SONAR is an acronym that stands for SOund NAvigation and Ranging. It is a technique that allows the operator of the SONAR transmitter/receiver to “see” underwater objects. SONAR equipment falls into two varieties – active and passive. In a passive system, an operator’s job is to merely listen to sounds generated by objects (1). In an active system, a pulse of sound energy is transmitted (propagated) from a transmitter (acoustic array). The sound energy moves through the water (speed of sound in water is about 1500 meters per second on average) strikes objects and is reflected back to the acoustic array where an operator is listening for the reflected energy (echo). The changes in the shape, return signal strength and frequency shifts in the reflected signal can be used to detect the presence of an object out of visual contact and gives an idea of both the size, bearing and range to the object being detected. For example, if it takes 6 seconds for an echo to be detected, the distance to the object producing the echo can be estimated by knowing the speed of sound in water and other variables such as temperature and salinity. Assuming it took 6 seconds to go and come back, it means that the “target” reflecting the sound energy is 3 seconds distant or 3 x 1500 meters away. The first SONAR sets were used in World War I for the purpose of detecting enemy submarines by the British. Towards the end of the war, there were active systems development by both the British and US Navy, although it was too late to have any effect on the outcome of the war – it would be actively used in WWII (2). The signals from different types of objects can be very distinctive and a trained operator can tell the differences between fish, whales, ships or other manmade objects (2).
Figure 1 – Basics of SONAR Theory (original graphic generated in Powerpoint)
SONAR technology has improved immensely since the early days. The first sets were passive SONARs - devices that just listened for signals. The first active sets were under development as early as the 1920's (1).From the first crude sets built, there are now multiple types, some of which have the transmitter and receiver in one antenna array, separate antenna arrays for each the transmitter and one for the receiver, and sets that include multiple transmitters and receivers. The arrays might be mounted on the hull of a vessel, might be towed behind on a device called a “fish” or be some combination of both approaches. The beam direction might run parallel to the path of the vessel or it might be perpendicular to its path.
As an example of a combined array is a device used in sport fishing – the “fish finder.” Many sport fishermen have simple sonar sets that they use for finding schools of fish – the purpose – to make it easier to detect the presence of “something” swimming between the hull of the boat and the seabottom. Simply put, if the fishermen see a return on a scrolling paper chart, they know that there may be fish to catch! (2) Some of these civilian sonars can be very elaborate in their outputs, even approaching military grade quality in the information provided. Military grade equipment typically has multiple beams to provide all-round cover while the simpler civilian ones only cover a narrow arc.
Figure 2 – Operation of a “Fish Finder.” (original graphic generated in Powerpoint)
The amount of energy reflected back from the seabed floor can vary greatly. It is a function of:
1. angle of incidence of the sound wave, (referred to as the grazing angle),
2. the topography of the seabed
3. the seabed sediment composition and
4. the frequency of the sonar signal
Sound energy is well reflected when it bounces off a flat surface normal to the sound waves path of travel. However at an oblique angle, much of the sound is reflected at a complementary angle away from the receiver. Similarly rough surfaces tend to scatter the sound energy in directions away from the source. This generally dissipates the received sound level, but can enhance it when the angle of interception with the surface would otherwise reflect most of the sound energy away.
Some of the sound energy is lost into the sea floor itself. The amount sound energy will propagate into the sea floor is highly dependent on the frequency and bottom composition. For a typical bottom type and nominal source level, frequencies above 10kHz penetrate very little. From 1kHz to 10kHz sound often penetrates to several meters of depth. From 100 Hz to 1kHz sound can penetrate to several 10s of meters or more. Below 100 Hz sound waves have been detected traveling at various depths in the earth's crust around the globe.
In the 1960’s, Dr. Harold Edgerton from MIT developed a method of using a SONAR array that sent pulses sideways rather than just vertically at frequencies between 100 and 500 KHz. A sidescan sonar is typically configured to be on a towed fish, although there are some versions that can be hull mounted. The pulses are transmitted both downwards and sideways and provide a view both downwards and to the sides of the fish. (3) The transducers look sideways, and receive the sound reflections from the seafloor and objects on the seafloor. This information is sent to a computer that displays an image based on the scan. This type of sonar is excellent for differentiating different types of seabottom, (mud or rocks). They provide decent returns in varying water conditions (turbidity). They can be used for accurate imaging of large areas of the bottom, and the resolution increases as the frequency increases. The best typically used can measure features on the bottom with up to 10 centimeters of resolution. (4) USGS uses sidescan in conjunction with other types for underwater surveys. In combination with a GPS, sidescan sonars can be used to map features over large areas of the ocean bottom. The typical configuration is in a towed array that is towed close to the bottom that radiates a fan like pattern up to 100 meters in width (5). They are very accurate for imaging large areas of the seafloor and can provide continuous datasets. In general, the higher the frequency, the better the resolution, but the shorter the range.
