The impact of run-of-river dams on sediment longitudinal connectivity and downstream channel equilibrium
Introduction
By disrupting the downstream flux of sediment and nutrients and blocking the upstream mobility of resident and diadromous fishes (Nilsson et al., 2005; Nilsson and Svedmark, 2002), large dams profoundly affect hydrologic, ecological, and geomorphic systems at local (Brandt, 2000; Church, 1995; Curtis et al., 2010; Gregory and Park, 1974; Juracek, 2000), watershed (Graf, 1999; Magilligan et al., 2008), and global (Syvitski et al., 2005; Vorosmarty et al., 2003) scales. However, in contrast to the considerable research on large, flow regulating dams, less attention has been given to smaller Run-of-River (RoR) dams, notwithstanding their greater occurrence across the landscape. In the United States (US), for example, the National Inventory of Dams (NID) documents >80,000 large dams, while the number of smaller dams may exceed two million (Smith et al., 2002).
Run-of-River dams span the width of the channel but do not impede discharge (Csiki and Rhoads, 2010). They have short hydraulic residence times, with reported values of <2 h (Ashley et al., 2006; Poff and Hart, 2002), and limited sediment trapping efficiency (Brune, 1953), especially at high flows (Csiki and Rhoads, 2014; Pearson et al., 2011). Where some sediment storage occurs, or sediment flux is partially attenuated, the downstream impacts of RoR dams mirror those of larger structures with statistically significant bed grain coarsening (Fencl et al., 2015; Magilligan et al., 2016a) or significant decreases in fine sediment (Csiki and Rhoads, 2010; Skalak et al., 2009). In other cases, RoR dams are not generally observed to create significant discontinuities in channel morphology or sediment character (Csiki and Rhoads, 2014).
The muted downstream impacts of RoR dams are likely caused by their limited sediment trapping efficiency, as evidenced by active scour and fill processes in the reservoir behind RoR dams (Csiki and Rhoads, 2014; Pearson et al., 2011), particularly after the reservoir storage capacity is filled (Major et al., 2012). Pearson et al. (2011) documented that reaches downstream of the RoR Merrimack Village Dam were commonly characterized by episodic deposition and remobilization of material (primarily sand) transported over the dam. Less is known about the transport of coarser material over RoR dams, but Pearson and Pizzuto (2015), based on field observations and 1D flow modeling, hypothesized that sediment deposition in the reservoir constructs a ramp up to the dam crest that facilitates transport of coarse material over the dam. They modeled the potential for coarse sediment transport over an impoundment with a ramp and estimated that clasts up to ~23 mm could move through the reservoir and ultimately over the dam crest. Yet they did not directly observe coarse sediment transport over the impoundment. Casserly et al. (2020), however, recently observed tracer particles greater than the median grain size travelling through and over two low-head dams in southeastern Ireland, and argued that the retention of coarser sediment after reservoir scour events led to transient storage behind the dam and resulted in transient supply-limited conditions downstream.
An important outstanding question is: when sediment flux disruption across RoR dams occurs, is it sufficient to cause channel morphological disequilibrium downstream of the dam? Although significant changes in channel bed grain size are observed downstream of some RoR dams (Csiki and Rhoads, 2010; Fencl et al., 2015; Magilligan et al., 2016a; Skalak et al., 2009), it is unknown if possible covariate adjustments in other channel parameters (i.e., flow depth, slope) are sufficient to compensate for these changes in channel bed grain size such that an equilibrium channel form is maintained even as grain size changes (Curtis et al., 2010; Dade et al., 2011; Renshaw et al., 2019).
Here we focus on the disruption of bedload transport because it is more likely to be affected by RoR dams than suspended sediment transport. To directly link the frequency that bedload sediment is transported over RoR dams with downstream channel (dis)equilibrium, we couple field monitoring of sediment transport over RoR dams using Radio Frequency Identification (RFID) PIT (passive integrated transponders) tags in coarse gravel and cobble clasts with detailed reach scale analyses upstream and downstream of the dams to characterize potential channel adjustments following impoundment. Importantly, the goal herein is as much to evaluate the efficacy of sediment transport across RoR dams as it is to document and quantify the ways in which stream channels might maintain equilibrium conditions relative to transient and/or sustained shifts in sediment supply.
