Environmental Monitoring and Incident Response
AAI's unique suite of remote sensing technologies and products can be effectively used to quickly size up the environmental scope and impact of an emergency situation, and systematically monitor the impact over time. First, images can often be acquired by overhead sensors at a critical early stage of the event, and AAI's Emergency Response products provide a unique broad area view that can enable a quick informed assessment of the area affected. This can occur even before disaster teams arrive, providing critical information for planning a more effective response. Furthermore, the fully automated quantitative analytical capabilities of the technology allow for fully remote unattended extraction of critical detailed actionable information for meaningful early analysis of possible immediate and long-term environmental and humanitarian impacts of the event. This can provide early insight that is crucial for development of an effective short- and long-term response.
Illustration 1: TVA Kingston Fossil Plant (Kingston, TN) Coal Fly Ash Slurry Spill - 22 December 2008
At around 1:00 am on 22 December 2008, the walls of a raised retention pond at the Tennessee Valley Authority Kingston Fossil Plant near Harriman, TN collapsed, dumping 525 million gallons of toxic fly-ash combustion waste from this coal-fired power plant into the Emory River (a tributary of the Tennessee River) and surrounding countryside. An aerial image of the containment breach is shown below.
AAI's Emergency Response products were used to first quickly assess the immediate apparent impact(s) of the event on the Tennessee River tributary system, and to identify what contamination-related processes were taking place. The Tennessee River is a principal water supply for millions of people, including the City of Chattanooga, and the spill went directly into the Emory River tributary. First, a LANDSAT 5 Thematic Mapper (ETM) image of the site, acquired on the day of the spill, was processed with AAI's environmental water quality technology to "see" through the disturbed water and retrieve the river bottom material characteristics. This identified the immediate spatial extent of fly ash contamination. The results (shown below) revealed that deposits of fly ash sludge (lighter shades of brown in this image) extended downstream in the Emory River from the spill to the confluence with the Clinch River (river entering from the east).
AAI's Material Identifier technology was next used to identify the extent of water quality disturbance immediately following the event. The results, shown below right, reveal that water quality was disturbed (medium blue) by the shock of the spill over a large area extending from just north of the spill site into the Clinch River to the south. The baseline conditions for an earlier pre-event image are shown on the left for comparison (undisturbed water in left and right figures is shown as dark blue).
AAI's QSC Water Quality technology revealed that at least some of the shock-induced water quality disturbance was due to an apparent sudden release of colored dissolved organic matter (see figure below), a decay product of submerged vegetation. Suspended sediments (not shown) also contributed. Suspended chlorophyll concentrations (not shown) further revealed a sharp drop in concentrations downstream of the spill, consistent with sediment entrapment of algae in the areas where the concentrations of organic material were high. The shock-induced release of organic material is consistent with the ongoing decay of sediment-entrapped algae and submerged vegetation, and the extent of the release suggests that there may have been substantial amounts of buried vegetative matter (including sediment-trapped algae) in the river bottom sediments at the time of the spill.
This is potentially a disturbing finding in light of the known chemistry of the coal fly ash sludge. The Kingston Fossil Plant produced Type F coal fly ash from bituminous coal, which not only introduced high concentrations of toxic substances, such as arsenic, barium, cadmium, chromium, lead, mercury, nickel and thallium, into the water. It also contained up to 4% sulphate (SO3) by weight (there was an estimated ~1000 tons of gypsum in the spill). The reason this is a potential problem is that the presence of dissolved gypsum in the water could lead to the production of deadly hydrogen sulfide (H2S) gas.
The bottom soils of rivers like those near and down-stream of the Kingston Fossil Plant with their system of dams and flow control tend to be anoxic (oxygen-starved), and are typically populated by organisms that continually generate gaseous byproducts. The gases are produced from decayed organic material like that revealed in the above image. The typical gaseous byproducts are non-toxic methane and CO2. The introduction of sulphate (dissolved gypsum) from the spill can change that, however. The introduction of sulphate tends to change the population of microorganisms in anoxic river bottom sediments to sulphate-reducing bacteria. Sulphate-reducing bacteria metabolize more efficiently and tend to become dominant under these conditions, deriving energy from oxidizing the trapped organic material to CO2, and reducing the sulphate to H2S gas.
