Early Pennsylvanian (309–318 Ma) paleocave sediments hosted in the Mississippian (345–359 Ma) Leadville Limestone were partly derived from long-distance (>2000 km) source areas. In addition to showing the importance of long-distant dust transport in cave sediments, because these paleocave deposits are derived from loess, their presence may document the earliest terrestrial signature of the late Paleozoic ice age in North America. The Leadville Limestone was subject to karst processes following late Mississippian eustatic sea-level fall, including formation of phreatic tubes, breakout domes, tower karst (kegelkarst), solution valleys (poljes), sinkholes (dolines), solution-enhanced joints (grikes), surficial flutes (rillenkarren), and solution pans (kamenitzas). In the Leadville Limestone, speleothems are interbedded with karst breccias and fluvial cave sediments. The overlying Pennsylvanian Molas Formation is a loessite (eolian siltstone) composed of angular quartz silt with ferruginous kaolinite rims. The U-Pb ages of accessory zircons indicate that the source areas for the eolian silt are from the peri-Gondwanan terranes and Grenville Province of eastern and southern North America, which are ~2000 km to the east. There is also a provenance signature from the rising Ancestral Rocky Mountains. The evidence suggests dust trapping on land surfaces by paleokarst topography, moisture, and vegetation. Weak paleosols in the Molas Formation suggest relatively rapid rates of dust accumulation. The high porosity and low bulk density of modern loess soils make them susceptible to groundwater piping. This mechanism may have facilitated redeposition of the Molas Formation loess into karst passageways, to be remobilized by later hydrologic events. The paleocave sediments in the Leadville Limestone can be linked to the overlying loess in the Molas Formation by compositional and textural matches. Facies analysis of the paleocave sediments documents episodic hydrologic events, producing a sequence of inundites and debrites separated by mud drapes with mud cracks. These event deposits are interbedded with flowstones and dripstones. Cave sediments are increasingly utilized as archives of geologic change. Recognition that dust is a significant component of cave sediments highlights the inherited properties from distant source areas, land-atmosphere transfer processes, land-surface deposition processes, and resedimentation processes into the karst system.

Abstract River restoration is a societal goal in the United States. This collection of 14 research papers focuses on our current understanding of the impacts of removing dams and the role of dam removal in the larger context of river restoration. The chapters are grouped by topic: (1) assessment of existing dams, strategies to determine impounded legacy sediments, and evaluating whether or not to remove the dams; (2) case studies of the hydrologic, sediment, and ecosystem impacts of recent dam removals; (3) assessment of river restoration by modifying flows or removing dams; and (4) the concept of river restoration in the context of historic changes in river systems.

Abstract In the northeastern United States, we have been removing dams for almost as long as we have been building them, yet many communities involved in current decisions to repair, replace, or remove a dam are not aware of this. This paper highlights some of the stories that have been recorded regarding the history of decision points for dams, including the colorful history of the Billerica Dam in Massachusetts, which has been removed and rebuilt numerous times and is now under consideration for removal for at least the sixth time in its 300 yr history. By understanding that dam removal is just one of the potential dam safety decisions that needs to be analyzed over the life cycle of a dam, and that dams are man-made structures with finite life spans, we can deconstruct the notion of dam removal as a radical concept. Dam removal is just one of many dam safety options that may be discussed over the course of a dam’s history. It is most commonly implemented when a dam no longer serves any economic purpose that justifies the expense of maintaining the dam structure. In the past, dams have been removed for many of the same reasons that we remove dams today; however, the procedures currently required to remove a dam are far more complex and highly regulated. This has led to increased documentation of dam removal efforts and now allows us to compare and categorize dam removal projects, such that the lessons learned from these projects can be incorporated into a more informed decision-making process in the future.

