No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
NOAA CoastWatch SAR applications and demonstration
Abstract
The goal of the National Oceanic and Atmospheric Administration (NOAA) CoastWatch Program is to provide satellite and other environmental data and products for near-real-time monitoring of U.S. coastal waters in support of environmental science, management, and hazard response. During the last few years, products available through CoastWatch have expanded beyond the original infrared, visible, and sea surface temperature images to include ocean color, scatterometer wind, and synthetic aperture radar (SAR) images. A NOAA research and development program with partners in government, academia, and industry has endeavored to develop coastal ocean SAR applications for CoastWatch. Some of these applications, in particular wind measurement and hard target (i.e., vessel) detection, were developed for a preoperational demonstration. Users include the NOAA National Weather Service, the Alaska Department of Fish and Game, and the U.S. Coast Guard.
... Use of remote sensing data to construct a climatology of wind events offshore of Central America has previously been made over a 9-month period using scatterometer data by Chelton et al. (2000). For this study SAR wind speed data from the Canadian Radarsat-1 satellite (Radarsat) was used (Pichel and Clemente-Colon 2000). These satellite data were collected in " wide-swath mode, " yielding a 450-km- wide image with 100–200-m resolution (Pichel and Clemente-Colon 2000). ...
... For this study SAR wind speed data from the Canadian Radarsat-1 satellite (Radarsat) was used (Pichel and Clemente-Colon 2000). These satellite data were collected in " wide-swath mode, " yielding a 450-km- wide image with 100–200-m resolution (Pichel and Clemente-Colon 2000). The higher resolution of SAR allows one to work closer to the coast than with scatterometers (Thompson et al. 2001), a key advantage when studying barrier jets. ...
This paper investigates the temporal and spatial climatology of coastal barrier jets in the Gulf of Alaska. The jets are divided into two categories based upon the origin of the air involved: “classic” barrier jets fed primarily by onshore flow and “hybrid” jets fed primarily by gap flow from the continental interior. The analyses are compiled from five years (1998–2003) of synthetic aperture radar images from the Radarsat-1 satellite totaling 3000 images. Thermodynamic and kinematic data from the NCEP reanalysis is used in the interpretation of the results.
The majority of coastal barrier jets occur during the cool season, with the coastline near Mount Fairweather and the Valdez–Cordova mountains experiencing the greatest number of barrier jets. Hybrid jets are even more strongly restricted to the cool season and are commonly found to the west of Cross Sound, Yakutat Bay, and Icy Bay. Some interannual variability in the total number of jets is observed.
Coastal barrier jet formation is associated with onshore wind directions and maximum terrain heights exceeding 2 km within 100 km of the coast, features that support low-level flow blocking. Hybrid jet formation requires the additional condition of an abnormally large offshore-directed pressure gradient force.
Half of the barrier and hybrid jets exhibit surface wind speeds in excess of 20 m s−1 (strong gale), although their widths are typically less than 100 km. The maximum speed of both types of jet tends to be 2–3 times that of the ambient synoptic flow. A small percentage of the jets detach from the coastline, with the typical detachment distance being 10 km.
... Use of remote sensing data to construct a climatology of wind events offshore of Central America has previously been made over a 9-month period using scatterometer data by Chelton et al. (2000). For this study SAR wind speed data from the Canadian Radarsat-1 satellite (Radarsat) was used (Pichel and Clemente-Colon 2000). These satellite data were collected in " wide-swath mode, " yielding a 450-km- wide image with 100–200-m resolution (Pichel and Clemente-Colon 2000). ...
... For this study SAR wind speed data from the Canadian Radarsat-1 satellite (Radarsat) was used (Pichel and Clemente-Colon 2000). These satellite data were collected in " wide-swath mode, " yielding a 450-km- wide image with 100–200-m resolution (Pichel and Clemente-Colon 2000). The higher resolution of SAR allows one to work closer to the coast than with scatterometers (Thompson et al. 2001), a key advantage when studying barrier jets. ...
Rain within the footprint of the SeaWinds scatterometer on the QuikSCAT satellite causes more significant errors than existed with its predecessor, the NASA scatterometer (NSCAT) on Advanced Earth Observing Satellite-I (ADEOS-I). Empirical relations are developed that show how the rain-induced errors in the scatterometer wind magnitude depend on both the rain rate and on the wind magnitude. These relations are developed with collocated National Data Bouy Center (NDBC) buoy measurements (to provide accurate sea surface winds) and simultaneous Next Generation Weather Radar (NEXRAD) observations of rain reflectivity. An analysis, based on electromagnetic scattering theory, interprets the dependence of the scatterometer wind errors on volumetric rain rate over a range of wind and rain conditions. These results demonstrate that the satellite scatterometer responds to rain in a manner similar to that of meteorological radars, with a Z-R relationship. These observations and results indicate that the combined (wind and rain) normalized radar cross section will lead to erroneously large wind estimates when the rain-related radar cross section exceeds a particular level that depends on the rain rate and surface wind speed.
... The SeaWinds scatterometer has a beamwidth-limited resolution of approximately 25 km, which is too coarse for many geophysical applications. An order-of-magnitude improvement in resolution is required to observe many mesoscale wind features associated with storms and in coastal waters [9], [23]. For land and ice applications, spatial resolution comparable with visible/IR imaging radiometers-such as the 1-5-km resolution Advanced Very High Resolution Radiometer (AVHRR)-is desired [3], [26], [31]. ...
... Quantitatively this interleaving constraint can be expressed as PRI PRI (22) where is an integer representing the number of pulses in flight; is the pulse length; is the round-trip delay between the inner and outer edges of the usable footprint [see (6)]; and is timing margin allowed for uncertainties in the precise value of . A corollary to (22) is a limit on the transmit pulse length that is given by PRI (23) For the wide-swath and high-incidence angles typically used for pencil-beam scatterometers, however, the constraints of (8) and (22) may be difficult to meet. This is particularly true if significant timing margins are required in order to allow for spacecraft attitude variations and for land surface topography. ...
Conically scanning pencil-beam scatterometer systems, such as the SeaWinds radar, constitute an important class of instruments for spaceborne climate observation. In addition to ocean winds, scatterometer data are being applied to a wide range of land and cryospheric applications. A key issue for future scatterometer missions is improved spatial resolution. Pencil-beam scatterometers to date have been real-aperture systems where only range discrimination is used, resulting in a relatively coarse resolution of approximately 25 km. In this paper, the addition of Doppler discrimination techniques is proposed to meet the need for higher resolution. The unique issues associated with the simultaneous application of range and Doppler processing to a conically scanning radar are addressed, and expressions for the theoretical measurement performance of such a system are derived. Important differences with side-looking imaging radars, which also may employ Doppler techniques, are highlighted. Conceptual design examples based on scatterometer missions of current interest are provided to illustrate this new high-resolution scatterometer approach. It is shown that spatial resolution of pencil-beam scatterometer systems can be improved by an order of magnitude by utilizing combined range/Doppler discrimination techniques, while maintaining the wide-swath and constant incidence angle needed for many geophysical measurements.
... Synthetic aperture radar (SAR) remote sensing of the ocean is frequently used to identify, map, and monitor winds, currents, shallow water bathymetry and internal waves, ships and icebergs, algal blooms and aquatic vegetation, sea ice, and marine pollution (e.g., Pichel and Clemente-Colón (2000), Monaldo et al. (2004), Kerbaol and Collard (2005), Brekke and Solberg (2005), Gens (2008), Caruso et al. (2013), and references therein). Many of these applications rely in whole or in part on identifying the scattering surface of interest based upon contrast with open seawater. ...
In remote sensing of the ocean, contrast in the measured intensity between clean water and other features is used to identify different objects on the ocean surface either directly or indirectly via alteration of the ocean wave spectrum. The damping ratio, a measure of contrast, is increasingly used for operational oil spill monitoring as an aid or alternative to visual inspections by trained personnel, and can in some cases identify thicker oil in a slick. A method is proposed for automatically calculating the contrast based upon the statistical properties of the measured intensity signals from the ocean surface, and shown to work well even for complex slick geometries. The algorithm is demonstrated using synthetic aperture radar (SAR) data from UAVSAR and Sentinel-1 to show that it can handle multi-frequency and medium-to-high resolution data. The algorithm's flexibility and computational simplicity makes it suitable for real-time processing to support oil spill response.
