Introduction
The activities of interacting species are often coordinated in time and any shifts in the timing of ontogenetic or phenological events are hypothesized to cause fitness declines in interacting species, affecting population and community dynamics (Cushing 1990; Miller-Rushing et al. 2010; Yang & Rudolf 2010). Evidence is mounting that climate change is altering the timing of environmental cues (e.g., temperature and precipitation) that species use to initiate key life history events, such as seasonal emergence or activity (Parmesan & Yohe 2003; Forrest 2016). Furthermore, differential responses across species within the same community indicate that the temporal coordination of species interactions is changing, which raises concerns that interacting species are becoming temporally “mismatched” (Visser & Both 2005; Bothet al. 2009; Kharouba et al. 2018). Such mismatches can lead to major changes in the outcome of species interactions and population dynamics, but our understanding of the mechanisms linking phenological asynchrony to their immediate and long-term consequences is unclear (Van der Putten et al. 2010).
The intricacies of how unequal shifts in emergence phenologies disrupt the timing of species interactions are critical in order to understand how climate change can decouple species interactions (Kerby et al. 2012; Johansson et al. 2015). For many species interactions the outcome depends, in part, on the developmental stages of interacting individuals. For instance, predators tend to attack specific developmental stages of their prey, and pollinators require plants to be in a certain flowering phase (Stacconi et al. 2015). Thus, the extent that interaction strength and resulting dynamics change with shifts in phenology depends on how much the outcome varies across developmental stages (e.g., the extent traits of individuals change during development) and how long individuals remain available to interact (e.g., window of vulnerability) (Memmott et al. 2007; Miller & Rudolf 2011; Johansson et al. 2015). Besides contributing to shifts in phenologies, warmer temperatures also accelerate growth rates, which shortens the amount of time spent within each developmental stage (Kingsolver et al. 2011). This ultimately narrows and shifts the window of vulnerability for predator-prey interactions, or magnifies size differences in competitive interactions (Benrey & Denno 1997; Rudolf & Singh 2013), potentially exacerbating the effects of shifts in the emergence phenology of interacting species (but see (Tuda & Shimada 1995). For example, faster development rates of prey in warmer temperatures can result in natural enemies missing their window of opportunity to attack, even if the relative phenological relationship remains unchanged (Klapwijk et al. 2010; Ren et al. 2020). Thus, abiotic or biotic environmental conditions that change growth rates, or the temporal availability of resources are likely to modify how the strength of species interactions scales with phenological shifts, though few studies have quantified these effects.
Phenological shifts are taking place across a variety of community contexts, several of which are known to alter growth rates and the temporal availability of interacting species. For example, strength of intraspecific competition (e.g. resource limitation or host density) and presence of alternative resource species, can affect growth rates and temporal availability at a similar magnitude to changes attributed to global warming (Benrey & Denno 1997; Wolf et al. 2017). Development rates often slow down when levels of resource competition are high, prolonging the number of days which prey species remain vulnerable to attack (Barker & Podger 1970; Benrey & Denno 1997). Similarly, the presence of alternative resource species, with complementary development rates or phenologies, can help prolong the total period during which resources remain available (Wolf et al.2017). However, most phenological studies have ignored community contexts beyond pairwise interactions, which does not reflect the reality that consumers typically utilize multiple resource taxa and that levels of resource competition can show high spatial and temporal variation (Nakazawa & Doi 2012; Revilla et al. 2014). Furthermore, ongoing global declines in insect diversity and abundance make these two aspects of community context particularly imperative to study (Forister et al. 2019; Salcido et al. 2020). Such community contexts may act as a buffer against severe changes in the strength of interactions and promote interaction persistence in the face of shifting phenologies (Yachi & Loreau 1999; Timberlake et al.2019; Olliff‐Yang et al. 2020), yet it’s uncertain how these processes will act in future temperatures predicted by climate change. Therefore, there is a pressing need for studies that integrate relative interaction timing with concurrent shifts in temperature and community context.
Here, we use laboratory experiments (Fig. 1) and dynamic simulations to examine how warming and community context together modify the effects of phenological shifts on the strength and dynamics of host-parasitoid interactions. Parasitoids play an important role in determining host population dynamics and are commonly used as biological control agents, yet few studies have investigated the impacts of climate change on the relative timing of host-parasitoid interactions (Klapwijket al. 2010; Dyer et al. 2013; Jeffs & Lewis 2013). Using a native Drosophila -parasitoid system from seasonal tropical forests of North Queensland, Australia (Jeffs et al. 2021), we experimentally delayed the emergence phenology of parasitoids relative to their hosts and assessed how resource competition and presence of an alternative host species modified the effect of delayed emergence in ambient (24°C) and predicted warming temperatures (28°C) (Shuklaet al. 2019). It is important to note, that previous phenological studies have focused on how increased temperature acts as the environmental cue that drives shifts in emergence phenology among interacting species. However, elevated temperatures often persist throughout growing seasons. Thus, temperature is not only the cue; it is also the context in which mismatched ecological interactions must proceed. Therefore, we decoupled the effects of temperature and phenology to quantify the effects of different phenological shifts in current ambient and predicted warming temperatures. Additionally, we applied parameters derived from the single-generation experiment to simulate host-parasitoid population dynamics over 100 generations under our experimental conditions using an age-structured Nicholson-Bailey host-parasitoid model to examine their impacts on long-term persistence. Together, this allowed us to assess (1) how temperature alters the effects of phenological shifts on the outcome of host-parasitoid interactions, (2) to what extent resource limitation and (3) the presence of an alternative host species modifies effects of phenological shifts under warming, and (4) how community contexts and temperature interact to affect long-term persistence of host-parasitoid interactions. Combining experimental and modeling approaches helped identify which conditions were favorable over a single generation, but detrimental for long-term persistence of host-parasitoid interactions. The results represent an important step toward understanding how warming and community context interact to modify the effects of phenological shifts on the strength and dynamics of species interactions, which is critical for predicting how ecological communities will respond to climate change.