Introduction
Although much of ecology has focused on seemingly large differences among organisms and ecosystems, especially in terms of their spatial and temporal variation, the discipline has been criticised for lacking general unifying principles (McGill et al. 2006; Allen & Hoekstra 2015; O’Connor et al. 2019). Macroecology seeks to discover ecological principles by averaging over finer-scale variation to reveal large-scale statistical relationships (Brown & Maurer 1989; Lawton 1999; Gaston & Blackburn 2008). One such unifying principle is the importance of body size. Body size plays a central role in the physiology and ecology of organisms, governing processes such as respiration, metabolism, movement and trophic interactions (Peters 1986; Woodward et al. 2005; Yvon‐Durocher et al. 2011). As a result, body size is increasingly used in ecosystem models to generalise traits across vast numbers of taxa, particularly in the marine environment (Blanchard et al. 2017). Better understanding global patterns of body size, and their drivers, will provide stronger unifying principles in ecology and help support development of ecosystem models.
One of the earliest relationships identified in ecology was formulated by Bergmann (1848) in German, although translation into English (Mayr 1956; James 1970) subsequently contributed to confusion surrounding its definition. Bergmann’s Rule has been tested and verified for many taxa, including mammals (Brown & Maurer 1989; Ashton et al. 2000), birds (James 1970; Ashton 2002), fish (Wilson 2009; Saunders & Tarling 2018), reptiles (Ashton & Feldman 2003; Angilletta et al. 2004), amphibians (Olalla‐Tárraga & Rodríguez 2007), phytoplankton (Sommeret al. 2017), nematodes (Van Voorhies 1996), insects (Chown & Gaston 2010; Osorio‐Canadas et al. 2016; Tseng & Soleimani Pari 2019), crustaceans (Manyak-Davis et al. 2013; Garzke et al. 2015; Leinaas et al. 2016) and planktonic ciliates (Wanget al. 2020). Yet the taxonomic level at which Bergmann’s Rule applies is commonly debated. Some theorists consider it to be intraspecific (Ashton et al. 2000; Olalla‐Tárraga 2011), whilst others consider it interspecific (Blackburn et al. 1999; Hessenet al. 2013), leading to some confusion in the literature. Further, since the pattern was first described in endotherms, some question its applicability to ectotherms (Pincheira-Donoso et al.2008; Watt et al. 2010). Here, we consider Bergmann’s Rule to be defined as (Bergmann 1848) himself defined it: species from the same taxonomic clade (here subclass) are generally smaller in warmer regions and larger in cooler regions. It is clear that regardless of the precise definition used, much can be gained by investigating Bergmann’s Rule (Olalla‐Tárraga 2011) because spatial patterns in body size at all taxonomic levels strongly influence the ecology of a system (Peters 1986).
Commonly, temperature is considered the primary driver of Bergmann’s Rule, although some have argued that other drivers might modify the anticipated patterns across taxonomic groups (James 1970; Millienet al. 2006; Yom‐Tov & Geffen 2011). This is because latitude (and therefore temperature) is confounded with light availability, oxygen concentration (in aquatic systems), predation rate and food availability (Ho et al. 2010). Of these, light is unlikely to be a direct driver of Bergmann’s Rule because many species that follow the pattern do not depend directly on light. However, primary productivity depends on light, which could in turn influence the food available for many groups. It is also likely that drivers of Bergmann’s Rule differ between endotherms and ectotherms. For endotherms, it is generally accepted that temperature is the selective mechanism driving Bergmann’s Rule, with species from cooler regions conserving heat by being larger and consequently having lower body surface-area-to-volume ratios (Mayr 1956; Hessen et al. 2013). But this is not true for ectotherms (Olalla‐Tárraga et al. 2006; Watt et al. 2010). Ectotherms might benefit at cooler temperatures from increased cell sizes (Van Voorhies 1996; Hessen et al. 2013; Leinaas et al. 2016), selective protection from mortality, or increased fecundity, all of which scale with size (Yampolsky & Scheiner 1996; Vinarski 2014). Alternatively, the pattern might emerge as a result of a confounded driver such as oxygen concentration. For example, the ‘oxygen (limitation) hypothesis’ suggests that the size of marine ectotherms is limited by concentrations of dissolved O2 (Chapelle & Peck 1999; Spicer & Morley 2019). Because gas solubility and water temperature are inversely correlated, this would predict larger sizes at cooler temperatures (Forster et al. 2012; Rollinson & Rowe 2018).
