285

volume 88 number 28 10 JulY 2007

A new integrated coastal observation system is providing preliminary data from the North American Great Lakes. This sys-tem can be implemented in other coastal regions. To date, it has been successfully deployed on Lakes Michigan, Huron, and Erie to make seabed to sea-surface mea-surements of chemical, biological, and physical parameters, which are transmit-ted wirelessly through buoys and perma-nent stations. Called the Real-Time Coastal Observation Network (ReCON), the new system leverages existing networking tech-nology to provide universal access to a wide variety of instrumentation through the use of an underwater Ethernet port server [Austin, 2002]. A team of NOAA engineers and scientists has completed the develop-ment and testing of this integrated coastal observation network.

Utility of the Network

An Ethernet-based coastal observation network design enables the creation of sys-tem components, such as sensor drivers, data transfer software, system control func-tions, database management, web display ,and archival functions, using standard web-design tools. The underwater, univer-sal hub easily allows the attachment of sen-sors at any time during the deployment period. Portable buoys and permanent sta-tions transferring data into network nodes distributed across broad coastal regions can be integrated at a central location using the Internet. This implementation of a coastal network providing real-time chemi-cal, biological, and physical observations has already benefited ecosystem research-ers, resource managers, forecasters, edu-cational institutions, and public users. Fur-ther, regional observations downloaded at time intervals required to describe par-

ticular ecosystem features and events can be presented to managers and operational forecasters through ad hoc web displays or to students and researchers through search-able database management systems.

Using this approach, an observation net-work can be deployed in any coastal region with Internet availability. Buoys need only be placed within antenna range of the shore station. Deployment range, depen-dent on the height of the shore antenna, can be as much as 32 kilometers allow-ing buoy placement anywhere within an approximate 1400 square kilometer area. Additional buoys or fixed stations can be used to extend range through the use of the relay capability inherent in wireless network devices. The observation net-work supplies enough throughput capac-ity to simultaneously support continuous measurements from both standard ocean-ographic and meteorological instrumenta-tion (such as wind and temperature mea-surements, current velocity profilers, and chemical sensors) and more advanced sur-face and underwater applications such as streaming imagery.

By leveraging existing internet technol-ogy for real-time data collection, NOAA’s observation infrastructure can be signifi-cantly upgraded to provide forecasters, researchers, coastal resource managers, and the public with the data necessary to make informed decisions in response to ecosystem change [Ocean.US, 2002]. The transition of this research and development effort to an operational coastal implementa-tion has the potential to improve forecasts and forecast verification, increase marine safety, and reduce public health risks while responding to established national goals [Ocean.US, 2006].

Network Specifications

The important aspects of the ReCON coastal internet-based network are the abil-ity to wirelessly connect buoys to shore at

distances up to 32 kilometers or to other buoys in an array, and to connect vessel-based data collection systems through off-shore buoys or directly to the shore. The ReCON universal hub provides a standard instrumentation interface for the deploy-ment of multi-sensor data collection plat-forms. Data archiving, data management, and system control are implemented through the Internet.

Portable buoys [Frye et al., 2000] and permanent stations are connected to shore through wireless internet radios capable of providing up to 1.5 megabytes per sec-ond of system bandwidth. Connections to underwater instrumentation are imple-mented through a marine-grade Ethernet network cable. This connection allows buoy computer access to data or imag-ery from multiple devices plugged into an underwater network hub using standard marine connectors. Data and imagery are first transferred to and stored on the buoy computer, then transferred at a scheduled time (typically hourly) to the shore com-puter over the wireless internet connection.

System Implementations in the Great Lakes

Permanent and buoy-based ReCON sys-tems have been deployed at coastal loca-

tions around the Great Lakes, beginning in 2003. A network computer located at the Great Lakes Environmental Research Laboratory (GLERL) in Ann Arbor, Mich., accesses the data from shore computers through the Internet. The buoy computer monitors the solar-battery power system allowing variable power cycling to con-serve energy during periods of limited solar exposure. Underwater sensors can be re-programmed from the central com-puter in Ann Arbor in anticipation of rap-idly changing ecosystem events. During such episodic events, specific system com-ponents can be placed in continuous oper-ation to enable high resolution temporal sampling.