... In the last two decades, efforts have been made by several organizations towards the development of semi-automated or fully automated systems for oil spill detection based on SAR imagery. Examples include the semi-automated systems such as Ocean Monitoring Workstation (OMW) at CIS [2], Alaska SAR Demonstration (AKDEMO) system [12], the European Commission Joint Research Centre (JRC) system [13], the Norwegian Defense Research Establishment (NDRE) system [14], and a fully-automated Kongsberg Satellite Services (KSAT)s oil spill detection system at Norway [15]. ...
The detection of marine oil spill candidate from synthetic aperture radar (SAR) images is largely hampered by SAR speckle noise and the complex marine environment. In this paper, we develop a thresholding-guided stochastic fully-connected conditional random field (TGSFCRF) model for inferring the binary label from SAR imagery. First, an intensity thresholding approach is used to estimate the initial labels of oil spill candidates and the background. Second, a Gaussian mixture model (GMM) is trained using all the pixels based on the initial labels. Last, based on the G-MM model, a graph-cut optimization approach is used for inferring the final labels. By using a threholding-guided approach, TGSFCRF can exploit the statistical characteristics of the two classes for better label inference. Moreover , by using a stochastic clique approach, TGSFCRF efficiently addresses the global-scale spatial correlation effect, and thereby can better resist the influence of SAR speckle noise and background heterogeneity. Experimental results on RADARSAT-1 ScanSAR imagery demonstrate that TGS-FCRF can accurately delineate oil spill candidates without committing too much false alarms.
... The goal of the project was to verify the applicability of satellite wind maps derived from passive microwave, altimeter,scatterometer and imaging SAR technologies for wind energy tools for wind resource 28 ANSWRS was developed jointly by The Johns Hopkins University Applied Physics Laboratory (APL) and NOAA. ANSWRS uses the CMOD-5 GMF(Pichel et al 2000) and was recently upgraded to ANSWRS2.0, which can process most SAR databases.mapping. The SAT-WIND project examined wind products from EnviSAT/ASAR, retrieved using three different GMFs and a box averaging method, and compared them to hourly average Horns Rev met mast data for wind speeds between 2-15 m/s(Christiansen et al 2006 b). ...
Efficient development of offshore wind power will require accurate information about the wind field on a wide range of spatial and temporal scales. Lidar and Satellite Microwave Radar/Radiometry have been evaluated and used for wind speed measurement. Numerous studies have been published that examine the error, bias, and performance characteristics of variants of both technologies under a range of conditions. This paper reviews recent research and technological advances and outlines strategies for applying the technologies to reduce costs, increase energy production and improve energy forecasting through advanced rotor controls and more accurate resource estimation and mapping.
A literature search was conducted to identify the most recent and relevant correlation and validation studies of Lidar, synthetic aperture radar, scatterometers, and radiometers used for estimating wind speed. Database queries were conducted to estimate inventory for satellite wind data. Estimates of the accuracy (bias and uncertainty) and availability (sample density) of these technologies were developed based on the literature search and database queries. Both “snapshot” wind speed and energy density estimates were compared for satellite microwave systems and Lidar technologies. Offshore, where turbulence is lower, Lidar is found to have very high accuracy and availability, comparable to cup anemometers at a range of up to 200m on fixed platforms. Floating Lidar is rapidly approaching the same level of accuracy and availability, and is easily re-positioned. However, the short time series of Lidar is less useful for long term indexing, and it is limited to a single site per sensor. Satellite microwave wind retrievals are available over a 20 year period and are found to have good time-averaged accuracy at 10 meters above sea level for wind speeds between 3 and 15 m/s, but are subject to minor bias (below +/- 0.2 m/s) from the use of inaccurate shear profiles, from diurnal effects, and from local metocean conditions. Three strategies for use of these technologies are outlined and evaluated.
• Siting and Resource Assessment - By processing all available satellite microwave data sets, calibrated with data from a one year field campaign using floating Lidar systems, cross-correlated through a parametric geophysical model function, bias and error of wind speeds generated from the satellite data can be reduced, and wind mapping can be significantly improved in resolution and accuracy.
• Energy Production Estimates- By using wind profile data from floating Lidars, deployed on site, and indexed to a 20 year time series from calibrated satellite wind data, Annual Energy Production estimates can be greatly improved by reducing uncertainty (and thus, the risk premium on financing). In the near future, this methodology can obviate the need for a met tower for resource assessment.
• Rotor Control - By using nacelle or hub mounted Lidar to look upstream, new Lidar-assisted control systems can adjust blade pitch and nacelle yaw pro-actively to match rapid changes in wind speed or direction. This can reduce fatigue and extreme gust loading on components, allowing longer blades and greater swept area. It can also improve efficiency be reducing yaw mis-alignment.
In addition to power production benefits, rough, first-order costs were developed to check economic justification, and the expected change in Breakeven Price was calculat-ed for two different build-out scenarios of the study area. The analysis indicates that the recommended strategies for improving Rotor Control and reducing uncertainty of AEP estimates can reduce the Breakeven Price of power for the base case wind farm by at least 4%.
The benefits of improved mapping are more difficult to monetize due to high levels of uncertainty in all the primary factors, so two different scenarios are considered. If the benefits of improved mapping and better siting (2% to 3% lower BP) are available to the first ten or twelve projects, and the mapping effort is federally funded, or the costs are somehow distributed industry-wide over full build-out of the study area, the Break-even Price for the first phase of wind farms could be reduced by a total of around 6% to 7%. If mapping benefits are assumed to diminish over time as the study area builds out, the long term, annualized reduction in Breakeven Price over the entire study area will be lower, at around 4% to 5%. In either case the mapping effort is justified, and the cost can be reduced by about $120 million using Lidar equipped met buoys.
Keywords: offshore wind power, wind resource assessment, satellite radar, Lidar, floating Lidar, AEP estimates, wind turbine pitch control, wind turbine yaw control, wind energy mapping
... In 1997, The National Oceanic and Atmospheric Administration (NOAA) National Environmental Satellite, Data, and Information Services (NESDIS) started the Alaska SAR Demonstration (AKDEMO) project [20]. For the AKDEMO, we acquire SAR data from the Canadian RADARSAT-1 satellite in near real time to demonstrate the use of SAR imagery, as well as SAR-derived products (sea-surface wind, vessel positions, etc.), in a preoperational environment. ...
Cook Inlet, Alaska is an extremely dynamic system, with tidal range up to 9 m, maximum current up to 12 knots, and an average maximum surface current of 3 knots. The water depth is mostly between 20 and 40 m in the upper inlet. Since Cook Inlet is elongated, the strong current system tends to generate axial convergence fronts. Synthetic Aperture Radar images from satellite RADARSAT-1 have shown frontal lines along the axial direction of Cook Inlet. These frontal lines can be more than 100 km in length. It also appears that the fronts vary in position from time to time. We hypothesize that the convergence fronts are a result of strong tidal current under the influence of strong bottom friction. The cross channel variation in bottom friction caused by significant depth variations tends to generate differences in both magnitude and phase of tidal flow. These cross channel differences in flow field can generate convergence along the channel. This hypothesis is tested with some preliminary studies using a diagnostic 2-D tidal model using realistic bathymetry. The model is driven by tidal data along Cook Inlet from NOAA stations. The model results have shown that the strong tidal current does produce significant convergence across the channel with significant depth variations. The maximum convergence calculated from the model is consistent with that from SAR images.
... The Gulf Stream can clearly be seen in SST images as a narrow (∼100 km width), high SST meandering oceanic jet. [5] Since 1997, NOAA has sponsored a multi‐year project to determine and demonstrate SAR‐derived quantitative and qualitative products in a pre‐operational environment for US coastal waters [Pichel and Clemente‐Colón, 2000]. ...
