Introduction
Identifying organisms that are vulnerable to ongoing climate warming is a challenging but critical task for ecologists and evolutionary biologists. Ectothermic reptiles are highly dependent on environmental temperature, and are thus extremely sensitive to climate warming (Hueyet al. 2012; Böhm et al. 2016). Squamate reptiles (lizards and snakes) exhibit two reproductive modes: oviparity and viviparity. Whether oviparous or viviparous species are more vulnerable to climate warming is a long-standing puzzle. Explaining this puzzle is critical for deriving conservation strategies (e.g. prioritizing species or zones for conservation) as well as understanding how historical thermal regimes have driven the transition between the two parity modes.
Although the evolutionary history and selective forces underlying the two parity modes in squamate species have received extensive attention (Shine 2014), few previous works have explored the sensitivity of oviparous and viviparous species differ to climate warming (Sinervoet al. 2010; Jara et al. 2019). However, viviparous species are thought to be more vulnerable to climate warming than oviparous species for several reasons. First, viviparous species are adaptively constrained to cold regions (high latitudes and elevations) where climate warming is predicted to be more prominent (Tinkle & Gibbons 1977; Pincheira-Donoso et al. 2013). Second, viviparous species in cold regions have relatively low body temperatures and heat tolerance, which constrains their activity as temperatures warm (Sinervoet al. 2010; Wang et al. 2017). This “cul-de-sac” hypothesis gained supports from two studies that compared highland vs. lowland lizards from Mexico and South America, because climate warming shrinks the cold regions to which they have adapted and constrains their activity more severely due to lower body temperatures (Sinervo et al. 2010; Jara et al. 2019).
However, distributions of oviparous and viviparous squamates are geographically widespread and often overlapping (Fig. 1). Specifically, distributions of 90.7% of oviparous species (2146 in 2366 analyzed) overlap with viviparous species, while distributions of 97.8% viviparous species (541 in 553 analyzed) overlap with oviparous species (see the methods). In addition, climate-driven range shifts may increase overlap between the species with different parity modes, because oviparous species’ suitable habitats shift to (currently) colder regions, where viviparous species are occupying (Jara et al.2019). Obviously, a comparison between viviparous species from a cold region and oviparous species from a warm region may not inform warming-driven dynamics of the two modes where distributions overlap. Moreover, understanding differential responses of sympatric oviparous and viviparous species requires incorporating additional organismal differences (e.g. on behavior, physiology and life history) that are ecologically important (Kearney & Porter 2009). Oviparous and viviparous squamates may be able to respond to warming in a similar manner, including shifting thermoregulation behavior and body temperature or by evolving shifted thermal performance curves (Yuanet al. 2016; Wang et al. 2017). Nonetheless, reproductive life history (egg-laying vs. live-bearing), the only fundamental and universal difference between the two parity modes, can respond to climate warming differentially, and therefore should be the focus when we compare differences between oviparous and viviparous species (Blackburn 1993; Andrews 2000).
To answer how oviparous and viviparous species would differ in responding to climate warming, we need an integrated consideration of between-lineage differences in reproductive life history from maternal investment to offspring survival. On one hand, oviparous mothers can have more clutches per year than do viviparous mothers (Mesquitaet al. 2016), because they carry the embryos only for one third of the embryonic development (Andrews & Mathies 2000). But on the other hand, as viviparous mothers thermoregulate themselves, they actively select developmental temperatures that differ from nest temperatures (much longer than do oviparous mothers, which only hold the eggs for a maximum of 1/3 of total embryonic development), substantially change developmental time and increase embryonic survival (Shine 2004; Maet al. 2018). These differences in embryonic survival and the time of hatching or birth, can furtherly lead to divergent number and energy budgets of offspring, and therefore population growth between species of the two parity modes (Buckley 2008; Levy et al.2016b). Moreover, the energetic costs of reproduction can also differ between oviparous and viviparous females. In particular, oviparous females spend relatively more energy on producing multiple clutches of eggs (Landwer 1994), whereas viviparous females spend more energy during prolonged gestation, because of large metabolic consumption when carrying large embryos (Beuchat & Vleck 1990). These different reproductive costs may shift differently under climate change, influencing the total energy balance of a population, affecting energy available for reproduction in the next year. In sum, climate warming may have different effects on oviparous and viviparous squamates by differentially affecting the reproductive timing, survival and energy budgets of mothers as well as the survival and energy budgets of each clutch/litter of offspring produced.
In this study, we used Sceloporus lizards in North America as a model system to predict the impact of climate warming on oviparous and viviparous squamates, because [1] this genus includes species of both parity modes, [2] this region has the highest density of sympatric squamates (including non-sceloporus species) of two parity modes (Fig. 1), and [3] the physiology and life history traits of these species are widely available (Table S1). We developed a biophysical model that translates microclimates (hourly estimates for 1980-2000 versus 2080-2100) to operative temperatures, and simulates the life history of lizards through thermoregulation, feeding (digestion), reproduction and embryonic development. We used the model to estimate yearly remaining energy after subtracting metabolic consumption and energy content of all clutches/litters (of fresh eggs; only for reproductive females) from energy intake of adults (termed “maternal energy budget”) as well as total energy of surviving offspring (termed “offspring survival”) under current and future climates (Fig. 2). Because the energy budgets of individuals determine the survival and reproductive potential in future (Iraeta et al. 2008), we compared “maternal energy budget” and “offspring survival” between oviparous and viviparous species to quantify the impact of climate warming, and thereby assess whether these modes will be harmed by or benefit from future warming.