Figure 3 - Illustration of a research vessel with a towed side-scan sonar system and a hull-mounted multibeam sonar. Courtesy: NOAA Coast Survey (4)
Multibeam sonar combines hull mounted arrays and towed arrays with inputs from differential GPS and the ships location vertically and horizontally and can provide good spatial accuracy up to 1 meter. Although the technology is constantly improving, a typical operational depth currently can be as deep as 6000 meters. Multiple signals are sent down and to the sides to create swath generating a huge amount of 3 dimensional data. This 3D data can be combined with elevation data to create accurate topographic maps of the seafloor.
Single beam transponders collect discrete samples of data - point data along some predetermined survey track. Multibeam arrays, on the other hand, are designed to collect continuous bathymetry data that is high resolution throughout the entire survey area (8). The evolution of the technology has been exponential with respect to the amount of data collected. The capability of the amount of data collection in Multibeam systems:
Single beam equipment arrays are cheaper and easier to deploy than multibeam systems. They can be configured to interface with other instruments to provide data about seafloor composition. Multibeam systems are really just a series of single beam that provide more detailed views based on the sheer volume of data collected. What is sacrificed in portability is more than made up for in the products produced - high resolution bathymetric maps of an entire area georeferenced by design (8) since they are coupled with extra navigation equipment and shipboard attitude sensors. While data from single beam arrays requires interpolation between sampled points (which results in only 5-10% of the survey area being mapped), the multibeam systems can provide continuous resolution as small as 1 meter in diameter.
Mapping techniques – To map (or chart) an area, it is necessary to follow a search pattern or sampling pattern that is predetermined well in advance. Calm seas and a relatively straight course are necessary for good quality collection. (6) Both sidescan and multibeam can be used for undersea mapping, but because multibeam is designed up front to be precisely georeferenced, the maps provided by using the multibeam technology are more accurate than by simple sidescan technology alone. The images in Figure 6 show side by side images using different technologies. The image on the right was obtained in 1993 using a sidescan towed array, while the image on the left was acquired in 1998 using a multibeam array. Adjacent survey lines were matched and the data was “stitched” together to produce the combined image. (7) USGS uses both sidescan in conjunction with multibeam to generate accurate bathymetric charts (4). Round trip travel time of the sound energy is multiplied by the speed of sound in the ambient water and divided by two. The individual values of seafloor depth are then contoured to produce bathymetric maps.
Figure 5 - Single Beam Hull Mounted Transponder vs. Multibeam Interferometric (Multibeam) Sonar http://woodshole.er.usgs.gov/operations/sfmapping/bathy.htm
(For more details on multibeam SONAR go to http://www.ldeo.columbia.edu/~vschmidt/Presentations/An%20Introduction%20to%20Multi-Beam%20Sonarv4_files/frame.htm for a great presentation from Columbia University.)
Figure 6 – Side by side products of Multiscan (left) vs. Sidescan alone. Image courtesy of USGS. Note the image on the right has been stitched together from multiple profiles. http://walrus.wr.usgs.gov/mamalabay/products/side_back_comparison.html
Other Helpful Resources From NOAA
Quick Facts (PDF):
When a survey vessel is on a track, the sonar array is moving continuously with respect to the seabed and the vessel's motion. The equipment must correct on the fly for the vessel's motion by contrasting the position of the vessel in 3D at any moment with some some known vertical reference. One method of doing this is by using inertial navigation (INS) inputs from one or a set of high-speed gyroscopes. It makes the data more accurate, but is an expensive addition.
