With more than 3,000 low-head dams devoid of hydroelectric powerhouses, the North Carolina Piedmont was a prime location for a feasibility study determining the realistic potential to add hydropower to the area.
By Christopher J. Sandt and Martin Doyle
In the North Carolina Piedmont region - the area of rolling hills between the coastal plain and Blue Ridge Mountains - streams and rivers are moderate in gradient and size and are peppered with more than 3,000 low-head dams. These dams provide the potential to be retrofit with powerhouses, yet no recent studies have been performed to determine the realistic potential. At the same time, electric utilities and the general public across the state and the entire U.S. are increasingly interested in developing additional sustainable domestic energy sources.
The authors conducted a feasibility study for installing hydropower infrastructure on low-head dams within four North Carolina Piedmont river basins. The focus of the study was on functionally obsolete dams - those that are not currently used for hydropower generation. These dams were selected for study because they may currently present few benefits with significant costs, including safety liability, fragmentation of river ecosystems and persistent economic burden induced on state agencies due to regular inspection requirements.
This research provides much-needed qualitative and quantitative data specific to micro-hydropower potential (i.e., <300 kW capacity) that may allow stakeholders from the dam safety sector, power sector and environmental conservation community to make more informed decisions on whether an existing low-head dam could be retrofit for local power generation - or whether removal of the structure is a more realistic long-term solution.
Qualitative and quantitative data specific to hydro potential at low-head dam sites are lacking both in this specific region and globally. These data are needed to make informed decisions regarding the viability of dam-use alternatives, as well as the long-term management of existing impoundment structures and affected aquatic ecosystems.
This study focused specifically on low-head dams not currently used for hydropower but that may have been at one time, for instance, in driving the early 20th century textile industry in North Carolina. Low-head dams are defined in this context as non-federal structures with hydraulic head of 15 feet to 35 feet. More than 3,000 low-head dams are on record in North Carolina based on publicly available dam inventories. Some of these structures date to the mid-18th century. Scores of additional low-head dams likely exist in North Carolina but are excluded from public inventories - most often due to low visibility and/or lack of adequate record-keeping for very old structures.
Many of the low-head dams in North Carolina were constructed to produce direct mechanical energy or generate electricity locally for use by industrial facilities. The majority of these facilities were grist mills for grinding grain, basic manufacturing mills, or particularly textile manufacturing mills that tended to locate at sources of natural waterpower. With the decline of the textile industry since the mid-20th century, hundreds of dams that once produced power are now remnant and functionally dormant. Some structures have been redeveloped as drinking water supply reservoirs, hubs for private residential subdivisions or public recreational facilities. Others are abandoned as possible safety hazards and environmental liabilities.
The U.S. Department of Energy (DOE) recently studied 84,000 existing dams across the country, identifying 12,600 MW of potential power capacity on 54,000 of these facilities.1 Conversely, low-head dam removal studies have gained momentum across the U.S. in recent years as a means for alleviating public safety concerns and mitigating some of the adverse impacts of these structures on river ecosystems.
As human population continues to grow, private dam owners, municipalities, conservation groups and regulators are cooperating to manage growth and restore lotic (i.e., swift-flowing) watersheds to pre-development conditions while balancing the concerns of aging infrastructure and drinking water supply redundancy. Some low-head dam structures may need to be considered for removal if they pose significant safety hazards or are not serving the owner in a beneficial manner. However, removing dams for safety reasons or ecological restoration credits may face significant financial, legal or regulatory hurdles, along with local social resistance because of a community's attachment to these historic structures. The recognized public interest in both domestic energy production and dam removal must be considered when evaluating the long-term viability of low-head facilities that no longer serve a useful purpose.
