Desert plant demography and climate change: more variable is not bad

Background/Question/Methods

Desert species respond strongly to infrequent, intense pulses of precipitation. Consequently, the native biota has developed a rich repertoire of life history strategies to deal with fluctuations in resource fluctuations. Examinations of how future climate change will affect desert biota have often forecast negative effects, but these –usually correlative– approaches overlook precipitation variation because they are based on averages. Here, we implement an innovative, mechanistic approach to understand the effects of changes in precipitation on desert plant population dynamics. The approach couples robust, super-high resolution climatic projections, including variable precipitation, with stochastic, stage-structured models constructed from long-term demographic datasets of the short-lived Cryptantha flava (1997-2011) in the Colorado Plateau Desert (USA), and the annual Carrichtera annua (2002-2009) in the Negev Desert (Israel). We built integral projection models (IPMs) for the former species and periodic matrix models for the latter and analyzed stochastic population growth rate as well as their underlying vital rates (e.g., survival, changes in size, seed bank, etc) under current and projected precipitation scenarios.

Results/Conclusions

Stochastic population growth rates in both species increased under projected scenarios of precipitaiton. In the case of Cryptantha flava, where wetter, longer growing seasons are expected, survival and drastic changes in size played a major role in the demographic dynamics. In the expected drier years for Carrichtera annua, a single vital rate, seed bank, was able to not only keep the population viable, but also improve its dynamics. Our results highlight the potential for demographic buffering –reassembly in vital rate values resulting in the maintenance of a viable population growth rate (λ > 1)– that both species might display under projected shifts in precipitation regime. Our findings add two remarkable exceptions to life history theory predictions for lower population growth rates in the light of more stochastic environments in deserts. We discuss the possible role of vital rate correlations and other mechanisms in our results and suggest future avenues of demographic research in deserts.