ABSTRACT

Aerial surveillance is becoming a foundation on the overall oil spill response strategy due to the ability to plan and tactically position response resources in the optimal areas of oil migration. It takes a complete multitasking approach to effectively respond to oil spills. While much of the regulatory focus to date has been on the resources on the sea - vessels, skimmers, dispersants - the reality is that they are only one of the components and not necessarily the most important in combating oil spills. It is imperative to determine the location of oil that is most recoverable, and give quantitative information - thickness, volume, area, classification - whether day or night.

Having the right information at the right time optimizes dramatically the use of all the response resources. And assess the effectiveness of the response and make an accurate natural resources damage assessment is critical and requires as well quantitative and timely information.

In the past the main effort has been directed towards developing airborne sensors with enhanced spill monitoring capability. Recently, more and more attention has been paid to the automated processing of oil spill data acquired by integrated airborne sensor platforms. Automated processing and real time relay of immediately usable information to the Incident Command Center is critical during all phases of response.

This paper focuses on advanced data processing and presents ways of improving the usability of airborne multi-sensor oil spill monitoring systems. In this context, is given an overview of currently existing oil spill remote sensing technology like infrared/ultraviolet line scanners, microwave radiometers, laser fluorosensors and radar system. The paper presents POSEIDON, a system for network-based real-time data acquisition, analysis and fusion of multi-sensor data. Also, a method for the distribution of oil spill data and related data products using web-based geographical information systems is described; automated generation of thematic maps of the oil spill scene along with their real-time web-based distribution is becoming more important in marine incident management.

INTRODUCTION

Oil spills have the potential to threaten human health and safety, the integrity of the environment, and the viability of local economies [1]. Industry and governments are actively looking for technologies to minimize the risk of spills and to deal effectively when they do occur.

The oil spill response community has been searching for innovative and cost-effective solutions to acquire real-time, quantitative information on an oil spill scene. Today, airborne remote sensing is becoming a foundation of the overall strategy to improve the ability to plan and position response resources in the optimal areas to respond to spills. It is also an effective and efficient method for natural resource damage assessment (NRDA) following the spill, providing objective and quantitative information. Another context where airborne remote sensing is widely used is on the periodic surveillance of coastal areas, with benefits ranging from the ability to provide early warning on accidental spills to provide a strong deterrent from illegal discharges [2, 3, 4, 5].

Industry and Governments are aware that aerial remote sensing and intelligence are fundamental elements in a response and issued several recommendations. Some recommendations focus on needs for development of special sensor (ICCOPR - Oil Pollution Research and Technology Plan) [6], other emphasize the importance of multi-sensors platforms, due to the multiplicity of strategic and tactic tasks during a spill (API - TR 1144,) [7], other promote pro-active surveillance programs, exercises and training to ensure familiarity with the equipment, real time data analysis and timely delivery of information (IPIECA – IOGP - Oil Spill Response Surveillance Monitoring Visualization JIP) [8]. In the aftermath of Deepwater Horizon Spill, several Workshops have been organized to bring together industry, governmental agencies and academic institutions to focus on needs for future technologies to improve oil observations and oil response [9, 10, 11, 12, 13]. The priorities identified show that for the oil responder community the main needs are: multi-sensors platforms for complementarity/redundancy of information; capability to classify oil targets as recoverable or non-recoverable; georeference the targets and track moving oil; real time information - for tactical and strategic use; data suitable for the Common Operating Picture; and ability to expand the operating window to low-light conditions.

Based on these needs and moved by the vision to protect the coastal ecosystem while supporting the vital energy industry, Fototerra recently introduced in service in the Gulf of Mexico POSEIDON, a twin turboprop Embraer EMB-110 aircraft equipped with an advanced data acquisition and processing system for airborne oil spill and pollution detection. The platform integrates multiple airborne remote sensors and mission system components into one network-based data acquisition, real-time processing and communication platform. Efforts have been directed toward developing integrated airborne sensors with enhanced spill monitoring capability, but more and more attention has been paid to the automated processing and relaying of oil spill information acquired by the platform. Automated processing and real-time relay of immediately usable information to responders is critical during all phases of an oil spill. POSEIDON was declared ready for operation in July 2016 and deployed in Houston, Texas.

Fig.1 -

POSEIDON – Multisensor Platform for Maritime Oil Pollution Surveillance

Fig.1 -

POSEIDON – Multisensor Platform for Maritime Oil Pollution Surveillance

INTELLIGENCE ON THE SCENE

The philosophy behind POSEIDON is to provide intelligence on the scene, which means understanding the environment, deciding on appropriate actions and controlling the results. Priority for the responders is to put timely and efficiently resources in the right position - day and night - to recover the oil and support the common operating picture with clear, usable, real-time information.

