If this thread was interesting to you, reach out! Otherwise, check out a preprint of this full study on bioRxiv!

www.biorxiv.org/content/10.1...

Successful management efforts must consider the potential negative effects of reduced patch size and overall suitability of the Northeast US continental shelf under warming conditions to cod recruitment and survival. Developing temperature-based strata for survey methods may be useful here.

Climate signals should be included in population assessment methods. VAST outputs indicate that a consequence of warming waters will be a condensed spatial distribution of cod into increasingly smaller patches of suitable habitat. Transboundary management may become more complicated as a result.

Bottom water temperatures across the Northeast US continental shelf have been warming, with changes in regional currents since ~2008 rapidly intensifying this trend. In the time since, cod stock centers of gravity have moved northeast and cod spatial ranges have condensed, especially for large cod.

Model results indicated strong relationships between cod presence/absence and depth and bottom temperature. While the spatial strata underpinning design-based indices of abundance are often defined to have relatively homogeneous depths, they do not yet consider bottom temperature.

There's a long history of speculation into environmental characteristics that shape cod spatiotemporal distribution. For my model, I explored using depth, bottom temperature, sea surface temperature, sediment characteristics, rugosity, and climate indices (NAO, AMO) as potential density covariates.

I've been developing these models for three size classes of cod (small: under 40 cm TL; medium: 40-70 cm TL; large: over 70 cm TL) at seasonal time steps (Spring: March-August; Fall: September-February) 1982-2021. I'm mostly interested in size-specific habitat associations and distribution shifts.

The approach--Vector Autoregressive SpatioTemporal (VAST) modeling--has the advantage of integrating these data to estimate density in areas with limited observations. Knowing both density and spatial area, you can then derive indices of relative abundance.