Feature: Sediment Mapping by Air
The Albuquerque Corps of Engineers contracts the development of geospatial data to monitor sedimentation in New Mexico’s Cochiti Reservoir.
By John Peterson, Daniel Paulsen, and Nathan Kempf
From its headwaters in Colorado, the Rio Grande bisects New Mexico on its way south to the Gulf of Mexico. Its waters have been the lifeblood of central New Mexico since pueblo peoples settled the region centuries ago. Historically prone to seasonal flooding, the river was known for its wild and erratic behavior, particularly as it emerged from its deep canyons in northern New Mexico and entered the broad valleys in the center of the state. Periodic high flows from spring snowmelt and monsoonal rains often caused flooding in these central valleys, which contain metropolitan Albuquerque, Indian pueblos, smaller communities, and agricultural regions. The seasonal nature of runoff also resulted in periods of low flow, which itself resulted in water shortages.
Cochiti Dam and Reservoir
In order to control these damaging floods and store water for a reliable supply during low flow, the U.S. Army Corps of Engineers
constructed Cochiti Dam and Reservoir on the Rio Grande below White Rock Canyon, the last of the deep canyons before the river enters the broad valleys to the south. Located 50 miles upstream of Albuquerque, the dam and reservoir lie within the lands of Pueblo de Cochiti, one of 19 sovereign Native American Pueblos in New Mexico.
Cochiti Dam is one of the 10 largest earthen dams in the United States. Over five miles long, it comprised 65 million cubic yards of earth. Its construction began in 1965 and was completed in 1975. Impoundment of Cochiti Reservoir began in 1973, collecting water from a 12,000-square-mile watershed. The reservoir is a popular recreational site providing camping, boating, and fishing.
Another important purpose for the dam is sediment control. Erosion within the watershed is a major problem in this semi-arid region, exacerbated by historic land use and wildfires. Sediment transported in the river is deposited on the riverbed as water flow diminishes, which is known as aggradation. It results in an elevation increase of the riverbed. Unless constrained by levees, aggradation can cause widening of the river channel and overflowing of its banks.
Ironically, degradation (river scouring and lack of sediment deposition) is a concern in the Albuquerque area. Monitoring and management of fluvial sedimentation is a principle responsibility of both the Corps of Engineers and the U.S. Bureau of Reclamation. With other dams in the watershed, Cochiti Dam serves as a sediment trap and is thus a critical component to region-wide sediment management efforts.
Paradoxically, the effectiveness of the dam for sediment control has led to a serious issue confronting the reservoir: sediment deposition is reducing reservoir storage capacity and causing significant aggradation upstream within the Rio Grande channel. Monitoring sediment volume, spatial distribution, and rate of deposition is of paramount concern to the Corps of Engineers. Consequences for the operation and life expectancy of Cochiti Dam and Reservoir are at stake.
Professional Services Needed
Because any hydrologic modeling and proposed remedial measures must be based on accurate, empirical geospatial information, the Albuquerque District of the Corps of Engineers contracted for the services of Wilson & Company, Inc
., Engineers & Architects, to perform professional survey, mapping, and geospatial information services to support the management of Cochiti Dam and Reservoir. The area of interest (AOI) is more than 20 miles long and includes the region occupied by the dam and reservoir, Cochiti Pueblo lands below the dam, and the Rio Grande and White Rock Canyon upstream of the reservoir. Wilson & Company was tasked with:
1) Researching the scope and availability of historical geospatial data, including conducting and acquiring:
- pre-dam and post-dam aerial photography,
- recent post-dam bathymetry of Cochiti Reservoir,
- pre-dam and post-dam river cross-section range lines,
- recent post-dam photogrammetric mapping and digital orthophotos, and
- field-surveyed ground control.
