstract

This paper examines the sensitivity of short-term forecasts of the western North Pacific subtropical high (WNPSH) and rainfall to sea surface temperature (SST) uncertainty using the Coupled Ocean–Atmosphere Mesoscale Prediction System (COAMPS). A comparison of analyzed SSTs with satellite observations of SST indicates that SST analysis errors are particularly pronounced on horizontal scales from 100 to 200 km, similar to the mesoscale eddy scales in the Kuroshio region. Since significant oceanic variations occur on these scales, it is of interest to examine the effects of representing this small-scale uncertainty with random, scale-dependent perturbations. An SST ensemble perturbation generation technique is used here that enables temporal and spatial correlations to be controlled and produces initial SST fields comparable to satellite observations. The atmospheric model develops large uncertainty in the Korea and Japan area due to the fluctuation in the horizontal pressure gradient caused by the location of the WNPSH. This, in turn, increases the variance of the low-level jet (LLJ) over southeast China, resulting in large differences in the moist transport flux from the tropical ocean and subsequent rainfall. Validation using bin-mean statistics shows that the ensemble forecast with the perturbed SST better distinguishes large forecast error variance from small forecast error variance. The results suggest that using the SST perturbation as a proxy for the ocean ensemble in a coupled atmosphere and ocean ensemble system is feasible and computationally efficient.

Abstract

An ensemble Kalman filter (EnKF) has been adopted and implemented at the Naval Research Laboratory (NRL) for mesoscale and storm-scale data assimilation to study the impact of ensemble assimilation of high-resolution observations, including those from Doppler radars, on storm prediction. The system has been improved during its implementation at NRL to further enhance its capability of assimilating various types of meteorological data. A parallel algorithm was also developed to increase the system’s computational efficiency on multiprocessor computers. The EnKF has been integrated into the NRL mesoscale data assimilation system and extensively tested to ensure that the system works appropriately with new observational data stream and forecast systems. An innovative procedure was developed to evaluate the impact of assimilated observations on ensemble analyses with no need to exclude any observations for independent validation (as required by the conventional evaluation based on data-denying experiments). The procedure was employed in this study to examine the impacts of ensemble size and localization on data assimilation and the results reveal a very interesting relationship between the ensemble size and the localization length scale. All the tests conducted in this study demonstrate the capabilities of the EnKF as a research tool for mesoscale and storm-scale data assimilation with potential operational applications.

Abstract

In this study, we use observational and numerical model data from the Coupled Air Sea Processes and Electromagnetic Ducting Research (CASPER) field campaign to describe the mean refractive conditions offshore Duck, North Carolina. The U.S. Navy operational numerical weather prediction model known as the Coupled Ocean–Atmosphere Mesoscale Prediction System (COAMPS) performed well forecasting large-scale conditions during the experiment, with an observed warm bias in SST and cold and dry biases in temperature and humidity in the lowest 2000 m. In general, COAMPS underpredicted the number of ducts, and they were weaker and at lower height than those seen in observations. It was found that there is a noticeable diurnal evolution of the ducts, more over land than over the ocean. Ducts were found to be more frequent over land but overall were stronger and deeper over the ocean. Also, the evaporative duct height increases as one moves offshore. A case study was chosen to describe the electromagnetic properties under different synoptic conditions. In this case the continental atmospheric boundary layer dominates and interacts with the marine atmospheric boundary layer. As a result, the latter moves around 80 km offshore and then back inland after 2 h.

This report discusses the design and implementation of a specialized forecasting system that was set up to support the observational component of the Labrador Sea Deep Convection Experiment. This ongoing experiment is a multidisciplinary program of observations, theory, and modeling aimed at improving our knowledge of the deep convection process in the ocean, and the air–sea interaction that forces it. The observational part of the program was centered around a cruise of the R/V Knorr during winter 1997, as well as several complementary meteorological research flights. To aid the planning of ship and aircraft operations a specially tailored mesoscale model was run over the Labrador Sea, with the model output postprocessed and transferred to a remote field base. The benefits of using a warm-start analysis cycle in the model are discussed. The utility of the forecasting system is illustrated through a description of the flight planning process for several cases. The forecasts proved to be invaluable both in ship operations and in putting the aircraft in the right place at the right time. In writing this narrative the authors hope to encourage the use of similar forecasting systems in the support of future field programs, something that is becoming increasingly possible with the rise in real-time numerical weather prediction.

