Characterization of Frequency and Intensity Rainfall Distributions and their Relationship with Physical Drivers : Application to French Polynesia

Intense rainfall events and associated flooding in the tropical high islands of French Polynesia located in the South Pacific represent a major threat for tens of thousands of people, in particular in coastal urban areas. In this context, our goal is to characterize the relationship between wet-day frequency and intensity distributions for multiple rain gauge sites located on coastal, inner valley and mountainous areas at various timescales (15 minutes to 1 day) during the period 1980-2022 and to identify the physical drivers of such distributions. This work provides indicators for integrated water resource management under high anthropogenic pressure, as well as flood risk anticipation and infrastructure design. In our daily scale analysis, we first superimpose the distributions of wet-day intensity of cumulative rainfall estimates for aggregated regions corresponding to each percentage of wet-day frequency that were derived from the global reanalysis datasets (ERA5 and MERRA2) by McErlich et al. (2023), to our estimates obtained for our each individual rain gauge data. Our regrouped distributions fit into the envelope drawn with the global distribution curves (frequency spanning from 75 to 99%), except a few cases with distributions exhibiting a higher extreme wet tail behavior, in particular for the high relief windward sector. Such evaluations are also led at subhourly to subdaily scales depending on available data. We also apply a classification of days into three weather types (WT1 referred to the Transition Season Trade Winds, WT2 the Dry Season Trade Winds and WT3 the South Pacific Convergence Zone weather conditions) that capture the main modes of variability of rainfall over the Island of Tahiti and Moorea, recently proposed by Benoit and Sichoix (2023), to evaluate the WT specific wet-day distributions. This reveals that over windward and inner valley zones the wet-day distributions are shifted toward larger extreme values for WT1 compared to WT3, whereas over leeward and coastal areas, we find more extreme values for WT3 than for WT1. In order to explain the possible links with the frequency and intensity rainfall distributions at daily to hourly scales, associated in particular with the occurrence of intense hydrometeorological events, we explore the influence of a selected set of atmospheric variables including dynamic and thermodynamic metrics extracted from the global reanalysis products.   Benoit, L., Sichoix, L. (2023): Sub-daily rainfall patterns in the mountainous regions of the Island of Tahiti: Insights from a one-year rain gauge network expansion. https://doi.org/10.1016/j.ejrh.2023.101559 McErlich,C., McDonald, A., Schuddeboom, A., Vishwanathan, G., Renwick, J.,Rana, S. (2023): Positive correlation between wet-day frequency and intensity linked to universal precipitation drivers. https://doi.org/10.1038/s41561-023-01177-4