2) Using this historical data to prepare comprehensive geospatial information, including to:
- develop new geospatial data through ground-survey and photogrammetric techniques,
- georeference all data into a common datum and projective coordinate system,
- integrate and merge data from multiple technologies, resolutions, accuracies, and time periods,
- format data for use in comparative GIS analysis,
- perform comparative volume calculations between pre- and post-dam DTMs,
- prepare descriptive graphics showing sediment distribution, and
- document and organize this multiple data into a file geodatabase to provide geospatial information for an enterprise GIS the Corps of Engineers is constructing.
Of particular interest to the Corps of Engineers was the potential to use aerial photography acquired before construction of the dam in 1965. Photogrammetric techniques could be used to digitize features such as the historic river channel and compile a DTM of pre-dam topography for comparative analysis. This pre-dam data would provide baseline geospatial information against which post-dam DTM and bathymetry could be compared. The comparison would yield visual, graphical, and quantitative information about the volume and spatial distribution of sediment deposition since the dam was completed.
Step 1: Research
The University of New Mexico had completed preliminary research on the availability of historical aerial photography and discovered photography was acquired in October, 1963, by the U.S. Forest Service. Wilson & Company photogrammetists determined the suitability of this photography for pre-dam mapping of the AOI. This photography had been collected with a calibrated metric aerial camera and had the requisite coverage and scale for photogrammetric mapping (see images, above).
The Corps of Engineers and Wilson & Company professionals were aware of additional historic geospatial data within the AOI obtained by the Corps over many years. Additional data (provided by the Corps or retrieved from the Wilson & Company archives) included:
- aerial photography acquired in April 1972, while the dam was under construction but before the river was impounded, which provided information at a critical time just before reservoir inundation and was used to supplement the baseline DTM developed from the 1963 photography,
- aerial photography acquired during historic high reservoir water level in 1987,
- aerial photography, planimetry, DTM, contours, and digital orthophotos from 2000 and 2004,
- range line data from 2005,
- surveyed ground control to support the 2000 and 2004 photography, and
- high-resolution bathymetry of Cochiti Reservoir acquired in 2004.
Step 2: Data Use
The pre-dam aerial photography from 1963 and 1972 would support DTM compilation for five-foot contour accuracy when combined with camera calibration data and ground control in an aerial triangulation solution. Aerial negatives were scanned for use in digital photogrammetric workstations. Existing ground control used to support aerial mapping in 2000 and 2004 was photo-transferred to the pre-dam photography.
In 2010, Wilson & Company dispatched survey crews to derive coordinates for additional photo-identifiable features. These features needed to be identifiable at the time of survey and on the pre-dam photography, creating a challenge for the photogrammetists and surveyors.
DTM, infrastructure, and hydrologic features such as roads, the river channel, and dams were digitized for the AOI from the 1963 photography and for the dam and reservoir area from the 1972 photography. This mapping provided data for the baseline geospatial information previously mentioned.
The Corps of Engineers provided existing DTM, two-foot contour, and planimetic feature mapping from 2000 as well as high-resolution bathymetry of Cochiti Reservoir from 2004. Wilson & Company supplemented this data derived from 2004 photography and merged the DTM with the bathymetry to create a composite DTM—termed Epoch 1—for comparison with the baseline.
Digital orthophotos were created from the 1963, 1972, and 2004 aerial photography. Existing digital orthophotos from 2000 were provided by the Corps of Engineers. In July of 1987, Cochiti Reservoir was filled to the highest pool elevation in its history. At 5,435 feet above sea level, this elevation was well below design maximum pool elevation of 5,482 feet but considerably higher than normal pool elevation of 5,335 feet.
Digital orthophotos were created from aerial photography acquired in 1987 to document this event. Orthophotos from the various dates provided georeferenced imagery for visualization and manipulations of the vector data. Informative visualizations included:
- Cochiti Dam and Reservoir on pre-dam orthophoto, (see 1963 orthophoto above),
- historic Rio Grande channel on current orthophoto,
- sediment deposition graphics on historical and current orthophoto, and
- spatial extent of simulated reservoir elevations derived from the DTM, such as design maximum pool elevation, on historical and current orthophoto.
