Using two different CTMs (OSLO CTM2 run by CICERO and LMDz-INCA run by CNRS) the response of the chemical composition of the atmosphere to 4 different localised emissions of CO and NOx was simulated. The emissions scenarios were:

1. CO fossil fuel surface emissions increased over Western Europe,
2. CO fossil fuel surface emissions increased over South East Asia,
3. NOx fossil fuel surface emissions increased over Western Europe,
4. NOx fossil fuel surface emissions increased over South East Asia,

The same amount of emissions results in a stronger ozone increase if the species (NOx or CO) are emitted in South East Asia.

The impact of geographically inhomogeneous perturbation of radiative active species was studied with two comprehensive GCMs (ECHAM4 run by DLR and LMDz run by CNRS) and one intermediate GCM (run by UREAD). In a first  phase perturbations impacting the long wave radiation were inserted, all of the same global mean rediative forcing of 1 W/m²:

1. a global homogeneous CO2 increase,
2. a CO2 increase confined to the tropics,
3. a CO2 increase confined to the extra-tropics,
4. a CO2 increase confined to the  northern extra-tropics.

Despite the fact that the models have different climate sensitivity parameters for a homogeneous CO2 perturbation, the normalised response in the global mean temperature, i.e., the temperature change normalised by the respective response for the homogeneous forcing, is astonishingly similar for all models. The tropical  and extra-tropical perturbations result in smaller and larger temperature changes, respectively.

Available metrics of climate change, in particular radiative forcing and global warming potential have been reviewed. There usefulness for "measuring" climate change was discussed and some limitations were shown. E.g. it was demonstrated that reductions in CO2 emissions and equivalent CH4 emissions (in terms of GWP) can results in quite different temperature responses.

Criteria for applicability of metrics for policy makers have been identified, e.g., regarding the procedural requirements at least three categories have to be considered:

(1) Simlicity and transparency: The metric should be easiliy to understand and to use.
(2) Flexibility: The methology should be open to advancements in scientific hkowledge and to changes in the negotiation process.
(3) Political feasibility: Uncertainties should be reduced to the maximum extent possible.

So far, 7 papers were published or have been submitted.