Constraining Hydrological Sensitivity in Tropical Oceans

Abstract

Precipitation is an important aspect of the climate and is expected to change from increasing greenhouse gases. However, model projections of precipitation changes are stubbornly uncertain, particularly in tropical regions. It has been long recognized that much of the uncertainty in tropical precipitation changes is rooted in the uncertainty in sea surface temperature (SST) changes. But unfortunately, constraining SST changes at regional scales has been quite difficult. Here, we advocate that instead of fixating on regional SST changes, it is much more productive to focus on constraining precipitation sensitivity to SST changes (i.e., hydrological sensitivity). We show that local hydrological sensitivity varies widely among the state-of-the-art global climate models, but it can be effectively constrained by honing in on specific aspects of the observed precipitation-SST relationships. We find that despite the large inter-model disagreement, regional hydrological sensitivity in tropical oceans is accurately projected by the CMIP6 multi-model averages.