Martin Leutbecher

Ensemble forecasting

Numerical weather prediction models as well as the atmosphere itself can be viewed as nonlinear dynamical systems in which the evolution depends sensitively on the initial conditions. The fact that estimates of the current state are inaccurate and that numerical models have inadequacies, leads to forecast errors that grow with increasing forecast lead time. The growth of errors depends on the flow itself. Ensemble forecasting aims at quantifying this flow-dependent forecast uncertainty. The sources of uncertainty in weather forecasting are discussed. Then, an overview is given on evaluating probabilistic forecasts and their usefulness compared with single forecasts. Thereafter, the representation of uncertainties in ensemble forecasts is reviewed with an emphasis on the initial condition perturbations. The review is complemented by a detailed description of the methodology to generate initial condition perturbations of the Ensemble Prediction System (EPS) of the European Centre for Medium-Range Weather Forecasts (ECMWF). These perturbations are based on the leading part of the singular value decomposition of the operator describing the linearised dynamics over a finite time interval. The perturbations are flow-dependent as the linearisation is performed with respect to a solution of the nonlinear forecast model. The extent to which the current ECMWF ensemble prediction system is capable of predicting flow-dependent variations in uncertainty is assessed for the large-scale flow in mid-latitudes.

A Spectral Stochastic Kinetic Energy Backscatter Scheme and Its Impact on Flow-Dependent Predictability in the ECMWF Ensemble Prediction System

Understanding model error in state-of-the-art numerical weather prediction models and representing its impact on flow-dependent predictability remains a complex and mostly unsolved problem. Here, a spectral stochastic kinetic energy backscatter scheme is used to simulate upscale-propagating errors caused by unresolved subgrid-scale processes. For this purpose, stochastic streamfunction perturbations are generated by autoregressive processes in spectral space and injected into regions where numerical integration schemes and parameterizations in the model lead to excessive systematic kinetic energy loss. It is demonstrated how output from coarse-grained high-resolution models can be used to inform the parameters of such a scheme. The performance of the spectral backscatter scheme is evaluated in the ensemble prediction system of the European Centre for Medium-Range Weather Forecasts. Its implementation in conjunction with reduced initial perturbations results in a better spread‐error relationship, more realistic kinetic-energy spectra, a better representation of forecast-error growth, improved flow-dependent predictability, improved rainfall forecasts, and better probabilistic skill. The improvement is most pronounced in the tropics and for largeanomaly events. It is found that whereas a simplified scheme assuming a constant dissipation rate already has some positive impact, the best results are obtained for flow-dependent formulations of the unresolved processes.

Increased stratospheric ozone depletion due to mountain-induced atmospheric waves

Chemical reactions on polar stratospheric cloud (PSC) particles are responsible for the production of reactive chlorine species (chlorine ‘activation’) which cause ozone destruction. Gas-phase deactivation of these chlorine species can take several weeks in the Arctic winter stratosphere, so that ozone destruction can be sustained even in air parcels that encounter PSCs only intermittently,. Chlorine activation during a PSC encounter proceeds much faster at low temperatures when cloud particle surface area and heterogeneous reaction rates are higher. Although mountain-induced atmospheric gravity waves are known to cause local reductions in stratospheric temperature of as much as 10–15 K (refs 5-9), and are often associated with mesoscale PSCs, their effect on chlorine activation and ozone depletion has not been considered. Here we describe aircraft observations of mountain-wave-induced mesoscale PSCs in which temperatures were 12 K lower than expected synoptically. Model calculations show that despite their localized nature, these PSCs can cause almost complete conversion of inactive chlorine species to ozone-destroying forms in air flowing through the clouds. Using a global mountain-wave model, we identify regions where mountain waves can develop, and show that they can cause frequent chlorine activation of air in the Arctic stratosphere. Such mesoscale processes offer a possible explanation for the underprediction of reactive chlorine concentrations and ozone depletion rates calculated by three-dimensional models of the Arctic stratosphere.