Bathymetric Data Sources by Agency
Data exists in many forms and has been collected by many agencies (more on other issues associated with this later). Below is a table of a few of these sources where data can be downloaded.
|NOAA Coastal Services Center||Benthic Survey Data||http://www.csc.noaa.gov/benthic/data/data.htm|
|NOAA Office of Coast Survey||Hydrographic Surveys||http://www.ngdc.noaa.gov/mgg/bathymetry/hydro.html or|
|NOAA Office of Coast Survey||Historical Maps and Coastal Charts||http://nauticalcharts.noaa.gov/csdl/ctp|
|National Geophysical Data Center (NGDC)||2-Minute Gridded Global Relief Data (ETOPO2v2)||http://www.ngdc.noaa.gov/mgg/announcements/announcements.html|
|National Geophysical Data Center (NGDC)||Bathymetry||http://www.ngdc.noaa.gov/mgg/bathymetry/relief.html|
|National Geophysical Data Center (NGDC)||Searchable access to all marine geology data archived by, and available through NGDC, regardless of data format. ESRI IMS interface||http://map.ngdc.noaa.gov/website/mgg/geolin/ or|
|National Geophysical Data Center (NGDC)||Coastal Bathymetric Charts||http://www.ngdc.noaa.gov/mgg/bathymetry/maps/nos_intro.html|
|Scripps Institute||Shuttle SRTM data (downloadable)||http://topex.ucsd.edu/WWW_html/srtm30_plus.html|
|Joint Airborne Lidar Bathymetry Technical Center||20 kHz topographic lidar data or 3 kHz bathymetric lidar data, each concurrent with digital RGB and hyperspectral imagery||http://shoals.sam.usace.army.mil/|
|University of Hawaii||Cable Route Survey : Bathymetry and Sidescan Imagery (among others)||http://imina.soest.hawaii.edu/HMRG/Data_Gallery/index.html|
|Woods Hole Science Center||Multiple projects in concert with USGS. List of projects, data and IMS site.||http://woodshole.er.usgs.gov/operations/sfmapping/ and|
|State of NY||IMS Site showing Map of Hudson River Channel Features||http://www.dec.ny.gov/imsmaps/benthic/viewer.htm|
Carleton University provides a website and software downloads that allow you to simulate planning, charting and sailing a bathymetric survey course either across the Atlantic or Pacific Oceans, make a sketch that shows an area of the seafloor in one of these areas, and then use the interface to simulate making a survey track and generating a contour map based on the track selected. Go to http://serc.carleton.edu/eet/seafloor/all_parts.html and follow the steps for downloading GeoMapApp and using it to further explore undersea mapping. What you will be instructed to do is a 4 step process:
Download and install GeoMapApp and open a dataset. Explore GeoMapApp's features and make a sketch of a visualization that shows an area of the seafloor.
Explore the Atlantic and Pacific ocean basins. Collect data on ocean depths across one of the basins along one or more latitudes.
Combine profiles from separate latitudes to build a whole-class contour map of an area of the seafloor.
Compare the contour map produced by hand with visualizations of the same area produced by GeoMapApp. Use GeoMapApp's Grid function to produce contour maps and profiles for other areas of the seafloor.
"GeoMapApp is a global topography database and stand-alone GUI application written in the Java™ programming language. Both the database and application are actively being developed. This page will be maintained to inform GeoMapApp users of changes to the software and underlying databases." (http://www.marine-geo.org/geomapapp/whatsnew.html)
(This site provides a good overview in the collection, display, and interpretation of bathymetric profile data.)
Personal Note 1: In the early 1980's I worked for an Ocean Engineering Consulting Company (AV Strock and Assoc) that specialized in coastal planning and beach erosion control projects. In the course of my work with them, I actually got hands on experience using SONAR equipment (a single beam fish finder - best technology available for a small firm at the time) running bathymetric survey profiles from a point 30' deep to the Mean High Water line, marrying up the marine profile with a Florida Dept of Natural Resources (DNR) or an Army Corps of Engineers (COE) survey monument. Basically, we would take elevation data on land along a profile line perpendicular to the beach and continue collecting elevation data with the hull mounted SONAR out to a depth of 30' or so. There were almost 300 profile lines measured in this manner in Palm Beach and Broward Counties. We were under contract from the state of Florida to try and track the amount of beach erosion occurring on an annual basis. Comparing the profiles to available historic data going back as far as the 1920's provided us with an estimate of exactly how much erosion (or accretion) was actually taking place. Our computer model attempted to accommodate for variables such as water temperature, salinity, (both of which significantly affect the speed of sound in water) tide tables to account for the mean sea level, and others. What it allowed us to do was use profile data to estimate the volume of sand lost to erosion with the idea of using a primitive GIS technique to select what was called a "borrow pit," an area in 40' of water or more with sand similar to the beach sand that could be dredged and pumped back onto the beach to restore the historical coastline. Being a very affluent area at the time (even more so today) provided the two counties with enough political clout to get the state to pay for a beach renourishment project to restore privately owned beachfront lost in major storm events. The model could actually generate a primitive bathymetric chart with the elevation data we collected. I actually had the pleasure of meeting and working with Dr. Edgerton from MIT, when he came down to Florida work with one of my professors testing new equipment on the East Florida Coast and in the Bahamas doing bathymetric surveys. The data was reduced by hand at the time, point by point, profile line by profile line. New tools exist to allow this to be done automatically today. Reading the SONAR Chart by hand point to point, loading the data by hand, and computing sand volumes by hand are no longer necessary. Most modern SONAR packages have the ability to record and process data on the fly, and an extension to ArcGIS can be downloaded from http://woodshole.er.usgs.gov/project-pages/DSAS/version3/ that automates much of the analysis work.
