Voudouri A.,Hellenic National Meteorological Service |
Khain P.,Israel Meteorological Service |
Carmona I.,Israel Meteorological Service |
Bellprat O.,Institute Catala Of Ciencies Del Clima Ic3 |
And 4 more authors.
Atmospheric Research | Year: 2017
Numerical weather prediction (NWP) and climate models use parameterization schemes for physical processes, which often include free or poorly confined parameters. Model developers normally calibrate the values of these parameters subjectively to improve the agreement of forecasts with available observations, a procedure referred as expert tuning. A practicable objective multi-variate calibration method build on a quadratic meta-model (MM), that has been applied for a regional climate model (RCM) has shown to be at least as good as expert tuning. Based on these results, an approach to implement the methodology to an NWP model is presented in this study. Challenges in transferring the methodology from RCM to NWP are not only restricted to the use of higher resolution and different time scales. The sensitivity of the NWP model quality with respect to the model parameter space has to be clarified, as well as optimize the overall procedure, in terms of required amount of computing resources for the calibration of an NWP model. Three free model parameters affecting mainly turbulence parameterization schemes were originally selected with respect to their influence on the variables associated to daily forecasts such as daily minimum and maximum 2 m temperature as well as 24 h accumulated precipitation. Preliminary results indicate that it is both affordable in terms of computer resources and meaningful in terms of improved forecast quality. In addition, the proposed methodology has the advantage of being a replicable procedure that can be applied when an updated model version is launched and/or customize the same model implementation over different climatological areas. © 2017 Elsevier B.V.
Rotach M.W.,University of Innsbruck |
Arpagaus M.,Federal Office for Meteorology and Climatology MeteoSwiss |
Dorninger M.,University of Vienna |
Hegg C.,Swiss Federal Institute of forest |
And 2 more authors.
Natural Hazards and Earth System Science | Year: 2012
D-PHASE was a Forecast Demonstration Project of theWorldWeather Research Programme (WWRP) related to the Mesoscale Alpine Programme (MAP). Its goal was to demonstrate the reliability and quality of operational forecasting of orographically influenced (determined) precipitation in the Alps and its consequences on the distribution of run-off characteristics. A special focus was, of course, on heavy-precipitation events. The D-PHASE Operations Period (DOP) ran from June to November 2007, during which an end-to-end forecasting system was operated covering many individual catchments in the Alps, with their water authorities, civil protection organizations or other end users. The forecasting system's core piece was a Visualization Platform where precipitation and flood warnings from some 30 atmospheric and 7 hydrological models (both deterministic and probabilistic) and corresponding model fields were displayed in uniform and comparable formats. Also, meteograms, nowcasting information and end user communication was made available to all the forecasters, users and end users. D-PHASE information was assessed and used by some 50 different groups ranging from atmospheric forecasters to civil protection authorities or water management bodies. In the present contribution, D-PHASE is briefly presented along with its outstanding scientific results and, in particular, the lessons learnt with respect to uncertainty propagation. A focus is thereby on the transfer of ensemble prediction information into the hydrological community and its use with respect to other aspects of societal impact. Objective verification of forecast quality is contrasted to subjective quality assessments during the project (end user workshops, questionnaires) and some general conclusions concerning forecast demonstration projects are drawn. © 2012 Author(s).
Bertotti L.,CNR Institute of Neuroscience |
Bidlot J.-R.,European Center for Medium Range Weather Forecasts |
Bunney C.,UK Met Office |
Cavaleri L.,CNR Institute of Neuroscience |
And 6 more authors.
Quarterly Journal of the Royal Meteorological Society | Year: 2012
We consider an exceptional storm-'Klaus' (January 2009)-its evolution on the Western Mediterranean Sea, and how the associated wind and wave conditions were modelled by seven of the major systems presently operational in this area. We intercompare the model results and then verify them and the related model ensemble versus the available measured data. Working with short-term forecasts (24 h) only, as expected, each model correctly anticipates the incoming of an exceptional storm. However, even at such limited range, we have found substantial differences among the results of the different models. The differences concern the time the storm should have entered the Western Mediterranean Sea, the peak values of wind speed and significant wave height, the general distribution of the fields, and the locations where the maxima were achieved. We have compared the model results versus the available measured data, wind from scatterometer, waves from altimeter, plus a few buoy data. We have found some inconsistencies in the results, model wind data being on average larger than the measured one, while the opposite was true for wave heights. However, the limited amount of data available and its different times and positions, at and off the centre of the storm, impede the drawing of any definite conclusion in this respect. On the whole we feel that our results, although related to a single storm, cast doubts on the reliability of a single forecast system to provide sufficiently reliable and accurate forecasts in case of an incoming exceptional storm. The results, both for wind and waves, have improved using an ensemble of the seven considered models. This suggests that there is no relevant systematic error in the used models except, as possibly suggested by our results, in the case of wave generation under very strong wind and very young sea conditions. © 2011 Royal Meteorological Society and British Crown, the Met Office.