While the 2011 DOE study includes many low-head dams, the results are specific to sites with potential capacity greater than 1 MW, and detailed cost analyses were not included due to scope constraints. The DOE study also identifies the need for refined site development models to help yield estimates of energy production that would be more realistic but less than the total energy available at a dam site.1
In contrast to the DOE study, the feasibility study detailed in this article assesses opportunities to leverage low-head structures as energy resources and thus as an alternative to their abandonment or removal. The primary objective of this research is to perform first-order conceptual site designs and financial viability evaluations on non-federal, low-head dams within the North Carolina Piedmont region. The conceptual site design methodology proposed leans toward low-impact, run-of-river micro-hydropower configurations that may cause fewer or less severe ecological impacts than dam removal. Run-of-river facilities also pose fewer ecological impacts because of their limited effects on flow regime.
The results of this study will provide valuable region-specific data for small (low-head) dam owners to reference when evaluating the potential for uprating their structures with micro hydropower equipment capable of distributing energy to the regional grid. Results will also inform broader considerations on dams that have limited or negligible long-term financial viability as energy sources.
Means and methods
We conducted a regional (7,193-square-mile area) study on dams contained within four North Carolina Piedmont river basins: Haw, Deep, Upper Cape Fear and Upper Neuse. More than 1,000 dams with gross hydraulic head of 15 feet to 35 feet were catalogued using a combination of ArcGIS-based data obtained through seven publicly-available dam inventories, the two most inclusive being the North Carolina State Dam Safety Office online database and U.S. Army Corps of Engineers' National Inventory of Dams (NID).
Dam inventory data were screened using ArcGIS analysis tools and manual observation of aerial photography to ensure accurate and independent geospatial location. Figure 1shows the geographical extent of the four river basins and the 1,049 low-head dams contained within the study area.
Due to the specified hydraulic head limitations and projected flow constraints for the relatively small watersheds behind these low-head dams, we estimated a potential power capacity range of 0 kW to 300 kW. This potential power capacity restriction qualifies these existing dams as "micro" hydropower sites that might be capable of local grid integration where applicable.
Drainage basin areas and aerial photography were further analyzed to select 49 test dam sites for detailed investigation. These sites were selected based on their providing a wide range of available flows and hydraulic heights, adequate powerhouse footprint areas, a relatively even split between public and private ownership, and a well-distributed geographic representation of the four rivers. Part of our goal was to conduct more detailed analyses than those done previously at large scales (e.g., DOE) but to do so in a way that could be readily transferred to other areas. Thus, we sought a publicly available software package that was both generalized and robust for the conditions of micro-hydro applications.
RETScreen4 software (available for free from Natural Resources CAMNET Energy Technology Centre, www.retscreen.net/ang/home.php) was used to estimate potential power capacity, capital costs, operating costs and energy revenues for these 49 test sites. RETScreen4 is an Excel-based software package consisting of a series of standardized worksheets designed to provide a means for low-cost preliminary assessments of renewable energy projects.
RETScreen4 uses a "pre-feasibility" analysis platform, allowing the user to compare a base case (typically the conventional energy production method for the region) to a proposed case that incorporates the renewable energy technology being studied. Sensitivity and risk assessments are performed to help the user estimate the relationships between important financial indicators and key technical input parameters.
Analyses were performed for the 49 test sites assuming a run-of-river configuration (i.e., using only the available flows that may be released over the spillway or through the existing drawdown structure when the impoundment is above normal pool). This type of configuration produces power capacity factors of 52% or less, as the turbine will only operate during the wetter periods of the year when adequate flow is available. Four basin-specific flow duration curves were generated using average daily flow data from 39 U.S. Geological Survey (USGS) gauges with at least 20 years of average daily flow measurements. The basin-specific flow duration curves were then normalized per unit area and applied to USGS drainage basin areas for each test dam site. We used the HUC12 (i.e., 12-digit hydrologic unit code) sub-watershed boundaries to define the USGS drainage basin areas. This normalization process produced estimates of flow duration curves that were specific to each dam site.