Deep-water (>500 m) oceanic bathymetric features are frequently observed in RADARSAT-1 SAR images in the Gulf Stream (GS) region. They are imaged apparently because of the unique environmental conditions in the region, oceanographically characterized by a strong GS current (2 ms−1) and favorable ocean stratification. SAR image analysis shows the basic characteristics of these bathymetric features. A coincident sea surface temperature image shows that the bathymetric feature is only “visible” by SAR within the GS pathway. The dominant wavelength of the wave-like feature is about 2.3 km and their crests are perpendicular to the GS axis. Shipboard sounding measurements confirm the SAR observation. A theoretical consideration of the ocean current and corrugated bathymetry interaction in a 3-layer ocean is presented. Using representative ocean density profile data and the GS current data, we analyze the requirements for the generation and upward propagation of the disturbance induced by the current-bathymetry interaction.
... For this study SAR wind speed data from the Canadian Radarsat-1 satellite (Radarsat) was used. The satellite data used were collected in wide-swath mode yielding 450 km wide spatial coverage and 100-200 m resolution (Pichel and Clemente-Colon 2000). ...
... As part of the NOAA/NESDIS Alaska SAR Demonstration Project (Pichel and Clemente-Colon, 2000), a multi-year demonstration of the production and use of RADARSAT SAR HH polarization imagery to generate products in a pre-operational environment, a wind product is created that automatically generates wind vectors over the coastal ocean. Two methods are used. ...
... The AKDEMO, sponsored by the U.S. National Environmental Satellite, Data, and Information Service (NESDIS) of the National Oceanic and Atmospheric Administration (NOAA) is a multi-year demonstration of the pre-operational production and use of SAR quantitative and qualitative products [1]. The AKDEMO region of interest includes the Gulf of Alaska, the Bering Sea, and portions of the Beaufort and Chukchi Seas along the Alaska coast (see Fig. 1). ...
The U.S. National Oceanic and Atmospheric Administration (NOAA) National Environmental Satellite, Data, and Information Service (NESDIS) ENVISAT project is focused on a pre-operational demonstration of wind and vessel position products using, predominately, the Wide-Swath Mode of the ENVISAT Advanced Synthetic Aperture Radar (ASAR). The necessary scientific algorithms, data management techniques, and product production and dissemination procedures are being prototyped using Canadian RADARSAT-1 SAR data. A near real-time demonstration of SAR product production, the Alaska SAR Demonstration (AKDEMO) has been underway since October 1999 for the waters surrounding Alaska. Wind speed, wind vector (with 180 degree ambiguity) and vessel position products are generated within about 6 hours of satellite acquisition and provided to operational agencies for evaluation and validation. Wind validation is accomplished by comparing SAR-derived winds with model output in Alaska and with buoy measurements from the NOAA moored meteorological buoys in the Atlantic off the U.S. East Coast. For validation of vessel positions, fishery observer reports are being paired with SAR-derived positions to ascertain vessel detection success. ENVISAT data will first be taken over the U.S. East Coast buoys to test and validate the wind algorithm. The vessel detection algorithm will be tailored for the ENVISAT ASAR imagery and tested as well. Once the algorithms are operating properly, ENVISAT data will then be taken over Alaska (in near real-time if possible) and made available to AKDEMO users. If RADARSAT or Japanese Advanced Land Observation Satellite (ALOS) SAR data are available as well, a two-satellite demonstration will be attempted.
... In 1997, The National Oceanic and Atmospheric Administration (NOAA) National Environmental Satellite, Data, and Information Services (NESDIS) started the Alaska SAR Demonstration (AKDEMO) project [20]. For the AKDEMO, we acquire SAR data from the Canadian RADARSAT-1 satellite in near real time to demonstrate the use of SAR imagery, as well as SAR-derived products (sea-surface wind, vessel positions, etc.), in a preoperational environment. ...
Axial fronts of tidal currents are observed in Cook Inlet, AK, on a RADARSAT-1 standard mode synthetic aperture radar (SAR) image taken at 16:31:47 coordinated universal time (UTC) on July 12, 2002. The longest front appears as a 100-km-long quasi-linear bright feature in the SAR image. This front is characterized by an increase in the normalized radar cross sec-tion (NRCS) of 7 dB in the C-band horizontal polarization (C-HH) RADARSAT-1 SAR image. Two other smaller fronts exist in the middle of the inlet. The NRCS modulations appear to be less, at about 5 dB. A diagnostic Cook Inlet tidal model is developed to calculate the current velocity fields of the inlet and to demonstrate that the variation in bottom friction caused by the bathymetry dis-tribution generates axial convergence at different tidal stages. The model, using the actual bathymetry, is driven by predicted tides from six tidal stations along the inlet coast. The model results show that the tidal current flowed into the inlet at the time the SAR image was obtained. Tidal current along two transects in the inlet is extracted to show that there is a significant cross-channel conver-gence of the along-channel velocity component, with a magnitude of 4 to 6 10 4 s 1 near the observed front positions. In general, a higher velocity convergence from the model corresponds to higher NRCS return areas in this SAR image.
... The resolution is high enough to resolve very fine scale wind variations associated with marine atmospheric boundary layer phenomena, including katabatic winds [11], roll vortices [12], atmospheric gravity waves [13], [14], and vortex streets [15], [16], among others. Since 1999, NOAA has conducted a demonstration of the production and the use of SAR quantitative and qualitative products in a preoperational environment [17]. Subkilometer high-resolution SAR wind products have been generated for U.S. coastal waters, primarily from the Canadian RADARSAT-1 SAR (e.g., Fig. 1). ...
In this letter, we generate a temporal/spatial matchup data set between QuikSCAT scatterometer and RADARSAT-1 synthetic aperture radar (SAR) wind products in offshore waters along the U.S. West Coast. Analysis of the resulting three-year database shows that, in general, the wind products from both sensors have characteristics similar to those reported in the literature. Then, we perform an error analysis in the space domain and find that there is significant discrepancy between the two wind products as the matchup points move closer to the coast. The root-mean-square error (rmse) and standard deviation (STD) between the two data sets increases markedly for points matched within about 100 km of the coastline. Beyond 100 km, the rmse, STD, and systematic bias become small and stable. In addition, an empirical relationship between QuikSCAT and SAR winds in coastal region is proposed. Thus, the bias and errors should be taken into account if the standard operational QuikSCAT wind products are used for forcing models in the coastal ocean.
... However, under actual conditions, the NRCS is also affected by wave current interactions, surface oil slicks, rains and air-sea boundary layer stabilities etc. By far those impact factors are not all taken into account in SAR wind retrieval [2]. ...
High spatial resolutions synthetic aperture radar (SAR) retrieved ocean surface wind field under actual conditions can be affected by several ocean surface features, such as artificial object, surface oil slicks and air-sea boundary layer stabilities etc. In this paper we present several case studies of those impacts, and a semi-empirical model to correct the air-sea boundary layer stabilities. We demonstrate that the new model helps to improve the wind retrieval accuracy in the Gulf Stream north wall areas. Index Terms—SAR, wind, SST
... Therefore, to derive the wind velocity, one must obtain the wind direction independently from another source, i.e., from operational weather model output [7], from finding the wind-aligned patterns in the SAR data itself [8], from scatterometer data, or from buoy measurements [9]. Since 1999, the National Oceanic and Atmospheric Administration has conducted a demonstration of the production and the use of SAR quantitative and qualitative products in a preoperational environment [10]. Subkilometer high-resolution SAR wind products have been generated for U.S. coastal waters. ...
In this paper, we perform a comparison of wind speed measurements from the ENVISAT Advanced Synthetic Aperture Radar (ASAR), the MetOp-A Advanced Scatterometer (ASCAT), the U.S. National Data Buoy Center's moored buoys, and the U.S. Navy Operational Global Atmospheric Prediction System (NOGAPS) model. These comparisons were made in near U.S. coast regions over a 17-month period from March 2009 to July 2010. The ASAR wind speed retrieval agreed well with the scatterometer and model estimates, with mean differences ranging from −0.69 to 0.85 m/s and standard deviations be- tween 1.16 and 1.77 m/s, depending upon the ASAR beam mode type. The results indicate that ASAR-derived ocean surface wind speeds are as accurate as the ASCAT and NOGAPS wind prod- ucts. Comparisons between ASCAT winds and synthetic aper- ture radar (SAR) winds averaged at different spatial resolutions show very little change. This demonstrates that it is suitable that the scatterometer wind retrieval geophysical model function, i.e., CMOD5, is used for SAR wind retrieval. The impact of C-band VV polarization SAR calibration error on wind retrieval is also discussed.
