First there was climategate, then glaciergate. In the first of a four-part series, Eloise Gibson talks to scientists about what they don't know, what they wish they knew and how they can find out more.

Scientists have admitted they know little about the climate change question most people want an answer to - what will happen in my city or region?

Local forecasts for towns and regions for the rest of the century have been named as one of the top areas climate scientists need to know more about by an international science journal.

New Zealand scientists agree it is a major issue - and say it is frustrating that they cannot give people hard answers about storms, droughts or rain in their regions.

The problem - singled out by the journal Nature this month as one of four top areas of uncertainty - is that models used to simulate Earth's climate work well on a global basis but can not "zoom in" to give accurate local forecasts.

Individual thunderstorms, for example, at 1km to 2km across are about a hundred times too small to be picked up. As for tracking changes that have already happened, Niwa principal climate scientist James Renwick said extreme events were so rare that even with climate change it could be 40 to 50 years before there was a good record of how they were changing.

Auckland University climate researcher Glenn McGregor said global climate models predicted more extreme events, including severe and prolonged periods of heat and drought, intense rainfall and strong wind storms.

"Knowing how climate change will play out at the regional/local scale is very important, but lack of robust information at this level continues to frustrate climate change impact assessments."

He said some climate processes were harder to model than others, especially in relation to clouds and rainfall.

Basic physical processes mean that as air warms up, it is expected to hold more moisture, meaning more rain.

Climate models suggest it will come in fewer, heavier bursts and that extra rainfall will not be evenly spread around the globe.

Dr Renwick said there were already signs of wet places getting wetter and dry places getting drier - meaning more rain for tropical regions close to the equator and high latitudes near the poles and less for the subtropical deserts.

But that is on a global scale.

"Teasing that out into how rainfall is going to change over the Wellington region versus Hawkes Bay, that is hard. We need basically faster computers ... then we can run the global model at very fine spatial scales. That will help," said Dr Renwick.

"But because there is so much chaos in the weather it is always going to be problematic.

"The smaller the scale you go to the more variable things are and the harder it is to pin down what is causing what - is this natural variability or is it climate change?"

Scientists know enough to say that more westerlies over New Zealand are likely to bring more rain to the west and less to the east in winter and spring, at least in the South Island.

Niwa is working on more detailed projections to furnish councils with better information for planning.

The Nature article named three other areas of uncertainty - actual rainfall patterns around the globe, the role of aerosols in warming and cooling, and palaeoclimate data, including why controversial tree-ring records diverge from real temperatures at some sites.