Introduction
Biodiversity is threatened by a multitude of anthropogenic factors, including ongoing climate change (IPCC 2014; Díaz et al. 2019). Under environmental change, species have two avenues for escaping decline and consequent extinction: adapting in situ through plastic or evolutionary responses, or moving to areas where conditions are more favorable (Davis et al. 2005). Phenology and range shifts are the most conspicuous responses of species to rapid environmental change, and they have been referred to as the fingerprints of climate change (Parmesan & Yohe 2003; Root et al. 2003). The spatial distribution of a species and the temporal manifestation of life-history events reflect its realized niche (Socolar et al. 2017). As shifts in distribution and phenology represent the mechanisms through which a spatial or temporal change in the utilization of the niche can be observed (Amano et al. 2014), they are expected to offer fitness benefits as means of adaptation under climate change. Species can adapt to climatic change by advancing their early-season phenology in accordance with warmer temperatures (Parmesan & Yohe 2003). Such shifts in phenology have often been shown to be related with positive population trends and increased demographic stability (Cleland et al. 2007; Møller et al. 2008; Saino et al. 2011; Frankset al. 2018). Range shifting towards cooler areas is the other main strategy for species to adapt to warmer climatic conditions, both under contemporary and past climate changes (Parmesan et al.1999; Davis & Shaw 2001; Donoghue 2008; Spence & Tingley 2020). Species that are able to shift their geographic distribution to track climate change and thus remain within their climatic niches are less likely to suffer population declines and local extinctions (Cooperet al. 2011; Devictor et al. 2012; Urban 2015).
The most straightforward expectation is that species would use either range or phenology shift to respond to changing climatic conditions (Hypothesis 1; Fig. 1a). Mobile species would not have as strong a need to advance their phenology as they can track suitable conditions in space. Less mobile species, however, will experience stronger selection pressure to adapt in situ and adjust their phenology in order to maintain suitable thermal conditions during critical life-history events (Amano et al. 2014; Socolar et al. 2017). Indications for such complementary thermal niche tracking via space or time have been found for both plants and birds (Amano et al. 2014; Socolaret al. 2017). Nevertheless, these strategies are not necessarily mutually exclusive and can provide enhancing benefits that together buffer species against climate warming and increased variation in extreme events. In such a case, species would benefit from usingboth range and phenology shift (Hypothesis 2; Fig. 1a). The ability to adapt in one dimension of niche utilization is likely to be correlated with high responsiveness to variability in others, not least through positive feedback loops that increase population persistence (Willis et al. 2010). Phenological timing is perhaps the most important aspect of life-history that affects species distributions (Chuine 2010), as it defines where and how successfully individuals of a population can proliferate. Hypothesis 2 is thus based on the assumption that adaptive in situ responses in phenology increase the fitness of individuals, leading to higher survival rates and more offspring (Cleland et al. 2007). Evidence suggests that stable or positive population trends, that is, no change or increase in abundance, are a prerequisite for species to expand their ranges (Mair et al.2014) as emigration is higher from larger populations (Pärn et al. 2012; Glorvigen et al. 2013). Also, the probability of successful establishment increases with the summed contribution of individuals from neighboring source populations (Hanski et al.1995; Hanski & Ovaskainen 2003).
In this study, we assess how 289 species of Lepidoptera in Finland have responded to nearly 30 years of climate change. Lepidoptera have been shown to be responsive to climate change, as exemplified by the observed range shifts towards higher latitudes (Parmesan et al. 1999; Pöyry et al. 2009; Chen et al. 2011) and altitudes (Konvicka et al. 2003; Wilson et al. 2005), as well as phenology shifts (e.g., advance in the date of first appearance [Roy & Sparks 2000; Stefanescu et al. 2003; Diamond et al.2011] and increased voltinism [Altermatt 2010; Pöyry et al.2011]). Previous studies have shown that the distribution of butterflies in Finland is mainly determined by climatic factors (Luotoet al. 2006) and that the phenology of moths tends to be controlled by temperature (Valtonen et al. 2014). During the last few decades, the mean temperature in Finland has risen by 0.2-0.4 °C/decade (Mikkonen et al. 2014; Fig. S1), with springs starting earlier and the timing of phenological events being advanced for many species (Helama et al. 2020; Hällfors et al. 2020). Therefore, we expect Lepidoptera to respond to changing climatic conditions by either shifting their ranges or adjusting their phenology in situ, or both. However, it is unclear how these different strategies influence species’ population trends. Here we ask whether Lepidoptera have advanced their mean flight period (=adaptive phenology shift) or shifted their northern range boundary (NRB) northwards (=adaptive range shift). Using these response estimates, we test two main hypotheses: do the same species tend to 1) either advance their mean flight period or shift their NRB northwards, or 2) both advance their mean flight period and shift their NRBs northwards (Fig. 1b). To gain understanding on the potential causes and effects of the response combinations, we further relate the responses to life-history traits and population trends (Fig. 1).