Abstract

Potential vorticity streamers (PVSs) are elongated filaments of high-PV air near the tropopause. In the warm season, anticyclonic Rossby wave breaking (AWB) produces enhanced PVS activity, which in turn modifies the equatorward tropical environment by enhancing vertical wind shear (VWS). This enhanced VWS can play an important role in suppressing nearby tropical cyclone (TC) activity. Given the important role that PVSs play in modifying their local environment, forecasts of PVS activity on subseasonal time scales may also influence forecasts of TC activity. This study uses Navy Earth System Prediction Capability (Navy ESPC) 45-day forecasts initialized during boreal summer 2009–15 to investigate subseasonal predictability of PVSs and TCs in the North Atlantic. PVSs are identified on the 350-K isentropic surface bounded by the 2 PV unit (PVU; 1 PVU = 10⁻⁶ K kg⁻¹ m² s⁻¹) contour and defined as the high-PV trough axis downstream of the AWB axis. TCs are identified in the forecasts using the TempestExtremes detection algorithm that tracks warm-core lows. PVS and TC activity metrics that sum the number and intensity of events for a given time period are also computed. We first use skill scores and mean-state biases to determine the typical predictability of PVS activity, and then subselect high- and low-PVS-activity forecasts to determine how PVS forecast errors impact TC activity forecast errors. Results show that PVS activity can modulate TC activity at subseasonal time scales, with over-forecasted PVS activity corresponding to underestimated forecasts of TC activity and vice versa. This inverse correlation is consistent with enhanced VWS occurring equatorward of PVS troughs in the high-PVS forecasts.

Abstract

The impact of stochastic convection on ensembles produced using the ensemble transform (ET) initial perturbation scheme is examined. This note compares the behavior of ensemble forecasts based only on initial ET perturbations with the behavior of ensemble forecasts based on the ET initial perturbations and forecasts that include stochastic convection. It is illustrated that despite the fact that stochastic convection occurs only after the forecast integrations have started, it induces changes in the initial perturbations as well. This is because the ET is a “cycling” scheme, in which previous short-term forecasts are used to produce the initial perturbations for the current forecast. The stochastic convection scheme induces rapid perturbation growth in regions where convection is active, primarily in the tropics. When combined with the ET scheme, this results in larger initial perturbation variance in the tropics, and, because of a global constraint on total initial perturbation variance, smaller initial perturbation variance in the extratropics. Thus, the inclusion of stochastic convection helps to mitigate a problem found in the practical implementation of the ET, namely, that of too little initial variance in the tropics and too much in the extratropics. Various skill scores show that stochastic convection improves ensemble performance in the tropics, with little impact to modest improvement in the extratropics. Experiments performed using the initial perturbations from the control ensemble run but forecast integrations using the stochastic convection scheme indicate that the improved performance of the stochastic convection ensemble at early forecast times is due to both “indirect” changes in the initial perturbations and “direct” changes in the forecast. At later forecast times, it appears that most of the improvement can be gained through stochastic convection alone.

Abstract

The ensemble transform (ET) scheme changes forecast perturbations into analysis perturbations whose amplitudes and directions are consistent with a user-provided estimate of analysis error covariance. A practical demonstration of the ET scheme was undertaken using Naval Research Laboratory (NRL) Atmospheric Variational Data Assimilation System (NAVDAS) analysis error variance estimates and the Navy Operational Global Atmospheric Prediction System (NOGAPS) numerical weather prediction (NWP) model. It was found that the ET scheme produced forecast ensembles that were comparable to or better in a variety of measures than those produced by the Fleet Numerical and Oceanography Center (FNMOC) bred-growing modes (BGM) scheme. Also, the demonstration showed that the introduction of stochastic perturbations into the ET forecast ensembles led to a substantial improvement in the agreement between the ET and NAVDAS analysis error variances. This finding is strong evidence that even a small-sized ET ensemble is capable of obtaining good agreement between the ET and NAVDAS analysis error variances, provided that NWP model deficiencies are accounted for. Last, since the NAVDAS analysis error covariance estimate is diagonal and hence ignores multivariate correlations, it was of interest to examine the ET analysis perturbations’ spatial correlation. Tests showed that the ET analysis perturbations exhibited statistically significant, realistic multivariate correlations.

