PCRIC’s excess rainfall product was first designed in 2023 in collaboration with the risk modelling firm JBA Risk Management. PCRIC’s approach is based on the number of people impacted by flooding as result of excess rainfall. JBA Risk Management, as a specialized flood modelling firm, serves as the calculation agent responsible for determining if a payout is due following an excess rainfall event.

In collaboration with JBA, the global (re)insurance broker WTW, and with financial and technical support from the World Bank Group, PCRIC’s methodology for the disaggregation of the population of each country was refined in 2024. The revised approach is used as an input into the excess rainfall, tropical cyclone and earthquake products to model the number of people impacted by each hazard.

The trigger for a payout is based on the number of people impacted by flooding resulting from an excess rainfall event.

Rainfall observation and defining an extreme rainfall event

PCRIC’s excess rainfall products are underpinned by state-of-the-art technology, utilising satellite data to capture comprehensive weather parameters. Advanced analytics merge this data with terrestrial observations, applying climatological adjustments to refine the accuracy of satellite-derived information.

Rainfall is measured across grid zones approximately 10 x 10 km in size (black squares in the figure below), with a unique risk profile developed for each grid cell. Rainfall is measured daily through PCRIC’s Rainfall Monitoring Tool, with an extreme rainfall event defined as the occurrence of any grid cell experiencing rainfall within a 24-hour period which exceeds the 5-year return period for that grid cell.

The level of rainfall associated with each return periods is defined scientifically, considering historical weather data and contemporary rainfall patterns. For example, in areas historically receiving less than 200mm of rain over a 24-hour period, any amount exceeding this threshold could constitute an extreme event, potentially impacting a large segment of the population and disrupting daily life. This determination is rooted in extensive research, including the analysis of past rainfall data, weather pattern modelling, and the study of the island’s topography.

 

Figure 1: Example output from PCRIC’s rainfall monitoring tool. Rainfall is measured across the 10km-by-10km calculation grid (black squares). The red grid cells indicate where daily rainfall has exceeded the 5-year return period rainfall for that cell.

 

If the 5-year return period rainfall threshold is met in any grid cell, the start date of the extreme rainfall event is defined as the day that either that grid cell or any connecting grid cell experienced rainfall greater than the 2-year return period rainfall for that cell. The event terminates once that grid cell or any connecting grid cell does not experience rainfall greater than the 2-year return period for any two consecutive days after the starting date.

The monitoring tool is used to calculate the highest daily rainfall in all relevant cells throughout the total duration of an extreme rainfall event. The final extreme rainfall event footprint is made of the highest daily rainfall in each cell for the duration of the event.

 

Figure 2: Forming the final extreme rainfall event footprint from the highest daily rainfall value in each grid cell.

From rainfall to people-impacted

For any extreme rainfall event as defined above, the calculation agent uses advanced flood modelling to determine if a payout is due as follows:

  1. Population disaggregation: The official population count from 2020, as developed by the New-Zealand research institute GNS Science for PCRIC using the most recent Pacific Islands censuses, is disaggregated into population points using the 30m population density map data of Meta for Good and road data from the Humanitarian OpenStreetMap Team (HOT).
  2. Modelling of flood extent and depth: JBA’s advanced flood model is used to correlate the final extreme rainfall event footprint with the resultant extent and depth of flooding. This is done by selecting an appropriate flood map for each cell which reflects the level of rainfall experienced by that cell.
  3. Determining the number of people impacted: The flood maps are used to determine the depth of flooding at each population point.  The number of people impacted is defined as the total number of individuals at population points with a modelled flood depth of 35cm or greater (a benchmark derived from scientific literature indicating the onset of significant mobility and safety challenges).

 

Figure 3: Example flood map within one grid cell (blue) with the disaggregated population (red dots)
  1. Triggering a payout: The number of people impacted is compared to a set of pre-determined thresholds, with a higher number of people impacted leading to a larger payout.
  2. Cumulative assessment during the policy term: PCRIC’s approach allows for the cumulative assessment of the number of people impacted across multiple extreme rainfall events within the policy term. The number of people impacted is summed across all extreme rainfall events which impact more than a specific minimum number of people, resulting in higher payouts if a country suffers multiple extreme rainfall events within the same policy year.