Abstract

In support of the Department of Defense's Gulf War Illness study, the Naval Research Laboratory (NRL) has performed global and mesoscale meteorological reanalyses to provide a quantitative atmospheric characterization of the Persian Gulf region during the period between 15 January and 15 March 1991. This paper presents a description of the mid- to late-winter synoptic conditions, mean statistical scores, and near-surface mean conditions of the Gulf War theater drawn from the 2-month reanalysis.

The reanalysis is conducted with the U.S. Navy's operational global and mesoscale analysis and prediction systems: the Navy Operational Global Atmospheric Prediction System (NOGAPS) and the Coupled Ocean–Atmosphere Mesoscale Prediction System (COAMPS*). The synoptic conditions for the 2-month period can be characterized as fairly typical for the northeast monsoon season, with only one significant precipitation event affecting the Persian Gulf region.

A comparison of error statistics to those from other mesoscale models with similar resolution covering complex terrains (though in different geographic locations) is performed. Results indicate similar if not smaller error statistics for the current study even though this 2-month reanalysis is conducted in an extremely data-sparse area, lending credence to the reanalysis dataset.

The mean near-surface conditions indicate that variability in the wind and temperature fields arises mainly because of the differential diurnal processes in the region characterized by complex surface characteristics and terrain height. The surface wind over lower elevation, interior, land regions is mostly light and variable, especially in the nocturnal surface layer. The strong signature of diurnal variation of sea–land as well as lake–land circulation is apparent, with convergence over the water during the night and divergence during the day. Likewise, the boundary layer is thus strongly modulated by the diurnal cycle near the surface. The low mean PBL height and light mean winds combine to yield very low ventilation efficiency over the Saudi and Iraqi plains.

Abstract

Two extreme heat events impacting the New York City (NYC), New York, metropolitan region during 7–10 June and 21–24 July 2011 are examined in detail using a combination of models and observations. The U.S. Navy's Coupled Ocean–Atmosphere Mesoscale Prediction System (COAMPS) produces real-time forecasts across the region on a 1-km resolution grid and employs an urban canopy parameterization to account for the influence of the city on the atmosphere. Forecasts from the National Weather Service's 12-km resolution North American Mesoscale (NAM) implementation of the Weather Research and Forecasting (WRF) model are also examined. The accuracy of the forecasts is evaluated using a land- and coastline-based observation network. Observed temperatures reached 39°C or more at central urban sites over several days and remained high overnight due to urban heat island (UHI) effects, with a typical nighttime urban–rural temperature difference of 4°–5°C. Examining model performance broadly over both heat events and 27 sites, COAMPS has temperature RMS errors averaging 1.9°C, while NAM has RMSEs of 2.5°C. COAMPS high-resolution wind and temperature predictions captured key features of the observations. For example, during the early summer June heat event, the Long Island south shore coastline experienced a more pronounced sea breeze than was observed for the July heat wave.

Abstract

Numerical simulations are conducted using the Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS) to investigate the impact of land–vegetation processes on the prediction of mesoscale convection observed on 24–25 May 2002 during the International H₂O Project (IHOP_2002). The control COAMPS configuration uses the Weather Research and Forecasting (WRF) model version of the Noah land surface model (LSM) initialized using a high-resolution land surface data assimilation system (HRLDAS). Physically consistent surface fields are ensured by an 18-month spinup time for HRLDAS, and physically consistent mesoscale fields are ensured by a 2-day data assimilation spinup for COAMPS. Sensitivity simulations are performed to assess the impact of land–vegetative processes by 1) replacing the Noah LSM with a simple slab soil model (SLAB), 2) adding a photosynthesis, canopy resistance/transpiration scheme [the gas exchange/photosynthesis-based evapotranspiration model (GEM)] to the Noah LSM, and 3) replacing the HRLDAS soil moisture with the National Centers for Environmental Prediction (NCEP) 40-km Eta Data Assimilation (EDAS) operational soil fields.

