ITR: Computational Environment Infrastructure with Applications to Mid Ocean Ridge Research: The 'Virtual Research Vessel' Prototype

东太平洋海隆（EPR）是目前我们研究得最好的快速扩展的洋中脊部分，产生了丰富的观测数据和跨越许多科学领域的结果。然而，这一信息尚未得到充分利用。其中大部分以非交互形式存在（例如期刊出版物）或作为不兼容的数据集和模型。为了充分利用这些数据，科学家将需要广泛的复杂编程支持，以协调分布式计算机网络中数据、计算工具和数值模型的使用。本项目将开发这种支持。从技术上讲，本项目将开发支持数据共享、工具组合和模型耦合所需的计算基础设施，以使用大规模、跨学科的数据档案。然后，它将应用该基础设施来构建一个特定领域的环境，称为虚拟研究船1（VRV-1）。VRV-1将促进与EPR相关的数据、地图和模型的使用。该项目将解决科学数据管理和计算科学的集成中间件的基本问题。特别是，它将通过合并三种技术来支持数据共享：地理信息系统、数据库管理系统和电子笔记本。它将通过扩展现有的电子笔记本来支持工具组合。最后，它将通过在快速原型环境中开发对在非常高的特定于领域的级别上探索模型相关性和关系的支持来支持模型耦合。

The East Pacific Rise (EPR) is currently our best-studied section of fast-spreading mid-ocean ridge, yielding a wealth of observational data and results spanning many scientific fields. However, this information has not yet been fully utilized. Most of it exists in noninteractive form (e.g. journal publications) or as incompatible datasets and models. To make the most of this data, scientists will need a wide range of sophisticated programming support to coordinate the use of data, computational tools, and numerical models across distributed networks of computers. This project will develop that support.Technically, this project will develop the computational infrastructure needed to support data sharing, tool composition, and model coupling for the use of large scale, interdisciplinary data archives. It will then apply that infrastructure to build a domain-specific environment called the Virtual Research Vessel 1 (VRV-1). The VRV-1 will facilitate the use of data, maps, and models related to the EPR. The project will address fundamental issues of integrated middleware for scientific data management and computational science. In particular, it will support data sharing by merging three technologies: geographic information systems, database management systems, and electronic notebooks. It will support tool composition through an extension of an existing electronic notebook. Finally, it will support model coupling by developing support for exploring model correlations and relationships at a very high, domain-specific level in a fast prototyping environment.