Food availability is another potential driver of Bergmann’s Rule, where more food can result in faster growth rates (Lin et al. 2013) and larger body sizes (Vidal 1980; Huston & Wolverton 2011; Andriuzzi & Wall 2018). Conversely, larger body sizes might be favoured in areas where food is scarce, because animals must forage further to find food (Belovsky 1997; Brown et al. 2017). Additionally, latitudinal variation in diet quality could further influence size (Berrigan & Charnov 1994; Ho et al. 2010).
Latitudinal variation in predation rate is also a plausible driver of Bergmann’s Rule within a taxonomic group (Wallerstein & Brusca 1982; Angilletta et al. 2004; Manyak-Davis et al. 2013) because predation rate tends to decline from the equator to the poles (Freestoneet al. 2011). Predation rate could affect the size and growth of communities in several ways: through evolution towards species that mature at varied sizes (Kiørboe 2011; Manyak-Davis et al. 2013); through selective predator behaviour (Kiørboe 2011); through predation-related mortality prior to maximum size (Angilletta et al. 2004); and through selective advantages of allocating energy to predator defences (Kiørboe 2011).
We focus on Bergmann’s Rule in marine pelagic copepods, arguably the most abundant multicellular organism on Earth (Schminke 2007). Copepods are the primary link between phytoplankton and fish in aquatic systems, and they play a central role in fisheries production (Verity & Smetacek 1996). They are crustaceans that swim weakly and thus drift in currents. Marine copepods are an ideal group for testing Bergmann’s Rule in ectotherms because they are widespread over diverse environments, from the poles to the equator. There are a large number of copepod species, facilitating a more robust test of Bergmann’s Rule. Moreover, copepods are sensitive to many plausible drivers of Bergmann’s Rule, including temperature (Vidal 1980; Miller & Wheeler 2012), food availability (Rutherford et al. 1999; Miller & Wheeler 2012), predation (Kiørboe 2011; Miller & Wheeler 2012), oxygen concentration (Rollinson & Rowe 2018) and latitude (Tseng & Soleimani Pari 2019).
A search for Bergmann’s Rule in the literature on copepods returned no previous studies; however, Brun et al. (2016) investigated what has been called the Temperature-Size Rule (Atkinson 1994) in marine copepods. This rule describes the plastic phenotypic response to temperature within a species, with warmer temperatures leading to smaller individuals (Diamond & Kingsolver 2009; Ghosh et al. 2013). As the study actually compared the mean sizes of species and not individual sizes within a species, it effectively tested Bergmann’s Rule. Based on a varied dataset collected using many different nets, Brun et al. (2016) found that body size declined weakly at warmer temperatures, but there was little or no effect of temperature between 10–30°C. However, the temperature effect was dwarfed by the effect of food, which surprisingly led to a decline in body size with increasing food availability. This contrasts with many other studies that have shown increasing body size with food concentration in copepods (Vidal 1980) and other ectotherms (Pafilis et al. 2009; Huston & Wolverton 2011; Andriuzzi & Wall 2018). In a recent study, Evans et al. (2020) found marine copepods in the North Atlantic conform to Bergmann’s Rule. Although they found that temperature (2–27°C) had a more profound relationship with body size than described in Brun et al. (2016), they did not consider the effect of food availability. Other studies have found the intraspecific version of Bergmann’s rule (often called Temperature-Size Rule) holds true for several marine copepods species (Garzke et al. 2015; Leinaaset al. 2016).
Here, we test whether food is a more important driver of body size than temperature in marine copepods, whilst considering other drivers of the potential relationships with body size, such as predation, and oxygen levels, which have not been investigated previously for copepods. We test these relationships simultaneously because they are likely partially confounded with one another, which could modify their perceived relationships with body size when tested independently. Further, we account for natural differences in size based on diet. These tests are facilitated at a near-global scale by virtue of the Continuous Plankton Recorder (CPR) survey dataset, the largest (>100,000 samples), most consistent (collected using the same device), global dataset on marine copepods (Richardson et al. 2006; Batten et al. 2019). We used a spatial comparative analysis to identify statistical relationships over environmental gradients across space (Brown 1995; Gaston & Blackburn 2008).