A ReCON system deployed on western Lake Erie on a U.S. Coast Guard navigation structure, in collaboration with the Univer-sity of Toledo’s Lake Erie Center (located in Oregon, Ohio), supported preliminary investigations into conditions contribut-ing to episodic hypoxia events and subse-quent effects on benthic organisms. Figure 1 shows data collected from this station during 2006. Water depth at the station was 8 meters. Sensors at the station included a bottom-mounted acoustic Doppler cur-rent profiler, a YSI 6600EDS multiprobe

A Wireless Real-Time Coastal Observation Network

S. RubeRg, S. bRandt, R. Muzzi, n. Hawley, t. bRidgeMan, g. leSHkevicH, J. lane, and t. MilleR

Fig. 1. (a) Wind speed in meters per second (blue) and current speed 1 meter above the bottom in centimeters per second (black). (b) Wind direction (blue) and current direction at 1 meter above bottom (black). The wind direction follows the meteorological convention, so the wind is from the direction indicated. The current direction follows the oceanographic convention so the flow is toward the direction shown. (C) Water temperature 1 meter above the bottom (black) and dissolved oxygen concentration in milligrams per liter 1 meter above the bottom.

Fig. 2. Lake Erie ice imagery, camera looking south at approximately 180°. (a) Image taken on 27 February 2007 showing snow ice with wind-rowed snow and areas of black ice (re-frozen leads or melt holes). (b) Image of 5 March 2007 showing pancake ice and open water.

Wireless Network cont. on page 286

286

EOS volume 88 number 28 10 JulY 2007

The carbon cycle science community has called for demonstrable progress in streamlining terrestrial-atmosphere obser-vations into a global network to enable syn-theses and regional to global modeling of the role of terrestrial ecosystems in climate change. The AmeriFlux network of about 100 research sites is the primary research group and data provider for large synthe-ses on terrestrial carbon cycling in the Americas. To broaden the value of the net-work, AmeriFlux has developed a database design and new guidelines for the submis-sion of micrometeorological, meteorologi-cal, and biological data to the archive.

The database, which includes atmo-spheric, plant, and soils data, was designed with significant input from Microsoft and the Berkeley Water Center (California). AmeriFlux is also working with interna-tional organizations such as CarboEurope and the Global Terrestrial Observing Sys-tem to standardize terrestrial observations and data products for global analyses.

The AmeriFlux network aims to quan-tify and explain the influences of climate and disturbances such as wildfire and land use on carbon, water, and energy exchange between terrestrial ecosystems and the atmosphere. An essential compo-nent that enables large-scale analyses is the consistency of the data. AmeriFlux guide-lines for data submission reflect input from data providers and users as well as recom-mendations from database designers and data managers to improve the efficiency of access and automation, especially for global syntheses and spatial model-ing activities that include data assimila-tion techniques. One of the fundamental requirements of network participants to maintain membership in the network is to submit micrometeorological and meteoro-logical data within 1 year of collection, and biological data within about 2 years of col-lection to the AmeriFlux data repository at the Carbon Dioxide Information Analysis Center (CDIAC), at the Oak Ridge National Laboratory in Tennessee.

Biological observations represent a par-ticular challenge in that they are collected at disparate scales, such as, for example, fast processes (photosynthesis, soil respi-ration) and slow processes (carbon pools, nitrogen content). The data submission requirements address this challenge by requesting variables in time and space scales that are most likely to be useful to a broad group of users. Controlled vocab-ularies are used for metadata and site characteristics such as disturbance histo-ries. The system has checks on expected ranges of variation, controlled vocabular-ies, and formats. The specified formats are being used to standardize existing data in the collection and new observations by investigators in AmeriFlux and the North American Carbon Program (NACP), in which AmeriFlux plays a major role.

The NACP research strategy recom-mends an integrated data and informa-tion management system that will enable researchers to access, understand, use, visualize, and analyze large volumes of diverse data at multiple thematic, tem-poral, and spatial scales. Managing and

integrating data for NACP requires an overarching data policy to ensure that par-ticipants have full, open, and timely access to data in order to promote the exchange of quality-controlled and quality-assured data. This is needed to protect intellectual property rights and to ensure that proper credit is given to data originators through authorship, citation, or acknowledgment. The development of the AmeriFlux data and information system will be an impor-tant asset to the NACP. Additional informa-tion on the NACP is located at http://www.nacarbon.org.

In an effort to make data broadly avail-able and yet allow data providers time to conduct original analyses within and among similar networks, AmeriFlux requests that users follow the Data Fair Use Policy posted on the AmeriFlux Web site. The AmeriFlux database is located at http://public.ornl.gov/ameriflux.

—Beverly e. law, AmeriFlux Network, Oregon State University, Corvallis; E-mail: bev.law@oregon-state.edu

This month, NASA will launch a coordi-nated field campaign centered in Costa Rica that will use satellite, ground-based, bal-loon, and aircraft measurements to investi-gate how chemicals in the air move into the stratosphere and how this movement might affect cloud formation, climate, and ozone.

Scheduled to run from 16 July through 8 August, the Tropical Composition, Cloud, and Climate Coupling (TC4) campaign “is an unprecedented opportunity to use NASA’s complete suite of satellite and air-borne Earth-observing capabilities to inves-tigate a largely unexplored region of the atmosphere,” said Michael Kurylo, a TC4 program scientist at NASA headquarters.