Figure 7 - Typical Features of a Beach Profile and Aerial View Showing Differences in Shoreline Based on Year and Placement of Profile Lines (left original slide created in Powerpoint; right image from http://woodshole.er.usgs.gov/project-pages/DSAS/version3/ (14) )
Figure 8 - Representation of an Extension of ArcGIS for Analysis of Shoreline Data http://woodshole.er.usgs.gov/project-pages/DSAS/version3/ (14)
Data Handling and Storage
With the explosion in the volume of data comes better, more detailed bathymetric maps. Accompanying this are all the issues that go with running a large geospatial database. The desired end to end processing that goes with this type of operation is not a simple process. Lets talk a moment about GIS and bathymetric end to end processing: What it is. What it isn’t. It is planning and providing for all the steps, especially collecting metadata, from the time a survey or cruise is planned to the final disposition of the processed data and the generation of information products from the results of the survey or cruise. This needs to be a flexible and evolutionary process, adapting to the ever-changing needs and requirements of the collectors, data centers, and data users. It is not a lock-step, fixed-in-time, set of unchanging, standard operating procedures designed as a one-size fits all methodology. To be such will simply drive users and collectors to other solutions. Merging of multiple types od datasets is making new visualization techniques possible - but is also driving up the data storage requirements.
Flythru movies of submarine topology are now possible to see in 3D http://www.ldeo.columbia.edu/res/pi/Hudson/research/bot_mapping/visualization.php (Example - this movie of the Hudson River Channel in New York) (15). One driver for the concern of handling and managing the data is Public Law 106-554, Section 515: policy on data quality. The Office of Management and Budget (OMB) directs "policy and procedural guidance to Federal agencies for ensuring and maximizing the quality, objectivity, utility, and integrity of information (including statistical information) disseminated by Federal agencies.''
Per the PL, Data and Information “Quality'' includes:
Utility – Usefulness of the information to the intended users.
Objectivity – Presentation of information in an accurate, clear, complete, and unbiased manner.
Integrity – Protection of information from unauthorized access or revision, to ensuring no compromise through corruption or falsification.
To quote a white paper from the Naval Research Lab (NRL), "Seafloor mapping to date has been highly inhomogeneous, a crazy-quilt of sounding lines to and from major ports and local areas mapped in detail for purposes of basic research, hydrocarbon exploration, or other purposes. The seafloors of the southern oceans are especially poorly known." (10) and combine it with a statement from then acting director of NOAA's National Geophysical Data Center (NGDC), Chris Fox, in a briefing (11) on data handling, "the current model of data flow and processing is a haphazard, loose connection of collectors, data centers, and users. Data Centers can only provide data which they have acquired and the flow of NOAA data to the NOAA Data Centers has not been a guaranteed proposition. While NGDC has historically provided data and information products to all of the multibeam components, they have not necessarily received multibeam data in return" paints a grim picture for the compliance with the law and protection of the geospatial bathymetric asset once it is collected. "Inhomogeneous" data collection plans combined with "haphazard" dissemination plans paint a grim picture for the current state of the art. A proposal from NRL that remains largely unfunded, is the Global Ocean Mapping Project (GoMap) would combine existing data sets from multiple projects, organizations and formats into a coherent managed project.
