A New Look at an Old Problem: The Diurnal Cycle of Rainfall over Tropical Oceans

The nature of the diurnal cycle of precipitation over tropical oceans and the possible causes for the observed peaks has been an active area of research for several decades. It has long been established that one rainfall peak is observed during the early morning hours prior to sunrise. Recent research has also suggested that a second rainfall peak is observed in the early evening. This second peak is thought to be due in part a diurnal cycle in surface fluxes driven by heating of the ocean in relatively light wind conditions. In addition, atmospheric absorption of solar energy by water vapor can also help to drive this peak. There are numerous explanations and much active debate on the cause for the rainfall maximum observed in the early morning hours.

In this study, we will in-situ and remote sensing data to investigate the mechanisms for the diurnal cycle of rainfall over the tropical western Pacific using in-situ and remote sensing data from two cruises by vessels operated by the Japan Marine Science and Technology Center. A unique aspect of this study is the rawinsonde data, which included well calibrated launches every three hours. These frequent launches allowed us to investigate the diurnal variation in variables used to diagnose the convective instability, including CAPE (Convective Available Potential Energy) and CIN (Convective Inhibition). These parameters are highly sensitive to changes within and just above the bounary layer. The accuracy of the data set allowed the diurnal changes to be clearly evident. We concentrated on suppressed conditions in an attempt to understand the diurnal changes in the ambient atmosphere relatively free of the local impacts of convection.

We found that the tropical atmosphere has two times of day where the ambient atmosphere is most unstable to convection. The first time of day is in the late afternoon and the second is during the early morning hours. These times match or just precede the two rainfall peaks described earlier. An implication of this finding is that one does not need to invoke direct cloud-radiation mechanisms to explain the two rainfall peaks as the ambient atmosphere is already susceptible to more vigorous convection at these two times. We use the sonde data together with high resolution lidar, radar, profiler and radiometer measurements made from the vessel to explain how various physical processes interplay to produce the observed behavior of the atmospheric stability. These findings show that it is only through the interplay of several of the proposed mechanisms (air-sea interactions, diurnal changes in the radiative flux divergence, "tidal" subsidence) that the observed diurnal cycle can be accurately understood.