Particle microphysics and chemistry in remotely observed mountain polar stratospheric clouds

Polar stratospheric clouds (PSCs) at 22–26 km were observed over the Norwegian mountains by airborne lidar on January 15, 1995. Simulations using a mesoscale model reveal that they were caused by mountain-induced gravity waves. The clouds had a highly detailed filamentary structure with bands as thin as 100 m in the vertical, and moved insignificantly over 4 hours, suggesting them to be quasi-stationary. The aircraft flight path was parallel or close to parallel with the wind at cloud level. Such a quasi-Lagrangian observation, together with the presence of distinct aerosol layers, allows an air parcel trajectory through the cloud to be constructed and enables the lidar images to be simulated using a microphysical box model and light scattering calculations. The results yield detailed information about particle evolution in PSCs and suggest that water ice nucleated directly from liquid HNO3/H2SO4/H2O droplets as much as 4 K below the ice frost point. The observation of solid nitric acid hydrate particles downwind of the mountains shows that such mesoscale events can generate solid PSC particles that can persist on the synoptic scale. We also draw attention to the possible role of mesoscale PSCs in chlorine activation and subsequent ozone destruction.

Relevance of mountain wave cooling for the formation of polar stratospheric clouds over Scandinavia: Mesoscale dynamics and observations for January 1997

The effect of mesoscale mountain wave-induced temperature anomalies on the formation potential of polar stratospheric clouds above northern Scandinavia is analyzed with a one-month mesoscale model integration. The simulation results are contrasted with synoptic-scale analyses and compared with remote sensing and in situ observations. The mesoscale mass flux of air parcels with temperatures below the threshold for cloud formation through a control volume is compared with its synoptic-scale counterpart. A classification of the synoptic-scale flow into periods of large and small mountain wave activity in the stratosphere is proposed. The derived classification will be used for a climatology of stratospheric mountain wave activity above Scandinavia.

Model-guided Lagrangian observation and simulation of mountain polar stratospheric clouds

Gravity-wave-induced polar stratospheric clouds (PSCs) were observed over the Scandinavian mountains by airborne lidar on January 9, 1997. Guided by the forecasts of a mesoscale dynamical model, a flight path was chosen to lead through the coldest predicted region parallel to the wind at the expected PSC level (23–26 km). Because of the nearly stationary nature of the wave-induced PSC the individual filaments visible in the backscatter data of the clouds can be interpreted as air parcel trajectories. Assuming dry adiabatic behavior and fixing the absolute temperature to the ice frost point in the ice part of the cloud enables detailed microphysical simulations of the whole life cycle of the cloud particles. Optical calculations are used to adjust open parameters in the microphysical model by optimizing the agreement with the multichannel lidar data. This case is compared with former work from the Arctic winter 1994/1995. The influence of the stratospheric H2SO4 content and the cooling rate on the type of cloud particles (liquid ternary solution droplets or solid nitric acid hydrates) released from the ice part of the cloud is evaluated.

Optimization of interference filters with genetic algorithms applied to silver-based heat mirrors

In the optimization of multilayer stacks for various optical filtering purposes not only the thicknesses of the thin films are to be optimized, but also the sequence of materials. Materials with very different optical properties, such as metals and dielectrics, may be combined. A genetic algorithm is introduced to search for the optimal sequence of materials along with their optical thicknesses. This procedure is applied to a heat mirror in combination with a blackbody absorber for thermal solar energy applications at elevated temperatures (250 °C). The heat mirror is based on silver films with antireflective dielectric layers. Seven dielectrics have been considered. For a five-layer stack the sequence (TiO(2)/Ag/TiO(2)/Ag/Y(2)O(3)) is found to be optimal.

A Reduced Rank Estimate of Forecast Error Variance Changes due to Intermittent Modifications of the Observing Network

A new tool for planning adaptive observations is introduced. Different modifications of the observing network can be compared prior to the time of modification. The method predicts the variance of forecast errors projected into a low-dimensional subspace. Tangent-linear error evolution is assumed and the contribution of model errors to the forecast error is neglected. Singular vectors of the propagator of the tangent-linear version of the forecast model are used to define a relevant subspace. The method employs the Hessian of the cost function of a variational assimilation scheme to obtain information on the distribution of initial errors. Thus, this technique for planning adaptive observations can be made consistent with operational variational assimilation schemes. The application of the method is currently limited to intermittent modifications of the observing network as changes of the background error distribution due to modifications of the network in previous assimilation cycles are not accounted for. The predicted changes of forecast error variance are identical to those that the ensemble transform Kalman filter method would yield if applied to a set of Hessian singular vectors. The reduced rank estimate has been implemented in the Integrated Forecasting System of the European Centre for Medium-Range Weather Forecasts. To illustrate the scope of the method, it is applied to the 2-day forecast of an extratropical cyclone. The expected reduction of the total energy of forecast error is computed for various hypothetical adaptive networks that differ by spatial coverage, observation density, and the type of observation.

