Upper Esopus Creek Tributary Bedload Pilot Study

Science Center Objects

Problem Sediment transport is a serious concern in the upper Esopus Creek watershed. The creek is a well-documented source of sediment and turbidity to the Ashokan Reservoir, which is part of the New York City water supply system. During the last 2 decades there has been a series of stream stabilization and sediment reduction projects completed in the upper Esopus Creek watershed intended to re...

Problem

Sediment transport is a serious concern in the upper Esopus Creek watershed. The creek is a well-documented source of sediment and turbidity to the Ashokan Reservoir, which is part of the New York City water supply system. During the last 2 decades there has been a series of stream stabilization and sediment reduction projects completed in the upper Esopus Creek watershed intended to reduce the suspended sediment load and turbidity levels delivered to the reservoir. During the last 7 years there has been a concerted effort to measure the effect of these projects on turbidity and suspended sediment. There is currently a large, long-term turbidity and suspended sediment study underway within the upper Esopus Creek watershed to quantify suspended sediment loads and turbidity levels throughout the watershed with a specific focus on the Stony Clove Creek watershed which has been identified as the tributary that contributes the largest amounts of suspended sediment to the Esopus.

Suspended sediment is an important component of sediment transport and turbidity is the constituent of regulation for the NYC drinking water supply, however sediment transport also includes the material transported as bedload. To date bedload has not been included in the sediment monitoring projects in the upper Esopus Creek watershed because bedload is difficult to measure and those measurements are subject to a great amount of uncertainty. The difficulty in measuring bedload also makes it expensive data to collect. Nonetheless, bedload is potentially an important component of the total sediment load in the upper Esopus Creek watershed. This study conducts a small scale pilot project to estimate the percentage of the total sediment load contributed by bedload at 2 locations within the upper Esopus Creek watershed (figure 1). The project will test the effectiveness of multiple bedload sampling and monitoring techniques to determine whether the additional expense of including a bedload monitoring component in the sediment monitoring is warranted. In addition, we will evaluate the effectiveness of two sediment transport models to estimate bedload transport.

A goal of this project is to produce sediment discharge rating curves with measurements of bedload and suspended sediment. The suspended sediment data will be provided by a project currently underway and funded by New York City Department of Environmental Protection  (NYCDEP). This project will supply the needed bedload data. Sediment transport estimates (suspended sediment and bedload) are a recommended element in designing, implementing, and evaluating stream restoration projects. To this point in time the data collected in support of NYCDEP stream restoration projects has included morphometric surveys, bed material characterization with pebble counts, stream water discharge and suspended sediment concentrations. Estimated bedload sediment transport used in design criteria has typically been derived using sediment transport functions. Obtaining bedload measurements will allow a full accounting of sediment transport within the reach measured, and if successful (1) could support the development of regional sediment transport curves that would allow sediment transport to be estimated in other watersheds with similar hydrophysiographic conditions that are not part of the current monitoring program; and (2) provide data to evaluate the efficacy of various sediment transport functions (equations) in predicting bedload transport through a stream restoration project reach.

Objectives and Scope

 

The primary objective of this pilot study is to determine the feasibility of measuring bedload in the upper Esopus Creek watershed at two locations of differing geomorphic conditions. The two study locations will be located at existing USGS stream gaging stations in order to leverage existing discharge and suspended sediment data. These locations have been determined after consultation with the cooperator and Danyelle Davis of the NYCDEP. The two sampling locations are the Stony Clove Creek at Janssen Rd at Lanesville (USGS ID 01362336) and at the Lasher Rd bridge downstream of the Birch Creek at Big Indian (USGS ID 013621955) streamgages (figure 1). Site investigations with the Ashokan Watershed Stream Management Program (AWSMP) and NYCDEP staff confirmed site selection. Multiple bedload measuring techniques will be tested. These include the use of standard and non-standard bedload samplers, active and passive tracers, and hydrophones to measure transport of bed material through a range of streamflow conditions. This project is designed as a 3-year study with two years of data collection and 1 year of data interpretation and publication production. This proposal, however, is for sample collection only because the current cooperative agreement is limited to the next two years. Specific objectives include:

1.      Collection of bedload samples during runoff from 3 high streamflows and 2 moderate streamflows during a two-year period at two locations.

2.      Measurement of the cumulative transport of bed material using active and passive tracers during runoff from 3 storms during a two-year period.

3.      Estimate transport of bed material using the results of the active and passive tracer data.

4.      Estimation of near-continuous bedload transport using acoustic data recorded by hydrophones as a surrogate.

5.      Assess the feasibility of bedload measurement and sediment rating curve development at two study locations in tributaries to the upper Esopus Creek Creek watershed.

