Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 46
Oceanography in 2025: Proceedings of a Workshop
Oceanography in 2025
John Orcutt*
Oceanography today is characterized by an increasing trend toward multi-institutional operation and management of seagoing resources that will accelerate in coming years for a variety of reasons. Examples today include UNOLS, Argo, OBSIP (U.S. National Ocean Bottom Seismometer Instrument Pool), HiSeasNet, OOSs, and the OOI. The effective management of such facilities is a major challenge; over a number of scientific and engineering disciplines, these collaborations have been characterized as Virtual Organizations.
Statistics and anecdotes both support this growth in virtual organizations. There are about 200 research universities in the country and publications in science and engineering involving multiple institutions increased by 48% between 1988 and 2001. Between 1990 and 2000, nearly 16% of all scientific publications involved international institutions, doubling over the decade 1990-2000. The rationale for this increased extra-institutional collaboration includes the trend toward more multidisciplinary research and the need to broaden the expertise to address increasingly complex problems. Practically, agencies and program managers often require multi-institutional collaboration. An NSF-sponsored study of virtual organizations, however, suggested that projects involving multiple universities produced fewer knowledge outcomes than those involving a single institution. While there are many reasons for these changes, finding effective methods for managing virtual organiza-
*
Scripps Institution of Oceanography, University of California, San Diego
OCR for page 47
Oceanography in 2025: Proceedings of a Workshop
tions is critical and must be considered more deeply than in the past. The NSF now often requests management plans for some programs and very extensive plans for projects such as OOI with budgets on the order of $400M. Another major driver in coming years for oceanography will be climate observations for both scientific purposes and in support of new treaties seeking to reduce the introduction of greenhouse gases and to mitigate changes associated with these gases.
The next major summit to develop the follow-on to Kyoto will begin on 30 November 2009. President Obama and his transition team believe that the replacement for Kyoto should not be a “Feel Good” treaty (an oft-used term to describe previous treaties like the United Nations Framework Convention on Climate Change). One of the hallmarks of a proper treaty is a monitoring regime to collect the data necessary to ensure that not only is the U.S. complying with the treaty, but that the other signatories are bearing their share of the burden. Monitoring systems for nuclear testing treaties are an example of previous efforts in this regard although the technical challenges for a climate treaty dwarf those encountered with nuclear testing! Significantly expanded monitoring on land and at sea will have to be undertaken, requiring new methodologies and broad collaborations between disciplines. Sixteen years lie between 2009 and 2025 and, with some certainty, it’s possible to estimate the capabilities that will be available in computing and digital communications to support the growing need for observations, observatories and related modeling.
Today, chip density on central processing units (CPUs) is doubling about every 18 months, network speeds double every eight months, and disk capacity doubles nearly annually with little or no increase in cost for a physical unit. For example, it’s now possible to buy a 1.5 TB disk drive for $166.00. In 2025, the same drive at the same cost will store 99 PB of data. The highest speed commercial network available today moves data at 10 GBps; in 2025 this will be 170 Pbps. An Apple Pro 8™ coe desktop computer can compute at 24 billion floating-point operations per second (GFlop). In 2025 the same kind of machine, if it still exists, will provide 24 TFlop. This past March Intel showed off an 80 core CPU and similar technologies may continue to push the envelope in 2025. In terms of supercomputing, the TeraGrid includes computers that can sustain 8 TFlop computations. Today, petascale computers are beginning to become available, a culmination of the TeraGrid that began eight years ago. If the NSF continues this pattern, there will be two more transitions before 2025, first to exascale and then to zettascale computing, 109 times faster than today’s TeraGrid. Geoscience models in 2025 will be computed at scales much more physically interesting than today and the management of the parent environmental data will be straightforward. The workflows needed to support the integrated management of modeling, data, data
OCR for page 48
Oceanography in 2025: Proceedings of a Workshop
assimilation and observational network control, however, will be a major challenge. Oceanography will only be able to occasionally make use of high-speed fiber optical networks, especially at a global scale.
The state-of-the-art in satellite communications today is the NASA Tracking and Data Relay Satellite System (TDRSS) that can support communications including a ground segment at speeds in excess of 150 Mbps using Ku-Band technology. Boeing™ and Lockheed Martin™ e now bidding on the Transformational Communication Satellite Program (TSAT) that will provide future communications for DOD. Like TDRSS, this is a multi-satellite, geosynchronous system that can relay data in space around the planet. Each satellite must be capable of providing multiple RF connections at >45 Mbps and laser communications at 10-100 s Gbps. These systems bound the upper end of communications available to oceanography. Today HiSeasNet, which I started a few years ago, provides shore->ship communications using C and Ku-Band antennae at 256 kbps to as many as five ships and individual ship->shore at 96 kbps. The highest data rate achieved by HiSeasNet was 19 Mbps although the bandwidth is limited by available funding and not by current technology. The system is presently installed on 16 UNOLS ships. Individual buoys, gliders or profilers generally use Iridium satellites with data rates of 2.4 kbps or multiples thereof. Oceanographers should seriously consider the use of commercial C-Band with spread-spectrum technology to provide continuous data rates significantly larger than this and at a fraction of the cost of Iridium.
Finally, long-term measurements of climate-scale phenomena are seriously lacking. If a 30-year climate time scale is used, several cycles must be observed to understand the complex system. The longest instrumental record may be the central England temperature composite going back to 1659. Useful global atmospheric records have been taken since World War I, global ocean measurements began in the 1990s, adequate ice measurements started five years ago, synoptic sea surface temperature measurements from satellites began 30 years ago, atmospheric CO2 measurements are fifty years old, satellite altimetry measurements started 15 years ago and deep ocean physical measurements have only begun. Too few scientists understand how to build long-term instruments capable of addressing climate questions, funding is generally limited, and faculties don’t reward such work. Finally, even fewer scientists and engineers understand how to extend measurements made with a single instrument and a single investigator to a global network required by science and policy over the coming 16 years.
