Short term climatic rescue of an endangered species? The interactive effects of climate and a pathogen on the demography of Menzies' wallflower

Background/Question/Methods: The Humboldt Bay population of the endangered plant, Menzies’ wallflower (Erysimum menziesii), is a perennial monocarp endemic to the dunes at Humboldt Bay, California. Flowering individuals are often infected by white crucifer disease that reduces fecundity. Moreover, survival, growth, and reproduction vary among individuals of different sizes, seasons and years. From 1991 to 2000, USFWS collected seasonal data of survival, individual growth, disease status, and reproduction of nearly 12 thousand marked individuals from emergence to either reproduction or death. To understand how climate and disease severity influence long-term demography of this endangered plant, we developed year and season specific integral projection models (IPMs). We used NOAA weather data to identify climatic correlates for the inter-annual and within-season variation in demographic rates of the IPMs. Using these environmental correlates, we developed a climate-based IPM which uses climatic information to make projections about population growth. We partially validated this climate-based IPM with population surveys from 1988 to 2006 and studied potential effects of changing climatic conditions on disease severity and the viability of Menzies' wallfower. Methods were developed to identify the sensitivity of the population growth rates to climatic conditions.

Results/Conclusions: This inter-annual variation in demography was predicted to cause population growth rates to vary from nearly 50% yearly reductions to 90% yearly increases. This variation in demographic rates was strongly correlated with deviations in seasonal temperatures from the historical averages. Notably, survival and growth of small-leafed individuals was lower in warmer non-winter seasons, and fecundity of flowering individuals was lower in years with warmer January to March temperatures. This latter correlation coincided with greater severity of white rust disease during years with warmer to January to March temperatures. Due to these correlates between temperature and demography, the models predicted that warmer seasons lead to lower population growth rates. While temperatures in this area exhibited a warming trend over the past century, there was a strong Pacific Decadal Oscillation signal around this trend. The warmest years, 1988 and 1992, occurred at the time Menzies' wallflower was listed as an endangered species. According to our model, warmer temperatures near this period were predicted to cause significant population declines. Since 1992, there has been a multi-decadal cooling trend which the models predicted would promote positive population growth consistent with population surveys from 1988 to 2006.