Large-Scale Meteorological Patterns Associated with Precipitation Extremes over Portland, Oregon, in Climate Models

Precipitation extremes are associated with substantial societal and ecological impacts across a spectrum of spatiotemporal scales and are projected to change globally as a result of anthropogenic climate warming. Despite this, considerable uncertainty exists in the magnitude and sign of changes in precipitation extremes at local-scales that are generally most relevant for resiliency decision-making. Adding to this uncertainty is the fact that most state-of-the-art climate models are challenged at resolving the fine scales on which precipitation extremes typically occur, particularly in regions of meteorologically-influential complex terrain. Here we introduce and apply methodology for evaluating key large-scale meteorological patterns (LSMPs) that are associated with local-scale extreme precipitation in a suite of climate model simulations of the recent climate over the Portland, Oregon, Metropolitan Region and surrounding watersheds. Because these patterns are synoptic in scale, their simulation is not subject to scale limitations imposed by relatively coarse resolution climate models, and therefore can be used as a proxy for local-scale extreme precipitation. We use a “top-down” approach that employs self-organizing maps (SOMs) to discern the range of archetypal LSMPs over the observational record, providing the foundation for observationally based model evaluation. By using a multivariate SOM input comprised of sea level pressure, 500 hPa geopotential height, and 250 hPa winds, we capture circulation near the surface, mid-troposphere, and upper-troposphere, illuminating dynamic mechanisms that produce extreme precipitation. The SOM nodes that are linked to extreme precipitation are then used as the target for a multi-climate model evaluation. Model simulated LSMPs are compared with observed LSMPs, allowing us to evaluate model performance in producing the synoptic patterns that lead to extreme precipitation. This will serve as a basis for constraining uncertainty in future projections of extreme precipitation over the Portland region in subsequent research.