C2-WP2 - Prediction of large regional climate anomalies

Project aims

The general goal of this work package is to quantify the capability of the regional climate model COSMO-CLM to provide reliable decadal forecasts for Europe, and to determine how far and why the forecasts vary on decadal time scales. Our aim is thus to uncover the processes underlying this predictability. With this aim, we analyse the predictive skill of the MiKlip hindcasts and identify the physical processes that lead to this predictive skill over Europe.
For this task, we follow two different routes: the first one concentrates on the statistics of the regional predictive skill depending on the large-scale weather-types or phases of the multi-decadal variability. The second one focuses on large climate anomalies related to an enhanced occurrence of extreme events like heat waves, droughts, floods and windstorms in the target region because of their high impact and importance to the society. Both approaches complement each other.
 

Project structure

Two groups (Deutscher Wetterdienst (DWD) and Karlsruhe Institute of Technology (KIT)) cooperate in work package C2-WP2. It is subdivided in four tasks with two thematic aspects: the conditional evaluation of the hindcasts and the analysis of large climate anomalies.

Tasks of the project

T1 Conditional evaluation of the hindcasts
The hindcasts will be stratified with respect to their weather conditions or state of multi-annual and multi-decadal oscillation (Atlantic multi-decadal Variability (AMV), North Atlantic Oscillation (NAO), etc.), to identify conditions with higher or lower forecast skill. First analysis of stratified hindcasts uncovered conditions with diverging forecast skill in Europe.

The stratified data will be statistically analysed with respect to the predictive skill in Europe. The correlation of precursors and climate elements to be predicted for each weather type will be determined. From the resulting correlations for the different classes (weather condition, states of the oscillations), conditions with higher or lower predictability can be deduced. Conditional climatologies depending on weather conditions will be contrasted against the mean climatology.


T2 Predictability of large climate anomalies
In a first step examples of potentially relevant events like distinct climate shifts from the 90s (warming of the Atlantic sub-polar gyre region) to the 2000s hiatus or other European large climate anomalies, e.g., like the series of heat waves in 2003 and 2006, the large flooding in 2002, the peak of windstorm activity in the early 1990s, or the anomaly of extreme precipitation events will be selected. The results from the conditional evaluation will be used to augment the selected cases with other exceptional conditions. A proper weather typing approach is selected and applied for each large climate anomaly.

Further, the mechanisms behind the predictive skill will be in focus. The correlation between the precursor defining the weather type and the climate elements to be predicted will be analysed to deduce the extent to what the large climate anomalies can be predicted or (partly) attributed to the multi-decadal variability pattern in the Arctic and North Atlantic climate system. In addition the signal leading to the large climate anomalies will be traced back in order to determine how far ahead the large climate anomaly can be detected.
 

Deliverables

D1a Methods for investigating the stratified hindcasts
D1b Results of the stratified evaluations
D2a Method for investigating the large climate anomalies
D2b Results of the large climate anomaly analysis
 

Progress so far

A suitable objective method of weather pattern classification is needed for the stratified evaluation as well as for the analysis of mechanisms and process chains behind the large climate anomalies (see example in Fig. 1 for the series of extreme storms in 1990 – Daria, Vivian, Wiebke). An appropriate weather type analysis for Europe is the Hess and Brezowsky Grosswetterlagen (GWL) classification. So far, an objective GWL classification has been applied to ERA40/ERA-Interim (Fig. 1) and the full available ensemble of the MiKLip decadal prediction system (baseline0, baseline1, prototype, historical runs).
A suitable objective method of weather pattern classification is needed for the stratified evaluation, as well as for the analysis of mechanisms and process chains behind large climate anomalies (see example in Fig. 1 for the series of extreme storms in 1990 – Daria, Vivian, Wiebke). An appropriate weather type analysis for Europe is the Hess and Brezowsky Grosswetterlagen (GWL) classification. So far, an objective GWL classification has been applied to ERA40/ERA-Interim (Fig. 1) and the full available ensemble of the MiKliP decadal prediction system (baseline0, baseline1, prototype, historical runs).

Fig. 1: Frequencies of GWL 1, 2, 4, and 6 from the original GWL catalogue and from the objective GWL classification.

For the stratified evaluation, data classes based on terciles of various large-scale circulation indices are generated. This means, that e.g. the global time series of the NAO index is used in order to determine negative, neutral and positive NAO phases and data of all variables to be analysed (temperature, precipitation, wind etc.) are split into the corresponding three classes of equal size regarding to their NAO phase. Since the circulation indices, obtained from the modeled data and the circulation index calculated from reanalysis, would only be equal for a perfect forecast, there are still two choices within each class: Cases, where the indices do not agree (different phase), and cases, where they do (same phase). An analysis of the latter cases provides an upper limit of the potential conditional forecast skill. Technically, the tool MurCSS from the Central Evaluation System (MiKliP-CES) was modified and tested in order to handle such conditional data. The updated version will be publically available soon.

In addition to the conditional evaluation, the decadal variability of climate phenomena is analysed based on climate indices. Focus is given to the relation between climate indices and potentially predictable climate variations like teleconnection patterns and weather types. For example, daily storm severity indices (SSI; as proxy for storm series) are calculated for 1961-2015 based on CCLM-ERA. The resulting time series are then correlated to time series of NAO, AMO and the Grosswetterlagen (Fig. 2). The results show a high correlation of SSI and NAO (0.72), while SSI is anti-correlated with AMO (-0.11). High positive correlations are also found for the sum of GWLs 1, 2 and 5. Similar analyses are done for droughts, heat waves and heavy precipitation.

Fig. 2: 4yr-running sum time series of SSI (red) for ERA (1961-2015) and 4yr-running mean time series of AMO (blue, left), NAO (blue, middle), and GWL 1+2+5 (blue, right).

The long-term variability of extreme indices and their teleconnection to the AMO index has been studied using the centennial downscaling experiments from work package C2-WP3, which cover the period since 1900. Several precipitation indices indicate a higher intensity in AMO+ phases. These findings still have to be evaluated carefully, since few reliable observational data are available before the mid of the 20th century.

Contact

Deutscher Wetterdienst - Climate and Environment Consultancy
Dr. Barbara Früh
Barbara.Frueh(at)nospamdwd.de
+49 (0)69 8062-2968

Deutscher Wetterdienst - Climate and Environment Consultancy
Dr. Sascha Brand
Sascha.brand(at)nospamdwd.de
+49 (0)69 8062-2967

Institute for Meteorology and Climate Research (IMK-TRO) Karlsruhe Institute of Technology (KIT)
Hendrik Feldmann
Hendrik.feldmann(at)nospamkit.edu
+49 (0)721 608-24942

Institute for Meteorology and Climate Research (IMK-TRO) Karlsruhe Institute of Technology (KIT)
Natalie Laube
Natalie.laube(at)nospamkit.edu
+49 (0)721 608-22844