Rossa A.,Centro Meteorologico Of Teolo |
Haase G.,SMHI |
Keil C.,German Aerospace Center |
Keil C.,Ludwig Maximilians University of Munich |
And 6 more authors.
Atmospheric Science Letters | Year: 2010
The COST-731 Action is focused on uncertainty propagation in hydrometeorological forecasting chains. The goals and activities of the ActionWorking Group 1 can be subdivided by (1) describing and studying the impact of imperfect observations, mostly from radar, (2) exploiting radar data assimilation as a promising avenue for improved short-range precipitation forecasts and (3) high-resolution ensemble forecasting. Activities of Working Group 1 are presented along with their possible significance for hydrological applications. Copyright © 2010 Royal Meteorological Society and Crown Copyright.
Willeit M.,University of Bologna |
Willeit M.,Potsdam Institute for Climate Impact Research |
Amorati R.,ARPA SIMC |
Montani A.,ARPA SIMC |
And 2 more authors.
Meteorology and Atmospheric Physics | Year: 2015
The spectral characteristics of hourly precipitation fields are studied for a number of meteorological events, classified depending on the precipitation process involved, namely stratiform, convective or mixed stratiform–convective. The study focuses on the comparison between spatial spectral characteristics of observed precipitation fields, obtained from hourly radar estimates, and modelled hourly predictions of the same field by the non-hydrostatic model COSMO. The results show that the power spectra of radar hourly precipitation are characterised by invariance within ranges of horizontal spatial scales that are different for the three classes of events. COSMO reproduces some basic characteristics of the radar spectra in all cases. Nevertheless, in stratiform events, COSMO spectra present a well-defined scale break at about 15 km with no counterpart in radar data. It is suggested that this model feature is related to the presence of spurious horizontal smoothing introduced by the semi-lagrangian advection scheme for precipitation. Smoothing affects all scales up to the maximum length scale of precipitation and horizontal advection by the wind. An analysis of wind intensity in the lower troposphere over the region supports this interpretation. Discrepancies in precipitation spectra between radar and COSMO data in convective events are interpreted as a consequence of the inadequacy of the model resolution for a correct representation of convection. © 2014, Springer-Verlag Wien.
Tomozeiu R.,ARPA SIMC |
Agrillo G.,ARPAL |
Cacciamani C.,ARPA SIMC |
Pavan V.,ARPA SIMC
Natural Hazards | Year: 2014
Future changes of seasonal minimum and maximum temperature over Northern Italy are assessed for the periods 2021-2050 and 2070-2099 against 1961-1990. A statistical downscaling technique, applied to the ENSEMBLES-Stream1 and CIRCE global simulations (A1B scenario), is used to reach this objective. The statistical scheme consists of a multivariate regression based on Canonical Correlation Analysis. The set-up of the statistical scheme is done using large-scale fields (predictors) derived from ERA40 reanalysis and seasonal mean minimum and maximum temperature (predictands) derived from observational data at around 75 stations, distributed over Northern Italy, over the period 1960-2002. A similar technique is also applied to the number of frost days and ice days at a reduced number of stations in order to construct projections on change of the selected extreme temperature indices for the two future periods. The evaluation of future projections for these extreme indices is relevant due to its impacts on transports, health, and agriculture. The downscaling scheme constructed using observed data is then applied to large-scale fields simulated by global models (A1B scenario), in order to construct scenarios on future change of seasonal temperature, mean and extreme indices, at local scale. The significance of changes is tested from the statistical point of view. The results show that significant increases could be expected to occur under scenario conditions in both minimum and maximum temperature, associated with a decrease in the number of frost and ice days in both periods and more intense to the end of the century. © 2013 Springer Science+Business Media Dordrecht.
Grazzini F.,ARPA SIMC |
Vitart F.,European Center for Medium Range Weather Forecasts
Quarterly Journal of the Royal Meteorological Society | Year: 2015
Historically, the objective identification of atmospheric wave-packets has been very elusive. However, interest in these important sources of atmospheric variability has recently increased, and some automated tracking methods have been proposed. The Rossby wave packet (RWP) tracking algorithms opened the way to different types of investigation, ranging from climatology and predictability to assessing the impact of climate change on wave packet characteristics. The present study investigates the relationship between predictability (intrinsic and practical, i.e. predictive skill in a numerical weather prediction model) and the properties of RWPs, such as temporal duration, spatial extension and their area of genesis. Results suggest a significant correlation between RWP length and medium-range skill over Europe and the Northern Hemisphere. Analysis of an ensemble system shows that the spread decreases when long-living RWPs are present in the forecast, supporting the hypothesis that part of the observed increase in skill could indeed be attributed to higher intrinsic predictability induced by RWPs. Higher than average medium-range forecast skill scores are often associated with the presence of long-lasting RWPs (duration of at least 8 days) in the initial conditions, with a source often located in the west Pacific. On the contrary, bad medium-range forecast skill scores tend to be associated with shorter RWPs coming from the central USA or western Atlantic. An analysis of the probabilistic skill scores confirms that predictive skill increases with the presence of long RWPs from the west Pacific, up to week 3. © 2015 Royal Meteorological Society.
