George Washington Bridge, between New York and New Jersey
Throgs Neck Bridge, New York
Sandy Hook, New Jersey
Citi Field, Queens, New York
Sandy Hook, New Jersey
Coney Island, New York
This is the bridge at Mantoloking (CR-528) in New Jersey. Barnegal Bay is to the West,and the Atlantic Ocean is to the East. This strip of land including the beach is barely 1000 ft wide.Image from the Esri ArcGIS Online base map, before Superstorm Sandy.
The same area after Superstorm Sandy, from NOAA Aerial Image Service, hosted by theEsri Disaster Response Team on ArcGIS Online.
Detailed and accurate charting of the sea bottom along coastlines is vital for environmental monitoring and remediation, shoreline construction, and coastal navigation. Each of these activities requires reliable depth data for planning and management purposes. However, sonar-equipped vessels conducting hydrographic surveys cannot navigate very shallow or rocky coastal areas or narrow inlets with high tidal ranges, and topographic airborne LiDAR mapping (ALM) systems cannot penetrate the water. Airborne LiDAR bathymetry (ALB) fills this gap: as a supplementary survey method, it provides a seamless transition between mapping the land and charting the sea.
Superstorm Sandy — technically, a post-tropical cyclone by the time it made landfall along the coastline of the United States on October 29 — affected 24 states, including the entire Eastern Seaboard from Florida to Maine, causing particularly severe damage in New Jersey and New York. Storm surges, made worse by the full moon, caused high tides to rise about 20 percent higher than normal and wreaked havoc along hundreds of miles of coastline. Its impact included widespreadflooding, erosion, and movement of millions of tons of coastal sediments with the extreme power of storm-driven water — thereby actually altering vast stretches of coastline.
In the wake of this devastation, dozens of federal, state, and local agencies, as well as many private companies, contributed to the response. A couple of weeks after the storm, the Joint Airborne LiDAR Bathymetry Technical Center of Expertise (JALBTCX) of the U.S. Army Corps of Engineers (USACE) flew its Coastal Zone Mapping and Imaging LiDAR (CZMIL) along several stretches of the northeast coast, collecting a consistent data set.
How Bathymetric LiDAR Works Bathymetry has traditionally been performed using echo sounders (sonars) mounted on ships or boats. The use of multi-beam echo sounders has allowed for highly detailed, accurate seabed charting. While these systems can measure depths even in very shallow water, boats cannot access areas with skerries, shallow reefs, long shallow beaches, large waves, and underwater currents. Furthermore, echo sounder systems are expensive and difficult to deploy at short notice.
ALM systems operated from fixed or rotary wing aircraft, on the other hand, can be deployed at short notice and enable accurate, cost-effective data collection, but use an infrared laser that has poor water penetration capabilities. Therefore, in coastal zones the combination of these two technologies is not always sufficient to create seamless mapping.
ALB systems are also operated from aircraft. However, unlike ALM systems, they transmit two light waves, one in the infrared and one in the green portion of the electromagnetic spectrum, and are capable of separately detecting the returns from each. The infrared band is quickly absorbed and is therefore used to detect the water surface. The green band, which is the optimum color to achieve maximum penetration in shallow water, is used to delineate the sea bottom.As the pulse of green laser light travels through the water and reflects off the seabed, it is subjected to refraction, scattering, and absorption, which attenuate it and limit the depth of water that it can measure. The greater the water’s particle content, the greater the backscattering, up to the point that it becomes impossible to distinguish the backscatter from the bottom return.
Therefore, the maximum depth for ALB is determined by the reflective characteristics of the seabed and by water clarity. Under ideal conditions, the theoretical maximum depth measurement is about 70 meters.In order to penetrate the denser medium of water and then minimize scattering, LiDAR bathymetry requires much higher power and longer laser pulses than topographic LiDAR. Therefore, bathymetric systems operate at a much slower rate and with much longer pulses than topographic ones. For example, Optech’s CZMIL has a measurement rate of 70 kHz when operating in topographic mode but of only 10 kHz when operating in hydrographic mode. This lower pulse frequency reduces the achievable point density. Even in relatively smooth air, all aircraft are subject to vibrations, sudden loss of attitude, and constant small changes in their pitch, roll and yaw. Therefore, knowing the range to a target is not sufficient to determine its position, which also requires knowledge of the aircraft’s exact location and attitude at the time each laser pulse is fired. Differential GPS provides the former and an inertial measurement unit (IMU) provides the latter.
ALB systems make it possible to survey in a single scan features and constructions both above and below the waterline. Typically, digital images are recorded at the same time, enabling their visual analysis and use with digital terrain models.