The campaign will focus on a region of the atmosphere called the tropical tropo-pause layer (TTL), the transition layer that lies between the troposphere and strato-sphere at about 12–17 kilometers in altitude. This layer, home to cirrus clouds, plays a key role in the movement of water vapor and other chemicals into the stratosphere, which can have implications for the greenhouse effect and global climate change, and for the depletion and recovery of the ozone layer.

The TTL is difficult to study because only a few research aircraft can reach it. Satellite observations have improved knowledge of the TTL, but they can be obscured by high clouds within the layer or otherwise affected

because the layer has less mass than the underlying atmospheric layers.

The coordinated TC4 campaign will pro-vide observations and measurements of gases, aerosols, ice crystals, and other chemicals as they move upwards from storms over the warm tropical ocean. These measurements will fuel future models of the atmosphere, Kurylo said.

NASA has kept the $12-million TC4 in plan-ning stages over the last few years, unable to move forward because of a lack of funds for research and analysis that are used for these types of missions, Kurylo said. However, the “silver lining” to the delay is that several new satellites have been launched during the wait, making their data and teams of scien-tists available for TC4, he said.

TC4 will take advantage of the “A Train” group of satellites, which includes Aqua and Aura and was expanded last year to include CALIPSO and CloudSat. However, relying on satellite data alone “is a little like the story of the group of blind men examining the elephant,” said TC4 mission scientist Brian Toon, of the University of Colorado, Boulder.

“None of them can see the entire pachy-derm. One goal of TC4 is to help the satel-lites understand the context of their data and how it relates to the clouds and the air they see so in the end we see an elephant and not a giraffe.”

In addition to the satellites, NASA will fly three aircraft at various altitudes, launch meteorological balloons from Panama, Costa Rica, and San Cristobal Island in the Galapagos Archipelago, and take ground measurements from Panama and Costa Rica. By using so many different sources of observations, project scientists will be able calibrate the satellite measurements and generate a comprehensive data set to answer various scientific questions about the TTL, said Hal Maring, a TC4 program sci-entist at NASA headquarters.

NASA is planning a second TC4 campaign for a later date in Guam. Additional informa-tion is available at http://www.espo.nasa.gov/tc4/

—Sarah ZielinSki, Staff Writer

286

TRANSACTIONS AMERICAN GEOPHYSICAL UNION

The Newspaper of the Earth and Space Sciences

Editors

John W. Geissman: Dept. of Earth and Plan-etary Science,University of New Mexico, Albu-querque, USA; jgeiss@unm.edu

Wendy S. Gordon: Texas Commission on Environmental Quality; wgordon@tceq.state.tx.us

Manuel Grande: University of Wales, Aberystwyth; mng@aber.ac.uk

Hassan Virji: START; hvirji@agu.org

Editor in Chief

A. F. Spilhaus, Jr.: AGU, Washington, D.C., USA; eos_fspilhaus@agu.org

StaffEditorial: Nina Tristani, Publisher; Randy Showstack, Managing Editor; Sarah Zielinski, News Writer/Editor; Mohi Kumar, Science Writer/Editor; Liz Castenson, Editor’s Assistant. Advertising Carla Fleming, Advertising Assistant; Tel: +1-202-777-7536; E-mail: advertising@agu.org Composition and Graphics: Valerie Bassett, Carole Say-lor, and Nancy Sims,Electronic Graphics Specialists

©2007 American Geophysical Union. Material in this issue may be photocopied by individual scientists for research or classroom use. Permission is also granted to use short quotes, figures, and tables for publication in scientific books and journals. For per-mission for any other uses,contact the AGU Publications Office.

Eos, Transactions, American Geophysical Union (ISSN 0096-3941) is published weekly by the American Geophysical Union, 2000 Florida Ave., NW, Washington, DC 20009 USA. Periodical Class postage paid at Wash-ington, D.C., and at additional mailing offices. POST-MASTER: Send address changes to Member Service Center, 2000 Florida Ave., NW, Washington, DC 20009 USA.Member Service Center 8:00 a.m.–7:00 p.m. Eastern time; Tel: +1-202-462-6900; Fax: +1-202-328-0566; Tel. orders in U.S.:1-800-966-2481; E-mail: service@agu.org. Information on institutional subscriptions is available from the Member Service Center.

Views expressed in this publication do not necessarily reflect official positions of the American Geophysical Union unless expressly stated.

www.agu.org/pubs/eos

(that included temperature and oxygen sen-sors) located 1 meter above the bottom, and a meteorological station. The clock-wise circulation observed through day 214 (August 1) is due to internal inertial waves. The data show that the oxygen concentra-tion decreased steadily until about noon on day 216 (August 3). Both the water tem-perature and the oxygen concentration then increased for about the next 24 hours, before declining again.