In order to provide a more accurate reservoir storage comparison between baseline DTM and Epoch 1, the DTM of Cochiti Dam (derived from current photography) was inserted into the 1963/1972 baseline DTM. Because this volume of space was never available for reservoir storage, insertion of this DTM effectively removed this space from pre-dam volume calculations.
Ground Control, Datums, and Coordinate System
All data was compiled in NAD83 NAVD88, New
Mexico State Plane central zone, U.S. feet. Ensuring the validity of comparisons between temporal vector and digital orthophoto data was critical. All ground control, orthophotos, bathymetry, and photogrammetric data was based upon permanent control monuments located on top of Cochiti Dam and georeferenced with common control points. Review of control as check points in the aerial triangulation and location of control in various data sets confirmed the geospatial consistency of the temporal data.
Cochiti Reservoir gaging is based upon NGVD29. Therefore, all data was translated into 27/29 for delivery along with 83/88.
Final deliverables to the Corps of Engineers included a temporal geodatabase for the periods of 1963, 1972, 1987, 2000, 2004, and 2010 in NAD83/NAVD88 and NAD27/NGVD29. Feature classes included survey control, contours, DTM, ArcTINs, bathymetry, infrastructure, water, and hydrology, as well as orthophotos of the multiple dates. Metadata was carefully prepared to thoroughly document the multiplicity of dates, scales, resolutions, ground control, technologies, and methodologies used for this composite geospatial project.
The Corps of Engineers identified specific areas defined by a uniform elevation contour in which volumes were compared between the baseline and Epoch 1. Accurate and complete DTMs will allow for comparison of any user-defined area. These DTMs have been combined into composite ArcTINs in which comparative profiles can be derived anywhere in the dataset. Various comparative graphics were prepared to better visualize spatial sediment distribution.
Quantitative comparisons show that maximum sediment accumulation of 80 feet occurs in the river upstream of the reservoir, increasing river baseline elevation significantly over historic conditions. Sediment deposition continues upstream for several miles before tapering off. These figures refer to change in river-surface elevation because only this can be derived from aerial photography. River depth averages three to five feet in most of the AOI and is not significant relative to total sediment depth. Range line data can be used in the future to model riverbed conditions.
Analysis of these temporal DTMs by Corps of Engineers’ hydrologists (in addition to providing geospatial information for GIS) will clarify spatial deposition patterns of sedimentation. Various hydrology-oriented software applications exist to further quantify and visualize this phenomenon. Hydrologic-flow modeling and forecasting can be augmented. Additional ancillary geospatial data, such as sediment core sampling and historic land use, can be georeferenced and interactively viewed in the GIS. This data will provide a base layer for a program being designed by the Corps of Engineers for use by all resource management agencies involved with the Rio Grande and its watershed above and below Cochiti Dam.
During the summer of 2011, the Las Conchas wildfire ravaged the watershed of the Rio Grande upstream of Cochiti Reservoir. The fire consumed 150,000 acres of the Santa Fe National Forest, Bandelier National Monument, and surrounding Pueblo and private lands. Subsequent monsoonal rains caused flash floods that flushed sediment and debris into the Rio Grande and Cochiti Reservoir. These floods caused considerable damage and compounded the effect of previous sediment deposition.
Wilson & Company collected multi-spectral, digital, aerial imagery of the affected area for the U.S. Forest Service and airborne lidar of the flooded canyons for the Corps of Engineers. This new data will provide post-fire analysis of sediment- and debris-deposition rates and distribution and will contribute valuable new information for the ongoing GIS effort.
John Peterson is the geospatial unit leader for the U.S. Army Corps of Engineers – Albuquerque District.
Daniel Paulsen is a geospatial project manager for Wilson & Company, Engineers & Architects, in Albuquerque.
Nathan Kempf is a geospatial data specialist for Wilson & Company.
» Back to our Aerial Mapping Spring '12 Issue