How would estimation of geographic range shifts of marine fishes be different when using occurrence and abundance data?

Abstract

Aim

Geographic range shifts are a common species' response to climate change. While occurrence data are commonly used to estimate species' geographical range shifts, ongoing debate suggests that local abundance data may be increasingly important for the estimates, but few studies have investigated differences between the above two types of data. We aimed to explore whether occurrence and abundance data would result in different patterns of geographic range shifts for marine fishes.

Location

Northeast US Continental Shelf, North Sea, and East Bering Sea.

Methods

We used bottom trawl datasets since 1968 in the three large marine communities to assess whether data types would affect estimated shifts in marine fish species. The range centroids of individual species were first estimated every year and linear regressions were fitted to estimate shift rates in both longitudinal and latitudinal directions. The average range centroids of the last 5 years were used to compare differences between the data types in species' shifts. We then grouped species by traits to overview species compositions.

Results

Significant differences in shift trends between regressions based on annual occurrence- and abundance-based range centroids were found in species' longitudinal shifts, particularly in the Northeast US Continental Shelf and North Sea. Approximately 38.5%–45.9% of fish species in the large marine communities had inconsistent shift directions when estimated by different data types. In comparison with the average range centroids of the last 5 years between the two data types, large changes were identified in the magnitudes of the shift distances towards the east and west. Fish species with inconsistent shifts between the two data types were mostly composed of commercial and demersal species.

Main Conclusions

The results provide observed differences over decades and suggest caution on the estimation of species' geographic range shifts using occurrence and abundance data and highlight the differences for future assessments of marine species shifts under climate change.