Nuissier O.,Meteo - France |
Marsigli C.,ARPA SIMC |
Vincendon B.,Meteo - France |
Hally A.,Meteo - France |
And 3 more authors.
Quarterly Journal of the Royal Meteorological Society | Year: 2016
This study compares and evaluates two convection-permitting ensemble systems based on Consortium for Small-scale Modeling (COSMO) and Applications of Research to Operations at Mesoscale (AROME) models in the HyMeX framework. The performance of both AROME-EPS and COSMO-H2-EPS (where EPS denotes Ensemble Prediction System) is assessed over the whole HyMeX special observation period SOP1. Afterwards, an analysis of the predictability of two heavy precipitation events observed during the intense observation period IOP16a of the first HyMeX special observation period SOP1 on 26 October 2012 is also undertaken. Ensemble discharge forecasts were carried out to reinforce the quantitative precipitation forecast evaluation. A probabilistic evaluation is conducted over a 53 day period of the HyMeX SOP1. AROME-EPS has a more discriminating behaviour than COSMO-H2-EPS, especially when comparing both ensembles over a verification domain including a strong variability in precipitation events. AROME-EPS still has slightly better reliability, but the statistical resolution is nearly the same for both convection-permitting ensemble prediction systems (CPEPS). For the specific case of heavy precipitation occurring over the Var region (southeastern France), the fine-scale surface precipitation prediction is strongly sensitive to the good behaviour of a surface low pressure simulated by the ensembles, focusing strong low-level moisture towards Var. The convergence between strong moistened southerly low-level inflow and northerly cold air blowing from the Po Valley is a key factor controlling the predictability of the heavy precipitation episode over the Liguria region (northwestern Italy). The two convection-permitting ensembles, though different in their characteristics, exhibit a good amount of probabilistic skill in forecasting heavy precipitation at a relatively high spatial and temporal resolution. Therefore they can be regarded as promising tools for operational forecasts. © 2016 Royal Meteorological Society
Pavan V.,ARPA SIMC |
Doblas-Reyes F.J.,Catalan Institution for Research and Advanced Studies |
Doblas-Reyes F.J.,Institute Catala Of Ciencies Del Clima Ic3
Climate Dynamics | Year: 2013
A statistical calibration scheme is applied to multi-model global seasonal ensemble reforecasts in order to predict the interannual variability of summer averaged surface maximum temperature over Italy. In some cases, this technique is shown to be able to improve the skill scores of the seasonal predictions during the last 35 years, with respect to the direct model output (DMO), using seasonal predictions initialised 1 month before the beginning of the season. It is shown that the presence of some skill in the DMO multi-model predictions is mostly due to the correct prediction of the observed secular trends in maximum temperature, and, partly, to the correct prediction of outliers, in particular, of the summer of 2003. At the same time, while the removal of trends produces a small reduction of skill in both the raw and calibrated predictions, the removal of outliers improves the performance of the calibration scheme. Once all trends and outliers are removed, the DMO predictions have no skill, while the calibrated predictions still present a detectable skill. The improvement introduced by the calibration are shown to be statistically significant by applying resampling techniques. It is shown that the reason of this partial success is linked to the fact that although the models present several shortcomings, some models can capture the existence of a weak large-scale signal, possibly linked with the presence of a summer teleconnection between the equatorial Pacific and Europe, with a spatial pattern substantially different from that associated with the temperature secular trend. The teleconnection is associated with a modulation of the quasi-stationary barotropic eddies in the Northern Hemisphere extra-tropics. © 2013 Springer-Verlag Berlin Heidelberg.
Marsigli C.,ARPA SIMC |
Montani A.,ARPA SIMC |
Paccagnella T.,ARPA SIMC
Nonlinear Processes in Geophysics | Year: 2014
The current resolution of the operational global models favours the possibility of driving convection-permitting limited-area model (LAM) simulations directly, sparing the necessity for an intermediate step with a coarser-resolution LAM. Though the resolution of global ensemble systems is generally lower than that of deterministic ones, it is also possible to consider this opportunity in the field of ensemble forecasting. The aim of this paper is to investigate the effect of this choice for driving a convection-permitting ensemble based on the COSMO model, for a specific application, namely the forecast of intense autumn precipitation events over Italy. The impact of the direct nesting in the ECMWF global ensemble is compared to a two-step nesting, which makes use of a LAM ensemble system with parametrised convection. Results show that the variability introduced in the geopotential field by the direct nesting is usually contained within the uncertainty described by the standard ensemble, and differences between pairs of members following different nesting approaches are generally smaller than the ensemble error, computed with respect to analysis. The relation between spread and error is even improved by the direct nesting approach. In terms of precipitation, it is found that the forecasts issued by members with different nesting approaches generally have differences at spatial scales between 16 and 180 km, depending on the case, hence not negligible. Nevertheless, the skill of the LAM ensemble precipitation forecasts, evaluated by means of an objective verification, is comparable. Therefore, the overall quality of the 2.8 km ensemble for the specific application is not deteriorated by the provision of lower resolution lateral boundary conditions directly from the global ensemble. © Author(s) 2014.