CONTROL, EDAS, and GEM develop convection along the dryline and frontal boundaries 2–3 h after observed, with synoptic-scale forcing determining the location and timing. SLAB convection along the boundaries is further delayed, indicating that detailed surface parameterization is necessary for a realistic model forecast. EDAS soils are generally drier and warmer than HRLDAS, resulting in more extensive development of convection along the dryline than for CONTROL. The inclusion of photosynthesis-based evapotranspiration (GEM) improves predictive skill for both air temperature and moisture. Biases in soil moisture and temperature (as well as air temperature and moisture during the prefrontal period) are larger for EDAS than HRLDAS, indicating land–vegetative processes in EDAS are forced by anomalously warmer and drier conditions than observed. Of the four simulations, the errors in SLAB predictions of these quantities are generally the largest.

By adding a sophisticated transpiration model, the atmospheric model is able to better respond to the more detailed representation of soil moisture and temperature. The sensitivity of the synoptically forced convection to soil and vegetative processes including transpiration indicates that detailed representation of land surface processes should be included in weather forecasting models, particularly for severe storm forecasting where local-scale information is important.

Abstract

A computationally inexpensive ensemble transform (ET) method for generating high-resolution initial perturbations for regional ensemble forecasts is introduced. The method provides initial perturbations that (i) have an initial variance consistent with the best available estimates of initial condition error variance, (ii) are dynamically conditioned by a process similar to that used in the breeding technique, (iii) add to zero at the initial time, (iv) are quasi-orthogonal and equally likely, and (v) partially respect mesoscale balance constraints by ensuring that each initial perturbation is a linear sum of forecast perturbations from the preceding forecast. The technique is tested using estimates of analysis error variance from the Naval Research Laboratory (NRL) Atmospheric Variational Data Assimilation System (NAVDAS) and the Navy’s regional Coupled Ocean–Atmosphere Mesoscale Prediction System (COAMPS) over a 3-week period during the summer of 2005. Lateral boundary conditions are provided by a global ET ensemble. The tests show that the ET regional ensemble has a skillful mean and a useful spread–skill relationship in mass, momentum, and precipitation variables. Diagnostics indicate that ensemble variance was close to, but probably a little less than, the forecast error variance for wind and temperature variables, while precipitation ensemble variance was significantly smaller than precipitation forecast error variance.

Abstract

Results are presented from a tracer-release modeling study designed to examine atmospheric transport and dispersion (“T&D”) behavior surrounding the complex coastal–urban region of New York City, New York, where air–sea interaction and urban influences are prominent. The puff-based Hazard Prediction Assessment Capability (HPAC, version 5) model is run for idealized conditions, and it is also linked with the urbanized COAMPS (1 km) meteorological model and the NAM (12 km) meteorological model. Results are compared with “control” plumes utilizing surface meteorological input from 22 weather stations. In all configurations, nighttime conditions result in plume predictions that are more sensitive to small changes in wind direction. Plume overlap is reduced by up to 70% when plumes are transported during the night. An analysis of vertical plume cross sections and the nature of the underlying transport and the dispersion equations both suggest that heat flux gradients and boundary layer height gradients determine vertical transport of pollutants across land–sea boundaries in the T&D model. As a consequence, in both idealized and realistic meteorological configurations, waterfront releases generate greater plume discrepancies relative to plumes transported over land/urban surfaces. For transport over water (northwest winds), the higher-fidelity meteorological model (COAMPS) generated plumes with overlap reduced by about one-half when compared with that of the coarser-resolution NAM model (13% vs 24% during the daytime and 11% vs 18% during the nighttime). This study highlights the need for more sophisticated treatment of land–sea transition zones in T&D calculations covering waterside releases.