An integrated multisensor environment, data processing and real-time communication system are the foundation of the POSEIDON platform.

Fig.2 -

Poseidon Platform

Fig.2 -

Poseidon Platform

Multisensor Environment

The raw data coming from each sensor system in a multisensor environment contain a lot of specific information, and the quality of the information is significantly enhanced by the real-time processing and fusion of the data. In a multisensor environment, the complementarity of the data extends the amount of information, while the redundancy assesses the validity of the data and provide an effective backup.

Data Processing

The fusion and analysis of multisensor and multitemporal data allow observers to derive quantitative information immediately useful to the responders instead of just showing the individual sensor images or tables of raw data.

The information that the POSEIDON platform provides from various sensors and real-time data processing and analysis is divided into far-range detection and near-range monitoring. Far-range detection is based on the use of radar systems, which usually cover swaths of several tens of kilometers. Suspicious structures detected by airborne radar are then investigated on-site using near range sensors. Near-range monitoring of oil spills includes visualization, quantification and classification of the type of oil, the area of the spill; the coordinates of the center of the oil spill or the patches of oil; definition of the polygon of interest; mapping of relative thickness; measurement of absolute thickness; volume of the spill; definition and localization of the hotspots, which are areas inside the spill where the bulk of oil is concentrated; and day and night (thermal imager) video recording of the scene.

Additionally, analyzing multitemporal data from different missions, or different approaches of the same oil spill during one mission, allows deriving of the dynamic properties of the spill, such as the drift and spreading of the oil on the sea surface. All the information is georeferenced and integrated with additional layers of data from a direction finder, AIS (Automatic Identification System) or satellite imagery, allowing the creation in real time of the complete scene. The direction finder monitors distress frequencies that support search and rescue (SAR) operations, and the AIS gives real-time positions and tracking of the vessels present on the scene.

Communication

All the data are made available in real time during the flight. This enables the operator to continuously report to the Incident Command Center to support the decision-making process with the most current data and report to the ship-based responders to improve the oil recovery process. POSEIDON’s primary communication system is based on an innovative microwave radio system that allows digital high-speed communication and data transfer between the aircraft and other assets on the scene. The system enables the operator on the aircraft to transmit the scene information on a high-speed, high-capacity and extremely robust communication network directly to the decision makers involved in the response. Real-time mission coordination is crucial for efficient and safe operations. This enables the responders to have the common operating picture of the spill which allows them to plan the best strategies and tactics for the response. Real time processed data can be easily uploaded in any GIS platform or shared in a dedicated website.

Readiness

With 5 hr. of mission time, POSEIDON can fly up to 900 naut. mi. (1620 km) in a single mission. This allows coverage of any block in the Gulf of Mexico in a single mission from the home base in Texas or Louisiana. From Houston, the aircraft can be repositioned within 10 hr. anywhere on the West or East Coast.

Readiness of the asset is also crucial. The aircraft and equipment are kept ready for deployment on a 24/7 basis, with an integrated team of pilots and operators highly trained and coordinated to conduct operations. This allows the aircraft to be ready for dispatch within 1 hr.

Fig.3 –

POSEIDON in flight with the impressive lineup of sensors. From the front: EO/IR, SLAR, MWR, DF, LFS, MBR antenna, VIS, IR/UV.

Fig.3 –

POSEIDON in flight with the impressive lineup of sensors. From the front: EO/IR, SLAR, MWR, DF, LFS, MBR antenna, VIS, IR/UV.

TECHNOLOGY

POSEIDON’s integrated multisensor platform is divided into far-range detection and near-range detection and analysis. Far-range detection is based on the use of a radar system, which usually covers swaths of several tens of miles. Targets detected by airborne radar are then investigated on site using near-range sensors. Near-range analysis of oil spills includes visualization, quantification and classification of the type of oil at a distance of 1,000–3,000 ft (300 – 900 m).

Sensors include far-range detection sensors: SLAR (side-looking airborne radar); near-range detection sensors: IR/UV (infrared ultraviolet) scanner, VIS (visual line) scanner, EO/IR (elecro-optical infrared), MWR (microwave radiometer), LFS (laser fluorosensor); SAR support—direction finder, AIS; mission control system for real-time data processing and analysis; and communication systems—broadband radio, satellite, satellite phone, VHF.

