Key Differences Between Lakes and Reservoirs Modify Climate Signals: A Case for a New Conceptual Model

Lakes and reservoirs are recognized as important sentinels of climate change, integrating catchment and atmospheric climate change drivers. Climate change conceptual models generally consider lakes and reservoirs together despite the possibility that these systems respond differently to climate-related drivers. Here, we synthesize differences between lake and reservoir characteristics that are likely important for predicting waterbody response to climate change. To better articulate these differences, we revised the energy mass flux framework, a conceptual model for the effects of climate change on lentic ecosystems, to explicitly consider the differential responses of lake versus reservoir ecosystems. The model predicts that catchment and management characteristics will be more important mediators of climate effects in reservoirs than in natural lakes. Given the increased reliance on reservoirs globally, we highlight current gaps in our understanding of these systems and suggest research directions to further characterize regional and continental differences among lakes and reservoirs. *Correspondence: [email protected] Present address: U.S. Geological Survey, Southwest Biological Science Center, Flagstaff, Arizona Author Contribution Statement: NMH and BRD co-led the manuscript effort and contributed equally. JRC and BRD conducted the statistical analyses. KES and JRC designed the lake pairing analysis. NMH, NRR, and KES developed the climate change conceptual model. This paper was a highly collaborative effort and all authors contributed equally to the development of the research question and study design as well as the writing of the paper. Data Availability Statement: Data are available in the Long Term Ecological Research Network Information System repository at https://dx.doi. org/10.6073/pasta/17cb7958c74f8bfc135f3e7f04ee944e (Corman et al. 2016). Nicole M. Hayes and Bridget R. Deemer are joint first authors. Additional Supporting Information may be found in the online version of this article. This is an open access article under the terms of the Creative Commons Attribution NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. Scientific Significance Statement Climate change poses a significant threat to freshwater ecosystems, though the exact nature of these threats can vary by waterbody type. An existing conceptual model describes how altered fluxes of mass and energy will affect standing waterbodies, but it does not differentiate reservoirs from lakes. Here, we synthesize evidence suggesting that lakes and reservoirs differ in fundamental ways that are likely to influence their response to climate change. We then present a revised conceptual model that contrasts climate change effects on reservoirs versus lakes. 47 Limnology and Oceanography Letters 2, 2017, 47–62 VC 2017 The Authors. Limnology and Oceanography Letters published by Wiley Periodicals, Inc. on behalf of Association for the Sciences of Limnology and Oceanography doi: 10.1002/lol2.10036 Climate change and freshwaters Climate change is one of the greatest threats to aquatic ecosystems (Blenckner 2005; Hayhoe et al. 2008). The effects of climate change range from direct changes in water level (Smol and Douglas 2007) and surface-water temperature (O’Reilly et al. 2015) to indirect, complex ecological shifts that alter trophic interactions (Winder and Schindler 2004), and have been observed in many different regions (e.g., Quayle et al. 2002; Schindler and Smol 2006; Schneider and Hook 2010). At the same time, increased human demand for water-related ecosystem services has resulted in the construction and operation of over 1 million dams globally (Lehner et al. 2011). As a result, human-made lakes (i.e., reservoirs) have come to comprise anywhere between 6% and 11% of global lentic surface area (Downing et al. 2006; Lehner et al. 2011; Verpoorter et al. 2014). While the global expansion of reservoirs has increased access to drinking water, irrigation, navigation, flood control, and hydropower, it has also fundamentally changed the movement of water, sediment, nutrients, and biota through aquatic networks. Globally, reservoirs are estimated to increase the standing stock of natural river water by over 700% (V€ or€ osmarty et al. 1997), reduce sediment flux to the ocean by over 1 billion metric tons of sediment per year (Syvitski et al. 2005), reduce phosphorus transport to the coast by approximately 12% (Maavara et al. 2015), contribute more than 30% of all lentic nitrogen and silica retention (Harrison et al. 2009, 2012), and emit methane at higher per area rates than any natural aquatic ecosystem (Deemer et al. 2016). These findings are consistent with the notion that inland waters are not “passive pipes” (Cole et al. 2007) and that the ecological role of reservoirs is unique from lakes. As low points on the landscape, lentic ecosystems also serve a unique role as integrators of atmospheric and catchment scale climate signals (Williamson et al. 2009). These signals, or sentinel responses, are shaped by a number of factors including large-scale geographic patterns and internal waterbody processes. Climate change conceptual frameworks have previously lumped reservoirs with natural lakes (Williamson et al. 2009) or excluded them from efforts to develop broadly applied sentinel response metrics (Adrian et al. 2009). Reservoirs and lakes are generally thought to share a number of similarities. Reservoirs are often divided into three zones for the purposes of ecological study: river, transitional, and lacustrine—with the lacustrine zone having slower water velocities and pronounced thermal stratification much like a lake ecosystem (Thornton et al. 1990). As a result, the lacustrine or lentic zone of a reservoir is thought to be similar to a lake in terms of planktic production, nutrient limitation of phytoplankton growth, and biogeochemical cycling (Wetzel 2001). Despite these similarities, reservoirs and lakes also differ in a number of ways that lead to differences in ecosystem functioning. Given the growing influence of reservoir ecosystems on the global hydrologic system (Zarfl et al. 2015), and recent evidence that reservoirs may serve ecological roles distinct from lakes even in the lacustrine zone (Beaulieu et al. 2013), we argue that these human-made systems should not be lumped with natural lakes in climate change conceptual models. An improved understanding of the interaction between reservoirs and climate may have broad scale implications for water quality from headwaters to coasts. Key differences between lakes and reservoirs Teasing apart the ecologically relevant differences between reservoirs and natural lakes can be a daunting task given the background variability in lentic ecosystem types. For example, one of the most common lake typological divisions is based on water source (i.e., relative contribution of groundwater versus surface water), of which reservoirs are a single lake type (Hutchinson 1957; Wetzel 2001; Figs. 1, 2). While natural lakes are generally subdivided based on hydrology (i.e., seepage, glacial, oxbow, intermittent, etc.), reservoirs are often categorized based on their size or their designed purpose (i.e., their primary reason for being constructed; Thornton et al. 1990; Poff and Hart 2002). Currently, the most common classification scheme for reservoirs divides these ecosystems into two groups: storage and runof-river (Poff and Hart 2002). Storage reservoirs typically store large volumes of water and have large hydraulic heads, long hydraulic residence times, and allow for relatively finetuned control over the rate at which water is released from the dam. Run-of-river reservoirs, on the other hand, typically store less water and have relatively small hydraulic heads, short hydraulic residence times, and little or no control over the rate that water is released from the dam (Poff and Hart 2002). In many ways, these categories represent two extremes on a spectrum of reservoirs with larger “lacustrine” zones that are more like lakes (storage) to reservoirs with larger “river” zones that are more like rivers (run-of-river). To foster a more detailed discussion of reservoir type, we have depicted reservoir types based on position within river network (Fig. 1). While reservoirs created by damming a preexisting lake are likely to share characteristics with the storage reservoir category, reservoirs created by damming a preexisting river can resemble either a storage or a run-of-river system. In addition, reservoirs created to store water outside of a river network (e.g., farm ponds, pump storage systems, etc.) may function quite differently than those receiving surface water inputs via stream or river inlets. In this article, we define reservoirs broadly as any humanmade lake, whether it be embedded within a river network or not (Fig. 1). However, for the purpose of our analysis, we restricted our comparison to reservoirs within river networks (Fig. 1). This selection was made to harmonize our reservoir comparison with the types of reservoirs considered in the Hayes et al. Lakes and reservoirs modify climate signals