‘But I can't preregister my research’: Improving the reproducibility and transparency of ecology and conservation with adaptive preregistration for model-based research

Preregistration is an open science practice which aims to improve research transparency and mitigate questionable research practices, like cherry-picking results. It helps protect against cognitive biases, like hindsight bias, that can influence how study outcomes are interpreted. There has been little uptake of preregistration in ecology and conservation, arguably because existing preregistration templates focus on null-hypothesis significance testing, whereas ecology and conservation often rely on different types of statistical modelling. 

We argue that preregistration in model-based research in ecology and conservation is both possible and beneficial, using templates adapted for domain-specific methodologies. We applied a user-centred design approach to translate the concept of preregistration into model-based research practice for ecology and conservation. 

To better align the internal logic of preregistration with the iterative and non-linear process of ecological modelling, we propose, test and evaluate a methodology for ‘Adaptive Preregistration’, using a case study of modelling managed water releases (‘environmental flows modelling’) in regulated rivers for maintaining riparian vegetation condition in Victoria, Australia. 

This research provides a template and methodology for implementing Adaptive Preregistration of ecological models. Although we focus on ecology and conservation in this paper, the concept of Adaptive Preregistration, and the templates developed here, could be applied to model-based research in other scientific disciplines. Modellers in ecology and conservation need no longer cry ‘but I can't preregister my research’.