Research Themes

The ReSC's research fell mainly into four main themes. More information about specific projects can be found through the links below, or on our Projects page. Almost all projects were highly collaborative, multidisciplinary and multi-institutional in nature.

Visualization

Visualization is a key technique for understanding and communicating complex datasets. However, as datasets grow exponentially in size, the process of visualizing data and interacting with it becomes ever harder. The ReSC addressed this problem by developing software that enables scientists and other data users to visualize large datasets interactively on the web, hiding all the complexity of generating visualizations behind a simple interface. Recently we have installed a videowall facility that allows very large datasets to be brought together, visualized and compared interactively using a large touchscreen interface.

Combining diverse sources of data

Environmental science is a highly multidisciplinary subject and scientists must increasingly combine data from different sources to solve a problem. End-users of environmental information (such as decision-makers in government and industry) must do the same, and usually need to use environmental data in concert with other sources of information such as population statistics and socioeconomic data. In the MELODIES project (coordinated by ReSC) we are using Linked Data and cloud computing techniques to develop new environmental services that combine multiple data sources. Open standards are the key to this work: if data are not presented in widely-understood formats through appropriate interfaces then the task of combining them becomes prohibitively difficult.

Sharing information between communities

A characteristic feature of the environmental sciences is that datasets can be used for many different purposes, and many different researchers will work collaboratively to understand the same datasets. Typically the results of these investigations are published in the peer-reviewed literature, but this body of knowledge contains only part of the knowledge accumulated within the community. Furthermore, the academic literature can be inaccessible (or incomprehensible) to many communities. The CHARMe project is collecting and sharing information (called "commentary information") about climate datasets from the literature, blogs, websites and other sources and linking it to data-sharing websites so that new users can more easily discover the strengths and weaknesses of the data for their purpose. Linked Data techniques and open Web standards are being used to connect diverse sources of information in a web of data. The project builds on previous work from the BlogMyData project.

Data quality and uncertainty

When deciding on which datasets are most appropriate for a given application, the user must understand the quality of the data. Data quality is a multifaceted area, including aspects such as positional accuracy, provenance (i.e. quality of the data processing chain), consistency and uncertainty. In the GeoViQua project, the ReSC developed new techniques for sharing and visualizing information about data uncertainty using open standards and the ncWMS software. These were applied to the Global Earth Observation System of Systems (GEOSS), which provides access to environmental data from all over the world.