Combining 2-m temperature nowcasting and short range ensemble forecasting

Abstract

Abstract. During recent years, numerical ensemble prediction systems have become an important tool for estimating the uncertainties of dynamical and physical processes as represented in numerical weather models. The latest generation of limited area ensemble prediction systems (LAM-EPSs) allows for probabilistic forecasts at high resolution in both space and time. However, these systems still suffer from systematic deficiencies. Especially for nowcasting (0-6 h) applications the ensemble spread is smaller than the actual forecast error. This paper tries to generate probabilistic short range 2-m temperature forecasts by combining a stateof-the-art nowcasting method and a limited area ensemble system, and compares the results with statistical methods. The Integrated Nowcasting Through Comprehensive Analysis (INCA) system, which has been in operation at the Central Institute for Meteorology and Geodynamics (ZAMG) since (Haiden et al., 2011, provides short range deterministic forecasts at high temporal (15 min-60 min) and spatial (1 km) resolution. An INCA Ensemble (INCA-EPS) of 2-m temperature forecasts is constructed by applying a dynamical approach, a statistical approach, and a combined dynamic-statistical method. The dynamical method takes uncertainty information (i.e. ensemble variance) from the operational limited area ensemble system ALADIN-LAEF (Aire Limitée Adaptation Dynamique Développement InterNational Limited Area Ensemble Forecasting) which is running operationally at ZAMG (Wang et al., 2011). The purely statistical method assumes a well-calibrated spread-skill relation and applies ensemble spread according to the skill of the INCA forecast of the most recent past. The combined dynamic-statistical approach adapts the ensemble variance gained from ALADIN-LAEF with non-homogeneous Gaussian regression (NGR) which yields a statistical correction Correspondence to: A. Kann (alexander.kann@zamg.ac.at) of the first and second moment (mean bias and dispersion) for Gaussian distributed continuous variables. Validation results indicate that all three methods produce sharp and reliable probabilistic 2-m temperature forecasts. However, the statistical and combined dynamic-statistical methods slightly outperform the pure dynamical approach, mainly due to the under-dispersive behavior of ALADIN-LAEF outside the nowcasting range. The training length does not have a pronounced impact on forecast skill, but a spread re-scaling improves the forecast skill substantially. Refinements of the statistical methods yield a slight further improvement.

Comparison of probabilistic post-processing approaches for improving numerical weather prediction-based daily and weekly reference evapotranspiration forecasts

“…For example, Hagedorn (2008) and Hagedorn et al (2008) showed gains in the lead time of between 2 and 4 d, with the gains being larger over areas where the raw forecast showed poor skill. Kann et al (2009Kann et al ( , 2011 used the NGR method to improve short-range ensemble forecasts of 2 m temperature. Recently, Scheuerer and Büermann (2014) provided a generalization of the original approach of Gneiting et al (2005) that produces spatially calibrated probabilistic temperature forecasts.…”

Experience of using the Kalman filter to correct numerical forecasts of surface air temperature

“…, a M ≥ 0 and the spread coefficients b 0 ≥ 0 and b 1 ≥ 0 need to be fitted from training data, with applying a minimum score approach. 8 Hagedorn (2008), Hagedorn, Hamill and, Kann et al (2009) and Kann, Haiden and Wittmann (2011), among others, have applied this approach to calibrate temperature forecasts.…”

“…2 While a majority of Member States has implemented postprocessing procedures, these are typically restricted to traditional bias correction of the HRES output, and to key variables such as 2m temperature, precipitation or 10m wind, using some form of model output statistics (MOS) technique, frequently implemented via Kalman filter (KF) algorithms, as described by Crochet (2004), and in some cases the perfect prog (PP) method. A few member countries have also considered ensemble calibration techniques, including Austria (Kann et al 2009;Kann, Haiden and Wittmann 2011), France and Hungary (Ihász et al 2010), 3 and some countries have implemented postprocessing efforts geared at binary probability forecasts, with Schmeits et al (2008) describing one such application to lightning.…”

A newly developed South American Mapping of Temperature with estimated lapse rate corrections

“…In this sense, the choice of the spatial interpolation method is of paramount importance in regions where data distribution is sparse. Thus, in order to minimize the interpolation errors associated with the lack of observations, data from numerical models (Fuentes and Raftery, 2005; Kann et al ., 2011), reanalysis (Essou et al ., 2017; Chen et al ., 2019) and satellite estimates (Rozante et al ., 2010; 2020) have been combined to surface observations in order to produce a more coherent meteorological fields. Many studies have been concentrated to address precipitation products over different regions such as South America (Vila et al ., 2009; Rozante et al ., 2010; 2020), China (Shuai Han and Shuai Han, 2019), Asia (Jia et al ., 2013), Ethiopia (Dinku et al ., 2014), Nigeria (Grimes et al ., 1999), Australia (Chappell et al ., 2013), India (Mitra et al ., 2013), among others.…”