... In October 1999, NOAA/NESDIS began a near real-time applications demonstration (called the Alaska SAR Demonstration, or AKDEMO) of SAR-derived environmental products, including SAR winds, in order to provide quasioperational experience with SAR products to selected government agencies including the National Weather Service [8]. RADARSAT-1 data are quick-look processed at the Alaska Satellite Facility (ASF) at the University of Alaska Fairbanks and sent via dedicated communications to the NOAA/NESDIS Comprehensive Large Array -data Stewardship System (CLASS), formerly known as the Satellite Active Archive. ...
High-resolution winds derived from RADARSAT-1 synthetic aperture radar (SAR) images have been produced for the waters around Alaska since 1999. Wind speed images show useful details of many meteorological phenomena of interest in weather analysis and forecasting. These include wakes, gap flows, lee waves, atmospheric fronts, cyclones, and barrier jets
... The Alaska SAR Demonstration (AKDEMO) is a multiyear demonstration of the production and use of SAR quantitative and qualitative products in a pre-operational environment [1]. The area of interest covers the Gulf of Alaska, the Bering Sea, and other seas bordering Alaska in the region 42 o N to 76 o N and 155 o E to 122 o W. AKDEMO products include high resolution ocean surface winds, vessel positions, and SAR imagery. ...
A summary of the interim results of the first two years of the authors' NASA RADARSAT-1 ADRO-2 SAR project is given. The Alaska SAR Demonstration (AKDEMO) is providing winds, vessel positions, and SAR imagery to users in Alaska for evaluation as to their utility to operational government agencies responsible for ocean and weather prediction and fisheries management/enforcement. The AKDEMO applications are maturing and their accuracy has been measured. A new demonstration, the Gulf of Mexico Experiment (GoMEx), is now underway to examine use of SAR data and products in hazardous algal bloom (HAB) and oil spill/seep monitoring. Initial results show correspondence of bloom signatures in ocean color and SAR data in areas of high HAB concentration as measured from ship water samples. In addition to these two applications demonstrations, research is underway in the use of SAR data to study upwelling, river plumes, ocean current boundaries, atmospheric boundary layer processes, and other applications. This research will continue into the third and final year of the ADRO-2 project.
Monitoring marine use is essential to effective management but is extremely challenging, particularly where capacity and resources are limited. To overcome these limitations, satellite imagery has emerged as a promising tool for monitoring marine vessel activities that are difficult to observe through publicly available vessel-tracking data. However, the broader use of satellite imagery is hindered by the lack of a clear understanding of where and when it would bring novel information to existing vessel-tracking data. Here, we outline an analytical framework to (1) automatically detect marine vessels in optical satellite imagery using deep learning and (2) statistically contrast geospatial distributions of vessels with the vessel-tracking data. As a proof of concept, we applied our framework to the coastal regions of Peru, where vessels without the Automatic Information System (AIS) are prevalent. Quantifying differences in spatial information between disparate datasets—satellite imagery and vessel-tracking data—offers insight into the biases of each dataset and the potential for additional knowledge through data integration. Our study lays the foundation for understanding how satellite imagery can complement existing vessel-tracking data to improve marine oversight and due diligence.
The sea surface roughness (SSR) research using synthetic aperture radar (SAR) at the national environmental satellite, data and information service (NESDIS) and national oceanic and atmospheric administration (NOAA) is discussed. Various plans for development of an operational SAR ocean products system are demonstrated. The SAR-derivable parameters which are included within the national polar-orbiting operational environmental satellite system (NPOESS) requirements documents are listed. It is expected that the international and domestic partnerships will led to the development of an automated operational SAR ocean products system.
APL and the Office of Research and Applications of the National Oceanic and Atmospheric Administration have developed a system to use near-real-time satellite synthetic aperture radar (SAR) data from the Radarsat-1 and Envisat satellites to produce high-resolution (subkilometer) maps of the ocean surface wind field in coastal areas. These maps have shown diverse meteorological phenomena, from gap flows to atmospheric roll vortices. In this article, we describe how SAR can measure wind over the ocean surface and then present examples illustrating how such measurements may be applied. The first application is a scientific one in which SAR wind fields are used to understand the dynamics and spatial variability of barrier jets off the west coast of Canada and the southern coast of Alaska. The second application is a practical one in which high-resolution SAR wind maps are used to determine the optimal placement of offshore wind turbines for generating electric power.
In 2013, the National Oceanic and Atmospheric Administration (NOAA) brought to operations a synthetic aperture radar (SAR)-derived subkilometer resolution wind speed product. This transition from research to operations comes 35 years after the 1978 launch of the US Seasat satellite, which demonstrated that radar backscatter from active microwave instruments in orbit can provide detailed information about ocean surface waves, winds, and sea surface height. NOAA's initial source of data for operational SAR winds is Radarsat-2, which was launched in 2007 by the Canadian Space Agency. In this paper, we discuss the history of our understanding of the relationship between microwave measurements, particularly SAR measurements, and wind speed, and how a spaceborne instrument first designed to measure ocean waves is now routinely used to derive wind speeds.
NOAA/NESDIS a initié le programme “Alaska SAR Demonstration” dont l'objectif est de faire la démonstration du potentiel des images RSO en bande C de RADARSAT-1 à fournir une information utile et en temps opportun sur l'environnement et pour la gestion des ressources pour des utilisateurs en Alaska. Un des produits développés dans le cadre du programme est une liste de localisations des navires. Cet article décrit l'algorithme développé pour générer ce produit par le biais de la détection automatique des navires basée sur des changements dans les statistiques locales. À l'aide d'images à basse résolution (100 mètres d'espacement), on démontre que l'on peut détecter des navires de dimension supérieure à 35 mètres (représentant 105 navires sur un total de 272 dans la zone test) avec un taux de fausse alerte de 0,01% pour une seule détection. Avec des images à haute résolution (50 mètres d'espacement), on peut détecter des navires d'une dimension supérieure à 32 mètres (représentant 124 navires sur 272) avec un taux de fausse alerte de 0,002% pour une seule détection. L'algorithme est entièrement automatisé et prend environ 10 minutes de temps-machine pour traiter une image ScanSAR en mode B large.
Ship surveillance is important for maritime security and safety. It plays important roles in many applications including ocean environment monitoring, search and rescue, anti-piracy and military reconnaissance. Among various sensors used for maritime surveillance, space-borne Synthetic Aperture Radar (SAR) is valued for its high resolution over wide swaths and all-weather working capabilities. However, the state-of-the-art algorithms for ship detection and identification do not always achieve a satisfactory performance. With the rapid development of space-borne Automatic Identification System (AIS), near real-time and global surveillance has become feasible. However, not all ships are equipped with or operate AIS. Space-borne SAR and AIS are considered to be complementary, and ship surveillance using an integrated combination has attracted much attention. In order to summarize the achievements and present references for further research, this paper attempts to explicitly review the developments in previous research as the basis of a brief introduction to space-borne SAR and AIS.
In this paper, we performed a comparison of wind speed from synthetic aperture radar (SAR), scatterometer, moored buoys and numerical model. These comparisons were made in near U.S. coast regions. The results indicate that SAR-derived ocean surface wind speeds are as accurate as the scatterometer and model wind products.
In 2013, the National Oceanic and Atmospheric Administration (NOAA) brought to operations a synthetic aperture radar (SAR)-derived subkilometer resolution wind speed product. This transition from research to operations comes 35 years after the 1978 launch of the US Seasat satellite, which demonstrated that radar backscatter from active microwave instruments in orbit can provide detailed information about ocean surface waves, winds, and sea surface height. NOAA's initial source of data for operational SAR winds is Radarsat-2, which was launched in 2007 by the Canadian Space Agency. In this paper, we discuss the history of our understanding of the relationship between microwave measurements, particularly SAR measurements, and wind speed, and how a spaceborne instrument first designed to measure ocean waves is now routinely used to derive wind speeds.