Far Range Detection

SLAR is an X-band microwave radar. When flying at an aircraft altitude between 1,000 and 6,000 ft. (300 – 1,800 m), SLAR has a cross-track swath between 35 and 50 naut. mi. (63 – 90 km) Oil spill detection by radar is based on the principle that oil spills and biogenic slicks or specific hydrodynamic effects may result in a reduction of the radar backscatter signal due to the dampening of surface wave structures [14]. SLAR is an all-weather sensor: Microwaves can penetrate clouds, fog, drizzle and rain. SLAR allows POSEIDON to cover more than 7,500 square nautical miles per hour (25,700 square km per hour), or the sensitive area of the Gulf of Mexico in 12 hr. of mission time.

Fig.4 -

SLAR output

Fig.4 -

SLAR output

Near-range analysis

IR/UV imaging devices are standard instruments for near-range monitoring of oil spills and have shown their reliability, stability and operability over decades. At the near-range operational altitude, the swath width is between 2,000 and 3,000 ft (600 – 900 m). Films of crude oil in water can be detected in thermal IR because oil has a lower integral emissivity than the surrounding water in the spectral region, and oil can be heated up through absorption of sunlight if there is enough thickness. Minimum detectable oil thickness in the thermal IR starts from 2 μm [15]; hot spots that are generated through absorption of sunlight appear in the thickness range up to 500 μm [14]. UV remote sensing of oil is based on the fact that the air/oil interface of the oil film has about twice the integral near UV reflectance of the water [14]. The UV can detect very thin sheens of oil; the minimum detectable thickness amounts to 0.01 μm [16]. The combined IR/UV device shows the area of large and intermediate film thickness, as well as the total extent of the oil spill.

VIS is used for acquisition of georeferenced RGB (red green blue) images. The RGB line scanner combines mapping and georeferencing accuracy with the acquisition of color information that allows thickness estimations based on oil appearance codes.

The MWR is a passive microwave remote sensor used to map oil layers exceeding a thickness of 50 μm [14, 17]. Oil spills appear as a brighter object in the microwave region relative to the oil-free surface. The MWR allows determination of the absolute thickness of the oil spill and, consequently, the volume of the spill. In addition, the difference of sensitivity between MWR and IR/UV allows determination of the hotspots of the spill where the bulk of the oil is concentrated. This feature is greatly important for the responders. MWR is an all-weather sensor. At the near-range operational altitude, the swath width is about 2,000 ft (600 m).

Fig.5 –

Near Range Analysis Sensors Output

Fig.5 –

Near Range Analysis Sensors Output

The LFS is based on a high-power UV laser that sends short pulses toward the water surface. The laser-induced fluorescence and backscatter are received by a telescope and separate spectrally into a number of monochromatic signals. The detected discrete emission spectrum is used to estimate the oil class and classify the oil [14]. LFS has also capability to measure absolute thickness in the 1 to 20 μm range and therefore the volume [18, 19, 20, 21]. The system has a pre-existing library of substances used to compare the actual detection.

EO/IR is an airborne observation system with HD sensors and HD video outputs. It is equipped with RGB and infrared thermal imaging cameras, multispectral zoom laser range finders, illuminators and pointers. The device delivers color and thermal images that allow situational awareness in all conditions day and night in critical SAR operations, navigation in challenging conditions, and clear images of the scene.

Data Processing

All the sensor data are collected through a fiber optic network to the Mission Control Unit. The mission control system computer provides sensor management, online visualization, real-time sensor fusion, data analysis and data storage in a central multimission unit. All operations are facilitated by an ergonomic graphical user interface (GUI). In contrast to traditional systems, which mainly perform visualization and storage of nongeoreferenced sensor images, the system allows real-time data fusion, georeferencing and analysis of multisensor oil spill data.

Fig.6 –

POSEIDON Mission Control Unit for real time data acquisition, analysis, communication.

Fig.6 –

POSEIDON Mission Control Unit for real time data acquisition, analysis, communication.

Communication

Information and data transfer is centered around a microwave communication system installed on the aircraft – the Maritime Broadband Radio (MBR). MBR is a so called ‘smart antenna’; smart antennas are antenna arrays with signal processing used to identify spatial signal signatures such as the direction of arrival of the signal, and use it to calculate beam-forming vectors, to track and locate the antenna beam on the target; this allow to increase the range of the signal and at the same time to improve substantially bandwidth utilization [22]. The system allows to send up to 15 Mbit of reliable data at ranges of 50–80 NM (90 – 145 km) depending with the flight altitude.

Concept of operations is based on several antenna nodes that can form a wireless network between aircraft, vessels and ground to support response operations. The POSEIDON communication system is also equipped with an Iridium Satellite voice/data system.

