DKRZ, the virtual laboratory for climate researchers/modellers

---

The earth cannot be experimented with.

An understanding of the human influence on the "system earth" and the impact of this global change on our living conditions is one of the Grand Challenges for science. The global environment, and in particular the climate, are extremely complex systems, whose dynamics and future development can only be understood through extensive investigations and complicated model computations.

Realistic climate and environment simulations demand highest computer performance and produce enormous amounts of data. Thus simulating climate change on a global scale requires super-computer with highest performance and massive data archiving facilities.

The DKRZ provides such resources to all groups investigating the earth system with complex numerical models. It is “the laboratory of the German earth system research”. Its dedicated high-performance computers and data servers provide the key tool for earth scientist to understand the natural processes better and to assess the impact of human activities. It enables the scientists to provide the public and policymakers with reliable information for a sustainable climate and environmental policy.

DKRZ, the virtual laboratory for climate researchers/modellers

---

Progress in the earth system and climatic research is limited by the availability of computing technology.

Four reasons, why we do need supercomputers:

Why we need supercomputers?

---

1. Complexity of the earth system

Schematic illustration of the most important components of the climate system. The more of these processes are considered in a climate model the more accurate becomes the result.

Climate modelling so far concentrated mainly on the interaction of physical components and their coupling, such as fluid dynamics, radiation balance, latent and sensitive heat exchange or wind effects on the ocean. The employment of modern supercomputers, however, permits to integrate also chemical and biological coupling mechanism as well as to investigate the interaction between the climate and the socio-economic system. The ultimate ambition of the international climate and global change research is the development of a complex model of the earth system comprehending all physical and biogeochemical interactions between the atmosphere, the ocean, the cryosphere and the continental biosphere as well as social issues.

The development of climatic models over the last 25 years showing how the different components are first developed separately and later coupled into comprehensive climate models (IPCC WG1, 2001)

 

Why we need supercomputers?

---

2. Spatial resolution of the models

Europe mapped in grids with different resolutions: distance between grid points (from above left downward right): 500km, 250km, 150km, 100km. The finer the resolution the more accurate the prognoses.

The accuracy of climate prognoses is substantially limited by the resolution of the models, respectively the grid size of the cells being the basis for the numerical simulations. A finer grid size results in more accurate prognoses of regional climate changes. But halving the distance between two grid points already produces four times as much grid points and requires for numerical reasons the tenfold compute performance. Therefore scientist had to wait for the new super computer to approach questions such as “Is it going to rain more often in middle Europe?”.

Why we need supercomputers?

---

3. Ensemble calculations

In this example of a forecast of the El Niño event several forecast runs with slightly changed initial data, so called ensembles, are started. (Quelle ECMWF)

Numerical models often react sensitively to slight disturbances of the initial data. For this reason the reliability of the models is sometimes questioned. The higher the computer performances become the more frequently ensemble calculations can be accomplished: The model computations are repeated many times with slight variations of the initial data so that random results can be distinguished from statistically proved trends.

Why we need supercomputers?

---

4. Integrations over many centuries

The different components of the climatic system react with very different time scales.

Many scientific investigations require simulations of the earth system over a period of several centuries. Such simulations make highest demands on computing and data archive capacities.

Deutsches Klimarechenzentrum GmbH | Impressum