The short duration of the hypoxic epi-sode illustrates the value of using real-time data provided by the ReCON buoys to trig-ger field sampling, since more traditional monitoring methods (using a small boat to go out once a week to measure the oxy-gen concentration, for example) would likely have missed the event. Recent stud-ies have suggested that even short term hypoxic events in western Lake Erie can have adverse effects on benthic organisms [Bridgeman et al., 2006].

The same Lake Erie ReCON station is also used to transfer hourly imagery. The pan and tilt internet camera, located 12.5 meters above the water level, has a hori-zontal field of view (FOV) of 71.3° and a vertical FOV of 53.1°. The camera is capa-ble of providing image resolution up to 1280 by 960 pixels. During winter, ice imag-ery (Figure 2) is used to develop ice cover estimates for use by Great Lakes shipping. GLERL and Jet Propulsion Laboratory scien-tists have also used this winter imagery to provide ‘ground truth’ information for the development and validation of algorithms for satellite synthetic aperture radar (SAR) and scatterometer ice cover classification and mapping. Great Lakes ice cover infor-mation—including spatial coverage, con-centration, ice type, thickness, freezeup and breakup dates, and ice duration—is a necessary input for ice control and ice breaking operations and ice forecasting and modeling efforts. In addition, ReCON provides real-time estimates of local condi-tions during cloud cover.

A separate ReCON node located on Lake Huron at the Thunder Bay National Marine Sanctuary (TBNMS) in Alpena, Mich., pro-vides buoy-based streaming imagery of underwater historic artifacts and conven-tional composite video from a remotely operated vehicle (ROV) operated from the deck of a research vessel. The webcam is plugged directly into the underwater net-work hub connected to a surface buoy located approximately 16 kilometers off-shore providing shipwreck observations to TBNMS visitors.

The system is also used to transfer audio and video over the Internet to multiple classrooms across the country for use in real-time ‘live dives’ providing historic and scientific educational opportunities. The composite video and audio signal from the ROV and scientific narrators on deck is digitally reduced in size using a hardware compression device. The compressed infor-mation is transferred over the Internet to host computers where it is restored to origi-nal quality for use with any video display device. Students and teachers in the class-room are then able to interact with onsite researchers using internet audio and text messaging techniques in wide use today.

Acknowledgments

The authors would like to thank the crew of the R/V Laurentian, Coast Guard Ninth District Group in Detroit, Mich., and Thun-der Bay National Marine Sanctuary staff members Jeff Gray and Russ Green for their valuable contribution to the success of

this research project. The work described here was funded by the NOAA Great Lakes Environmental Research Laboratory, the High Performance Computing program, the Office of Ocean Exploration, and the Inte-grated Ocean Observing Systems project. This is Great Lakes Environmental Research Laboratory contribution number 1434.

References

Austin, T., J. Edson, W. McGillis, C. von Alt, M. Purcell, R. Petitt, M. McElroy, C. Grant, J. Ware, and S. Hurst (2002), A network-based telemetry architecture developed for the Martha’s Vineyard Coastal Observatory, IEEE J. Oceanic Eng., 27(2), 228–234.

Bridgeman, T. B., D. W. Schloesser, and A. E. Krause (2006), Recruitment ofRecruitment of Hexagenia mayfly nymphs in western Lake Erie linked to environ-mental variability, Ecol. Appl., 16, 601–611.

Frye, D., B. Butman, M. Johnson, K. von der Heydt, and S. Lerner (2000), Portable coastal observato-ries, Oceanography, 13(2), 24–31.

Ocean.US (2002), An integrated and sustained ocean observing system (IOOS) for the

United States: Design and implementation, 21 pp., Arlington, Va. (Available at http://www.ocean.us/oceanus_publications)

Ocean.US (2006), IOOS: The system, Natl. Off. for Integrated and Sustained Ocean Obs., Arlington, Va. (Available at http://www.ocean.us/what_is_ioos)

Author Information

Steven Ruberg, Stephen Brandt, Ronald Muzzi, Nathan Hawley, George Leshkevich, John Lane, and Terry Miller, NOAA Great Lakes Environmental Research Laboratory, Ann Arbor, Mich.; E-mail: steve.ruberg@noaa.gov; and Thomas Bridgeman, University of Toledo, Lake Erie Center, Oregon, Ohio.

Wireless Network cont. frompage 285

newsExamining Chemical Transport Into the Stratosphere

AmeriFlux Network Aids Global Synthesis

In MemoriamGeorge A. Guy, 92, 23 February 2007; Atmospheric Sciences, 1948Takeo Kosugi, 57, 26 November 2006; Solar and Heliospheric Physics, 1994Elizabeth Sulzman, 40, 10 June 2007; Biogeosciences, 2000

G E O P H Y S I C I S T S