Data Distribution

Information and data from POSEIDON are sent as GIS (Geographic Information System) usable formats like shapefiles (lines, polygons, points) and GeoTiff (images) to be immediately utilized on the Common Operating Picture platforms. Data received from POSEIDON are made continuously available in a WebMap server.

Fig.7 –

Data Distribution in GIS platform.

Fig.7 –

Data Distribution in GIS platform.

OPERATION METHODOLOGY

The operational approach adopted during oil spill detection includes three basic steps: Step-1: Synoptic overview of the spill, Step-2: spill approach and Near Range Analysis, Step-3: Data Processing and Data Communication.

Step-1: POSEIDON approaches the Spill Area at an altitude of 3,000ft (900 m), scanning the area with the SLAR. At that altitude the Swath is about 50NM (80 km) and the total area scanned is about 7,500 square nautical miles (25,700 square km per hour). The EO/IR sensor is also a good support in this phase giving the operator a night and day vision of the scene. During every operation POSEIDON monitors marine and aviation distress frequencies with the Direction Finder, and ship traffic with the AIS.

After the identification, the target is analyzed and georeferenced, to allow for an immediate localization on the navigation map and successive detailed analysis with the near range sensors. In this stage a polygon is designed around the target and information such as area, coverage, overall dimensions and coordinates of the spill are determined. As the aircraft approaches the spill for the overflights with the near-range sensors, the operator records the overall spill with the EO/IR to have a visual, or thermal at night, overview of the spill.

Step-2: Based on the far-range overall images of the spill, the operator determines the best approach to be used for the subsequent low level overflights in order to make the best detailed analysis of the spill. When the flight pattern is determined, the aircraft overflies the spill at an altitude of 1,000ft (300 m) operating all the near range sensor that at that altitude have a swath of about 2,000ft (600 m). Near range sensors include: IR, UV, VIS, MWR, LFS. EO/IR and Direction Finder at this stage are still operational giving the operator support where necessary. The operator sets targets or polygons as significant areas of the spill appear.

Step-3: The operator performs the analysis of the targets and significant areas of the spill to prepare the package of information that shall be sent to ground for the response. Analysis includes preparation of polygons of the entire spill of patches with metadata including position, area, thickness, composition, volume; georeferenced images of the spill including polygons, targets or sensor images; fusion of information from different sensors to improve the capability to understand the scenario, images and videos where necessary. Data are then distributed through the MBR or the satellite link.

BENEFITS

The benefits of airborne remote sensing during an oil spill response are numerous. The technology allows responders to have a timely and clear picture of where the oil is, how much oil there is, where it is heading and where the hotspots are.

Aerial observation is the only means of obtaining a clear, realistic picture. It is the first link in a chain of decisions for incident management. Real-time, processed, quantitative information allows responders to manage the response based on facts, support the common operating picture, and improve safety while significantly reducing costs through an efficient utilization of resources.

Also, multisensor platforms with complementary/redundant sensors and data processing capabilities facilitate natural resource damage assessments, allowing to quantify with high precision the features of the spill, including the overall extent of the thinnest sheens, the pollutant composition and the total volume. High definition imagery and thermal imagery support also identification of natural resources at risk, such as marine mammals.

During the preparedness phase too, aerial remote sensing plays an essential role. Proactive surveillance programs provide response staff with familiarity of the capabilities and limitations of the methods employed. Exercises and drills involving airborne platforms help to insure readiness and increase confidence that the responder community is prepared to respond immediately and effectively in the event of a spill.

Periodic and planned surveillance flights can allow assessment and quantification of oil spills in the early stages, resulting in substantial savings in the overall cost of the recovery. Periodic and planned surveillance flights are also a strong deterrent for illegal discharge, as the international experience (Europe, Canada) has shown during recent years.

Aerial remote sensing allows independent monitoring of coastal water or infrastructure discharges, giving confidence and peace of mind to the public, media and governments. and managing of potential significant risks for human health and the environment

DEPLOYMENT

POSEIDON’s first mission has been a recent test campaign funded by the Bureau of Safety and Environmental Enforcement’s (BSEE) and organized by BSEE and the National Oceanographic and Atmospheric Administration (NOAA) in an effort to enhance the ability to detect and measure oil spills offshore. Researchers have the possibility to compare multiple remote sensing systems to determine how well each detects oil/water emulsion mixtures and measures oil thickness. Other platforms for the test include remote sensors mounted on unmanned aerial vehicles, helicopters and satellites. The tests have been performed in controlled environment at the BSEE Ohmsett facility and offshore at the Mississippi Canyon Block 20 (MC-20) in the Gulf of Mexico.