Abstract The removal of a large dam requires special engineering considerations not normally required for the removal of smaller dams. Large dams generally provide greater project benefits, but they represent a higher downstream hazard in the event of failure and greater challenges to fish passage. The reservoirs associated with large dams can impound more sediment and affect downstream water quality to a greater degree. The environmental compliance and decision-making process for a large dam removal project can take many years and will normally require the evaluation of a full range of project alternatives with estimated costs, the performance of a comprehensive environmental impact analysis, and the identification and implementation of extensive mitigation measures. Streamflow diversion and demolition plans for the removal of a large dam must ensure acceptable construction risks from start to finish, and produce reservoir drawdown at a controlled rate for sediment management purposes and to prevent instability of natural or embankment slopes. Large dams require more time for removal, at a higher cost, and contain greater volumes of materials for which disposal sites must be found.

Abstract Flood-control reservoirs designed and built by federal agencies have been extremely effective in reducing the ravages of floods nationwide. Yet some structures are being removed for a variety of reasons, while others are aging rapidly and require either rehabilitation or decommissioning. The focus of the paper is to summarize collaborative research activities to assess sedimentation issues within aging flood-control reservoirs and to provide guidance on such tools and technologies. Ten flood-control reservoirs located in Oklahoma, Mississippi, and Wisconsin have been examined using vibracoring, stratigraphic, geochronologic, geophysical, chemical, and geochemical techniques and analyses. These techniques and analyses facilitated: (1) the demarcation of the pre-reservoir sediment horizon within the deposited reservoir sediment, (2) definition of the textural and stratigraphic characteristics of the sediment over time and space, (3) the accurate determination of the remaining reservoir storage capacity, (4) the quantification of sediment quality with respect to agrichemicals and environmentally important trace elements over both time and space, and (5) the determination of geochemical conditions within the deposited sediment and the potential mobility of associated elements. The techniques employed and discussed here have proven to be successful in the assessment of sediment deposited within aging flood-control reservoirs, and it is envisioned that these same approaches could be adopted by federal agencies as part of their national reservoir management programs.

Abstract We investigated the viability of ground-penetrating radar (GPR) as a method to estimate the quantity of sediment stored behind the Merrimack Village Dam on the Souhegan River in southern New Hampshire. If the predam riverbed can be imaged, the thickness and volume of the reservoir deposit can be calculated without sampling. Such estimates are necessary to plan sediment management after dam deconstruction. In May 2008, we surveyed six cross sections with a Mala Geosciences ProEx 100 MHz GPR. In a related study, topographic surveys were conducted in 2008–2009 to monitor the sediment flux associated with the removal of the Merrimack Village Dam in August 2008. Within a month of the removal, these surveys mapped the predam riverbed in the uppermost cross sections in the former impoundment. We compared these surveys to our interpreted GPR images for one cross section to determine a calibrated velocity for the impounded sand of 0.043 ± 0.020 m/ns. We also estimated the radar velocity of the deposit by analyzing hyperbolic reflections in the GPR images, and found a similar result (0.039 m/ns). Using the calibrated velocity, we estimated a total volume of sediment stored behind the Merrimack Village Dam of 66,900 ± 9900 m 3 , which compares well to a previous estimate (62,000 m3) based on a depth-to-refusal survey. Our findings indicate that GPR is a useful technique for quantifying impounded sediment prior to dam removal in reservoirs containing 1–10 m of sand overlying a coarser predam riverbed, but it may be less effective in settings with finer and/or thicker impounded sediment.

Abstract The accurate prediction of sediment erosion after dam removal is critical to quantifying the impact of dam removal on the reservoir and downstream environment. A variety of methods can be used to estimate this impact, and one of the most common is to use a one-dimensional mobile bed sediment transport model. I describe a one-dimensional sediment transport model (SRH-1D) and use it to simulate a laboratory experiment of incision through a reservoir delta deposit. The model allows the user to specify the erosion width through the deposit as a function of the flow rate. The model is shown to predict the vertical incision and downstream sediment load with reasonable accuracy if the erosion width is specified. Sensitivity tests to the transport equation parameters, erosion width, and angle of repose are conducted. The sediment loads exiting the dam are shown to be sensitive to the critical shear stress, but they are relatively insensitive to changes to the erosion width and angle of repose. One-dimensional models are shown to require the specification of the erosion width, but the results are not considered to be extremely sensitive to its value, so long as it is approximately equal to the observed river width under the same flow conditions. Further work on modeling of bank erosion is necessary to more accurately predict the long-term evolution of reservoir deposits.