This has the potential of becoming a serious problem in the affected area. H2S gas in low concentrations produces a very noxious "rotten egg" odor. A more serious problem can arise, however, when H2S reaches higher concentrations. The gas is highly toxic, flammable, and heavier than air. It is a broad-spectrum poison, affecting many bodily systems at once, and its toxicity is comparable to that of deadly hydrogen cyanide gas. At 50-100ppm, for example, it can produce eye damage; at 150-250ppm it can paralyze the olfactory nerve; at 320-530ppm it can trigger pulmonary edema with possibility of death; and at 530-1000ppm it can induce strong stimulation of the central nervous system, stopping breathing. 800ppm is considered a lethal concentration for 50% of humans after 5 minutes of exposure. Because it is heavier than air, locally high concentrations can accumulate in protected areas, including the layer of air next to the surface of the water. There it can be highly poisonous to people, e.g., in recreational boats, and to animals, as well as to plants and aquatic life. H2S in water can also be highly corrosive to infrastructure, threatening the structural integrity of metal work, highway bridges, and dams.
This episodic entrainment and down-stream re-deposition of sediments apparently resulted in a progressive down-stream redistribution of sludge into the Clinch River, as illustrated in the image below from 21 March 2009. The retrieved river bottom materials were dominated by sludge (shades of grey in this image), with very little native sediments (tan) exposed. Some earlier and later images (not shown) revealed only partial exposure of sludge on the river bottom, suggesting that the sludge deposits may be episodically covered with native deposits. This suggests that the river bottom material may consist of layered deposits of coal fly ash sludge alternating with layers of native sediment.
A zoomed-out view of the 21 March image (see figure below) reveals that the fly ash sediment deposits on the river bottom (grey) appear to have spread by episodic deposition more than 40km down-stream by 21 March 2009. If microorganisms in the river bottom sediments respond to the presence of co-deposited sulphate, and the population shifts partially or fully to a sulphate-reducing type, the expanding area of contamination suggests the microorganism population density and H2S production rate could progressively build up over time, at least until the sulphate is depleted. With such a large and expanding area apparently contaminated, this could potentially pose a significant problem that will need to be closely monitored.
These products produced by AAI were groundbreaking. The ability to periodically measure key water quality parameters, and to "see" through the impaired water to characterize the bottom sediments over broad areas provided a highly valuable new insight into the character and evolution of a toxic chemical sludge spill in a dynamic waterway environment. They also demonstrated a potentially critical new operational support tool for emergency response to chemical spills.
Illustration 2: Oil Spill Off Unalaska Island - 8 December 2004
On 8 December 2004 the Malaysian Freighter Selendang Ayu ran aground just north of Skan Bay off Unalaska Island in the Aleutian Island chain. The ship had been adrift in the stormy Bering Sea and broke in half when it ran aground, spilling its cargo of soybeans and some of the 424,000 gallons of intermediate fuel oil and 18,000 gallons of diesel fuel it was carrying. Unalaska Island is almost entirely a wildlife refuge, and the spill threatened to impact the highly environmentally sensitive regions along the island's northwest coast where large numbers of sea birds, sea lions, otters, seals, salmon and other coastal and inland wildlife visited and were living. Because the seas were rough, and the waters were cold, the leaking oil could not form a visible slick. Instead it broke up into a floating mass of partially submerged tarballs, making it difficult to spot the oil from an aircraft or helicopter. This was made even more difficult by the high sea state and low light levels at this high northern latitude in December. Needing a quick assessment of where the wildlife would likely be impacted, AAI was requested to attempt to locate the oil using satellite imagery and its specialized technologies.
It took several days before the oil started spilling, but it then spread quickly in the turbulent waters. A QuickBird image was requested and acquired five days after the accident, catching the early stages of the spill event. Light levels were very low and the waters were dark, but AAI's iCee™ technology was able to automatically correct for the atmosphere, sea state, and low illumination levels, and produce an accurate reflectance image. AAI's spectral analysis technologies could then be applied to the image to search for the oil. The results are shown in the figures below.
