Generally, only valuable or iconic species have data sets suited to stock assessment models. However, in abalone, high levels of spatial heterogeneity in biological characters associated with productivity means obtaining representative data concerning growth or maturity is difficult. Stock assessment models for abalone are thus highly uncertain, which is why they are managed using empirical harvest strategies (HS). Defining an empiricalHS that drives a fishery towards a target and away from a limit reference point is hard. They should always be tested using management strategy evaluation (MSE). An R-package, ‘aMSE’, enables MSE of empirical abalone HS for zones consisting of one-to-many spatial assessment units (SAU), each of which can contain one-to-very-many almost isolated populations. A size-based model, ‘sizemod’ is used in the conditioning of each SAU. The ‘aMSE’ framework is being used to compare alternative HS, explore potential impacts of such events as virus outbreaks and marine heatwaves, and explore options for management controls such as legal minimum lengths. These, and associated R packages, will be on GitHub when FRDC 2019-118 finishes. Typical MSE runs will be illustrated, as will the structural uncertainty associated with the size-based model to demonstrate why abalone still require empirical HS rather than model-based HS.