Abstract The removal of obsolete and unsafe dams for safety, environmental, or economic purposes frequently involves the exploration of sediments trapped within the impoundment and the subsequent assessment of sediment management needs and techniques. Sediment management planning requires a thorough understanding of the watershed’s surficial geology, topography, land cover, land use, and hydrology. The behavior of sediments is influenced by their age, consolidation, and stratigraphy. All watersheds have a history that helps forecast sediment loads, quality, gradation, and stratigraphy. Impounded sediment deposits may include coarse deltas and foreset slopes, fine or coarse bottom deposits, cohesive or organic matter, and wedge deposits immediately behind the dam. Some watersheds have anthropogenic pollutants from agricultural activities, mining, industries, or urban runoff. The volume and rate of sediment release during and after small dam removal can be limited by active management plans to reduce potential downstream impacts. Management strategies include natural erosion, phased breaches and drawdowns, natural revegetation of sediment surfaces, pre-excavation of an upstream channel, hazardous waste removal or containment, flow bypass plans, and sediment dredging.

Abstract The 2005 removal of the 3.6-m-high Munroe Falls Dam from the middle Cuyahoga River, Ohio, provided an opportunity to assess dam removal channel-evolution models and to anticipate impacts from additional dam removals on the Cuyahoga River. Preremoval geomorphic and sedimentologic conditions were characterized. Monitoring of the river response to dam removal has continued for 5 yr. The dam removal lowered base level and increased flow velocity upstream of the former dam site. Postremoval, the initial channel response was rapid incision to the predam substrate, followed by rapid lateral erosion of the exposed impoundment fill. Four to nine months after removal, dewatering and vegetation of the exposed impoundment fill greatly reduced the rate of lateral erosion. For 2.5 yr post -removal, sandy bar forms were present upstream of the former dam, and sand was transported under all flow conditions of the year. Subsequently, the bed has become armored with gravel. Downstream of the former dam site, the channel aggraded with sand, causing flow to occupy meander bend chutes that had formerly only been active during high flow. A sandy deltaic feature has accumulated 3.3 km downstream in the impoundment created by the Le Fever Dam. The impacts of the Munroe Falls Dam removal are generally well described by published channel-evolution models with minor exceptions due to local geology and hydrology. The similarities between the Munroe Falls and Le Fever Dam impoundments suggest that this study can aid in understanding the impacts of the possible future removal of the Le Fever Dam.

Abstract Before a dam removal project is implemented, engineers are often asked to estimate the potential for impacts from the release of reservoir sediment. Field measurements, numerical models, and physical models are typically used to develop sediment impact estimates. This information helps decision makers to make informed decisions about when and how to remove the dam, whether to allow the river to erode the reservoir sediment, or to remove or stabilize the reservoir sediment prior to dam removal, or whether mitigation of the effects is needed. Although numerous dams have been removed, mostly small in size, few case studies on sediment impacts have been documented. Because there are limited case studies, dam removal regulators and stake-holders often err on the side of caution when selecting the level of pre removal analysis or determining whether the reservoir sediment needs to be removed prior to dam removal. The purpose of this paper is to increase our knowledge base for application to future dam removals. The chapter discusses sediment impacts associated with the removal of the 11.9-m-high Savage Rapids Dam on the Rogue River near Grants Pass, Oregon. A unique factor to the Savage Rapids project was the construction and operation of a new diversion facility and water intake located immediately downstream of the dam, which introduced additional consequences associated with the release of reservoir sediment.

