On the Large Range of Rain Rates Observed over the Tropical Warm Pool during Periods of High Column-Integrated Relative Humidity

The relationship between tropospheric humidity and precipitation intensity in the Tropics has been extensively documented. Ample lower-tropospheric humidity in particular supports the growth and sustenance of deep convection. Generally, precipitation rate may be described as exponentially increasing as a function of CRH. For example, following Rushley et al. (2018), predicted rain rate in an environment given CRH of 40, 60, and 80% is, respectively, 0.02, 0.47, and 8.97 mm day^-1. However in observations, while rain rate is usually near zero for CRH < 50%, a wide range of rain rates occurs at high CRH.

This study explores reasons for the spread in rain rates at high CRH in the Tropics. A combination of ground-based precipitation radar data at locations over the Indo-Pacific warm pool characterized the areal coverage, rain-type (convective, stratiform, or isolated) and rain rate associated with precipitation echoes and rawinsonde data described the thermodynamic and dynamic structure of the convective environment. The observations mirror previously documented exponential relationships between CRH and rain rate. We then examined observations only when CRH ≥ 80%. Typical mean rain rates within the radar domain during these times ranged from 0–4 mm hr^-1. At these times, the mean rain rate within the radar domain was a linear function of the areal coverage of intense convective echoes, regardless of the total areal coverage of rainfall when including less intense stratiform precipitation. Therefore, at least on the spatial scale of a radar domain (~70000 km²), observed rainfall at high CRH may depend on whether convection consists of many intense convective elements or has evolved into wide, but weakly raining, stratiform regions.

Second, there were times when almost no echo was observed despite presence of high CRH. Using rawinsonde data, we compared profiles of zonal and meridional wind, temperature, and relative humidity during periods of high and low areal coverage of radar echoes. Statistically significant differences existed only in the profiles of temperature and relative humidity: 1) The upper-troposphere was moister during periods of high areal coverage of precipitation. 2) The lapse rate between 2–4 km was about 0.25 K km^-1 higher (i.e. less static stability) during periods of high areal echo coverage. Differences are consistent with, although larger in magnitude than, those in profiles derived from ERA-Interim reanalysis. Additionally, satellite based observations of sea surface temperature (SST) indicate no differences in SST or SST gradient between periods of no convection and widespread convection. The results hint that changes in the large-scale environmental subsidence, which alter static stability in the lower troposphere, may contribute to the spread in observed rain rates at high CRH.