In the United States, satellite Synthetic Aperture Radar (SAR) imagery from the Canadian RADARSAT-1 satellite are being used routinely by the National Ice Center (NIC) in the production of sea ice analysis charts. The routine production of other SAR ocean products is also being demonstrated by the National Oceanic and Atmospheric Administration (NOAA) National Environmental Satellite, Data, and Information Service (NESDIS). Since 1998, NOAA/NESDIS has supported the development and demonstration of SAR oceanographic and hydrologic applications. In particular, automated algorithms to determine ocean surface wind speed and direction, to detect vessel location, and to map sea ice have been developed and are being evaluated by U.S. operational agencies. In addition, SAR images are being analyzed routinely for the detection of polar mesoscale cyclones, oil spills, and river ice breakup and associated ice jams and flooding in the spring. Efforts are underway to develop an international consensus on the best tools for operational use of present and future SAR data streams. The approach includes the selection and/or merging of the most suitable algorithms for efficient operational processing leading to a robust products system that can be incorporated into national environmental monitoring programs. SAR data streams under consideration include the ENVISAT Advanced SAR (ASAR) as well as the upcoming Advanced Land Observing Satellite (ALOS) Phased Array L-band SAR (PALSAR) and the RADARSAT-2 SAR.
The U.S. National Oceanic and Atmospheric Administration (NOAA) National Environmental Satellite, Data, and Information Service (NESDIS) ENVISAT project is focused on developing coastal and marine applications of ENVISAT Advanced Synthetic Aperture Radar (ASAR) data and then demonstrating these applications in an operational environment. Applications include surface winds, automated vessel detection, oil spill and ocean feature detection, and river ice monitoring. This project builds upon product development accomplished with ERS-1/2 and RADARSAT-1 data and upon a successful near-real-time applications demonstration conducted in Alaska since 1999, called the Alaska SAR Demonstration (AKDEMO). Products being developed primarily use the ASAR Wide Swath mode; however, the Alternating Polarization mode is also being tested, particularly for improvement in ocean feature discrimination. Geographic regions under study include selected regions of the Atlantic Ocean, Gulf of Mexico, the Northeast Pacific Ocean and the waters around Alaska. This paper covers primarily the applications development effort for this project.
Two cases of upstream propagation of atmospheric solitons generated by atmospheric flow over topography were identified on two RADARSAT-1 synthetic aperture radar (SAR) images acquired near St. Lawrence Island in the Bering Sea on 7 September 1997 and 6 June 2001, respectively. In both cases, a group of solitons was shown as three dark-bright linear features on the SAR images. The soliton width and peak-to-peak distance measured from the two SAR images are about 3 km and 4.5 km, respectively. Simultaneous radiosonde measurements, a surface weather map and operational weather model results confirm that these waves did not propagate on the lee side of the island as is commonly observed with island lee waves, but indeed propagated against the flow in the upstream direction. In the second case, the same group of solitons was also identified as a wave-like cloud pattern on a MODIS image taken about 4.5 hours later. In this MODIS image, the soliton train propagated further upstream with the leading crest reaching about 30 km north of the island, and the number of crest-trough features increased from three to seven. In this study, we describe the generation and evolution of upstream atmospheric solitons using the Force KdV (fKdV) model with radiosonde and island topography data. The numerical solution of the fKdV exhibits a sequence of solitons propagating upstream of the island. Both temporal and spatial scales of the solitons are in good agreement with that estimated from the successive SAR and MODIS satellite images.
Examining the use of satellite data in the retrieval of oceanic physical and biological properties, this study presents examples of the kinds of data that can be acquired and recounts their oceanographic application. It also describes the national and international programs in satellite oceanography of the past two decades, and reviews current and future programs up to 2019. The textbook, designed for graduate and senior undergraduate courses in satellite oceanography, will prepare students and interested scientists to use satellite data in oceanographic research.
The ever-changing weather and lack of in situ data in the Bering Sea warrants experimentation with new meteorological observing systems for this region. Spaceborne synthetic aperture radar (SAR) is well suited for observing the sea surface footprints of marine meteorological phenomena because its radiation is sensitive to centimeter-scale sea surface roughness, regardless of the time of day or cloud conditions. The near-surface wind field generates this sea surface roughness. Therefore, the sea surface footprints of meteorological phenomena are often revealed by SAR imagery when the main modulator of sea surface roughness is the wind. These attributes, in addition to the relatively high resolution of SAR products, make this instrument an excellent candidate for filling the meteorological observing needs over the Bering Sea. This study demonstrates the potential usefulness of SAR for observing Bering Sea meteorology by focusing on its ability to image the sea surface footprints of polar mesoscale cyclones (PMCs). These storms can form unexpectedly and are threatening to maritime interests. In this demonstration, a veteran meteorologist at the Anchorage National Weather Service Forecast Office is asked to produce a surface reanalysis for three separate archived cases when SAR imaged a PMC but the original analysis, produced without the aid of SAR data, did not display it. The results show that in these three cases the inclusion of SAR data in the analysis procedure leads to large differences between the original surface analysis and the reanalysis. Of particular interest is that, in each case, the PMC is added into the reanalysis. It is argued that the reanalysis more accurately portray the near-surface meteorology for each case.
The synthetic aperture radar (SAR) has now successfully demonstrated its capacity to uniquely provide valuable high-resolution information for coastal applications (oil-spill monitoring, ship detection, shallow-water bathymetry mapping, sea-ice monitoring, high-resolution wind fields, coastal wave fields). However, it appears that the operational use of SAR-derived products still remains limited, particularly in Europe. Although costs and sampling rate are often invoked to explain this limitation, it also appears that the SAR-instrument capabilities are generally poorly known within the operational community. Consequently, no real initiative currently involves the sustainable use of SAR in the main European projects for operational oceanography nor meteorology. Conversely, other countries such as Norway and United States are now moving into the use of SAR on an operational basis for coastal ocean monitoring. Significant efforts are being led by these countries to develop and harmonize such a monitoring system and extend the number of locations. In order to promote the potential of SAR to routinely scrutinize our coastal environment, the objective of this paper is to provide an overview of current SAR-related issues, including a brief technical system description (coverage, revisit time, etc.) and qualitative and quantitative descriptions of operational marine products. The prospects for achieving true operational usage and improving these products will then be considered in terms of requirements (satellite receiving station, revisit time, low data costs).
An important aspect of operational meteorology in and around the Great Lakes region of the United States and Canada in the winter months is the forecasting of lake-effect precipitation. While the synoptic- and mesoscale processes that govern the development of lake-effect precipitation have been well understood for many years, problems observing these bands remain because of the limited boundary layer coverage provided by the Weather Surveillance Radar- 1988 Doppler (WSR-88D) network. While traditional visible and infrared satellite imagery helps alleviate these coverage limitations, overcast conditions often negate this advantage. Here, a new method for observing lake-effect bands by using synthetic aperture radar (SAR) to identify and characterize their surface signatures is presented. SAR is a remote sensing tool that images surface roughness. Over water, this roughness is related to the surface wind stress and, hence, surface wind field. Here, three cases are documented where the SAR aboard the Canadian Radar Satellite-1 imaged the footprints of precipitating bands over the Great Lakes: one case with multiple snowbands west of one main band over Lake Superior, and two cases with shore-parallel bands over each of Lakes Ontario and Michigan. These cases are first documented using traditional observing methods: infrared satellite imagery, WSR-88D, and surface observations. Then, each SAR image is interpreted based upon the traditional observations. The ultimate goal is to demonstrate that SAR is capable of detecting the surface signatures associated with Great Lakes precipitation bands that could be of value to forecasters when data from traditional observation platforms are unavailable.
Sea surface imprints of coastal katabatic winds are observed on a RADARSAT-1 synthetic aperture radar (SAR) standard mode image off the U.S. west coast taken on January 21, 2003. The katabatic wind pattern is shown in the SAR image as a striking finger-like bright-dark region that mirrors the coastal mountain topography. We simulate the low-level atmospheric circulation using the numerical weather model MM5 with a triply nested-grid technique. The model simulation reproduces the finger-like katabatic wind pattern at the SAR imaging time. Katabatic wind is strongest in the morning when the surface temperature gradient between ocean and land reaches a maximum. The sea surface wind simulated by the MM5 model and that derived from the calibrated SAR normalized radar cross section are in good agreement. The katabatic wind variation along the coastline is between 5–8 ms−1, and the longest wind finger stretches over 20 km offshore.