The median value of the 30% met or exceeded stream flow (i.e., the stream flow amount that is available at least 30% of the time during the year) was used as the base design flow for our dam sites. The powerhouse configurations included a simplex, cross-flow turbine and an asynchronous generator housed in a non-insulated cinder block powerhouse on an engineered concrete slab. The penstock configuration was assumed to include a buried PVC pipe and custom-fitted, hydraulically-engineered, double-valved pipe fitting connected to the existing dam drainage valve or sluice gate. The assumed penstock configuration allows the micro-hydropower system to capture the maximum available hydraulic head while continuing to provide a means for draining the impoundment as needed for routine dam maintenance.
A life cycle cost analysis was conducted over a 30-year project life to determine whether the investments required to up-rate these facilities with hydropower infrastructure were financially viable. While most hydro facilities will easily surpass a useful project life of 30 years, this temporal length was assumed in order to represent the conservative low end of the standard Federal Energy Regulatory Commission license term - which ranges from 30 years to 50 years.
Realistic potential for hydropower generation and financial viability based on net present value (NPV) was refined for the 49 test sites using three electricity export rates: $0.075, $0.115 and $0.180 per kWh. These rates were selected to represent three theoretical scenarios inclusive of regional energy pricing and renewable energy credits for micro hydropower. Hydropower renewable energy credits are trading in North Carolina for $6/MWh to $12/MWh. The first electricity export rate scenario is specific to current hydropower export rates, the second scenario to electricity export rates similar to the North Carolina state average, and the third to theoretical electricity export rates similar to current solar renewable energy credits as per Progress Energy's Solar Sense Program.
RETScreen4 models for each of the test dam sites were run to isolate and identify the theoretical electricity export rate that would be required to provide an NPV of zero. Existing dam sites with positive NPV were recognized as worthy of further financial analysis via a more detailed feasibility study. A sensitivity analysis and a risk analysis using Monte-Carlo simulation was also performed on all 49 test sites to determine site-specific relationships between design parameters and the key financial indicator (NPV). Information is available on the base model assumptions, sensitivity analysis details, and risk assessment details.2
Twenty-one of the 49 test dam sites are publicly-owned, either by municipalities or public water utilities. The remaining 28 sites are privately-owned, by homeowner associations, private land owners, community groups or private developers. Accordingly, supplemental RETScreen4 model iterations are being performed to incorporate the two primary differences in state and federal income tax scenarios for public and private entities. Specifically, models are being conducted that include an assumed 15% income tax burden on revenues derived from electricity export sales and theoretical CO2 reduction credits for privately-owned facilities. As with the base model runs and down payment sensitivity runs, the theoretical electricity export rate required to produce positive NPV will be isolated and quantified for each dam site using these modified income tax provisions.
Additional qualitative data for test dam sites was collected through consultation with federal and state regulators, as well as low-head hydropower specialists. Case study analyses were also conducted on operational low-head hydro facilities through site visits to the 15 most promising test dam sites to provide supplemental qualitative data for NPV estimates. Data such as owner knowledge of FERC requirements, federal/state/local regulatory jurisdiction, permitting history, upstream and downstream water use, and electrical utility constraints were evaluated in conjunction with the quantitative data derived from the physical characteristics of each test dam site. These data are summarized in the full study report.2
Preliminary results suggest that most of the low-head dam sites in this study are not financially viable for electricity production with current hydropower energy buyback rates (+/-$0.075/kWh) and available funding incentives. Initial findings also indicate that feedback from dam owners related to micro-hydropower up-rating was mixed. Distinct similarities were noted depending on whether the dam site was publicly or privately owned.
The two underlying themes observed throughout the ongoing data collection stages are specific to financial capital deficiency concerns (for publicly-owned facilities) and considerable lack of knowledge regarding the FERC regulatory processes and micro hydropower permitting/licensing (for privately-owned facilities).
The majority of public and private dam owners have a favorable opinion on the benefits of producing hydropower on their dams - yet of the 23 dam owner representatives we spoke with, only two had any knowledge of FERC requirements for hydropower licensing. These preliminary results reveal a substantial disconnect between low-head dam owners' long-term dam management goals and the reality of current economic and regulatory constraints on micro hydropower generation at these dam sites.