The processed 13 December 2004 QuickBird image of the Selendang Ayu freighter accident in Skan Bay, Unalaska, AK is shown above. This was the only view of the character and spatial extent of the oil spill available to the responders (Alaska Dept. of Environmental Conservation Unified Command, Greenpeace, and others). Signatures of the oil were successfully derived from the image, even though they were not visually apparent in the image. Upon processing, several streamers of oil could be seen emerging from the ship, remaining consolidated over a significant distance, rather than forming a slick or dispersing, due to the low temperatures. A consolidated streamer of black oil (magenta) can be seen flowing north from the bow half of the broken-up ship. The escaping cargo of soybeans (white) can also be seen emerging from the hull and flowing toward the shoreline, where it then migrated and beached along the shoreline toward the southwest. Other streamers of oil (magenta) can be seen flowing with the soybeans from the hull, and then independently breaking south toward the shoreline. Although some black oil stayed intact long enough to impact the beach, much of it dispersed (gold and orange areas) to form tar balls, weathered to mousse. The areas shown in gold are relatively dispersed (by wave action) into a dull sheen/mousse state, while the areas shown in orange are in a more compressed (by wind and current) dull sheen/mousse state. The area impacted by beached oil can be clearly seen behind the flow of soybeans along the shoreline.
A zoomed-out view of the spill site (same image as above) is shown below. The accident site can be seen toward the middle of the figure, and the dispersed spill can be seen spreading to the east. The pattern of physical states of the dispersed oil is consistent with the expected wind- and current-driven compression of the tar balls (orange and magenta) along the advancing edge of the spill toward the shoreline, and enhanced dispersion (gold) along the trailing edge due to agitation by the wave action. The beached oil in the vicinity of the compressed orange and pink areas is clearly evident, and beaches (beyond the image boundary) further to the east of the large concentrations of tar balls were later confirmed to be some of the most heavily impacted with the greatest loss of wildlife. The fisheries in these areas were also some of the most heavily damaged in the region.
These products produced by AAI were the first of their kind. They provided a highly valuable new insight into the character and evolution of generally poorly understood dispersed oil spills in low-temperature environments, and they demonstrated a potentially critical new operational support tool for emergency response to future oil spills of this type.
Illustration 3: Post-Tsunami US Naval Rescue Mission Hazard Assessment Meulaboh, Indonesia
07 Jan 2005
On December 26, 2004, the second strongest (magnitude 9.1 - 9.3) and longest duration (8.3 - 10 minutes) earthquake ever recorded occurred off the coast of Sumatra. This undersea subduction-type megathrust Sumatra-Andaman earthquake triggered the massive Asian Tsunami with a tidal wave up to 30 meters (100 feet) high, killing over 230,000 people in coastal communities across 11 countries.
A QuickBird image of one of those devastated communities, Meulaboh, Indonesia, is shown below. Meulaboh was the closest population center to the epicenter, and it suffered severe building damage from the earthquake. The image, acquired on 7 January 2005, reveals a nearly totally devastated city and a deep coastal swath left barren by the tidal wave of all structures and vegetation. The muddy river behind the city continued to carry mud and debris from both the earthquake and tidal wave back into the harbor and out to sea.
A close-up view of three of the naval rescue ships is shown in the image below.
The imagery revealed a variety of important characteristics about the condition of the harbor waters. First is the major shoal area to the north of the ships that formed where the river flows into the harbor. While most of the sediment deposits are submerged, some areas are emergent and pose a potential navigational hazard. Other new shoal areas are evident along the shoreline.
Next is the substantial reduction in suspended sediment concentrations in the wake of the ships. The waters appear blue in each wake relative to their surroundings, and the image retrieved suspended minerals concentrations were a factor of 2 - 5 lower there than in the surrounding waters (results not shown). This revealed that the suspended sediments occurred in a relatively thin surface layer, consistent with an expected persistent laminar flow of the relatively warm (lower density) inland waters on top of the cooler harbor waters. Appreciable vertical mixing was not occurring until well out into the harbor. The harbor waters beneath this thin surface layer in the vicinity of the ships were relatively free of suspended sediment from the river, measuring only 2-10g/m3, suggesting that new shoals were probably not still forming and posing new hazards in the immediate vicinity of the ships at the time of the rescue mission.
A third important observation was the occurrence of significant floating debris in the vicinity of the ships. This is more apparent in the close-up view below.