The unique combination of terrain and dynamic weather systems makes the coastal waters of lower Cook Inlet and Kodiak Island a complex and difficult place to forecast coastal winds. In addition, there are only a few real- time in situ observations available. However, this region has a large concentration of maritime operations that require accurate, high resolution wind information. Spaceborne synthetic aperture radar (SAR) - based surface wind speed fields, at 200 m resolution, provides detailed location and character of the wind patterns, regardless of time of day or cloud conditions. Some case studies are presented to demonstrate the usefulness of the SAR-based winds to: 1) greatly improve operational marine forecasts issued for an area where synoptic and mesoscale meteorological events coexist; 2) provide detailed validation of the forecasted coastal winds; and 3) give insight on complex wind patterns and extreme winds that occur in a data sparse area.
A new approach for automatically estimating wind vectors for SAR imagery is presented, which relies on a projection operation to generate wind direction estimates. A threshold can be applied to the projection results in order to eliminate estimation of wind directions in regions where the SAR image contains no visible features. The process also allows multiple possible directions to be generated at a given location using the image features, which are then resolved to a single direction based on uniformity across local regions of the image. The algorithm has been validated using 137 comparisons to in situ buoy observations, giving a direction RMSE of 39° (mean error of 10°, error standard deviation of 38°) and a wind speed RMSE of 2.2 m/s (mean error = -1.4 m/s, standard deviation of the error = 1.7 m/s). The largest errors come from automatically utilizing strong features in the image that are not aligned with the local wind, such as atmospheric lee waves are current fronts. If we manually eliminate these from the comparisons, then the direction RMSE is 31° (mean error of 20°, error standard deviation of 20°) and the wind speed RMSE is 2.1 m/s (mean error = -1.2 m/s, standard deviation of the error = 1.7 m/s).
This paper represents a consensus on the state-of-the-art in wind field retrievals using synthetic aperture radar (SAR) as it emerged during the 2nd Workshop on Coastal and Marine Applications of SAR in Longyearbyen, Spits-bergen, Norway, held on 8–12 September 2003. This work is the result of the efforts of many co-authors. The length of the author list a compels us to include it as a footnote, 1 but the contributions of these co-authors rep-resent far more than a footnote to this paper These con-tributions were critical to completeness and accuracy of this paper. This paper documents the substantial progress that has been accomplished in SAR wind speed retrieval. The consensus is that SARs can estimate wind speed to better than 2 m/s and wind direction to 25 ¡ with clear evidence that these values can be improved upon.
Satellite oceanography within the National Oceanic and Atmospheric Administration (NOAA) National Environmental Satellite, Data, and Information Service (NESDIS) focuses on observation retrieval and applications to address the NOAA missions of environmental assessment, prediction, and stewardship. The Satellite Oceanography Division encompasses three functional areas: satellite ocean sensors, ocean dynamics / data assimilation, and marine ecosystems / climate. The breadth of scientific investigation includes sea-surface temperature, sea-surface height, sea-surface roughness, ocean color, surface vector winds, sea ice, data assimilation, and operational oceanography.
This study uses the high-resolution all-weather surface wind field mapping capability of SAR to detect and quantitatively describe the strong mountain-parallel, low-level mesoscale wind maxima, commonly called barrier jets that occur along the synoptically windward slopes of Alaska's coastal mountain ranges. SAR reveals that there are two classes of this phenomenon: pure barrier jets caused by terrain blocking of on-shore (i.e. upslope) flow and hybrid jets resulting from the cyclonic turning of off-shore gap flow until it becomes shore-parallel. This paper includes a five-year study from May 1998 through April 2003 of coastal barrier and hybrid jet occurrence in the Gulf of Alaska. Temporal and spatial distributions of banner jet and hybrid occurrence highlight the favored seasons and locations for jet formation and explanations for shape of each distribution are given. The climatological values of structural characteristics of barrier and hybrid jets including strength and width are also computed
The change in accuracy of synthetic aperture radar (SAR) near-surface wind measurements as a function of increasing wind speed is assessed by matching satellite SAR measurements with wintertime buoy measurements in the Bering Sea. For wind speeds less than 15 m/s, SAR-buoy wind speed comparisons show biases less than 1 m/s with standard deviations less than 2.5 m/s. For wind speeds from 15 to 25 m/s, SAR wind measurements have a positive bias above 2 m/s and somewhat more scatter when compared to buoy winds. For winds larger than 25 m/s, the accuracy of SAR winds is not well known. Very few buoy matches are available. Comparisons of SAR winds with hurricane model winds and aircraft measurements are still in a preliminary stage, but indicate that improved algorithms are needed.
Optimal search strategies for conducting reconnaissance, surveillance or search and rescue operations with limited assets are of significant interest to military decision makers. Multiple search platforms with varying capabilities can be deployed individually or simultaneously for these operations (e.g., helicopters, fixed wing or satellite). Due to the timeliness required in these operations, efficient use of available search platforms is critical to the success of such missions. Designing optimal search strategies over multiple search platforms can be modeled and solved as a multiple traveling salesman problem (MTSP). This paper demonstrates how simultaneous generalized hill climbing algorithms (SGHC) can be used to determine optimal search strategies over multiple search platforms for the MTSP. Computational results with SGHC algorithms applied to the MTSP are reported. These results demonstrate that when limited computing budgets are available, optimal/near-optimal search strategies over multiple search platforms can be obtained more efficiently using SGHC algorithms compared to other generalized hill climbing algorithms. Applications and extensions of this research to other military applications are also discussed.
Biodegradable/bioeliminable, core-cross-linked, block co-polymer nanoparticles have been synthesized as a potential anti-tumour drug-delivery system. Methacrylate-modified poly(ethylene glycol)-b-poly(D,L-lactide) (PEG-b-PDLLA) composed of low-molecular-weight polymer blocks (<5 kg/mol) were synthesized by ring-opening polymerization and post-polymerization chemical modifications. Nanoparticles with a diameter of 110 ± 20 nm were formed from methacrylate-modified PEG(45)-b-PDLLA(41) in a THF/water mixture (1:3). The particles were then core-cross-linked using a new, highly acid-labile ketal cross-linker. The cross-linked particles had a hydrodynamic diameter of 104 ± 20 nm (in THF/water, 1:3), as determined by DLS. The particles in THF exhibited a similar hydrodynamic diameter. Doxorubicin as a model anti-tumour drug was loaded into the nanoparticles (25-31 wt%). The particles released 50% of the loaded drug slowly approximately in 2 days at pH 5.5 and in 5 days at pH 7.4. The particles degraded to bioeliminable polymer fragments (<40 kg/mol) after the hydrolysis of the ketal cross-links at pH 5.5 in seven days, as determined by GPC. Doxorubicin-loaded cross-linked particles (9.3 μM doxorubicin/2.5 μM polymer) inhibited the viability of human neuroblastoma SH-EP cells, whilst the particles without drug at the same concentration were non-toxic, as determined by an Alamar Blue assay. Flow cytometry experiments revealed that the doxorubicin-loaded cross-linked particles were taken up by SH-EP cells in quantities comparable with free doxorubicin. Overall the results support the value of the cross-linked particles for further investigation as a carrier for anti-tumour drugs.
