Premise: Shifts in the timing of life history events, or phenology, have been recorded across many taxa and biomes in response to global change. These phenological changes are often studied in a single species context, but considering the community context is essential for anticipating the cascading effects on biotic interactions that are likely to occur. Focusing on an annual grassland plant community, we examined how experimental changes in precipitation affect flowering phenology in a community context and explore the implications of these shifts for competitive interactions and species coexistence.

Methods: We experimentally manipulated rainfall with rainout shelters and recorded detailed flowering phenology data for seven annual species including two grasses and five forbs. We assessed how their first and peak flowering days were affected by changes in rainfall and explored how flowering overlap between competing species changed.

Results: Changes in rainfall shifted flowering phenology of some species, but sensitivity differed among neighboring species. Four of the seven species studied started and/or peaked flowering earlier in response to reduced water availability. The idiosyncratic shifts in flowering phenology have the potential to alter existing temporal dynamics that may be maintaining coexistence, such as temporal separation of resource-use among neighbors.

Conclusions: Our results show how species-specific phenological consequences of global change can impact community dynamics and competition between neighboring plants and warrant future research.

The experiment was conducted at the University of California, Santa Barbara’s Sedgwick Reserve in Santa Barbara County, USA. Twenty 0.75m*0.75m plots were established and seeded in October 2019 in a fenced area designed to exclude deer and gophers, the two primary mammalian herbivores in the system. One hundred seeds of seventeen cooccuring annual grasses and forbs were mixed and hand sown in each plot. The plots were paired into ten blocks. Rain reduction shelters designed to divert 50% of incoming rain were placed over five of the blocks and therefore half of the plots on February 8, 2020. This timing allowed all plants to germinate and establish in late December and January with the same ambient rainfall conditions, resulting in the rainfall exclusion treatment impacting just the growth and reproduction phase of the plants’ life cycle. The seven most abundent species that germinated were tracked over the course of their lifetime. We used daily photographs from cameras mounted at each plot to identify the flowering window for each species (i.e. first and last days of flowering). We then counted the number of flowers in the photos over that time period, avoiding images that were obscured by condensation, rain, or wind-driven camera and plant movement. This resulted in counts of open flowers every 1 to 5 days for each species during their flowering period. The plots were also visited in person every two weeks and every flowering individual was identified and counted.