Abstract Dams on rivers modify habitat and water chemistry, resulting in degradation of fish and macroinvertebrate community integrity within and, in some cases, downstream of the dam pools. Thus, removal of a dam is usually accompanied by the expectation of improved habitat quality and biotic integrity. The Ohio Environmental Protection Agency applies a Qualitative Habitat Evaluation Index, an Index of Biotic Integrity (IBI, fishes), and an Invertebrate Community Index (ICI) to assess stream habitat quality and the habitat-dependent structural and functional integrity of the fish and invertebrate communities. Our objective was to demonstrate that these three indices reliably detect differences in the quality of habitat and fish and macroinvertebrate communities between dam pools and free-flowing reaches and that they are sensitive to changes in habitat and biotic condition following dam removal. Data from 21 stream reaches in Ohio containing dams showed that habitat and biota in dam pools possess lower quality than nearby upstream and downstream reaches. Case studies of dam removals on the Cuyahoga, Olentangy, and Sandusky Rivers confirmed that the indices are sensitive to the rapid changes in habitat and biotic communities that accompany return of dam pools to free-flowing conditions. IBI and ICI scores indicated that the former dam pools had met or exceeded the designated aquatic life use criteria within 1 yr following dam removal. We conclude that the IBI and ICI are valuable tools for measuring the rapidity and extent of changes in the fish and macroinvertebrate communities, respectively, following dam removal.

Abstract Gravel-bed river floodplains are dynamic landscapes that support a high level of ecosystem biocomplexity and biodiversity in large part because of the continual physical turnover of habitat. We evaluated the potential of a gravel-bed river to do geomorphic work on a series of floodplains below a dam by linking airborne hyperspectral imagery with corresponding groundtruth measures of flow velocity, water depth, floodplain surface topography, and vegetative cover. These data were analyzed in a geographic information system to map the spatial distribution of potential stream power over a range of discharge regimes. Nodes of flow separation at specific discharges that co-occurred with zones of high stream power were used as a metric to determine potential geomorphic threshold levels and location of channel avulsions. In order to address discharge duration as a factor affecting geomorphic change on the study floodplains, we established the relationship between discharge and total cumulative power applied to a single key floodplain and then used that relationship to examine historical discharge records and changes in flow release in terms of total cumulative power. We used the assumption that similar levels of total cumulative stream power, above a minimum geomorphic threshold, would produce similar levels of geomorphic work as a higher-magnitude, short-duration flood event. These results form the basis of an objective approach to evaluating flow releases needed from dams to maintain the dynamic structure and ecological function of gravel-bed river flood-plains. Moreover, the methodologies presented herein lend themselves to quantitative investigation of potential geomorphic changes related to complete dam removal and return of normalized flow of water and materials through the river system.

Abstract The stream restoration potential of recreational modifications made to lowhead dams on an urban reach of Buck Creek, in Springfield, Ohio, is dependent on constraints imposed by the urban infrastructure on stream grade. A privately led initiative to improve the recreational potential of a 9 km reach of Buck Creek and its tributary Beaver Creek includes the modification of four lowhead dams. The hydraulic heights of these dams will be replaced with a series of v-shaped drop structures engineered to create hydraulics conducive to kayak play. The drop structure is a constructed channel constriction composed of a hard step in the long stream profile immediately upstream of a scour pool, forming a morphologic sequence of constriction, step, and pool. In this study, we assess the potential benefits of these changes for urban stream restoration. Two of the dams have been modified to date. Stream quality, as measured by the qualitative habitat evaluation index (QHEI), dissolved oxygen of surface and substrate water, and the pollution tolerance index (PTI), increased at the Snyder Park site but decreased at the Art Museum site. Stream quality increased at the Snyder Park site, where stream grade could be lowered upstream of the lowhead dam, but decreased at the Art Museum site, where grade upstream of the lowhead dam had to be maintained because of water and wastewater utilities buried in the channel bed. Where stream grade is lowered in the former impoundment, sand and gravel deposits upstream of the constriction are not embedded with finer particles and organic matter. Increased QHEI values, particularly the substrate metric, and greater abundance and diversity of pollution-intolerant macroinvertebrates, supported by higher dissolved oxygen in the substrate water, characterize the Snyder Park site. At the Art Museum site, the v-shaped constriction increased the upstream impounded area. The substrate has become embedded with fine sands, silts, and organics, lowering QHEI values, dissolved oxygen is critically low in the substrate, and macroinvertebrate populations are more pollution-tolerant. The results highlight the significance of stream grade if stream restoration is to be incorporated into the engineering design of in-stream recreational features.