The spectral signatures of the debris were retrieved and analyzed. The materials were found to be floating, with a portion of each object above water and a portion partially submerged. The spectral properties of the emergent portion matched those of several standing buildings in the city, consistent with them being building debris. The large sizes of the objects and strong current would have been potentially serious hazards for the numerous small craft that were present. The continued emergence of debris from the river 11 days after the tsunami further suggests that there could be substantial accumulated debris deposits on the harbor floor, creating potential submerged obstruction hazards for the ships, particularly in the shallower waters (including possible earlier-formed post-tsunami sediment shoals) down-current from the river.
It is important to note that this potentially dangerous debris could not be distinguished from the clutter of surface reflections in the imagery prior to glint removal, and it could easily have escaped detection without that process. The dark color of the debris may also have caused much of it to escape detection by either the helicopter pilots or the crew on board the ships and small craft, where surface reflections characteristically present an even greater clutter problem than from the perspective of a nadir-viewing satellite. AAI's products demonstrated an important new imagery-based operational support role for emergency response missions in harbors following catastrophic events.
Illustration 4: Post-Tsunami Water Contamination Assessment, Porto Novo, India - 29 December 2004
Among the many devastating impacts of the great Asian Tsunami was the widespread contamination of water supplies and coastal fisheries. The destruction of dwellings and other infrastructure by the tidal wave in the coastal communities caused the advancing and retreating tidal wave waters to be contaminated by raw sewage and other hazardous materials exposed by the destruction. Of immediate concern was how much of the surface water supply was contaminated by these waters, and whether there was any surface water that may have escaped contamination. Also of concern was how much of the inland aquaculture and offshore fisheries areas were contaminated by the retreating waters from the land. Aquaculture and fishing are two of the principal livelihoods in the region, and it was critical to identify which aquaculture facilities and fishing areas had become contaminated and off-limits.
AAI's water quality technologies were used to quickly assess the extent of water contamination in a coastal community near Parangipettai (Porto Novo) on the southeast coast of India. An IKONOS image was acquired on 29 December 2004, only three days after the arrival of the tidal wave. The image is shown below. A portion of the Porto Novo fishing village can be seen in the northwest corner of the image just north of the Vellar Estuary, which emerges from the west and empties into the Bay of Bengal just south of he village. The Killai backwaters with their aquaculture and Mangrove swamps extend to the south along the coast behind the deep beach. The area was largely denuded of vegetation (brown coloration) by the tidal wave. The principal exception was the large mangrove swamp in the Killai backwaters just below the center of the image, which was left largely intact.
AAI's water quality technology was applied to the image to first retrieve the concentration of suspended minerals across the image. Any sediments in the water could be presumed to have been entrained during the violent advance and retreat of the tidal wave, and those sediments could have come in contact with the raw sewage and other hazardous materials exposed by the destruction. In the absence of ground samples to demonstrate otherwise, any sediment-laden waters therefore needed to be considered as likely contaminated. The retrieved concentrations of suspended minerals were transformed into a "Contamination Potential," presented in the figure below. Dark blue represents "safe," largely uncontaminated waters. Medium and light blue can be considered at or near the threshold of contamination, and the cyan through red colors represent progressively higher contamination potentials.
In the set of three figures below are shown close-up views of representative large aquaculture facilities along the Vellar Estuary (upper right) and within the Killai backwaters (upper left and bottom figure), and each reveals significant apparent contamination from the tidal wave. The offshore fisheries were also adversely impacted, with presumed sewage-contaminated sediments extending approximately 2 km offshore. These directly impacted the fringing coral reefs, associated marine life, and fisheries in this coastal swath that extend from Porto Novo south.
It is notable that the waters within the Mangrove swamp were largely protected from contamination by the tidal sediments. A zoomed-in view of the waterways within the swamp is shown in the figure below. A river that connects the larger contaminated waterways bounding the swamp was contaminated by flow between them, but the smaller tributaries and independent waterways within the swamp appear to be largely uncontaminated by the tidal sediments.
The waterways were then searched for areas where water clarity was relatively high relative to the immediate surroundings, indicating possible dilution by an incoming underground fresh water source of potable water. A location showing such an area of anomalously "clear" water is shown in the zoomed-in image below. It occurred just inside the northwest boundary of the Mangrove swamp, showing up in the river connecting the contaminated waters bounding the swamp. This suggests that there may be a source of potable groundwater in the vicinity (preferably far enough away from the river to avoid contamination) that could be tapped as a potential water supply. The specific location would need to be identified by a field visit, and the water would need to be tested to be sure it had adequate purity. The results significantly narrow the area that would need to be searched, however, saving valuable time at such a critical phase of the rescue and recovery mission.
