There are several known differences between trends in the reanalyses products (observations constrained by models), and trends in historical climate model runs, and given the limitations of both products, it is not clear which is closer to the truth. A mechanistic understanding of how moisture fluxes, and thus other parts of the hydrological cycle, change in the reanalyses and in climate models, will hopefully provide some constraints which will help understand modelreanalyses biases and narrow the uncertainty in future model projections. This is specifically important for Arctic Amplification, for which the relative contribution of changes in poleward moisture fluxes is uncertain.

In Franzke and Harnik (2023) we find a significant weakening of both bandpass and low-pass eddy moisture fluxes, in contrast to GCM projections and most idealized models and current theoretical projections. A possible cause for this is the observed significant drying of the subtropics also found in JRA55, which is not accounted for by the most basic theoretical models. It is also possible that these features of the reanalysis are driven by biases in the reanalysis system.

A current project aims to better understand the processes governing changes in the transport of water vapor through the atmosphere, as the climate system will warm, and their contribution to Arctic surface warming.

​

Franzke C. and N. Harnik, 2023: Long-Term Trends of the Atmospheric Circulation and Moist Static Energy Budget in the JRA-55 Reanalysis. J. Climate, 36, 2959–2984.