This report is a review of the publicly available literature on algorithms for detecting ships in Synthetic Aperture Radar (SAR) imagery, with the aim of recommending algorithms for inclusion in the Analysts' Detection Support System (ADSS). Our intention is not just to summarise the current state of affairs in the literature. We also aim to outline the motivations, theory and justifications of the various approaches so that educated comparisons and evaluations can be made. The driving requirement at DSTO for ship detection in SAR imagery is wide area surveillance of the oceans surrounding Australia and especially to the North where tropical atmospheric conditions hinder surveillance with electro-optic sensors. Modern SAR sensors can generate large amounts of data in a short period of time and there is an obvious need for automatic detection. This is especially true in the context of ship surveillance where much of the imagery contains only open ocean. The ADSS is a software system being developed at DSTO for the automatic detection of targets in SAR imagery. It was designed with the intention of reducing imagery analysts' workloads in wide area surveillance operations by directing their attention to areas in the imagery of likely interest. The launch of the Seasat SAR system in 1978, verified that ships could be imaged with space-based SAR systems. It also revealed an unexpected richness of oceanographic phenomena. Seasat was followed by several other space- based SAR systems including ERS1, ERS2, Radarsat and SIR-C/XSAR. These systems have made high quality SAR imagery of the ocean readily available and operational surveillance of ocean ship traffic is now viable. Several near real-time operational ship monitoring systems using SAR imagery are now In place or are currently being developed. The prIme examples are the Ocean Monitoring Workstation (OMW), the Alaska SAR Demonstration (AKDEMO) system, the European Community Joint Research Centre (JRC) system, and the Qinetiq's MaST system. The Norwegian Defence Research Establishment (NDRE) is an important player too and was behind the system Eldhuset reported on in 1996. Given the long history and ongoing interest there is an extensive literature on algorithms for ship detection in SAR imagery. While many papers begin with a brief review of the literature relevant to their application, only one general review has been found. That review, by Fingas and Brown, notes the lack of previous reviews but is itself limited. This report aims to fill that gap. There are two fundamentally different means of detecting ships in SAR imagery: de- tection of the ship target itself and detection of the ship wake. While it is clearly desirable to make use of both possibilities (and many ship detection systems do), in this report, we only focus on algorithms which detect the ship itself. This decision was made in order to limit the size of the report. It is anticipated that the detection of ship wakes will be the topic of a companion report. As a justification of this stance, we note that most current research on ship detection is based on detecting the ship rather than its wake. Among the reasons for this are: ship backscatter is robust and less dependent on sea state; stationary ships have no wake; wakes are often not visible and especially at large radar incidence angles; and wake detection is more computationally expensive. We also comment that there is a bias in the open literature towards space based rather than airborne SAR systems. In theory this should make little difference to detection algorithms since the imaging principles for both platforms are the same. However, we note that space-based SAR imagery is generally single channel and lower resolution. Thus most ship detection algorithms in the literature are aimed at locating bright point-like targets in single channel imagery. Space based SARs also tend to have lower incidence angles which makes the detection problem harder. We add however, that the next generation of space-based SARs offer multi-channel polarimetric data and consequently there is a shift towards research on algorithms which make use of multi-channel data. Past research efforts on automatic target detection in SAR imagery have clearly demonstrated that no single detection algorithm will produce satisfactory results and a hierarchical system of algorithms is needed. Thus ship detection systems generally consist of several stages: land masking; preprocessing; prescreening; and discrimination. In this report, we consider each of these stages in turn and discuss the various algorithms which have been used. Recommendations on algorithms for the ADSS are made in the final section. There are two main difficulties in making recommendations. First, different types of imagery require different algorithms. Resolution is of prime importance here but other factors such as polarisation and image processing are relevant too. Second, there is a lack of rigorous performance evaluation in the literature. Often the imagery is visually ground- truthed and some algorithms are very poorly tested with results only being presented for a handfull of images. This makes comparison of the various approaches difficult. Given this, our first recommendation is to: .specify a surveillance scenario; .decide on appropriate radar parameters and geometry; .set up well ground-truthed benchmark tests; .compare the different detection algorithms. The only complete systems to have extensive and convincing publicly available reports of their performance are the OMW, AKDEMO, SUMO and Eldhuset systems. On this basis we recommend the following basic system for low to medium resolution single channel SAR imagery: .land masking using ashore-line database with a buffer zone included; .a simple moving window adaptive threshold algorithm (with the two parameter CFAR being the preferred option); .clustering of detected pixels; .discrimination of false alarms using elementary target measurements; .human supervision and discrimination of the final detections; We note that this system can be applied to higher resolution imagery if an initial down sampling step is included. Once this basic system is in place, improvements and alternatives can be considered. We discuss the details of this basic system and make suggestions for improvements and alternatives. We believe that while many of the suggested refinements and alternatives may result in improved performance, the most significant gains are likely to come from future research on polarimetric imagery and better false alarm discrimination. Much of this research will be based on Envisat ASAR and Radarsat-2 data. This is is a review of the publicly available literature on algorithms for detecting ships in Synthetic Aperture Radar (SAR) imagery, with the aim of recommending algorithms for inclusion in the Analysts' Detection Support System (ADSS). The ADSS is a software system being developed at DSTO for the automatic detection of targets in SAR imagery. The author outlines the motivations, theory and justifications of the various approaches so that educated comparisons and evaluations can be made. Most current research on ship detection is based on detecting the ship itself rather than its wake. Detection of ship wakes is not addressed in any depth. Ship detection systems generally consist of several stages: land masking; preprocessing; prescreening; and discrimination. Each of these stages are considered in turn and the various algorithms which have been used are discussed. A basic system for low to medium resolution single channel SAR imagery is recommended and suggestions for improvements and alternatives are made. DGISREW
The National Oceanic and Atmospheric Administration (NOAA) National Environment Satellite, Data, and Information Service (NESDIS) provides synthetic aperture radar (SAR) derived products under a demonstration project named the Alaska SAR Demonstration (AKDEMO) to the US government community. The AKDEMO near real-time data and products include SAR wind images and vectors, hard target locations, and ancillary data. The hard target locations are available for use in fishery management by agencies such as the Alaska Department of Fish and Games (ADF&G), the National Marine Fisheries Service ( NMFS) and the United States Coast Guard (USCG). Vessel positions are obtained form hard target signatures through the use of a constant false alarm rate (CFAR) vessel detection algorithm developed by Veridian Systems Division. This algorithm has gone through testing and validation, using fleet information and vessel observer reports, during the Red King Crab fisheries in Alaska in 1999 and 2000. The goal was to maximize the number of ships found while minimizing the number or false alarms. Using general fleet location information, it was found that the minimum vessel size detected by the CFAR algorithm was 36 m using RADARSAT-1 ScanSAR wide mode data with a nominal spatial resolution of 100 m. Still, when comparing the CFAR results with the actual positions reported by the ship observers, vessels over 36 m were not always detected. This led to the hypothesis that the heading and perhaps wind conditions may have affected the ability of the SAR to detect the vessels. In 2001, vessel observers again reported their positions during SAR overpasses, this time also reporting heading and wind conditions. Unfortunately, due to high winds and waves, SAR was not able to detect the fishing fleet. In 2002, this was repeated, resulting in 3 days during the fishery opening when RADARSAT-1 was able to image the fishing fleet in the ScanSAR wide B mode. Approximately twenty ships each day in the a rea covered by the RADARSAT-1 data reported their position and heading. Results showing the dependence between RADARSAT-1 vessel detection and vessel heading are presented using the GIS platform.
The National Oeanic and Atmospheric Administration (NOAA)/National
Environmental Satellite, Data, And Information Service (NESDIS) is in
the second year of a two-year demonstration of Synthetic Aperture Radar
(SAR) derived products called the Alaska SAR Demonstration (AKDEMO).
