A natural history model of New England salt marsh die-off.

Natural history gave birth to ecology and evolutionary biology, but today its importance is sometimes marginalized. Natural history provides context for ecological research, a concept that we illustrate using a consumer-driven vegetation die-off case study. For three decades, local predator depletion promoted the formation of high-density crab (Sesarma reticulatum) grazing and burrowing fronts, resulting in the spread of vegetation die-off through southern New England and Long Island marshes. We review results from a decade of research on this phenomenon and synthesize these findings with new field surveys, experiments, and historical reconstructions to test the hypothesis that the locations and processes of vegetation die-off and recovery are spatially predictable. We discovered that crab-driven die-off consistently begins on marsh creek heads, where peat and high flow conditions overlap, before spreading to inner creeks following peat availability, stunted cordgrass, and flow. Eventually, die-off eliminates most low marsh vegetation, leaving behind unvegetated substrate too soft to support burrows. Vegetation recovery exhibits the reverse patterns of die-off; it consistently begins in the low marsh within inner creeks, where soft substrate and low flow conditions overlap, before spreading to creek heads. This spatially explicit, substrate-dependent recovery eventually leads to ungrazed cordgrass abutting grazed cordgrass on the high marsh border. We present a conceptual model of die-off through recovery progression to provide managers and landowners with a diagnostic tool for identifying marsh die-off and recovery status. Collectively, this work illustrates the fundamental importance of long-term, natural history-based investigations of ecosystem dynamics in informing ecology, conservation, and management practices.