Figure 9. Image Courtesy of Naval Research Lab ( http://mp-www.nrl.navy.mil/marine_physics_branch/current_trk_cov.jpg ) (10)
Included in the GoMap project would be efforts using acoustic technologies such as multibeam as well as radar altimetry. In my opinion, private companies will not likely fill the voids where the data doesn't currently exist. While the trans-Atlantic cable routes are still increasing in demand for bandwidth, other areas are having difficulties getting subscribers to sign up for capacity. This has a direct bearing on the question of which technology to use (sidescan or multibeam) when collecting new bathymetric data useable for the cable laying industry (12). Combined arrays of multibeam and sidescan technologies obviously provide the best data quality - however - how good is "good enough" is as much a business decision as it is a technology capability question. For years, sounding lines were used and the effort, while primitive, worked. It has to be difficult to sell expensive solutions to companies that may be only marginally profitable from cables lying dormant for large blocks of time.(12) One can only hope that governmental entities in some sort of Technology Consortium will emerge as the means of providing us at least the same degree of resolution about the surface of our own world that we have when discussing Mars, the Moon and the surface of Venus (better than 10m resolution), but as we know from our studies about data collection on land, large sets are expensive to acquire. If it is doubtful that we will ever have complete coverage of the land in detailed resolution, it must be even more true for that 3/4 of the earth's surface covered in water. (10) For simple purposes of just cable laying, sidescan sonar is a capable solution that has been used for decades, but so much more could be learned if the multibeam technology (and others) can be employed in the seafloor mapping effort as well. To sum up - sidescan provides excellent resolution. Multibeam is great but loses energy at greater depths. Since sidescan is towed, the amount of cable can overcome this. So which technology is better for cable route mapping? It depends on a number of factors (depth of water's effects on resolution being one) and ultimately the business model and customer being served will play a big part (13) - there are advantages and disadvantages to both. Basically, sidescan can provide better resolution at deeper depths, but multibeam provides the better positional accuracy.
Figure 10. Single (left) and multibeam (right) echo sounding of the seafloor. Single beam systems typically have beam widths of 10-30 degrees and estimate depth by measuring the shortest slant range to the seafloor within the main beam. Multibeam (swath sonar) systems provide a series of slant range and elevation angle estimates along a fixed azimuth. This method is preferred because it measures an entire area rather than a single line on the seafloor. (University of New Brunswick via Image and Text Courtesy of NOAA (13)
Personal Note 2: The Naval Research Lab located at the Stennis Space Center has developed a program (Thin and Thick Client versions) that can access and display many types of data concurrently in a single view. It is called the GeoSpatial Information Database (GIDB) and among the datasets it accesses is bathymetric data from several different organizations using varying technologies - many other than sonar. I was actually the sponsor and participant for some of this research and development effort from 2001-2005. A list of papers regarding this effort appears elsewhere on this website, as well as a systems architecture diagram. It should be noted that the data available for viewing comes from not only sonar data surveys, but from other technical sources as well. The expansion of the worldwide bathymetric dataset is being generated from many sources - ones outside the scope of this particular project explanation.
Figure 9. Screen Capture of NRL Thin Client
2. Type in "bathymetric" into the Search space.
3. The following layers will appear:
|Click a link below to load the layer:
4. With respect to the area in the viewer, the software will load multiple datasets provided they are from the same AOI being viewed.
5. Now go to http://dmap.nrlssc.navy.mil/dmap/gidb2/thick_client.jsp and explore the layers available with the Thick Client - they should be the same, except the Thick Client enables a user to load their own data into the viewer as well. A list of the Data Formats the tool will accept is located on the page.
Personal Note 3 - I wouldn't consider the project complete without a bathymetric chart and a few aerials of my "home waters" downloaded from http://historicals.ncd.noaa.gov/historicals/histmap.asp and loaded into GoogleEarth.
Figure 11 - Bathymetric Chart of The waters around Pensacola, Florida. Image courtesy of NOAA.
Figure 12 - Channel Leading from the Gulf of Mexico into Pensacola Bay - Surveyed After Hurricane Ivan - Approximately 23000'
Figure 13 - Channel Leading from the Gulf of Mexico into Pensacola Bay - Surveyed After Hurricane Ivan - Approximately 41000'
This document is published in fulfillment of an assignment by a student enrolled in an educational offering of The Pennsylvania State University. The student, named above, retains all rights to the document and responsibility for its accuracy and originality.