A HYDRATE DATABASE: VITAL TO THE TECHNICAL COMMUNITY

Natural gas hydrates may contain more energy than all the combined other fossil fuels, causing hydrates to be a potentially vital aspect of both energy and climate change. This article is an overview of the motivation, history, and future of hydrate data management using a CODATA vehicle to connect international hydrate databases. The basis is an introduction to the Gas Hydrate Markup Language (GHML) to connect various hydrate databases. The accompanying four articles on laboratory hydrate data by Smith et al., on field hydrate data by Löwner et al., on hydrate modeling by Wang et al., and on construction of a Chinese gas hydrate system by Xiao et al. provide details of GHML in their respective areas .


INTRODUCTION: MOTIVATION FOR A HYDRATE DATABASE
Recent analyses by Bernstein (2004) and Sachs (2005) suggest that technology is a major driver of societal economies. In turn energy is a foundation of technology, and efficient access to data will enable effective use of energy. The thesis of this work is that an international hydrate database will enable efficient access to an important new energy source, and that the development of Gas Hydrate Markup Language (GMKL) is the first step in connecting the international databases.
Hydrate knowledge is expanding exponentially, as shown in the semi-logarithmic plot of Figure 1 (Sloan, 2004). This plot gives the number of refereed hydrate publications in the last century, by decade; there were two publications from 1900-1910 and 3010 publications from 1990-2000. A semi-logarithmic extrapolation into the first decade of the 21 st century leads to a staggering 7,500 publications -slightly more than two refereed publications per day, seven days per week.  (Sloan, 2004) With the above motivation and the expanding knowledge base, it is apparent that we need an efficient means of managing hydrate data. Through its multi-disciplinary worldwide network, CODATA has provided a convenient, internationally-sanctioned means of dealing with such information.

CODATA HISTORY OF HYDRATE DATABASE DEVELOPMENT
In 2000 CODATA authorized the first hydrate database Task Group, chaired by F. Kuznetsov. During this phase the Task Group met twice to organize the project and publicized the first local database, GASHYDAT (2001) developed by Drs. Klerkx and Dimitrov, with European Union funding. While the latter effort represents a landmark in hydrate databases, the database was initially small; it was limited by the funding, and work stopped in 2001. The GASHYDAT website domain has recently expired.
In 2002 a second CODATA Hydrate Task Group was again authorized, chaired by F. Kuznetsov. During the ensuing two years the Task Group held six world regional meetings shown in Table 1 to educate the hydrate community regarding the need for a database and to gain worldwide acceptance of the idea of an internationallydistributed database. The project was supported by several National and International bodies such as ICSU 1 .  Dallimore and Collett (2005). At the top of Figure 2 is a Client who may request some information about hydrates. This client query is sent to a Portal, which communicates with the various databases through a common language -the Gas Hydrate Markup Language. Thus the client's query and response is communicated with each participating database, so that a number of data responses are obtained.

HOW DOES THE GAS HYDRATE MARKUP LANGUAGE WORK
At this point, the reader may be asking, "Why do we need another Markup Language? Isn't this a fairly specific or niche application -too narrow to be general?" It was desirable to use an existing Markup Language, but the field of hydrates is so general that it incorporates segments of several markup languages, without the entire language of any other. For example, when we studied the IUPAC standard Markup Language for thermodynamic data (ThermoML, as described by Frenkel, et al., 2006) there was no capability for inserting geological and geophysical data, such as well logs, which seemed vital to hydrates. Similarly other Markup Languages, were missing components for thermodynamic and kinetic data, so in the end, a new language GHML was justified. However, a study of the existing Markup Languages provided valuable guidelines for our construction of GHML.

PUBLISHING GHML AND THE WAY FORWARD
The initial outline of the Gas Hydrate Markup Language was generated at the Computer Network Information Center at the Chinese Academy of Science ( (1) a laboratory data schema by Smith et al. (2007), (2) a field data schema by Löwner et al. (2007), and (3) a modelling/simulation schema by Wang and Moridis et al. (2007). The reader is referred to these three accompanying articles for the schema details of GHML.
For the first six months of 2007, comments from the user community will be solicited by placing the GHML, including a glossary and documentation, on the CODATA website. Revisions will be made to address the comments received. We anticipate an electronic, ever green publication of GHML, in which each revision will update, but not outdate, the earlier versions.
The next step for this project is to generate the Portal shown in Figure 2, as a means of connecting the client(s) to the databases via GHML. That is the intention of the CODATA Hydrate Task Group for the calendar year 2007. At the same time, national databases are continuing to be developed in parallel to this effort. The development of the GHML and Portal by CODATA should act to both enable and encourage these databases.

CONCLUSION
A Beta version of the Gas Hydrate Markup Language (GHML) has been developed and published in three parts: (a) laboratory data, (b) field data, and (c) simulations. This is a first step, to be followed in 2007 by the development of a Portal to connect the hydrate databases which are growing in various parts of the world. Individual database developments are proceeding in parallel to this connection effort, with CODATA encouragement to use GHML as a common language.