Approach

 

The objectives of this project will be accomplished by collecting bedload samples and measuring movement of bed materials through a range in streamflow conditions, ideally from moderate streamflow to streamflows exceeding bankfull discharge. There is no guarantee that the watershed will experience bankfull discharge during the 2-year study period. Therefore, sample collection will target three high and two moderate streamflow conditions to provide data that can be used to assess the feasibility of bedload monitoring and allow for development of sediment rating curves throughout the range of streamflow conditions experienced during the study. We will target samples throughout the hydrograph since there can be considerable differences in sediment concentrations during  similar streamflow conditions depending on whether samples are collected on the rising or the falling limb of the hydrograph. Samples will also be collected during 2 moderate streamflows, that is, approximately 0.5 estimated bankfull streamflow, so as to better define the sediment rating curves at each site. These moderate streamflow samples will be targeted to capture streamflow conditions not sampled during the 3 high streamflow sampled.

To improve estimates of sediment transport, a surrogate technique using acoustic data recorded by hydrophones will be used. Hydrophones detect the sounds generated by the collisions of sediment particles (primarily gravel and cobble) as the particles roll and saltate along the bed. The sounds produced by the sediment collisions are referred to as sediment-generated noise (SGN). The hydrophone uses a piezoelectric element to convert the sound pressure waves to an electrical signal. The electrical signal is proportional to the pressure acting on the piezoelectric element, which is assumed to be proportional to the magnitude of bedload transport occurring near the hydrophone. The acoustic data is then recorded electronically and later processed.   

Recent work using hydrophones to monitor bedload transport has exhibited good results on the Trinity River, California (Marineau et al., 2016). In that study, bedload was measured at two sites during a restoration streamflow release. Sediment-generated noise was recorded using hydrophones and calibrated to those physical bedload measurements to create a continuous time series of bedload transport. A key advantage of using a surrogate method for bedload monitoring is that it can provide near-continuous measurements. An acoustic-based monitoring station can collect measurements at high frequency (for example, 15-minute intervals) during an entire storm event.

Data Collection

 

All bedload and bed material data for this project will be collected using methods approved by the U.S. Geological Survey (Edwards and Glysson 1999) or modified when necessary because of specific site conditions using technically supportable and citable rationale. Sediment samples will be analyzed at the Kentucky Water Science Center Sediment Laboratory using approved sediment methods.  Streamflow and suspended-sediment data will be available from an existing project funded by the NYCDEP and USGS.

Bedload sampling will be conducted using the equal width-depth integrated method and pressure-difference bedload samplers (Edwards and Glysson, 1999). Bedload samples will be collected using Federal Interagency Sedimentation Project approved bedload samplers with a deployment method appropriate for the bed-material and site characteristics. A full size-distribution determination will be conducted on each composite bedload sample. Particle-size distributions of bed material will be determined using Wolman pebble counts at 5 locations in the upstream vicinity of each monitoring location.

Passive tracers of bedload movement will be constructed, deployed, and recovered in one sampling reach using methods established in peer-reviewed literature (Olinde and Johnson, 2015; Phillips and Jerolmack, 2014). Passive tracers will be constructed using radio frequency identification technology that will uniquely identify each rock. This entails choosing rocks from the study reach that are representative of the size distribution found in the channel, but greater than the effective sampling size of the bedload samplers. Each rock will have a ¼ inch hole drilled in it, a passive RFID tag inserted, and the hole will be sealed with epoxy. The rocks will then be returned to their original location in the stream channel and the coordinates of the location recorded using GPS. After the occurrence of each bedload storm sampling event, a back pack locator unit will be used to locate the tagged rocks and the coordinates of the new location will be recorded again. The tracers will be left in place and re-located after the next bedload sampling event. The cumulative transport distance during the interval between deployment and recovery of the passive tracer will then be calculated.

Ten active tracers, “smart rocks”, of bedload movement will be constructed, deployed, and recovered using methods established in peer-reviewed literature (Olinde and Johnson, 2015). Ideally the number of active tracers deployed would be larger, however this is beyond the scope of this pilot study. Ten smart-rock containers are available from a manufacturer that successfully produced them for a previous bedload study and they can be obtained at a reduced price for this study. The smart rocks are produced by injection molding using a copper filled PA Alloy at 3.3g/cm3. When the hollow interior that houses the electronics is taken into account, the containers duplicate the density of sandstone. An accelerometer and RFID tag will be placed in the center of the smart rock container. The active tracers will then be deployed and recovered using the same methods as those of the passive tracers. The accelerometer data will be used to determine transport distance of the bed material during individual storm events, to relate the transport distance to streamflow, and to determine the initiation of bedload transport (timing and streamflow).

