x

Detection of state-dependent prediction skill using an adaptable, machine learning-based approach

Figure caption: Illustration of the neural network architecture, including SST inputs (left) and the region over which a conditional Gaussian distribution of 500 mb heights is predicted (right).

Check out this new Research Highlight, written by Kyle Shackelford (Colorado State University), that summarizes research presented in the Journal of Geophysical Research: Atmospheres article, A regimes-based approach to identifying seasonal state-dependent prediction skill. Shackelford et al. (2025) present a framework for identifying state-dependent prediction skill using neural networks and self-organizing maps (SOMs), and offer an example application to prediction of North Atlantic atmospheric circulation.

↑