Identifying equilibrium is often difficult in fluvial systems as the signal to noise ratio can be muted by the inherent variability of hydrologic and geomorphic processes. At the extreme end where significant sediment trapping and hydrologic alterations occur at larger, flow-regulating dams, it is easier to identify channel dis-equilibrium following impoundment (Magilligan et al., 2013). However, for RoR dams where streamflow is minimally altered and sediment trapping is similarly less affected, traditional metrics for evaluating geomorphic change, such as channel width or cross-sectional area (Csiki and Rhoads, 2014), may not be sufficiently sensitive to identify geomorphic dis-equilibrium. To enhance the sensitivity of our analyses, we obviate a singular focus on channel parameters by representing the possible adjustments in terms of the Shields parameter and bed shear stress that simultaneously represent variations in characteristic flow depth, slope, and, in the case of the Shields parameter, grain size, to characterize (dis)equilibrium (Curtis et al., 2010; Dade et al., 2011; Renshaw et al., 2019). Our research questions are: (1) How frequently, if at all, do RoR dams trap bedload sediment; (2) given the frequency of sediment trapping, what is the expected average residence time of sediment within RoR reservoirs; (3) are the downstream channels able to maintain equilibrium vis-à-vis these changes in sediment flux? By taking advantage of the natural experiments provided by RoR dam emplacements that chronically perturb sediment flux in a manner that is fixed in time and space, our research questions seek to improve our understanding of the covariate channel evolution following a disturbance and thereby advance our understanding of the fundamental controls on the equilibrium form of a river channel.
Section snippets
Study sites and background
During 2016–2017, we sampled eight dammed sites (Fig. 1, Fig. 2, Table 1) within the Upper Connecticut River watershed, one of the most dammed regions in the US (Graf, 1999; Magilligan et al., 2016b). These dams, ranging in age from 60 to 97 yr old, represent a range of dam heights, slopes, flow regimes, and drainage areas (Table 1). Morphologically, the study reaches are dominated by riffles, pools, and plane bed units (Montgomery and Buffington, 1997). The channel bed surface grain size is
Equilibrium metric
Mackin (1948) first qualitatively linked equilibrium channel form to sediment transport dynamics. To quantify this concept, Dade and Friend (1998) and Dade et al. (2011) noted that conventional sediment transport equations often define sediment flux as a function of the Shields parameter, θ, a dimensionless ratio of bed shear stress τ to submerged particle weight:where ∆ρ = ρs − ρf, where ρs and ρf are the solid and fluid densities, g is the gravitational acceleration, d is the
Tracer rock surveys
The percentage of tracers recovered varies by site, from just under half of all tracers at the site with the largest drainage area (MAS) to almost all tracers at two with the smallest drainage areas (CBR and FBR) (Table 2). All sites had tracers that did not move (between 10%–29%). Of the 791 total tagged rocks placed at all five sites, 503 tracers were relocated after being transported from their deployment location. Among those tracers that moved and were relocated, the maximum cumulative
Discussion
Unlike larger flow-regulating dams with higher trapping efficiencies and greater accommodation space to store sediment, these results demonstrate that RoR dams intermittently permit coarse sediment mobilization into, and out of, their reservoirs, with maximum grain sizes transported over the dam crest exceeding the median grain size of the upstream sediment distributions and as large at 73 mm. Similar to that observed by Pearson and Pizzuto (2015) at a RoR dam in Delaware (USA) and Casserly et
Conclusion
For the scale of RoR dams we sampled, our results indicate that over decadal time spans they are similar to natural morphological traps in a river reach like deep pools or large woody debris (LWD) that have limited sediment residence times (Fisher et al., 2010; Lisle and Hilton, 1992). During the study period, which included both low and moderate hydrologic regimes, bedload tracer particles moved through the reservoir on the order of days to weeks. Through the construction of a sediment ramp up
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This research was supported in part by the National Science Foundation (BCS-1636415), a Dartmouth College Provost Seed Grant, a CompX Grant from Dartmouth's Neukom Institute for Computational Science, and a Geological Society of America Graduate Student Research Grant. We would like to thank Will Shafer, Brad Geismar, Olivia Lantz, and Andrew Crutchfield for help with field work, and both Jonathan Chipman and James Dietrich for geospatial guidance. We greatly appreciate the comments and
References (71)
Classification of geomorphological effects downstream of dams
Catena
(2000)- et al.
