Forecast uncertainty—ex Ante and ex Post: U.S. inflation and output growth

Survey respondents who make point predictions and histogram forecasts of macro-variables reveal both how uncertain they believe the future to be, ex ante, as well as their ex post performance. Macroeconomic forecasters tend to be overconfident at horizons of a year or more, but overestimate (i.e., are underconfident regarding) the uncertainty surrounding their predictions at short horizons. Ex ante uncertainty remains at a high level compared to the ex post measure as the forecast horizon shortens. There is little evidence of a link between individuals’ ex post forecast accuracy and their ex ante subjective assessments.

Robust approaches to forecasting

We investigate alternative robust approaches to forecasting, using a new class of robust devices, contrasted with equilibrium-correction models. Their forecasting properties are derived facing a range of likely empirical problems at the forecast origin, including measurement errors, impulses, omitted variables, unanticipated location shifts and incorrectly included variables that experience a shift. We derive the resulting forecast biases and error variances, and indicate when the methods are likely to perform well. The robust methods are applied to forecasting US GDP using autoregressive models, and also to autoregressive models with factors extracted from a large dataset of macroeconomic variables. We consider forecasting performance over the Great Recession, and over an earlier more quiescent period.

Estimate of the cutoff errors in the Ewald summation for dipolar systems

Theoretical estimates for the cutoff errors in the Ewald summation method for dipolar systems are derived. Absolute errors in the total energy, forces and torques, both for the real and reciprocal space parts, are considered. The applicability of the estimates is tested and confirmed in several numerical examples. We demonstrate that these estimates can be used easily in determining the optimal parameters of the dipolar Ewald summation in the sense that they minimize the computation time for a predefined, user set, accuracy.

Evaluation of boundary-layer type in a weather forecast model utilising long-term Doppler lidar observations

Many studies evaluating model boundary-layer schemes focus either on near-surface parameters or on short-term observational campaigns. This reflects the observational datasets that are widely available for use in model evaluation. In this paper we show how surface and long-term Doppler lidar observations, combined in a way to match model representation of the boundary layer as closely as possible, can be used to evaluate the skill of boundary-layer forecasts. We use a 2-year observational dataset from a rural site in the UK to evaluate a climatology of boundary layer type forecast by the UK Met Office Unified Model. In addition, we demonstrate the use of a binary skill score (Symmetric Extremal Dependence Index) to investigate the dependence of forecast skill on season, horizontal resolution and forecast leadtime. A clear diurnal and seasonal cycle can be seen in the climatology of both the model and observations, with the main discrepancies being the model overpredicting cumulus capped and decoupled stratocumulus capped boundary-layers and underpredicting well mixed boundary-layers. Using the SEDI skill score the model is most skillful at predicting the surface stability. The skill of the model in predicting cumulus capped and stratocumulus capped stable boundary layer forecasts is low but greater than a 24 hr persistence forecast. In contrast, the prediction of decoupled boundary-layers and boundary-layers with multiple cloud layers is lower than persistence. This process based evaluation approach has the potential to be applied to other boundary-layer parameterisation schemes with similar decision structures.

Representation errors and retrievals in linear and nonlinear data assimilation

This article shows how one can formulate the representation problem starting from Bayes’ theorem. The purpose of this article is to raise awareness of the formal solutions,so that approximations can be placed in a proper context. The representation errors appear in the likelihood, and the different possibilities for the representation of reality in model and observations are discussed, including nonlinear representation probability density functions. Specifically, the assumptions needed in the usual procedure to add a representation error covariance to the error covariance of the observations are discussed,and it is shown that, when several sub-grid observations are present, their mean still has a representation error ; socalled ‘superobbing’ does not resolve the issue. Connection is made to the off-line or on-line retrieval problem, providing a new simple proof of the equivalence of assimilating linear retrievals and original observations. Furthermore, it is shown how nonlinear retrievals can be assimilated without loss of information. Finally we discuss how errors in the observation operator model can be treated consistently in the Bayesian framework, connecting to previous work in this area.

Will Arctic sea ice thickness initialization improve seasonal forecast skill?

Arctic sea ice thickness is thought to be an important predictor of Arctic sea ice extent. However, coupled seasonal forecast systems do not generally use sea ice thickness observations in their initialization and are therefore missing a potentially important source of additional skill. To investigate how large this source is, a set of ensemble potential predictability experiments with a global climate model, initialized with and without knowledge of the sea ice thickness initial state, have been run. These experiments show that accurate knowledge of the sea ice thickness field is crucially important for sea ice concentration and extent forecasts up to 8 months ahead, especially in summer. Perturbing sea ice thickness also has a significant impact on the forecast error in Arctic 2 m temperature a few months ahead. These results suggest that advancing capabilities to observe and assimilate sea ice thickness into coupled forecast systems could significantly increase skill.