CONCLUSIONS

Today airborne remote sensing is becoming a foundation of the overall strategy to improve the ability to plan and position response resources in the optimal areas to respond to spills, it is also an effective and efficient support technique for natural resource damage assessment (NRDA) following the spill, and the preferred method for periodic surveillance of coastal areas against illegal or accidental discharge. The most important features that such platforms should have are: multi-sensors system for complementarity/redundancy of information; capability to classify oil targets as Recoverable or Non-recoverable; capability to georeference the targets and track moving oil; real time information - for tactical and strategic use; data suitable for the Common Operating Picture; ability to expand the operating window to low-light conditions. In July 2016 POSEIDON was introduced into service to answer to the needs of the responders community. The platform integrates multiple airborne remote sensors and mission system components into one network-based data acquisition, real-time processing and communication platform. The philosophy behind POSEIDON is to provide intelligence on the scene, which means understanding the environment, deciding on appropriate tactical-strategic actions and controlling the results. Real-time, processed, quantitative information allows responders to manage the scenario based on facts, support the common operating picture, and improve safety while facilitating the efficient utilization of resources.

REFERENCES

REFERENCES
[1]
Partington
,
K.
,
An assessment of Surface Surveillance Capabilities for Oil Spill Response using Airborne Remote Sensing
-
PIL-4000-38-TR-1.1 Provided for IPIECA and OGP, 2015
.
[2]
IMO, IPIECA and OGP
,
Aerial observation of oil spills at sea
,
2015
.
[3]
Bonn Agreement, Aerial Operations Handbook
,
2009
.
[4]
Transport Canada, National Aerial Surveillance Program (NASP)
,
2017
.
[5]
CleanSeasNet Service
,
European Maritime Safety Agency, EMSA
2017
.
[6]
ICCOPR - Oil Pollution Research and Technology Plan
,
2015
.
[7]
API - Technical Report 1144
,
2013
.
[8]
IPIECA – IOGP - Oil Spill Response Surveillance, Modelling & Visualization Joint Industry Program Work Package 2
,
2015
.
[9]
Coastal Response Research Center – Oil Observing Tools Workshop Report
,
2015
.
[10]
Leifer
,
I.
et al
.,
State of the art satellite and airborne marine oil spill remote sensing: Application to the BP Deepwater Horizon oil spill
,
2012
.
[11]
OSPR/Chevron
,
Oil Spill Response Technology Workshop Reports
,
2015
.
[12]
Brown
,
C.
,
Environment Canada Oil Spill Remote Sensing, NOAA-Environment Canada Workshop
,
2012
.
[13]
The Pacific States/British Columbia, Oil Spill Task Force, 2014–2019 Strategic Plan
,
2014
.
[14]
Robbe
,
N.
,
Hengstermann
,
T.
Remote sensing of marine oil spills from airborne platforms using multi-sensor systems, OPTIMARE Sensorsysteme AG
,
2006
.
[15]
Robbe
,
N.
,
Airborne Oil Spill Remote Sensing: Modelling, Analysis and Fusion of Multispectral Data
.
Ph.D. thesis, Institute of Experimental Physics, University of Hamburg
,
2005
.
[16]
Fingas
,
M.F.
&
Brown
,
C.E.
,
Review of Oil Spill Remote Sensing
.
Spill Science & Technology Bulletin
,
4
(
4
),
pp
.
199
208
,
1997
.
[17]
Ulaby
,
F.T.
,
Moore
,
R.K.
&
Fung
,
A.K.
,
Microwave Remote Sensing: Active and Passive, volume III
.
Artech House
,
1986
.
[18]
Fantasia
,
J.F.
,
Hard
,
T.M.
&
Ingrao
,
H.C.
,
An investigation of oil fluorescence as a technique for the remote sensing of oil spills
.
Technical report, DOT/Transportation Systems Center U.S. Coast Guard, Report TSCUSCG- 71-7
,
1971
.
[19]
Hengstermann
,
T.
&
Reuter
,
R.
,
Lidar fluorosensing of mineral oil spills on the sea surface
.
Applied Optics
,
29
(
22
),
pp
.
3218
3227
,
1990
.
[20]
Hengstermann
,
T.
,
Untersuchungen zur Laserfernerkundung mariner Ölverschmutzungen
.
Ph.D. thesis, Department of Physics, University of Oldenburg
,
1992
.
[21]
Hoge
,
F.E.
&
Swift
,
R.N.
,
Oil film thickness measurement using airborne laser-induced water raman backscatter
.
Applied Optics
,
19
(
19
),
pp
.
3269
3281
,
1980
.
[22]
Kongsberg – Maritime Broadband Radio Product Sheet
,
2016
.