JALBTCX is a joint center of government agencies working together to advance LiDAR bathymetry and complementary technologies for coastal mapping, explains Jennifer Wozencraft, the center’s director. It is made up of USACE, the U.S. Naval Oceanographic Office, the Remote Sensing Division of the National Geodetic Survey of the National Oceanographic and Atmospheric Administration (NOAA), and the U.S. Geological Survey’s Coastal and Marine Geology Program. “We own and operate, through contract, LiDAR bathymetry and airborne remote sensing sensors for coastal mapping for the USACE and the Navy, and then with the other agencies we do research and development to advance hardware and software and to demonstrate the different applications of the data that we collect,” says Wozencraft.The center funds companies to develop hardware for it — primarily airborne sensors. “It all started with the SHOALS program, back in the late 1980s,” Wozencraft recalls, referring to the Scanning Hydrographic Operational Airborne LiDAR Survey. “One of the main goals of that program was to develop a capability that could be commercialized to make it more widely available for the USACE and our other partners.”
Immediately following Sandy, JALBTCX was tasked by USACE’s North Atlantic Division to collect LiDAR data for the districts that fall within that division’s jurisdiction. The center used its own planes to collect the data for the New York District and contracted with private companies to collect data for the New England District, the Philadelphia District, and the Norfolk District.
The flights took place November 11-24, as soon as the center had been able to gather the requirements from the impacted areas. “The primary requirement,” says Wozencraft, “was classified topographic LiDAR data as well as bare earth DEMs generated for mapping. There was a coordinated effort in advance of the survey with other agencies to see what everyone else was doing to try to collect a consistent dataset and a vertical accuracy of 12.5 centimeters RMSE (root mean square error). We typically fly about 1,000 meters off shore or to laser extinction, whichever happens first. In this case, I think it was probably less than that, maybe even only 250 meters off shore.”
JALBTCX flew from its base in New Haven, Connecticut along the New Jersey coast line, the Long Island shoreline, New York’s inner harbor, Sandy Hook, Staten Island, and the East River.
“Typically, after a storm, you have access issues,” Wozencraft explains, “and we did experience those getting the GPS way stations and ground truth locations and such. Ours was the only bathymetric LiDAR that was deployed. We fly lower than everybody else, so we had some air traffic control issues that forced us to fly most of our flights at night, which is different for us. We usually do our work in the summer, when the water is nice and clear. This time, of course, it was much colder than usual. It wasn’t particularly a challenge, but different than it usually is, and the wave climate was a little more energetic than it is in the summer.”
Normally, JALBTCX collects concurrent RGB and hyperspectral imagery, which it was not able to do in this case due to having to fly at night. Instead, it collected that imagery subsequently, when it was able to fly during the day.
“Another part of the work that JALBTCX did and is currently doing for the North Atlantic Division is volumetric change analysis of beach projects,” says Wozencraft. “The main operational program that JALBTCX does for the USACE is called the National Coastal Mapping Program. We collect data around the coast of the United States on a cyclical basis. We had collected data in this area in 2010, so we are comparing all of that data to the post-event data for volumetric change analysis.”
The CZMIL ALB sensor that JALBTCX used for its post-Sandy flights was built by Optech Inc., the U.S. subsidiary of the Canadian company Optech Incorporated, for the U.S. Navy, which then loaned it to the center. It is an airborne coastal zone mapping system that produces simultaneous high-resolution 3D data and imagery of the beach and shallow water seafloor, including coastal topography, benthic classification, and water column characterization. According to Optech, CZMIL performs particularly well in shallow, turbid waters. Its bathymetric LiDAR is integrated with a hyperspectral imaging system and a digital metric camera.
The CZMIL requires no ground truthing. “It is connected to a GPS receiver and an IMU,” says Max Elbaz, Optech Inc.’s president, “so all collected data is automatically georeferenced and co-registered with initial geometric calibration procedures. When we built the system, before delivery, we compared CZMIL data with ground truth data to characterize the system performance. When these very stringent specifications are being met, the data is deemed within required accuracies.”
According to Elbaz, his company’s history in using LiDAR for bathymetry and environmental applications for rapid response and disaster management scenarios began in the 1970s and the company helped pioneer airborne LiDAR throughout the 1980s. “We followed up with the SHOALS LiDAR bathymeter system in the 1990s and after,” he says, “adding a camera and a hyperspectral sensor. Then, last year, we introduced the Optech CZMIL, which includes an RGB camera, a hyperspectral sensor, and a topographic LiDAR, as well as the bathymetric LiDAR. To date we have delivered three CZMIL systems: one to the U.S. Army Corps of Engineers (USACE), one to the U.S. Navy, and one to the Japanese Coast Guard.”
The nominal height of CZMIL operation is 400 meters, Elbaz explains, but JALBTCX flew it mostly at 570 meters to cover a wider area. “They were not interested in surveying deeper water; they wanted to fly only over shallow areas.” Because the device is a joint topographic and bathymetric LiDAR system, it can be used to probe land, shallow water, and deep water up to about 50 meters, he points out. “Beyond 50 meters it isn’t as essential because you can use sonar on ships.”