Basic concepts for convection parameterization in weather forecast and climate models: COST Action ES0905 final report

The research network “Basic Concepts for Convection Parameterization in Weather Forecast and Climate Models” was organized with European funding (COST Action ES0905) for the period of 2010–2014. Its extensive brainstorming suggests how the subgrid-scale parameterization problem in atmospheric modeling, especially for convection, can be examined and developed from the point of view of a robust theoretical basis. Our main cautions are current emphasis on massive observational data analyses and process studies. The closure and the entrainment–detrainment problems are identified as the two highest priorities for convection parameterization under the mass–flux formulation. The need for a drastic change of the current European research culture as concerns policies and funding in order not to further deplete the visions of the European researchers focusing on those basic issues is emphasized.

A spatial view of ensemble spread in convection permitting ensembles

With movement toward kilometer-scale ensembles, new techniques are needed for their characterization. A new methodology is presented for detailed spatial ensemble characterization using the fractions skill score (FSS). To evaluate spatial forecast differences, the average and standard deviation are taken of the FSS calculated over all ensemble member–member pairs at different scales and lead times. These methods were found to give important information about the ensemble behavior allowing the identification of useful spatial scales, spinup times for the model, and upscale growth of errors and forecast differences. The ensemble spread was found to be highly dependent on the spatial scales considered and the threshold applied to the field. High thresholds picked out localized and intense values that gave large temporal variability in ensemble spread: local processes and undersampling dominate for these thresholds. For lower thresholds the ensemble spread increases with time as differences between the ensemble members upscale. Two convective cases were investigated based on the Met Office United Model run at 2.2-km resolution. Different ensemble types were considered: ensembles produced using the Met Office Global and Regional Ensemble Prediction System (MOGREPS) and an ensemble produced using different model physics configurations. Comparison of the MOGREPS and multiphysics ensembles demonstrated the utility of spatial ensemble evaluation techniques for assessing the impact of different perturbation strategies and the need for assessing spread at different, believable, spatial scales.

HESS Opinions: On forecast (in)consistency in a hydro-meteorological chain: curse or blessing?

Flood forecasting increasingly relies on numerical weather prediction forecasts to achieve longer lead times. One of the key difficulties that is emerging in constructing a decision framework for these flood forecasts is what to dowhen consecutive forecasts are so different that they lead to different conclusions regarding the issuing of warnings or triggering other action. In this opinion paper we explore some of the issues surrounding such forecast inconsistency (also known as "Jumpiness", "Turning points", "Continuity" or number of "Swings"). In thsi opinion paper we define forecast inconsistency; discuss the reasons why forecasts might be inconsistent; how we should analyse inconsistency; and what we should do about it; how we should communicate it and whether it is a totally undesirable property. The property of consistency is increasingly emerging as a hot topic in many forecasting environments.

The DYMECS project: a statistical approach for the evaluation of convective storms in high-resolution NWP models

A new frontier in weather forecasting is emerging by operational forecast models now being run at convection-permitting resolutions at many national weather services. However, this is not a panacea; significant systematic errors remain in the character of convective storms and rainfall distributions. The DYMECS project (Dynamical and Microphysical Evolution of Convective Storms) is taking a fundamentally new approach to evaluate and improve such models: rather than relying on a limited number of cases, which may not be representative, we have gathered a large database of 3D storm structures on 40 convective days using the Chilbolton radar in southern England. We have related these structures to storm life-cycles derived by tracking features in the rainfall from the UK radar network, and compared them statistically to storm structures in the Met Office model, which we ran at horizontal grid length between 1.5 km and 100 m, including simulations with different subgrid mixing length. We also evaluated the scale and intensity of convective updrafts using a new radar technique. We find that the horizontal size of simulated convective storms and the updrafts within them is much too large at 1.5-km resolution, such that the convective mass flux of individual updrafts can be too large by an order of magnitude. The scale of precipitation cores and updrafts decreases steadily with decreasing grid lengths, as does the typical storm lifetime. The 200-m grid-length simulation with standard mixing length performs best over all diagnostics, although a greater mixing length improves the representation of deep convective storms.

Long term individual load forecast under different electrical vehicles uptake scenarios

More and more households are purchasing electric vehicles (EVs), and this will continue as we move towards a low carbon future. There are various projections as to the rate of EV uptake, but all predict an increase over the next ten years. Charging these EVs will produce one of the biggest loads on the low voltage network. To manage the network, we must not only take into account the number of EVs taken up, but where on the network they are charging, and at what time. To simulate the impact on the network from high, medium and low EV uptake (as outlined by the UK government), we present an agent-based model. We initialise the model to assign an EV to a household based on either random distribution or social influences - that is, a neighbour of an EV owner is more likely to also purchase an EV. Additionally, we examine the effect of peak behaviour on the network when charging is at day-time, night-time, or a mix of both. The model is implemented on a neighbourhood in south-east England using smart meter data (half hourly electricity readings) and real life charging patterns from an EV trial. Our results indicate that social influence can increase the peak demand on a local level (street or feeder), meaning that medium EV uptake can create higher peak demand than currently expected.

How is the balance of a forecast ensemble affected by adaptive and non-adaptive localization schemes?

