Theme session A addresses three topics:
- Novelty in habitat model methods: address improvements in species distribution models and the extension to community-level models, including innovations in characterizing the error structure associated with model output, evaluating the importance of predictor variables, and examining the performance of model ensembles
- From model to assessment: examine examples where habitat model output informs multistock assessment advice or where habitat models guide the development and interpretation of fishery-dependent and -independent survey data
- From assessment to management: consider studies that use habitat models to guide ecosystem-level management, spatial planning, and the management of fishing and industrial sectors
There is a growing consensus that marine resources be assessed using a more integrative, ecosystem-based approach. Habitat models have emerged as both a standalone and an accessory tool for the stock assessment process. Recently, improvements to traditional single species models have been developed that seek to predict the spatial and temporal change in habitat and provide insights into the mechanisms driving change at the species and community levels. Theme session A will survey the state of the art in the construction and application of habitat models in ICES and other regions.
Practitioners apply a wide range of modelling approaches to estimate habitat including regression, classification techniques, and machine learning. Habitat models are well suited to assimilate a wide range of environmental and biological data and presents a solution to the issue of bringing ecosystem information into the advisory process. Different modelling approaches often produce differing results, making model comparison, ensemble estimates, and model validation areas of active innovation.
Habitat-modelling output has been used to design and interpret surveys, guide dynamic stratification of the ecosystem, calibrate catchability of survey gears, and improve stock size indices, among other applications, resulting in an enhanced scientific basis for assessment advice and helping stakeholders and managers understand the expected impacts of management policies.
Habitat models also play a pivotal role in ecosystem assessment and marine spatial planning. The spatial output from these models can be queried to examine the effect of habitat change throughout the ecosystem and especially how the structure of the community may be affected by climate-driven shifts in the distribution of species.
