Identifying coral refugia from observationally weighted climate model ensembles

Reef-building corals face a variety of mounting anthropogenic stressors originating both at the global scale (increasing sea surface temperatures (SSTs), rising sea levels, and ocean acidification) and the local scale (such as destructive fishing, overfishing, sedimentation, invasive species, nutrient over-enrichment, and chemical pollutants). We are working to produce more accurate and better validated downscaled projections of likely 21st century refugia from global stressors (with robust uncertainty estimates) to guide conservation management at local scales. We begin by assigning skill-based weights to CMIP model projections of SST by comparing hindcasts to gridded instrumental SST records. Our weighting method decomposes the model and observational time series into climate (low frequency) and weather (high frequency) components, and measures the similarity between the two time series with the p-value of a test of the null hypothesis that the wavelet coefficients describing the climate signals (observed and modeled) are the same. At each location, we produce a time series of weighted probability distributions of the forecast projections from the entire model ensemble and statistically downscale the weighted SST projections using satellite observations, in order to estimate the year after which SST conditions annually exceed a severe bleaching threshold of 8°C-weeks.