This paper investigates the effect on balance of a number of Schur product-type localization schemes which have been designed with the primary function of reducing spurious far-field correlations in forecast error statistics. The localization schemes studied comprise a non-adaptive scheme (where the moderation matrix is decomposed in a spectral basis), and two adaptive schemes, namely a simplified version of SENCORP (Smoothed ENsemble COrrelations Raised to a Power) and ECO-RAP (Ensemble COrrelations Raised to A Power). The paper shows, we believe for the first time, how the degree of balance (geostrophic and hydrostatic) implied by the error covariance matrices localized by these schemes can be diagnosed. Here it is considered that an effective localization scheme is one that reduces spurious correlations adequately but also minimizes disruption of balance (where the 'correct' degree of balance or imbalance is assumed to be possessed by the unlocalized ensemble). By varying free parameters that describe each scheme (e.g. the degree of truncation in the schemes that use the spectral basis, the 'order' of each scheme, and the degree of ensemble smoothing), it is found that a particular configuration of the ECO-RAP scheme is best suited to the convective-scale system studied. According to our diagnostics this ECO-RAP configuration still weakens geostrophic and hydrostatic balance, but overall this is less so than for other schemes.

Origin and impact of initialisation shocks in coupled atmosphere-ocean forecasts

Current methods for initialising coupled atmosphere-ocean forecasts often rely on the use of separate atmosphere and ocean analyses, the combination of which can leave the coupled system imbalanced at the beginning of the forecast, potentially accelerating the development of errors. Using a series of experiments with the European Centre for Medium-range Weather Forecasts coupled system, the magnitude and extent of these so-called initialisation shocks is quantified, and their impact on forecast skill measured. It is found that forecasts initialised by separate ocean and atmospheric analyses do exhibit initialisation shocks in lower atmospheric temperature, when compared to forecasts initialised using a coupled data assimilation method. These shocks result in as much as a doubling of root-mean-square error on the first day of the forecast in some regions, and in increases that are sustained for the duration of the 10-day forecasts performed here. However, the impacts of this choice of initialisation on forecast skill, assessed using independent datasets, were found to be negligible, at least over the limited period studied. Larger initialisation shocks are found to follow a change in either the atmospheric or ocean model component between the analysis and forecast phases: changes in the ocean component can lead to sea surface temperature shocks of more than 0.5K in some equatorial regions during the first day of the forecast. Implications for the development of coupled forecast systems, particularly with respect to coupled data assimilation methods, are discussed.

Towards improving the framework for probabilistic forecast evaluation

The evaluation of forecast performance plays a central role both in the interpretation and use of forecast systems and in their development. Different evaluation measures (scores) are available, often quantifying different characteristics of forecast performance. The properties of several proper scores for probabilistic forecast evaluation are contrasted and then used to interpret decadal probability hindcasts of global mean temperature. The Continuous Ranked Probability Score (CRPS), Proper Linear (PL) score, and IJ Good’s logarithmic score (also referred to as Ignorance) are compared; although information from all three may be useful, the logarithmic score has an immediate interpretation and is not insensitive to forecast busts. Neither CRPS nor PL is local; this is shown to produce counter intuitive evaluations by CRPS. Benchmark forecasts from empirical models like Dynamic Climatology place the scores in context. Comparing scores for forecast systems based on physical models (in this case HadCM3, from the CMIP5 decadal archive) against such benchmarks is more informative than internal comparison systems based on similar physical simulation models with each other. It is shown that a forecast system based on HadCM3 out performs Dynamic Climatology in decadal global mean temperature hindcasts; Dynamic Climatology previously outperformed a forecast system based upon HadGEM2 and reasons for these results are suggested. Forecasts of aggregate data (5-year means of global mean temperature) are, of course, narrower than forecasts of annual averages due to the suppression of variance; while the average “distance” between the forecasts and a target may be expected to decrease, little if any discernible improvement in probabilistic skill is achieved.

Validation of a priori CME arrival predictions made using real-time heliospheric imager observations

Between December 2010 and March 2013, volunteers for the Solar Stormwatch (SSW) Citizen Science project have identified and analyzed coronal mass ejections (CMEs) in the near real-time Solar Terrestrial Relations Observatory Heliospheric Imager observations, in order to make “Fearless Forecasts” of CME arrival times and speeds at Earth. Of the 60 predictions of Earth-directed CMEs, 20 resulted in an identifiable Interplanetary CME (ICME) at Earth within 1.5–6 days, with an average error in predicted transit time of 22 h, and average transit time of 82.3 h. The average error in predicting arrival speed is 151 km s−1, with an average arrival speed of 425km s−1. In the same time period, there were 44 CMEs for which there are no corresponding SSW predictions, and there were 600 days on which there was neither a CME predicted nor observed. A number of metrics show that the SSW predictions do have useful forecast skill; however, there is still much room for improvement. We investigate potential improvements by using SSW inputs in three models of ICME propagation: two of constant acceleration and one of aerodynamic drag. We find that taking account of interplanetary acceleration can improve the average errors of transit time to 19 h and arrival speed to 77 km s−1.