1 Introduction
Global climate change affects animals in many ways, from phenology, geographic distribution, and phenotypic traits to distributions and population dynamics (Walther et al. 2002; Parmesan 2006; Gardner et al. 2011). Changes in phenotypic traits correlated with global warming mainly involve temporal trends in body mass and size (Gardner et al. 2011), two traits that are pivotal for individual life histories (Sauer and Slade 1987). Importantly, although mass and size are often used interchangeably, their change arises from different underlying physiological mechanisms. Mass is more sensitive to short-term environmental conditions affecting fat stores and the mass of the gastrointestinal tract and other organs (Piersma et al. 1999; Hume et al. 2002; Canale et al. 2016).
How climate influences body size is often described based on Bergmann’s rule, which in its commonly used version describes a positive relationship between body size and latitude within species. Individuals of many homeothermic species are smaller at lower latitudes where temperatures are higher (e.g., Ashton et al. 2000; Ashton 2002). This implies that a similar relationship should occur in response to temperature change over time. It has been postulated that a decrease in body size is the third universal response to contemporary global warming in addition to changes in distribution and phenology (Gardner et al. 2011). Although the majority of studies did not find significant temporal changes in the sizes of birds and mammals, there are many examples of long-term body size and body mass changes (recorded over one to several decades) correlated with global climate change. While a decrease in size is indeed the major response in birds, the situation is much less clear in mammals, where many species increase in size (Teplitsky and Millien 2014; Naya et al. 2017; Nengovhela et al. 2020). This increase in size is usually explained as a reaction to increased food availability as a result of increasing temperatures (Yom-Tov and Geffen 2011; Boutin and Lane 2014 and references cited therein).
Soricine (red-toothed) shrews, especially Sorex shrews, are an excellent study system to investigate climate change-induced patterns in species-level body size changes. They have extremely high metabolic rates, much higher than expected for their body mass (Taylor 1998; Ochocińska and Taylor 2005). Consequently, shrews require a constant high food supply (Hanski 1994; Keicher et al. 2017). In addition, the body size of several Sorex species, represented by skull length, shows positive correlations with temperature and negative correlations with latitude across their distribution range, contrary to Bergmann’s rule (Ochocińska and Taylor 2003). It was suggested that food scarcity during winter in cold northern climates is a major factor selecting for small body size in shrews. This is in line with the “Resource Rule” of McNab (2010), which posits that “mammalian species will become larger or smaller depending on the size, abundance and availability of resources”. A decrease in body size with latitude and in cold areas was also found in the masked shrew, Sorex cinereus , in Alaska and was also interpreted as related to better food resources in warmer areas (Yom-Tov and Yom-Tov 2005). In agreement with this, the body size ofS. cinereus in Alaska increased during the second half of the twentieth century along with increasing temperature, presumably due to increasing food availability in winter (Yom-Tov and Yom-Tov 2005). However, an increasing temperature, especially when coupled with a decrease in precipitation, can also lead to a decrease in food resources. Drought has a negative impact on the abundance of the shrews’ invertebrate prey (Coyle et al. 2017). This includes earthworms (Lumbricidae), the main food of the common shrew, S.araneus(Shchipanov et al. 2019), the availability of which depends on high soil moisture (Coyle et al. 2017; Singh et al. 2019).
The importance of small size in soricine shrews under unfavourable conditions (caused directly or indirectly by climate) is also visible in the remarkable seasonal morphological changes they exhibit. Individual skeletal size in endotherms remains more or less inert once full size is reached, which occurs rapidly in small mammals, and climate change causes size changes in the skeleton at the species or population level. However, soricine shrews undergo a profound seasonal and reversible transformation: individual seasonal shrinkage and regrowth of the skull and brain along with other organs and tissues (Dechmann et al. 2017; Lázaro et al. 2017). This change in the skull shape and height as well as the body mass of soricine shrews and some other small high-metabolic animals is called Dehnel’s phenomenon (Dehnel 1949; Pucek 1963, 1970; Dechmann et al. 2017). After juveniles are fully grown, their brain mass decreases by 21% on average, and skull height decreases by 13% in S. araneus . Their brain mass reaches a minimum value in winter and then partially regrows in spring (Lázaro et al. 2021). Body mass, in contrast, decreases in anticipation of winter and then almost doubles in spring as the animals achieve sexual maturity. S. araneus, born in early summer, has a maximum lifespan of approximately 14 months, and almost all individuals die before the second winter (Pucek 1981). The size decrease in winter is thought to be an adaptation to harsh climatic conditions in these nonhibernating, high-metabolic animals. The low body mass in shrews in winter has been hypothesized to reduce absolute food requirements when food availability is limited (Mezhzherin 1964; McNab 1991; Taylor et al. 2013). Accordingly, small winter animals have the same mass-corrected energy consumption as larger, first and second summer shrews, even under ambient conditions with temperatures differing by as much as 30 °C (Schaeffer et al. 2020). This results in absolute energy savings in winter and correlates with reduced food requirements. Disproportionally reducing the size of the brain, an energetically expensive tissue (Aiello and Wheeler 1995; Isler and Schaik 2006), may lead to further energy savings.
The magnitude of the decline in skull height in S. araneus , a commonly used proxy for Dehnel’s phenomenon, increases towards the northeast and is linked to large-scale environmental conditions and probably also the local habitat structure (Pucek 1970; Lázaro et al. 2021). It is positively correlated with temperature seasonality, annual temperature range, and other climate parameters, although no such relationships have been observed with the skull height regrowth (Lázaro et al. 2021), which has led to the hypothesis that the shrinking may be the result of different evolutionary drivers than the regrowth. In line with these observations, the individual decrease in skull height inS. araneus is flexible and modulated by ambient temperature (Lázaro et al. 2019).
Our aim was to confirm that the change in climate over several decades has had an impact on 1) the overall size of S. araneus as measured by skull dimensions and 2) the intensity of the reversible seasonal size change (Dehnel’s phenomenon). We combined climate data and skull size measurements of S. araneus from a 52-year series of specimens collected in the Białowieża primeval forest, NE Poland, from 1953 to 2004.
We first hypothesized that, because they are larger at lower, warmer latitudes, the body size of S. araneus increased over the years as a response to increasing temperatures or decreased as a response to a dryer climate leading to a decreasing soil water level and thus lower earthworm availability in the Białowieża Forest. We further hypothesized that there should be a negative relationship between the winter skull height decrease and winter temperatures. Conversely, the winter skull height decrease should become greater if climate change is combined with increasingly less favourable food availability in winter than in summer.