Abstract Small dams are often situated on low-order tributaries that drain grazed hill-slopes in dry regions of the western United States. In this paper, we use remote-sensing techniques in a case study to explore the effects of multiple small stock-pond dams on tributaries to Chileno Creek, a coastal watershed in central California. Dam density, or number of dams per drainage area, is 0.76 dams per km 2 , with most of the tributaries containing one or more dams. The total eroding channel length downstream of dams is ∼11% greater than total eroding length upstream. The relatively high density of the small stock-pond dams leads to cumulative effects that elevate the magnitude of (1) increased downstream erosion and (2) fragmentation of longitudinal connectivity between tributary headwaters and the main channel. The total headwater area producing sediment blocked by small dams equals 30% of the Chileno watershed. From these data, we infer that reduced sediment load due to the presence of the dams slows downstream riparian recovery. Results of the Chileno Creek case study emphasize that basin-scale management approaches and restoration strategies to restore connectivity are imperative in watersheds with high dam density. Uncertainty related to the biophysical effects of small dams and their removal may be investigated through analysis of baseline and long-term monitoring data, and adaptive assessment and management.

Abstract The concept of “passive” river restoration after dam removal is to allow the river to restore itself, within constraints such as localized bank erosion defense where infrastructure or property boundaries are at risk. This restoration strategy encounters difficulties in an urban environment where virtually the entire stream corridor is spatially constrained, and stream-bank protection is widely required. This raises the question of the meaning of river restoration in urbanized settings. In such cases, the sedimentary record can document paleohydrologic or paleogeomorphic evolution of the river system to better understand long-term response to the removal of the dam. Secor Dam was a low-head weir on the Ottawa River flowing through the City of Toledo, Ohio, and its outlying suburbs. The dam was constructed in 1928 and removed in 2007 to enhance aquatic ecosystems, improve water quality, and avoid liability concerns. Predam removal feasibility studies predicted the hydrological and sedimentological responses for the dam removal and determined that reservoir sediments were not significantly contaminated. Postdam removal studies included trenching, sediment coring, geochronology, and surveying. The buried, pre-1928 channel was located and showed that watershed urbanization resulted in channel armoring. Incision in the former reservoir exhumed a woody peat layer that was subsequently shown to be a presettlement hydromorphic paleosol currently buried beneath 1.7 m of legacy sediments, mostly deposited since ca. 1959. Today, the river flows through an incised channel between fill terraces composed of legacy sediments. Additional coring and survey work documented that the channel lateral migration rates averaged 0.32 m/yr over the past ∼80 yr, and that the meander wavelength is increasing in response to dam removal. Using sediment budget concepts, significant channel bank erosion and lateral channel migration should be expected until this river system reworks and removes accumulated legacy sediments currently in floodplain storage. In this dam removal project, “active” restoration practices, such as riparian wetland restoration, would have been more in accord with scientific understandings. That did not happen in this case because of disagreements among different constituencies and because of limitations of funding mechanisms.

Abstract For safety and environmental reasons, removal of aging dams is an increasingly common practice, but it also can lead to channel incision, bank erosion, and increased sediment loads downstream. The morphological and sedimentological effects of dam removal are not well understood, and few studies have tracked a reservoir for more than a year or two after dam breaching. Breaching and removal of obsolete milldams over the last century have caused widespread channel entrenchment and stream bank erosion in the Mid-Atlantic region, even along un-urbanized, forested stream reaches. We document here that rates of stream bank erosion in breached millponds remain relatively high for at least several decades after dam breaching. Cohesive, fine-grained banks remain near vertical and retreat laterally across the coarse-grained pre- reservoir substrate, leading to an increased channel width-to-depth ratio for high-stage flow in the stream corridor with time. Bank erosion rates in breached reservoirs decelerate with time, similar to recent observations of sediment flushing after the Marmot Dam removal in Oregon. Whereas mass movement plays an important role in bank failure, particularly immediately after dam breaching, we find that freeze-thaw processes play a major role in bank retreat during winter months for decades after dam removal. The implication of these findings is that this newly recognized source of sediment stored behind breached historic dams is sufficient to account for much of the high loads of fine-grained sediment carried in suspension in Mid-Atlantic Piedmont streams and contributed to the Chesapeake Bay.