Influence of four run-of-river dams on channel morphology and sediment characteristics in Illinois, USA
Geomorphology
(2014) - et al.
Temporal and spatial scales of geomorphic adjustments to reduced competency following flow regulation in bedload-dominated systems
Geomorphology
(2010) - et al.
Sediment transport constraints on river response to regulation
Geomorphology
(2011) - et al.
Estimating and restoring bedload transport through a run-of-river reservoir
Sci. Total Environ.
(2019) - et al.
Flow regulation by dams
Treatise on Geomophology
(2013) - et al.
Immediate changes in stream channel geomorphology, aquatic habitat, and fish assemblages following dam removal in a small upland catchment
Geomorphology
(2016) - et al.
A comparison of past small dam removals in highly sediment-impacted systems in the U.S
Geomorphology
(2012) - et al.
Effects of weir height and reservoir widening on sediment continuity at run-of-river hydropower plants in gravel bed rivers
Geomorphology
(2017) - et al.
Distribution and significance of small, artificial water bodies across the United States landscape
Sci. Total Environ.
(2002)
Anthropogenic sediment retention: major global impact from registered river impoundments
Glob. Planet. Chang.
The effects of small dam removal on the distribution of sedimentary contaminants
Environ. Monit. Assess.
Empirical predictors of annual bed load travel distance, and implications for salmonid habitat restoration and protection
Earth Surf. Process. Landf.
Trap efficiency of reservoirs
Trans Amer Geophy Union
Geomorphic response to river flow regulation - case-studies and time-scales
Regulated Rivers-Research & Management
Hydraulic and geomorphological effects of run-of-river dams
Prog. Phys. Geogr.
Grain size, sediment-transport regime and channel slope in alluvial rivers
J. Geol.
The role of chronic and episodic disturbances on channel-hillslope coupling: the persistence and legacy of extreme floods
Earth Surf. Process. Landf.
A New Method for In-situ Measurement of the Erosion Threshold of River Channels
How big of an effect do small dams have? Using geomorphological footprints to quantify spatial impact of low-head dams and identify patterns of across-dam variation
PLoS One
Dam removal: Listening in
Water Resour. Res.
Denudation rate of central New England determined from estuarine sedimentation
Am. J. Sci.
Dam nation: a geographic census of American dams and their large-scale hydrologic impacts
Water Resour. Res.
The remains of the dam: what have we learned from 15 years of US dam removals?
Adjustment of river channel capacity downstream from a reservoir
Water Resour. Res.
The historic influence of dams on diadromous fish habitat with a focus on river herring and hydrologic longitudinal connectivity
Landsc. Ecol.
Restoring aquatic ecosystem connectivity requires expanding inventories of both dams and road crossings
Front. Ecol. Environ.
Channel stability downstream from a dam assessed using aerial photographs and stream-gage information
J. Am. Water Resour. Assoc.
Stable channels with gravel-paved beds
J of the Waterways and Harbors Division
Is the critical Shields stress for incipient sediment motion dependent on channel-bed slope?
Journal of Geophysical Research-Earth Surface
Fluvial Processes in Geomorphology
Variable Shields number model for river bankfull geometry: bankfull shear velocity is viscosity-dependent but grain size-independent
J. Hydraul. Res.
Cited by (16)
Impacts of climate change and human activities on sediment load in Longchuan River Basin, China
2024, Journal of Hydrology: Regional StudiesMulti time scale influence of dams on bedload transport
2023, Journal of Environmental ManagementSlowing the flow for climate resilience in human-dominated riverine landscapes
2023, Resilience and Riverine LandscapesDense and long-term monitoring of earth surface processes with passive RFID — a review
2022, Earth-Science Reviews