Abstract

Singular vectors (SVs) are used to study the sensitivity of 2-day forecasts of recurving tropical cyclones (TCs) in the western North Pacific to changes in the initial state. The SVs are calculated using the tangent and adjoint models of the Navy Operational Global Atmospheric Prediction System (NOGAPS) for 72 forecasts for 18 TCs in the western North Pacific during 2006. In addition to the linear SV calculation, nonlinear perturbation experiments are also performed in order to examine 1) the similarity between nonlinear and linear perturbation growth and 2) the downstream impacts over the North Pacific and North America that result from changes to the 2-day TC forecast. Both nonrecurving and recurving 2-day storm forecasts are sensitive to changes in the initial state in the near-storm environment (in an annulus approximately 500 km from the storm center). During recurvature, sensitivity develops to the northwest of the storm, usually associated with a trough moving in from the west. These upstream sensitivities can occur as far as 4000 km to the northwest of the storm, over the Asian mainland, which has implications for adaptive observations. Nonlinear perturbation experiments indicate that the linear calculations reflect case-to-case variability in actual nonlinear perturbation growth fairly well, especially when the growth is large. The nonlinear perturbations show that for recurving tropical cyclones, small initial perturbations optimized to change the 2-day TC forecast can grow and propagate downstream quickly, reaching North America in 5 days. The fastest 5-day perturbation growth is associated with recurving storm forecasts that occur when the baroclinic instability over the North Pacific is relatively large. These results suggest that nonlinear forecasts perturbed using TC SVs may have utility for predicting the downstream impact of TC forecast errors over the North Pacific and North America.

Abstract

The impacts of assimilating dropwindsonde data and enhanced atmospheric motion vectors (AMVs) on tropical cyclone track forecasts are examined using the Navy global data assimilation and forecasting systems. Enhanced AMVs have the largest impact on eastern Pacific storms, while dropwindsonde data have the largest impact on Atlantic storms. Results in the western Pacific are mixed. Two western Pacific storms, Nuri and Jangmi, are examined in detail. For Nuri, dropwindsonde data and enhanced AMVs are at least as likely to degrade as to improve forecasts. For Jangmi, additional data improve track forecasts in most cases. An erroneous weakening of the forecasted subtropical high appears to contribute to the track errors for Nuri and Jangmi. Assimilation of enhanced AMVs systematically increases the analyzed heights in this region, counteracting this model bias. However, the impact of enhanced AMVs decreases rapidly as the model biases saturate at similar levels for experiments with and without the enhanced AMVs after the first few forecast days. Experiments are also conducted in which the errors assigned to synthetic tropical cyclone observations are increased. Moderate increases in the assigned errors improve track forecasts on average, but larger increases in the assigned errors produce mixed results. Both experiments allow for reductions in innovations and residuals when compared to dropwindsonde observations. These experiments suggest that a reformulation of the synthetic tropical cyclone observation scheme may lead to improved forecasts as more in situ and remote observations become available.

Abstract

The initial-state sensitivity and interactions between a tropical cyclone and atmospheric equatorial Kelvin waves associated with the Madden–Julian oscillation (MJO) during the DYNAMO field campaign are explored using adjoint-based tools from the Coupled Ocean–Atmosphere Mesoscale Prediction System (COAMPS). The development of Tropical Cyclone 5 (TC05) coincided with the passage of an equatorial Kelvin wave (KW) and westerly wind burst associated with an MJO that developed in the Indian Ocean in late November 2011. COAMPS 18-h adjoint sensitivities of low-level kinetic energy to changes in initial state winds, temperature, and water vapor are analyzed for both TC05 and the KW to document when the evolution of each system is sensitive to the other. Time series of sensitivity patterns confirm that TC05 and the KW low-level westerlies are sensitive to each other when the KW is to the southwest and south of TC05. While TC05 is not sensitive to the KW after this, the KW low-level westerlies remain sensitive to TC05 until it enters the far eastern Indian Ocean. Vertical profiles of both TC05 and KW sensitivity indicate lower-tropospheric maxima in temperature, wind, and moisture, with KW sensitivity typically 20% smaller than TC05 sensitivity. The magnitude of the sensitivity for both systems is greatest just prior to, and during, their closest proximity. A case study examination reveals that adjoint-based optimal perturbations grow and expand quickly through a dynamic response to decreased static stability. The evolution of moist-only and dry-only initial perturbations illustrates that the moist component is primarily responsible for the initial rapid growth, but that subsequent growth rates are similar.

Abstract

The three-dimensional structure of random error growth in the National Meteorological Center's Medium-Range Forecast Model is investigated in an effort to identify the sources of error growth. The random error growth is partitioned into two types: external error growth, which is due to model deficiencies, and internal error growth, which is the self-growth of errors in the initial conditions. Forecasts from winter 1987, summer 1990, and winter 1992 are compared to assess seasonal variations in regional error growth as well as forecast model improvement. The following is found:

1. In the tropics, large external error growth at the 200-mb level is closely associated with deep convection. There is evidence of significant model improvements in the tropics at the 850-mb level between 1987 and 1992.

2. The spatial structure of the external error growth in the midlatitudes suggests that the representation of orography in the model, especially over Antarctica and the Rockies, is a significant source of errors.

3. Internal error growth in the midlatitudes is greater over the Atlantic and European regions than over the Pacific region and appears to be associated with blocking phenomena, especially over the North Atlantic and Europe. The Northern Hemisphere exhibits a seasonal cycle in the magnitude of error growth, but the Southern Hemisphere does not.

