Incorporation of colonization pressure into the propagule pressure-based global ballast water standard

In 2024 cargo vessels must meet the International Maritime Organization’s global ballast water discharge standards (IMO D-2) that limit the concentration of living organisms. D-2 focuses on reducing invasion risk by reducing ‘community propagule pressure (CPP)’, though it does not consider colonization pressure (CP). We modeled risk differences in IMO D-2-compliant discharges (10 ind. m-3) for communities that had inverse patterns of CP and species’ individual propagule pressures (IPP). Secondly, we determined the effect on risk of varying CPP and CP. As part of this we tested whether the IMO D-2 standard for zooplankton-sized organisms of <10 individuals m-3 was an optimal choice. Risk was defined as probability of at least one species invading using four risk-release models. Risk differed strongly at the D-2 limit based on community composition. At low CPP (<25 ind. m-3), risk was strongly affected by CP for hyperbolic and linear risk-release models and weakly for exponential an..., Data was collected through a mechanistic model. The model considered real-world Ballast water volumes from Bradie, Rolla, Bailey & MacIsaac (2022), which consisted of 10000 recorded ballast water discharges that ranged from 20 to 1060620 m3 (13871 ± 25738, mean ± standard deviation). Using these volumes we then set community propagule pressure and colonization pressure while varying individual propagule pressure to determine the risk of the establishment (1 - risk of extinction). Next, we set individual propagule pressure, colonization pressure, and community propagule pressure and determined the risk of establishment. Each combination of community propagule pressure, individual propagule pressure, and colonization pressure was run for 100 iterations to simulate 100 trips. 
The risk of establishments per species across 100 trips was used to calculate the probability of at least one species establishing following Wonham, Byers, Grosholz & Leung (2013). Additionally, the probabili..., All data files are provided in comma-separated value (.csv) format. Owing to the large file size, we recommend files be opened in statistical software (e.g. R, Python).