This demonstration provides near real-time SAR data and derived
products, including wind images and vectors, hard target locations,
along with ancillary data, to specific users in the government
community. One of the derived products are vessel positions obtained
from a constant false alarm rate (CFAR) vessel detection algorithm
developed by Veridian ERIM. This algorithm has been tested and validated
to maximize the number of ships found while minimizing the number or
false alarms on one SAR image of the Red King Crab fishery in Bristol
Bay on October 18, 1999. This resulted in using a detection statistic
threshold of about 5.5, depending on image resolution used. Until now,
this validation has been done with only general knowledge of fishing
fleet size and location, but no in situ vessel information. This paper
presents the results of a validation of the SAR vessel detection
algorithm using observer reported vessel positions along with
information on vessel size and local wind speed
Satellite synthetic aperture radar (SAR) instruments such as those
carried on the Canadian Space Agency RADARSAT-I satellite and the
European Space Agency European Remote Sensing Satellite 2 (ERS-2)
provide a wealth of information on the ocean surface expression of
atmospheric and oceanic phenomena. However, since these SAR instruments
are both single frequency, single polarization, it is often difficult to
unambiguously analyze the myriad features in an image without consulting
other types of data. A near real-time demonstration of SAR applications
in Alaska, the Alaska SAR Demonstration, provides products to users who
have minimal acquaintance with SAR imagery. To aid them in interpreting
the SAR signatures, an analysis system called the World Wide Web
Interactive Processing Environment (WIPE) is used to provide a
capability for analyzing SAR imagery together with other satellite, in
situ and model data. This system is then used to interpret SAR
signatures of oceanic and atmospheric fronts, storms, upwelling, oil
slicks, etc. This paper describes the AKDEMO and its products as well as
the WIPE system used to display and analyze the SAR imagery. An example
illustrating the synergy of using visible satellite radiometer data
along with SAR imagery for ocean feature analysis is then
presented
We performed a systematic comparison of wind speed measurements from the SeaWinds QuikSCAT scatterometer and wind speeds computed from RADARSAT-1 synthetic aperture radar (SAR) normalized radar cross section measurements. These comparisons were made over in the Gulf of Alaska and extended over a two-year period, 2000 and 2001. The SAR wind speed estimates require a wind direction to initialize the retrieval. Here, we first used wind directions from the Navy Operational Global Atmospheric Prediction System (NOGAPS) model. For these retrievals, the standard deviation between the resulting SAR and QuikSCAT wind speed measurements was 1.78 m/s. When we used the QuikSCAT-measured wind directions to initialize the inversion, comparisons improve to a standard deviation of 1.36 m/s. We used these SAR-scatterometer comparisons to generate a new C-band horizontal polarization model function. With this new model function, the wind speed inversion improves to a standard deviation of 1.24 m/s with no mean bias. These results strongly suggest that SAR and QuikSCAT measurements can be combined to make better high-resolution wind measurements than either instrument could alone in coastal areas.
River ice jams which form during break-up have been responsible for the largest flood at many Alaska villages. Since ice jams can occur at any location along a river, ice reconnaissance flights are a major source of river ice and ice jam flooding information during break-up. The flights are expensive, can be delayed by weather, and are generally restricted to the Yukon and Kuskokwim Rivers; an additional method of monitoring these and other rivers is very desirable. To determine the usefulness of Synthetic Aperture Radar (SAR) imagery for break-up monitoring and forecasting and for flood monitoring, the Alaska River Forecast Center received near-real-time SAR images during the 1997 break-up season and during a flood in August 1997. Initial analysis indicates that SAR imagery is useful for break-up monitoring and forecasting on larger rivers.
An applications demonstration of the use of Synthetic Aperture Radar (SAR) data in an operational selling is being conducted by the National Oceanic and Atmospheric Administration (NOAA) CoastWatch Program. In the development phase of this demonstration, case studies were conducted to assess the utility of SAR data for monitoring coastal ice in the Bering Sea, icebergs from calving glaciers in Prince William Sound, and lake ice in the Great Lakes. ERS-l SAR data was used in these studies. Results showed that depending on size and sea state icebergs could be detected from background and computer enhanced in the imagery, thaI SAR data can supplement and enhance the utility of satellite visible and infrared data sources for coastal ice monitoring, and that Greal Lakes ice cover can be classified by ice type and mapped in the SAR data using image processing techniques. Cloud cover was a common problem. Based on the further development of automated analysis algorithms and the increase in frequency of SAR coverage, the all-weather, day/night viewing capabilities of SAR make it a unique and valuable tool for operational ice detection and monitoring.
Synthetic aperture radar (SAR) data from the Canadian RADARSAT
satellite, along with infrared and visible data from operational weather
satellites, is being collected over the ocean off the U.S. East, Gulf of
Mexico, Gulf of Alaska, and Bering Sea coasts. Mesoscale features
observed in the SAR data are identified using coincident Geostationary
Operational Environment Satellite (GOES) data to decide whether the
signature is atmospheric or oceanic in origin. Combining the two data
sources allows information in the marine atmospheric boundary layer
(MABL) to be observed. This technique of feature detection using
multiple data sets is referred to as data fusion. This paper focuses on
a single storm event that is imaged by RADARSAT SAR in the Bering Sea on
February 5, 1998 at 06:00 UTM. From the fusion of SAR and GOES data,
frontal boundaries can be deciphered
The ocean surface wind field is observed from space operationally
using scatterometry. The European Space Agency's (ESAs) ERS-1 satellite
scatterometer routinely produces a wind product that is assimilated into
forecast models. Scatterometry cannot give accurate wind estimates close
to land, however, because the field of view of a spaceborne
scatterometer is on the order of 50 km. Side lobe contamination, due to
the large contrast in backscatter between land and water, compounds the
problem. Synthetic aperture radar (SAR) can provide wind speed and
direction estimates on a finer scale, so that high-resolution wind
fields can be constructed near shore. An algorithm has been developed
that uses the spectral expression of wind in SAR imagery to estimate
wind direction and calibrated backscatter to estimate wind strength.
Three versions, based on C-band scatterometer algorithms, are evaluated
for accuracy in potential operational use. Algorithm estimates are
compared with wind measurements from buoys in the Gulf of Alaska, Bering
Strait, and off the Pacific Northwest coast by using a data set of 61
near-coincident buoy and ERS-1 SAR observations. Representative figures
for the accuracy of the algorithm are ±2 m/s for wind speed and
±37° for wind direction at a 25-km spatial resolution
CoastWatch Applications of Synthetic Aperture Radar Imagery
- W Pichel
- P Clemente-Colón
- K Friedman
- A Lunsford
- G Leshkevich
Pichel, W., Clemente-Colón, P., Friedman, K., Lunsford, A., Leshkevich, G.,
et al., "CoastWatch Applications of Synthetic Aperture Radar Imagery," J.
Adv. Marine Sci. Tech. Soc. (in press, 1999).
The Use of Synthetic Aperture Radar Observations as Indicators of Fishing Activity in the Bering Sea
- P Clemente-Colón
- D R Montgomery
- W Pichel
- K Friedman
Clemente-Colón, P., Montgomery, D. R., Pichel, W., and Friedman, K., "The
Use of Synthetic Aperture Radar Observations as Indicators of Fishing
Activity in the Bering Sea," J. Adv. Marine Sci. Tech. Soc. (in press, 1999).
The authors wish to thank those involved in developing the Alaska SAR Demonstration
ACKNOWLEDGMENTS: The sections describing the mission of the National
Weather Service, Alaska Region, and the Coast Guard 17th District were taken
from presentations given, respectively, by Dr. Gary Hufford and Capt. Dennis Egan
at the JHU/APL Symposium on Emerging Coastal and Marine Applications of
Wide Swath Synthetic Aperture Radar, Laurel, MD (23-25 March 1999).
The authors wish to thank those involved in developing the Alaska SAR
Demonstration, specifically Robert Beal, Francis Monaldo, and Donald Thompson
of JHU/APL; Christopher Wackerman, Robert Shuchman, and Michael True of
ERIM International, Inc.; Erick Malaret of Applied Coherent Technologies Corp.;
Rick Guritz, Christopher Wyatt, and Verne Kaupp of the Alaska SAR Facility;
Xiaofeng Li of Research and Data Systems Corp
- Karen Friedman Of Caelum
- Corp
Karen Friedman of Caelum Corp.; Xiaofeng Li of Research and Data Systems Corp.;
William Tseng, Robert Stone, and Ralph Meiggs of NOAA/NESDIS;
Funding for the Alaska SAR Demonstration and research preceding the demonstration were provided by the NOAA/NESDIS Ocean Remote Sensing Program. Radarsat data were provided under NASA Radarsat Applications Development and Research Opportunity (ADRO) Project 396
- Gregory Busch Of The
- U S Coast
Gregory Busch of the U.S. Coast Guard 17th District; and Fritz Funk and Katherine
Rowell of the Alaska Department of Fish and Game. The SAR data system used for
the demonstration was developed in partnership with the National Ice Center, in
particular David Benner, Cheryl Bertoia, and John Dowdell.
Funding for the Alaska SAR Demonstration and research preceding the
demonstration were provided by the NOAA/NESDIS Ocean Remote Sensing
Program. Radarsat data were provided under NASA Radarsat Applications Development and Research Opportunity (ADRO) Project 396. Mention of a commercial
entity or product does not constitute endorsement by the U.S. government.