The results for the external and internal error growth rates were obtained using a parameterization of the correlation between forecasts and the verifying analyses. The parameterization is based on the assumption that linear random error growth is caused primarily by model deficiencies, and the validity of this assumption is examined. The results suggest that, in the tropics, significant increases in forecast skill may be obtainable through both model and analysis improvement. In the midlatitudes, however, there is less potential for increases in forecast skill through model improvement, and decreasing the analysis error becomes more important. The parameterization yields results that are physically meaningful and in agreement with previous predictability studies, and that provide quantitative estimates of the spatial and temporal distribution of the sources of forecast errors.

Abstract

The sensitivity of tropical cyclogenesis and subsequent intensification is explored by applying small perturbations to the initial state in the presence of organized mesoscale convection and synoptic-scale forcing using the adjoint and tangent linear models for the Coupled Ocean–Atmosphere Mesoscale Prediction System (COAMPS). The forward, adjoint, and tangent linear models are used to compare and contrast predictability characteristics for the disturbance that became Typhoon Nuri and a nondeveloping organized convective cluster in the western Pacific during The Observing System Research and Predictability Experiment (THORPEX) Pacific Asian Regional Campaign (T-PARC) and the Tropical Cyclone Structure-2008 (TCS-08) experiments.

The adjoint diagnostics indicate that the intensity (e.g., maximum surface wind speed, minimum surface pressure) of a tropical disturbance is very sensitive to perturbations in the moisture and temperature fields and to a lesser degree the wind fields. The highest-resolution adjoint simulations (grid increment of 13 km) indicate that the most efficient intensification is through moistening in the lower and middle levels and heating in banded regions that are coincident with vorticity maxima in the initial state. Optimal adjoint perturbations exhibit rapid growth for the Nuri case and only modest growth for the nondeveloping system. The adjoint results suggest that Nuri was near the threshold for development, indicative of low predictability. The low-level sensitivity maximum and tendency for optimal perturbation growth to extend vertically through the troposphere are consistent with a “bottom up” development process of TC genesis, although a secondary midlevel sensitivity maximum is present as well. Growth originates at small scales and projects onto the scale of the vortex, a manifestation of perturbations that project onto organized convection embedded in regions of cyclonic vorticity.

Abstract

An exact closed form expression for the infinite time analysis and forecast error covariances of a Kalman filter is used to investigate how the locations of fixed observing platforms such as radiosonde stations affect global distributions of analysis and forecast error variance. The solution pertains to a system with no model error, time-independent nondefective unstable dynamics, time-independent observation operator, and time-independent observation error covariance. As far as the authors are aware, the solutions are new. It is shown that only nondecaying normal modes (eigenvectors of the dynamics operator) are required to represent the infinite time error covariance matrices. Consequently, once a complete set of nondecaying eigenvectors has been obtained, the solution allows for the rapid assessment of the error-reducing potential of any observational network that bounds error variance.

Atmospherically relevant time-independent basic states and their corresponding tangent linear propagators are obtained with the help of a (T21L3) quasigeostrophic global model. The closed form solution allows for an examination of the sensitivity of the error variances to many different observing configurations. It is also feasible to determine the optimal location of one additional observation given a fixed observing network, which, through repetition, can be used to build effective observing networks.

Effective observing networks result in error variances several times smaller than other types of networks with the same number of column observations, such as equally spaced or land-based networks. The impact of the observing network configuration on global error variance is greater when the observing network is less dense. The impact of observations at different pressure levels is also examined. It is found that upper-level observations are more effective at reducing globally averaged error variance, but midlevel observations are more effective at reducing forecast error variance at and downstream of the baroclinic regions associated with midlatitude jets.

Abstract

The initial state sensitivity of high-impact extratropical cyclones over the North Atlantic and United Kingdom is investigated using an adjoint modeling system that includes moist processes. The adjoint analysis indicates that the 48-h forecast of precipitation and high winds associated with the extratropical cyclone “Desmond” was highly sensitive to mesoscale regions of moisture at the initial time. Mesoscale moisture and potential vorticity structures along the poleward edge of an atmospheric river at the initialization time had a large impact on the development of Desmond as demonstrated with precipitation, kinetic energy, and potential vorticity response functions. Adjoint-based optimal perturbations introduced into the initial state exhibit rapidly growing amplitudes through moist energetic processes over the 48-h forecast. The sensitivity manifests as an upshear-tilted structure positioned along the cold and warm fronts. Perturbations introduced into the nonlinear and tangent linear models quickly expand vertically and interact with potential vorticity anomalies in the mid- and upper levels. Analysis of adjoint sensitivity results for the winter 2013/14 show that the moisture sensitivity magnitude at the initial time is well correlated with the kinetic energy error at the 36-h forecast time, which supports the physical significance and importance of the mesoscale regions of high moisture sensitivities.