Relevance of mountain waves for the formation of polar stratospheric clouds over Scandinavia: A 20 year climatology

A climatology of meteorological conditions necessary for the existence of polar stratospheric clouds (PSCs) over Scandinavia is presented. The frequency of low enough temperatures for synoptic-scale PSCs is compared with the frequency of additional events when mesoscale PSCs might be induced by the adiabatic cooling in mountain waves. The climatology is based on European Centre for Medium-Range Weather Forecasts analyses of the 20 winter seasons 1979-1980 to 1998-1999 and a simple parameterization of mesoscale stratospheric mountain wave activity. The synoptic-scale formation potential of stratospheric clouds is determined as the fraction of time when the stratospheric minimum temperature is less than the thresholds T NAT (PSC of type I) and T frost (PSC of type II). The potential of additional mesoscale formation is computed by reducing the synoptic-scale temperature on dates when stratospheric mountain wave activity is predicted. This simple approach allows an estimate of the climatological relevance of mountain waves for the existence conditions of PSCs above Scandinavia. In the climatological mean, maximum stratospheric mountain wave activity is found in January, and the maximum seasonal winter average for various levels amounts to 11% of all analysis times (i.e., approximately 3 days per month). Based on the temperature analyses the existence conditions for PSCs of type I are dominated by synoptic-scale processes (maximum 36% of all analysis times are conducive to their existence) whereas the additional mesoscale fraction is less than 6%. Suitable conditions of ice PSCs occur less frequently. It is found that episodes of temperatures low enough for the existence of PSC of type II are controlled by mesoscale temperature anomalies induced by stratospheric mountain waves; on the synoptic scale the stratosphere above Scandinavia is almost always warmer than the threshold T frost .

The Propagation of Mountain Waves into the Stratosphere: Quantitative Evaluation of Three-Dimensional Simulations

On 6 January 1992 measurements of a mountain wave with significant amplitude were taken over the southern tip of Greenland during an ER-2 flight at an altitude of about 20 km. This work focuses on 3D numerical simulations of the wave generation and its propagation into the stratosphere during this event. The sensitivity of the simulated mountain wave to surface friction and horizontal resolution is explored. A nonhydrostatic model is used for experiments with horizontal resolutions of 12, 4, and 1.3 km. In all simulations the flow over the southern tip of Greenland generates a mountain wave, which propagates into the stratosphere. Changes of surface friction and horizontal resolution affect mostly the amplitude of the mountain wave. Increasing surface friction on the slopes reduces the amplitude of the excited orographic gravity wave. Horizontal diffusion required for numerical stability attenuates gravity waves during their propagation into the stratosphere. Increasing the horizontal resolution permits a smaller diffusion and thereby results in larger stratospheric wave amplitudes. The experiment with increased surface friction at 1.3-km horizontal resolution shows the best agreement with the observational data of the wave in the stratosphere. The differences between the simulated and measured amplitudes of vertical displacement and temperature anomaly are less than about 20%. The disparity in vertical velocity is larger ; downward velocities were observed up to 4.8 m s21 and simulated up to 2.7 m s21. In the experiments with lower surface friction at 4-km resolution, the accuracy regarding the amplitude of vertical displacement and temperature anomalies is similar, but the simulated maximum downdraft is even weaker. The other experiments with increased surface friction at 4-km resolution and normal friction at 12-km resolution significantly underestimate the wave amplitude. The results of the experiments suggest that the generation of orographic gravity waves and their propagation into the stratosphere can be simulated in three dimensions in a realistic manner provided that the magnitude of the parameterized surface friction is in a realistic range and the horizontal resolution is sufficient.