Site viability for energy production for the 49 test dam sites ranged from as low as 0.98 mi2 to as high as 254.63 mi2, with power generation potential ranging from 1 kW to 168 kW, respectively. Annual power capacity factors ranged from 46% to 52.2% based on using the available stream flows that may overtop or bypass an existing low-head dam during the wetter periods of the year. The majority of the dams studied were located on low-order tributaries. Annual electricity export ranged from 6 MWh to 742 MWh.
The preliminary cut-off relationship for realizing a positive NPV when evaluating reasonable electricity export rates ($0.075/kWh to $0.180/kWh) ranged from a higher head, lower flow scenario of 35 feet and 15 mi2 drainage basin area to a lower head, higher flow scenario of 15 ft and 45 mi2. Annual energy export rates were evaluated in relation to the hydraulic heights and drainage areas of individual test dam sites. Preliminary results suggest that flow availability is the dominant independent variable for energy production and subsequent financial viability of low-head dam sites, as anticipated.
Dam owner sentiment
The majority of public dam owners consulted were water resource directors or public utility directors, but other representatives ranged from town council member level to mayoral level. Most public dam owner representatives supported the idea of micro hydropower; however, they approached the possible installation of micro-hydropower infrastructure on their facilities with great trepidation. Their primary concerns were specific to the potential threat of any changes to their primary duty - providing reliable and affordable drinking water services to their customer base.
All of the public dam owners articulated that flow limitations were their most pressing concerns. Some public representatives expressed interest in installing micro-hydropower facilities on their dam structures if deemed financially feasible (i.e., positive NPV in most cases). These same representatives also stated that the additional operational demands required for micro-hydropower facilities would likely not be feasible due to annual monetary constraints, the need to maintain full capacity and/or redundant water volume behind the impoundment at all times, and personnel training issues.
The private dam owners consulted ranged from HOA board members to facility maintenance managers. Most private dam owners responded to the idea of micro hydropower installation with great interest. Rather than perceiving micro hydropower operations as a liability, the majority of private dam owners stated that they would consider installing a micro hydropower facility as an asset assuming a viable NPV and adequate access to capital. All private dam owners expressed their concerns over maintenance/operational requirements and sufficient access to free or low-interest capital via federal/state/local incentive programs. Many private dam owners stated that federal grants and tax deductions were preferable over loan programs.
Preliminary discussion and applications
Sensitivity analyses suggest that a surprising number of low-head dam sites - 15 out of 49 in this regional study - may be viable for production if the owner is provided funding opportunities that are comparable to alternative sources of renewable energy. There may be hundreds of existing low-head dam sites across North Carolina that would become financially feasible for micro hydropower production if provided electricity export rates comparable to those currently provided to the wind and solar markets.
The short- and long-term ecological impacts that may derive from retrofitting the low-head dam sites evaluated in this study must be fully understood prior to making a decision, rather than relying solely on financial viability. This discrepancy, combined with the limited data available for understanding the full financial and ecological costs of removing low-head dams such as these, supports the need for further research in this arena.
1Boulem, H. et al., "An Assessment of Energy Potential at Non-Powered Dams in the United States," U.S. Department of Energy, Wind and Water Program, Prepared by Oak Ridge National Laboratory, Oak Ridge, Tennessee, 2012, http://www1.eere.energy.gov/water/pdfs/npd_report.pdf
2Sandt, C., "Micro Hydropower in the North Carolina Piedmont: Hydrologic, Economic and Political Constraints on Retrofitting Existing Low-Head Dams," Master's Thesis Report. Department of Environmental Sciences and Engineering. University of North Carolina, Chapel Hill, North Carolina, 2011.
Christopher Sandt, P.E., is an engineer at DC Water, a public authority in Washington, D.C. Martin Doyle, Ph.D., is a professor at the Nicholas School of the Environment at Duke University.