SORS: Improving decadal climate predictions by ocean ensemble dispersion filtering (EDF) and an efficient systematic evaluation framework (Freva)

Date: 30/May/2019 Time: 11:00

Place:

Sala d'actes de la FiB

Primary tabs

15th December 2022. 10:30 am. UPC's Vèrtex Building Auditorium + Gardens & Virtual

We are looking forward to celebrating the BSC staff annual meeting this year.

As in the previous edition, the event will be online and shorter.

The event will begin on December 15th at 10:30 am. We will send you a link nearer the time so you can connect.

Even more than normal, we hope that as many BSC staff as possible will be able to attend and participate.

You can participate by telling us:

What you would like to know about?

Let us know which subjects you find interesting to be incorporated into the BSC directors’ presentations.
Deadline exceeded

What would you like to listen to?

What you would like to share?

We would like to know more about your work, so we have booked one hour of the Annual Meeting to listen to 6 BSC employees that will explain what they do. We encourage you to be one of them.

What I would like to explain

Coming soon

 BSC Talks

We have booked time for 6 employees to explain in 6 minutes what you do at BSC.

The event will be VIRTUAL but the BSC Talks will be presented from the studio where the event is taking place.
 

We would like you to make short, attractive presentations (6 minutes) so all the BSC staff know what you do.

You could be one of those to go up the stage, and we will help you to prepare it.

What will you need to do?

  1. Let us know about what you would like to explain.
    • Send us a short video (landscape) of 45 seconds’ maximum explaining what you would like to present at annualmeeting2021@bsc.es
    • State your name, department, research group and a provisional title for your talk.
    • Deadline: November 29th.
  2. All the BSC staff will vote for the 6 most interesting proposals.
    • All videos will be available at the web page and the voting process will be open. Each employee will be able to vote up to 3 proposals.
  3. If your proposal is one of the chosen ones:
    • A training session will be organized for the 6 most voted.
  4. Prepare a 6 minutes’ presentation.
  5. The best presentation of December 15th will be awarded!

Directors’ presentation of 2022

We would like to know what subjects you are interested in so as to incorporate them in the directors’ presentation of the center. Fill in this form with your questions.

Objectives

To download the presentation, please click here

Abstract: Decadal climate predictions have the objective to predict the development of the climate for the following years to decades. Numerical Earth system models are initialized with observational values, similar to the methodology applied in weather forecasting. Additionally, they are forced by boundary conditions, like greenhouse gas scenarios, to project the long term development. This talk investigates decadal climate predictions with Earth system models and their further improvement with new strategies in research and technologies.

A modern decadal prediction system is evaluated and investigated for sources of potential skill. Hence a systematic evaluation strategy is developed. It contains the assessment of accuracy of the ensemble mean and the ensemble spread, and compares decadal experiments with climatology, observations, and climate projections. The entire assessment is performed within a novel evaluation system called Free Evaluation System Framework (Freva). This system is designed to complement climate modeling by a systematic and efficient assessment. Freva serves as a resource-efficient process framework between the data generation and its evaluation, to detect climate research potential. Freva runs on high performance computers to handle customizable evaluation systems of research projects, institutes or universities - to connect scientists. It combines different software technologies into one common hybrid infrastructure, including all features present in the shell and web environment.

A new prediction technique called ’Ensemble Dispersion Filter’ is developed. It exploits two important climate prediction paradigms: the ocean’s heat capacity and the advantage of the ensemble mean. The Ensemble Dispersion Filter averages the ocean temperatures of the ensemble members every three months, uses this ensemble mean as a restart condition for each member, and further executes the prediction. The evaluation by the new verification framework Freva shows that the Ensemble Dispersion Filter results in a significant improvement in the predictive skill compared to the unfiltered reference system. Even in comparison with prediction systems of a larger ensemble size and higher resolution, the Ensemble Dispersion Filter system performs better. In particular, the prediction of the global average temperature of the forecast years 2 to 5 shows a significant skill improvement.

Bio: Christopher Kadow is an expert in climate modeling, prediction, and verification. His research and scientific achievements are primarily centred around decadal climate predictions and IT interfaces in climate science. He is the project coordinator and leading scientist of the project ‘INTEGRAted data and evaluation system for decadal scale predicTION’ within the major project on decadal climate prediction ‘Mittelfristige Klimaprognosen’ (1) at the Freie Universität Berlin in Germany. His project INTEGRATION is responsible for the development of the central evaluation system in MiKlip (2). This central evaluation system is based on the Free Evaluation System Framework (Freva) and it includes data and software interfaces for a common developer base on high performance computers in web and shell. Since 2017 Christophr is leading the forecast group for the official German decadal forecast. Therefore his team developed a state-of-art forecast website as modern climate service for climate predictions (3). Furthermore he developed a novel forecast technique for decadal prediction (4). This technique – namely the ensemble dispersion filter – exploits the benefits of an ensemble forecast in combination with the ocean decadal memory to improve the multi-annual prediction skill. Besides decadal prediction he is involved in many climate modeling activities like for example the German contribution to CMIP6 (5) and climate modeling and verification teaching at the Freie Universität Berlin (6) - where Freva is also applied.

1 fona-miklip.de, 2 www-miklip.dkrz.de, 3 fona-miklip.de, 4 Kadow et al. 2017, 5 cmip-eval.dkrz.de, 6 freva.met.fu-berlin.de

 

Speakers

Christopher Kadow, expert in climate modeling, prediction, and verification.