This example provides an illustration of the value of AAI's products for providing rapid remote assessments of areas impacted by catastrophic flooding. Areas of potential water contamination, and locations of potential fresh drinking water sources can be identified before response teams arrive. Such information can provide critical early emergency response planning support, saving valuable time and maximizing the utilization and effectiveness of available resources during the rescue and recovery phases.
Illustration 5: Urban flood (New Orleans, Hurricane Katrina) hazard assessment - 3 September 2005
On 28 August 2005 Hurricane Katrina made landfall in St. Barnard and Plaquemines parishes, just east of the City of New Orleans. The resultant storm surge in New Orleans caused more than 50 failures of the city's drainage and navigational canal levees. By 31 August 2005, about 80% of the city was flooded, with some parts under as much as 15 feet of water. Although the evacuation was one of the most successful in history, about ten percent of the residents remained behind and became stranded by the floodwaters. Numerous health concerns were raised over fears of toxic floodwater contamination, and AAI's QSC2 Water Quality technologies were used to search for early evidence of contamination, and identify apparent sources.
Although images were acquired immediately after the flooding began, it was decided to process QuickBird images from 3 September 2005. This brief delay was needed to provide sufficient time for contaminants, such as petrochemicals and sewage, to generate high enough concentrations of the Colored Dissolved Organic Carbon (CDOC) indicator byproduct to be detected by QSC2, allowing the extent and apparent source locations of contamination to be assessed. Two images of New Orleans (shown below) were processed. One covered the western (left) and the other the eastern (right) portions of the city.
The submerged areas were first automatically identified and converted to shapefiles (cyan) using AAI's Land Water Interface to enable: 1) a detailed early assessment of the areal extent of flooding; and 2) a direct comparison with digital maps and other GIS data for identification of specific street-level infrastructure affected. The water shapefiles are shown below.
An area of immediate concern was the Agriculture Street Landfill, a 95-acre Superfund site storing 50 years worth of municipal garbage and industrial waste. It contained large amounts of lead, arsenic, dioxin, and carcinogenic hydrocarbons, which were later sprayed with the pesticide DDT. It was eventually remediated by being fenced-off and covered with a mat barrier and two feet of clean soil. Early reports of these contaminants in the floodwaters prompted a concern that the floodwaters may have caused the contaminants to escape the landfill and migrate into the surrounding flooded areas.
The imagery was processed with AAI's QSC Water Quality technology to search for patterns indicative of organic carbon-based contamination emerging from the site. One of the useful image-retrieved water quality parameters for this purpose is the concentration of Colored Dissolved Organic Carbon (CDOC). CDOC can be used as an effective indicator of organic carbon-based contaminants, such as raw sewage, petrochemicals, and other hydrocarbons. In water, these contaminants partially dissolve into CDOC compounds, providing a detectable byproduct indicator.
A zoomed-in view of the CDOC results for the area surrounding the Agriculture Street Landfill Superfund site is shown below (the site is outlined in red). The CDOC showed a streaky pattern in the open waters, particularly to the north of the site, with areas of higher concentration generally aligned in a NE-SW orientation. It also appeared to be adhering to protruding objects, such as trees and buildings. This pattern is characteristic of a floating layer of material being swept by the wind and adhering to objects, and it indicates that the much of the carbon-based component of the contaminant in this area was likely in the form of floating petrochemical slicks.
Note that some of the highest concentrations of CDOC (~11mg/m3) occurred inside and around the southern part of the landfill site (area indicated by the black oval), suggesting that contaminants may indeed have been emanating from the landfill, as feared.
This possibility was further explored using the image-retrieved pattern of suspended mineral concentrations at the site. By 3 September, the extent of flooding had reached its maximum and water flow was minimal. This means that non-floating suspended sediments would have already settled out. As a result, the image-retrieved concentration of suspended minerals could be effectively used to search for evidence of local entrainment of sediments associated with turbulence arising from upward migration of water escaping from the interior of the landfill. The results for the suspended mineral concentrations at the site are shown in the figure below.