A total of four hydrophone-based underwater sound recording systems (Marineau et al.; 2015; 2016) will be constructed and programmed with the goal of installing two systems at each of two monitoring sites (one system on each bank). Each system will comprise two hydrophones, one liquid-detection sensor, and one recorder. At each study site, two systems will be installed (one on each bank) to monitor spatial variability in transport and to provide redundancy in data collection in case equipment is damaged by flood debris. The hydrophones will be mounted to rebar which will be driven into the bed and the recorders will be secured to the bank. A liquid-detection sensor will also be installed. The recorders will be programmed to periodically check the sensor to determine if the water level has risen. When water levels are low, the recorders will remain in standby mode to preserve power. Once activated, the recorders will record 1-minute of audio data at 15-minute intervals. The data will be stored on a removable micro-SD card. The recorders will be serviced periodically to replace batteries and the SD cards.

The data stored on the removable micro-SD card will be post-processed using Matlab following the procedures described in Marineau et al. (2015, 2016) using a Fast-Fourier Transform to calculate sound level at various frequency ranges. The sound level over specific frequency ranges that have been shown to correlate well with coarse bedload transport will be averaged together. The acoustic data and physical bedload samples will be used to develop an acoustic-based sediment rating curve and, if possible, a time-series of estimated bedload transport rates for one or more size ranges (e.g. >16 mm, 4-16 mm). These data will be released on USGS NWISweb or in ScienceBase. 

 
Data Interpretation
 

Data interpretation is not included in the current funding cycle, but it is anticipated that funding will be available through a future amendment to this project. A description of the future data interpretation follows.

The goal of this project is to assess the feasibility of bedload measurement and sediment rating curve development at two locations in the upper Esopus Creek watershed. Project data will be used to compare estimates of bed material transport using standard sampling techniques, active and passive tracers, and acoustics as a surrogate. The data will also be used to create sediment rating curves for the study sites and to determine the initiation and cessation of bed material movement (timing and streamflow). All data collected by the USGS will be incorporated into our National Water Information System database and will be publically available after they have been quality assured.

Sediment Transport Models

 

Sediment transport will be modeled using the computer models SEDDISCH (Stevens and Yang, 1989) and BAGS (Bedload Assessment in Gravel Bedded Streams) (Wilcock and others, 2009). We will use the two sediment transport models, which both contain multiple sediment transport equations, to evaluate how well the models estimate sediment transport and particularly the bedload portion of that transport. The equations used will be determined after discussions with Stream Management Program personnel. The results of the modeling component of this study will help to evaluate how well these models can estimate bedload in other similar watersheds throughout the Catskill Mountains. Furthermore, in the event that no bankfull storms occur during the data collection period, bedload streamflow could be estimated using the computer models. While it is our intent to provide physical measurements of bedload transport at a range of streamflows at the 2 study sites, we cannot control the frequency of bankfull storms and so bedload data collected during smaller storms may need to be used as input to calibrate the sediment transport models and to extrapolate bedload at greater streamflows.

The sediment transport models used for this study require some or all of the following parameters: channel geomorphic data, streamflow, water-surface slope, bed-material size-grain distribution, and bed load sample data. Much of the required data are already being collected as part of other projects being conducted in the upper Esopus Creek watershed. This project will provide the bedload and bed material data and will supplement any channel geomorphic data that is currently unavailable. 

References

 

Edwards, T. K., and Glysson, G. D., 1999, Field Methods for Measurement of Fluvial Sediment. Techniques of Water-Resources Investigations of the U.S. Geological Survey Book, Book 3 Applications of Hydraulics, Chapter C2, 97 p

Marineau, M.D., Minear, J.T., Wright, S.A., 2015. Using Hydrophones as a surrogate monitoring technology to detect temporal and spatial variability in bedload transport. Proceedings of the 10th Federal Interagency Sedimentation Conference, Reno NV, April 19-23, 2015

Marineau, M.D., Wright, S.A., and Gaeuman, D., 2016, Calibration of sediment-generated noise measured using hydrophones to bedload transport in the Trinity River, California, USA, in Constantinescu, G., Garcia M., and Hanes, D., eds., Proceedings of the 8th International Conference on Fluvial Hydraulics River Flow 2016, St. Louis, USA, 11–14 July 2016: London, UK, Taylor & Francis Group, Paper No 222, p. 1519–1526.

Olinde, L. and Johnson, J.P., 2015, Using RFID and accelerometer‐embedded tracers to measure probabilities of bed load transport, step lengths, and rest times in a mountain stream, Water Resources Research 51, 7572-7589.

Phillips, C.B., and Jerolmack, D.J., 2014, Dynamics and mechanics of bed-load tracer particles, Earth Surface Dynamics 2, 513.

Stevens, H.H., and Yang, Chih Ted, 1989, Summary and use of selected fluvial sediment-discharge formulas: U.S. Geological Survey Water-Resources Investigations Report 89-4026, 121 p

 

Wilcock, Peter; Pitlick, John; Cui, Yantao. 2009. Sediment transport primer: estimating bed-material transport in gravel-bed rivers. Gen. Tech. Rep. RMRS-GTR-226. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 78 p.

Project
Location by County

Catskill Region: Delaware County, NY, Greene County, NY, Schoharie
County, NY, Sullivan County, NY, Ulster County, NY