Geographic variation in flowering phenology in an endemic California annual wildflower, Clarkia unguiculata
Deciphering and forecasting seasonal plant activities are key parts of managing natural resources and anticipating natural hazards to ecological communities and human health (Enquist et al. 2014). As a result, understanding species’ phenological parameters and their responses to climatic variation has become a pressing objective in ecology, evolution, and natural resource management (Haggerty et al. 2013; Mazer et al. 2015). One way of evaluating species’ phenological associations with climate is with a space-for-time approach (Pickett 1989). Elevational gradients, for example, can make steep climatic gradients over short distances, allowing statistical associations to be evaluated between local climatic conditions and population-level mean plant phenology (Korner 2007, Etterson et al. 2016). The recent availability of extensive gridded climatic data through central databases (e.g., PRISM Climate Group) enables researchers to evaluate climatic influences on the phenology of populations and species across complex landscapes. Consequently, assessing geographic patterns of intraspecific phenological variation and its co-variation with climatic conditions is now a viable approach to forming a foundation from which to evaluate future changes in phenology and its potential effects on ecological communities and human health.
A substantial amount of our knowledge on the geographic variation of phenological traits in plants comes from studies on the onset of reproduction, in particular the first flowering date (FFD) of angiosperms. As a result, FFD is commonly used as a proxy for an individual’s entire flowering and reproductive season and has become a standard metric of comparison across studies – so much so that it has been identified by scientists and policy makers as a key indicator by which to assess and compare species’ long-term rates of phenological responses to climate change (EPA 2014). Because of this tremendous focus on FFD, the dynamics of the entire flowering season are often overlooked, and as a result far less is known about other parameters characterizing flowering phenology, specifically individual lifetime flowering duration and the overlap (synchrony) of flowering among population members. Thus, while initiating flowering early relative to surrounding conspecifics is widely understood to be of high adaptive value and show predictable geographic variation, far less is understood about whether the duration and synchrony of flowering exhibits similar geographic variation and confers adaptive value.
Understanding the quantitative relationships among these phenological parameters and their associations with climatic conditions is required to forecast the effects of climate change on these ecologically important reproductive attributes. Particularly in semi-arid regions where plant growth and reproduction are limited by both water and temperature, it is critical that we improve our understanding of how multifaceted climatic conditions influence phenological parameters.
Here, I present two chapters examining geographic variation in plant phenology. In the first chapter, I review the literature reporting geographic variation in intraspecific plant phenology as well as the literature reporting on the adaptive significance of flowering time with a focus on flowering onset, duration, and synchrony. In addition, using gridded climatic data, I explore the potential for temperature and precipitation gradients to co-influence growing season conditions across a 1000m elevational gradient in the semi-arid ecosystem of California’s Sierra Nevada. In the second chapter, I run an experiment to investigate whether long-term winter-spring climatic conditions may have influenced the evolution of flowering onset, duration, and synchrony in an annual wildflower, Clarkia unguiculata, which is found in these semi-arid habitats. I detected extensive genetically based differences among populations for each phenological parameter. When grown in a common environment, populations originating from low latitudes and elevations characterized by relatively warm and dry winter-spring conditions flowered significantly earlier, for a longer duration, and with lower synchrony than populations originating from higher latitudes and elevations characterized by relatively cool and mesic conditions. Overall, latitudinal and elevational clines in flowering phenology mirrored latitudinal and elevational gradients in long-term climatic conditions. However, variation in flowering duration was best explained by days to flowering, and variation in synchrony was best explained by duration. If these geographic patterns reflect the outcome of adaptive evolution on flowering time, then the warm and dry conditions forecasted for California in the coming decades are likely to exert direct selective pressures on flowering time, which may cause the evolution of earlier onset of flowering, longer potential flowering duration, and lower flowering synchrony.