Introduction
Many fundamental aspects of an organism’s biology are controlled by body size, including metabolic rate, life history characteristics, diet breadth, and trophic position (Brown et al. 2004; Woodwardet al. 2005; White et al. 2007). Communities, particularly in aquatic environments, are often size structured and characterized by a strong negative relationship between the abundance or biomass of individuals and body size, known as the size spectrum (White et al. 2007; Blanchard et al. 2009). The slope of this relationship is related to trophic transfer efficiency (Trebilco et al. 2013; Sprules & Barth 2015). Therefore, understanding the distribution of biomass within communities connects individual- and population-level traits to community structure, function, and ecosystem dynamics (Dossenaet al. 2012; O’Gorman et al. 2012; Yvon-Durocher & Allen 2012; Trebilco et al. 2013).
Size spectra are one of the few well documented organizing principles in ecology. A large body of literature has demonstrated the consistency of size-spectra relationships in diverse ecosystems (Jennings & Blanchard 2004; Trebilco et al. 2013; Blanchard et al. 2017; Mazurkiewicz et al. 2019, 2020). A strongly negative relationship between the abundance or biomass of individuals for a given body size range has been consistently documented in aquatic communities, to the point that size-spectra parameters have been recommended as a potential “universal” indicator of ecological health (Petchey & Belgrano 2010).
However, while estimated slope coefficients are universally consistent in sign (i.e. always negative), their specific values can vary in response to both natural and anthropogenic environmental drivers (Yvon-Durocher et al. 2011; Dossena et al. 2012; O’Gormanet al. 2012; McGarvey & Kirk 2018; Pomeranz et al. 2019). For example, because metabolic rates increase with both body size and temperature, environmental warming may have asymmetrical effects on community structure and biomass distributions (Brown et al. 2004; Brose et al. 2012). Bergmann’s and James’ rule predict that warmer regions will tend to have smaller species, or smaller-sized populations within a species, respectively (Bergmann 1847; James 1970). Further, the temperature-size rule states that warmer temperatures cause smaller individual body sizes in ectotherm species (Atkinson 1994). Finally, small body size is a predicted response to global warming (Daufresne et al. 2009; Gardner et al. 2011). Therefore, temperature is hypothesized to be one of the main drivers of variation in size spectra (O’Gorman et al. 2012, 2017). Other anthropogenic impacts, including land use (Martínez et al. 2016), acid mine drainage (Pomeranz et al. 2019), as well as natural variation, including seasonal variation (McGarvey & Kirk 2018), and resource subsidies (Perkins et al. 2018), have also caused slope estimates to vary.
While these and other studies suggest that environmental conditions can alter size spectra relationships, it is difficult to know how those effects scale up to broader patterns in size-spectra at the continental scale and across multiple years. A major limiting factor in large-scale studies of size spectra is the logistical challenge of obtaining consistently collected, processed, and analyzed data across a large spatiotemporal scale. As a result, empirical studies of size spectra share a fundamental limitation of small geographic scales and limited sampling through time (typically a single sample). To overcome this limitation, we estimated size spectra across stream sites in the National Ecological Observatory Network (NEON). We had two primary objectives in this study: 1) assess the broad geographical consistency of size spectra across North American streams, 2) test the hypothesis that size spectra vary as a function of temperature (measured as mean annual temperature).