Fig. 4From: Projections of temperature-attributable premature deaths in 209 U.S. cities using a cluster-based Poisson approachMonth and cluster differences in temperature mortality effects. Legend: This figure shows the different premature mortality response to temperature by month and cluster. The kink in the response line for a cluster is at the median temperature for that cluster in that month based on 2003–2006 weather dataBack to article page