We used to throw this one around back in the day...

He uses statistics like a drunk uses a lamp post, for support rather than illumination.

We are now thinking about generalising this method to the whole world so it can be used for the next IPCC projections and help coastal decision makers around the world rather than only along the Dutch coast.
os.copernicus.org/articles/21/...

Depending on the answer to this question more or less models are selected to make projections. We present different selection options and discuss how they relate to the needs of users of sea-level scenarios.

Then there is the usual tough question: how much of what we observe is forced by climate change and how much is natural variability?

We therefore decided to select models that perform well in the past. This is not straight forward because sea-level change is measured with tide gauges and satellite altimeters but ocean dynamic sea-level change is only a part of total sea level and it can't be measured directly.

We looked at "ocean dynamic sea-level change", a difficult process pushing sea level up or down depending on the regions, as a result of changes in the ocean circulation. In the models, changes in dynamic sea-level in the past 30 years is correlated with the change during the 21st century.

However, we touch on important questions for climate projections in general:
- How can we assess if climate models are fit for the purpose of making projections?
- If they are not yet good enough what do we do?