The processing results revealed enhanced concentrations of suspended minerals (~3gm-3) along the northern and southern boundaries of the central mound (green area), and extending along the northern boundary of the southern section (white ovals). This pattern was consistent with turbulence and local entrainment of sediment along these boundaries, likely associated with the upward flow and injection of material from interior of the landfill into the floodwaters above. The escape of material from the excavation below would have been an expected consequence of the seepage of floodwaters into and eventual filling of the landfill cavity. The seepage of water into the landfill cavity would have displaced and driven trapped air out, and the jets of escaping material would have propelled water and associated contaminants into the floodwaters above, forcing bubbles and sediments into suspension.
The injection of fluid into the floodwaters at these locations can also explain the pockets of low CDOC concentration there. A zoomed-in view of the CDOC concentrations at the landfill site, shown below, revealed localized areas of anomalously low CDOC concentrations at the same locations where the sediments were being entrained. This pattern was consistent with an injection of water from below at those locations, the turbulence from which would be expected to have displaced the surface layer containing the CDOC generated by the petrochemical slicks.
AAI's products provided valuable early evidence, based on the CDOC concentration patterns, that the Agriculture Street Landfill may indeed have been a source of the reported toxic floodwater contamination, as feared. The results indicated that the neighborhoods to the immediate north, east, and south of that Superfund site were likely the most affected. Although the actual toxic materials were not directly detected, the CDOC served as a valuable surrogate indicator of the areas likely contaminated, areas that should be among the first ones field-sampled and assessed for early response and mitigation.
Illustration 6: Rural flood inundation mapping (Mississippi River near St. Louis, MO) - 18 July 1993
AAI's Land Water Interface technology provides a means for rapid wide area assessment and mapping of the extent of floodwater inundation in rural areas as well. A LANDSAT Thematic Mapper image of the flooding Mississippi, Missouri, and Illinois rivers near St. Louis, MO during the Great Flood of 1993 is shown below (left). This image was calibrated to units of material reflectance using AAI's iCee™ Atmospheric Correction technology, and it was then processed to automatically delineate water from the other non-water cover materials (right) using AAI's Material Identifier technology.
A zoomed-in view of a section of the flooded area, which reveals additional spatial detail of the affected area, is shown below. Some of the areas with very shallow inundation (less than half-meter depths) are shown as dark brown.
The water (blue) was then automatically converted to a shapefile, shown in the two figures below, which could be used to create a detailed map of inundation. The shapefiles can be directly compared to maps of property boundaries to determine specific fractions of parcels inundated for rapid assessments of property damage, likely insurance claims, and economic loss.
The shapefiles of the inundated area can be compared to shapefiles of the river during more typical conditions to obtain a detailed map and accurate quantitative measurement of the specific area flooded during 1993. This is illustrated in the figures below, where the river during more typical late spring conditions (retrieved from a LANDSAT Thematic Mapper image from 18 May, 1994) is shown in light blue, and the flooded area (retrieved from the 18 July 1993 image) is shown in dark blue.
The total area (upper figure) covered by water in the image during the 1993 flood was 1298.34 km2, compared to only 654.98 during the more typical conditions of the following Spring. This corresponded to 643.36 km2 of inundated property in this 183km x 170km area near St. Louis. Most of the inundated property was productive cropland, which can be seen in the figures below. The inundated area (left, dark blue) corresponded almost exclusively to floodplain cropland (right, light tan), with few exceptions. The figures also clearly identify which floodplain croplands were successfully protected by the levees and escaped inundation.
AAI's products can provide valuable rural flood damage assessment information. The shapefiles can be directly compared to maps of property boundaries to determine which specific parcels were affected, and to measure specific fractions of parcels inundated. This enables rapid assessments of property damage, predictions and verifications of insurance claims, and projections of economic loss.
Illustration 7: Deepwater Horizon Blowout Disaster in the Gulf of Mexico - Spring 2010
21 Apr 2010 (U.S. Coast Guard Photo)
On 22 April 2010, the Deepwater Horizon offshore oil drilling platform burned and collapsed after a catastrophic blowout two days earlier. Located about 65 miles off the Louisiana coast, the collapse of the rig ruptured the well pipe at BP's Macondo well over 1500 meters below the surface. This caused an uncontrolled gusher of oil and gas to surge into the Gulf waters from the
seafloor. AAI is applying its unique technologies to commercial satellite imagery acquired of the Gulf to "see" below the surface and characterize the submerged oil and its effects on the water column, seafloor, and beaches.
>> more