What caused the increased rainfall during the Pliocene in today’s hyper-arid Atacama Desert?

The Atacama Desert in Northern Chile is considered to be the driest warm desert on Earth. However, its hyper-aridity was punctuated by more humid phases in the past. A prominent paleoclimate period of increased rainfall is the warm Pliocene (ca. 5.3 to 2.6 Ma before present). The processes leading to this rainfall increase are not well understood. While recent studies propose regions Southwest or East of the Atacama Desert as moisture sources for enhanced rainfall, the tropical South-eastern Pacific has so far been overlooked as a potential source. The goal of this study is to identify potential key drivers for the increased rainfall in the Pliocene. With this aim, we have downscaled PMIP4-CMIP6 mid-Pliocene (3.2Ma) and historical (1985-2014) experiments using the regional climate model WRF with a spatial resolution of 10 km. The mid-Pliocene WRF simulation exhibits increased mean annual rainfall in the hyper-arid core of the Atacama Desert when compared with the historical period consistent with paleo-records. This increase can be attributed to frequent intense rainfall events in austral winter during the mid-Pliocene, often associated with strong upper-level moisture conveyer belts (MCBs) originating in the tropical Southeast Pacific in front of mid-tropospheric troughs. In present-day climate, such MCBs are much weaker and mostly originate from the relatively dry subtropics producing less moisture advection. Our clustering of upper-level moisture fluxes uncovers systematic differences between MCBs in the mid-Pliocene and the present-day climate: for the mid-Pliocene we found clusters of strong MCBs from the tropical Pacific, that do not occur under present-day conditions. This is due to stronger troughs off the Atacama Desert paired with warmer sea-surface temperatures in the tropical Eastern Pacific and along the northwest coast of South America during the mid-Pliocene, which are favourable for the development of MCBs. The winter rainfall amount associated with MCBs in the mid-Pliocene far exceeds the present-day total rainfall amount. We therefore conclude that tropical MCBs are a key driver for increased rainfall in the Atacama Desert and should be assessed for future climate. This will require kilometre-scale simulations to resolve the dynamical processes involved.