“Our team in Kiln, Mississippi, where CZMIL was developed and built, supported the mission from data acquisition through to data processing,” says Elbaz. “However, we did not send personnel out; instead we supported the mission remotely, using CZMIL’s Web monitoring and diagnostic tools. It’s a new system on an important mission, so our team made sure USACE had all the support it needed.”
To process the data from its postSandy mission, USACE used Optech’s HydroFusion post-processing software suite, which handles data from all three CZMIL sensors, from mission planning to the fused LiDAR and imagery data set. “It’s a single user interface that replaces the seven to eight different pieces of software that our customers previously used to get to the same level,” says Elbaz. HydroFusion handles mission planning, georeferencing of topographic and bathymetric LiDAR data and of hyperspectral and digital camera image data, image mosaic processing, the generation of LiDAR sea floor reflectance data, the extraction of sea floor spectral data using data fusion techniques, classification of the sea floor, and extraction of the shoreline, explains Joong Yong Park, Optech Inc.’s Software Development Manager, who manages a group for post-processing software and algorithms development.
“Once the data are downloaded, several steps of processing are required to extract interesting points from the waveform,” says Park. The process involves filtering the input waveform, converting the optical signal to an electric signal, detecting the peaks or half peaks in LiDAR wave format at the water surface and the sea floor, and georeferencing the data to compute the coordinates at the water surface and the sea floor in ellipsoid height.
Unlike in the typical TIM-based processing used for topographic LiDAR, Park explains, bathymetric LiDAR requires waveform processing for each point in order to extract the time and position at the water surface and the sea floor. “Since the CZMIL is a bathymetric LiDAR system, even though it has 70 kHz topo/shallow returns, all point clouds were processed by waveform analysis techniques. Therefore CZMIL point clouds are from multiple returns from one waveform, up to 31 returns. Optech and JALBTCX worked very closely to keep the CZMIL system in operation and to resolve data processing problems on a daily base.”
The key recent advance in LiDAR bathymetry has been the development of new ‘shallow water’ high-resolution sensors, according to Edwin Danson, a Chartered Surveyor and Past-President and Fellow of the Chartered Institution of Civil Engineering Surveyors in the United Kingdom. “This technology,” he says, “offers significant advantages and matches well with the commercial transition zone between traditional vessel-based swathe systems and airborne. In my opinion, this holds a more interesting future for the technique.”
Wozencraft sees a couple of paths for developments looking forward. “At JALBTCX,” she says, “we are focusing on a comprehensive system for coastal mapping that includes LiDAR bathymetry, topography, aerial photography, and hyperspectral imagery, along with exploitation software that allows us to generate products beyond point clouds, such as land cover and sea floor classifications. This information, with the point cloud elevation data, enables us to begin to truly characterize the coastal zone in terms of both geomorphology and environmental resources.” In industry, she adds, “there is a push for smaller bathymetric LiDAR sensors that work well in water that is shallow (less than 10 meters) and very clear, to make for more cost-effective surveys of clear water beaches, coral reefs, and clear, shallow rivers and streams.”
“In the future,” Wozencraft predicts, “we’ll continue to see advancements in data fusion and exploitation to really take advantage of all the information contained in the LiDAR returns and imagery data. The JALBTCX and industry efforts will focus a lot in the near term on data processing strategies for these new systems as we begin to understand the new data better.”
“Bathymetric LiDAR systems are not only for bathymetry anymore,” Park points out. “Using an optical signal from a laser through a water column now makes sea floor reflectance imagery available for environmental monitoring. LiDAR system data will be fused with data from other sensors and used for oceanographic and environmental monitoring. The main interest area for bathymetry will shift to shallow and turbid water near coastlines. The system will be lighter and smaller for small aircraft and UAS, with an advanced laser and digitizer.”
ArcGIS Also Used for Analysis
DATA ACQUIRED USING USACE’S COASTAL ZONE MAPPING AND IMAGING LIDAR (CZMIL) was also analyzed in ESRI ArcGIS to lend insight into storm-driven coastal geomorphology change, and to produce information products critical to USACE emergency response requirements. In the past, workflows for vector, lidar, and imagery have been collected on a project-by-project basis. ArcGIS enables users such as USACE to manage these massive collections of data for more than one purpose and extend the content into 3D. These 3D models, lidar, and imagery can be cataloged and distributed for access in multiple collections as either raster data, raw elevation data, or Esri webscenes. If data is not necessary, it can be easily filtered out for a clean representation. By managing massive 3D models and point clouds with ArcGIS, data can be more easily visualized and analyzed.
EDITOR’S NOTESAll images, except the last two, are of point clouds collected with an Optech CZMIL sensor and are courtesy of JALBTCX and Optech.