Climate Exchange

[

] 90

ater

A steering group is monitoring, evaluating and readjusting the

service where needed. Decisions at the operational level are made

by the responsible project managers together with the users of

the services and scientific partners.

Data and models

The meteorological and hydrological data and models of KLIWAS

are linked in a complex model chain which mimics the complex

structure of the system under investigation, namely the major

river systems and the coastal areas of Germany and the neigh-

bouring countries. At each link of the chain, multiple models are

involved to give an idea of the uncertainty associated with each

modelling step.

In the framework of KLIWAS DWD acquires climate model

outputs of global and regional climate models from various

research groups. The model data are then further processed for

use in KLIWAS impact models. This post-processing includes the

verification of control runs, the statistical analysis of the climate

projections, further downscaling of regional climate model outputs

by means of statistical methods, and the bias correction of model

output. In addition, DWD compiles and generates meteorological

reference data sets for the entire basins of the international rivers

under investigation, as well as for the North Sea, the Baltic Sea, and

the adjacent North Atlantic. The reference data sets support the

verification of climate model data and KLIWAS impact models as,

hydrological models for example, and are used for bias correction

of climate model output.

At the BfG and the BSH the simulated and observed meteoro-

logical data are used as input for hydrological and oceanographic

models to assess the impacts of climate variability on stream

flows, sea level etc. The climate model data are bias-corrected

to be applicable in the impact models. Here too, multiple

correction methods are compared to demonstrate the specific

modifications of model results.

Methods and (hydrological) models are set up and/or evalu-

ated in cooperation with water research institutes and water

managers working on the same river basins as KLIWAS. All

institutions and persons bring in their own hydro-

logical expertise and data. For example, the model

used for short term forecasts is used in the same

version in Germany and the Netherlands. Thus, the

quality of the models and data is double-checked

and consistent across the border.

The KLIWAS programme does not end with infor-

mation on water quantity and hydrodynamics. It also

tailors specific information for different user groups.

For example, further models are coupled to assess

and monetize the effects of climate and hydrologi-

cal change on inland waterway transport. Specific

information on future transport costs and capacities

under climate change conditions are offered – includ-

ing an uncertainty statement. Using this information,

the users (for example, the BAW) evaluate adaptation

measures such as innovative steering techniques for

the vessels travelling on the River Rhine.

Example service: discharge scenarios for policy

development

As a consequence of the close stakeholder involvement,

information provided by KLIWAS is already in use in

policy development. For example, in 2010 KLIWAS and

network partners agreed on a general evaluation frame-

work designed to select and process climate model data

for the purpose of impact modelling.

This framework

consists in an agreed set of evaluation periods, evaluation

regions, and evaluation statistics as given in Appendix A.

This procedure resulted in a reduced ensemble (20 out

of 26 members) of bias-corrected climate simulations,

which are regarded as suitable for hydrological modelling.

On that basis, a set of hydrological indicators was gener-

ated, which characterises the potential future hydrological

regimes of the River Rhine in the middle and at the end of

the 21st century. The indicators were chosen to meet the

user needs that were identified; i.e. they indicate a span of

results for relevant hydrological statistics representing the

current modelling uncertainty as sampled by an ensemble

of quality-checked and bias-corrected climate simulations.

This work was coordinated, documented and published

online by the International Commission for the Hydrology

of the Rhine basin (CHR).

The hydrological statistics are

given in Appendix B. The bandwidth is characterized by

so-called scenario corridors. These corridors are defined

as clusters within the ensemble there where many projec-

tions produce similar values (Figure on page 91). They

represent the characterstics of the ensemble much better

than mean values or percentiles

These indicators were discussed by the International

Commission for the Protection of the Rhine (ICPR),

an advisory board that coordinates the work of repre-

sentatives of the governments of Switzerland, France,

Germany, Luxembourg, Netherlands and the European

Commission. Guidance on how to read the indicators

was given during specific workshops for representatives

dealing with floods, ecology, sediments and strategic

questions. In 2011, the ICPR members adopted the

indicators as scenarios to prepare the development of

This graph on the effect of projected climate change on transport capacity of cargo

on the River Rhine is an example of user-specific information generated by KLIWAS

Source: Development Centre for Ship Technology and Transport Systems