Interview with Kevin Trenberth

Q: We’ve seen that scientists don’t all have the same opinion on the repercussion of global warming on hurricanes behavior, where do you stand in this controversy?

A: The changes in the water cycle and storms (very much part of my research) is complex and not well understood by many.  With global warming there is more energy available for storms in the form of a warmer and moister environment and higher ocean temperatures.  This leads to more activity. The question is how that activity is manifested.  E.g. it could be in a lot more thunderstorms.  Possibilities include more storms, bigger storms, longer lasting storms, more intense storms etc.  Current understanding is that storms become more intense, bigger in size but there are fewer of them.  This is because one big more intense storm can take the place, so to speak, of several smaller storms.  For instance, hurricanes leave a cold wake behind in the ocean: bigger and stronger storms amplify that and make the environment less favorable for the next storm.  Heavier rainfalls and risk of flooding is one of the clearest and often less appreciated affect.

Q: Would you attest that hurricanes change of behavior is linked to a climate change? If yes, what is for you the evidence of a substantial human contribution in the change of hurricanes behavior?

A: Undoubtedly climate change affects hurricanes.  They can not be unaffected.  Unfortunately we do not have good records of hurricane behavior.  The best record is in the Atlantic where there has been a substantial increase in activity since 1994 in association with higher sea surface temperatures.  Moreover the land falling hurricanes are few and have a large chance element related to the weather pattern at the time.  So the statistics are not robust.  However we have excellent evidence for increases in extreme rainfall events that are well understood to be associated with the capacity of the atmosphere to hold more water vapor when it is warmer.  So the theoretical understanding is at least partly in place to understand the nature of expected changes.

Q: What are the tools to measure the change in hurricane behavior?

A: This is a problem.  The best views are from satellite but they involve interpretation and that has changed over time.  Of value is the estimate of central sea level pressure and the wind strength(s).  The problem is these change rapidly with time and location.  Often a storm is labeled by the maximum wind (cat. 4 etc) but it is the wind everywhere, say within 500 km of the eye, at all levels, that matter and which relate to energy.  The latter are generally not known.  The size of storms is not included in the records.  It is not just the peak value but the ongoing value and how long they last etc that matter.  Rainfall values are much needed and are now possible with satellite borne radar.    As well as measurements, we need models to assimilate the information and make predictions.

Q:  Which models do you use? How do you translate the data into a model?

A: At NCAR we use a variation of WRF model.  We also use global climate models run at various resolutions. Data can be assimilated as is standard practice for numerical weather prediction, although tropical storms often require extra attention.