This work aimed to investigate and show how the application of new internet-based technologies - specifically cloud computing - and web portals can facilitate the vision of the virtual observatory. This involved integrating a wide variety of information sources, together with associated information tools and services, to provide answers to big environmental science questions. More specifically, the challenge was to define the architecture and associated principles underpinning the Environment Virtual Observatory portal (EVOp), supporting multi-scale experimentation through an open, extensible infrastructure, and harnessing existing resources such as data and models.
A bespoke cyber-infrastructure was developed for EVO. This followed a hybrid model of both owned and leased hardware resources. This cloud-based architecture offers a number of advantages that translate to low operational costs at the infrastructure level and very high flexibility at the application level.
Such flexibility is key in order to support a collection of highly customised web-based tools designed to enable users from different backgrounds to access data related to different environmental issues. This allows subject specialists to compose workflows without any programming prerequisites, and also supports access for non-specialist users through a feature-rich web portal.
The importance of the cyber-infrastructure is that it enables a new kind of science: i) a more integrative science that encourages the bringing together of environmental assets, such as consuming datasets from different sources and at various scales, and combining models to create more sophisticated and collaborative software services; and ii) a more open science where such assets can be accessed and added to by all stakeholders including the general public.
This work built an infrastructure that demonstrates the potential of a virtual observatory in the field of environmental sciences. Besides the infrastructure, the portal also provided a virtual space that showcased – through the use of well-defined storyboards – how facilitating different scientific techniques and datasets might provide tools that are useful for decision and policy makers, as well as for local communities and members of the general public who are interested in environmental issues. More importantly, the lessons learned in assembling the infrastructure and developing the end-user tools are of benefit to other communities looking to assemble similar virtual spaces.
Future work will combine cloud computing with an emerging “Internet of Things” technology to provide a comprehensive cyber-infrastructure enhanced with access to a variety of real-time data sources. The work is also feeding into a £13m NERC Environmental Big Data Initiative to provide a cloud infrastructure for UK Environmental Science.