Teleconnections Reveal That Drivers of Inter-Annual Growth Can Vary From Local to Ocean Basin Scales in Tropical Snappers

Individual growth rate is one of the key traits that determine the productivity of populations. Chronological approaches that relate time series of growth and climate information present the opportunity to identify important climatic drivers of demography and thereby understand the likely impact of climate change. We constructed otolith chronologies (a proxy for somatic growth) to examine synchrony of growth patterns within and between two mesopredatory fishes (Lutjanus bohar and L. gibbus) in the remote Chagos Archipelago, Indian Ocean. We then used mixed-model and pathway analysis to relate growth responses to a suite of climatic and environmental factors to determine the extent to which variation in inter-annual growth could be predicted at individual and population levels. Our models explained up to half the variance associated with annual growth at the level of populations. Significant environmental drivers of growth differed between species, as did the spatial scale of these drivers: L. gibbus exhibited a strong relationship with regional ocean temperature, whereas growth of L. bohar was correlated with the Pacific Decadal Oscillation, suggesting influential teleconnections between ocean basins as an underlying predictor of productivity of fish populations. Our results demonstrate that (1) synchronous growth stemming from relationships with climate factors may be suppressed at very low latitudes; (2) closely related species may respond to very different environmental stimuli; and (3) within the same environment, the scale of influential drivers may be local in nature or reflect oceanographic processes stretching across entire ocean basins. We demonstrate that biochronological approaches are effective tools for reconstructing relationships between climate variability and fish growth even in tropical regions where seasonality is low, and these methods can be valuable for forecasting population-level responses to projected climate change.

DOI: 10.1007/s00338-020-01903-z