Details of Award
NERC Reference : NE/N018176/1
Sampling the environment: design and analysis for efficient and robust collection of data.
Training Grant Award
- Lead Supervisor:
- Professor RM Lark, British Geological Survey, Environmental Modelling
- Grant held at:
- British Geological Survey, Environmental Modelling
- Science Area:
- Atmospheric
- Earth
- Freshwater
- Marine
- Terrestrial
- Overall Classification:
- Atmospheric
- ENRIs:
- Biodiversity
- Environmental Risks and Hazards
- Global Change
- Natural Resource Management
- Pollution and Waste
- Science Topics:
- Applied Statistics
- Environmental Statistics
- Sampling
- Spatial Statistics
- Statistical Estimation
- Statistical Uncertainty
- Statistics & Appl. Probability
- Abstract:
- If data are to be collected in environmental science a sampling strategy is needed: a formal procedure to decide where and when to collect observations. If the analysis of the resulting data is to be valid, then it must be consistent with the sampling strategy. For example, many common statistical methods require that sample sites are selected at random and independently of each other. This is not consistent with sampling on a regular grid, a useful strategy for some problems. Sampling strategies must therefore be chosen in the light of the problem that is to be addressed. They must also be efficient, since the costs of visiting sample sites in the landscape, and undertaking analysis, can quickly mount up. Modern statistical approaches to sample design allow us to choose efficient strategies which use no more effort than necessary. These issues are acute ones for organizations that sponsor sampling by scientists. For example the Environment Agency (EA) sponsors research projects and monitoring exercises. They want to obtain reliable information, but they also want to know that the considerable resource which is put into collecting it is deployed as efficiently as possible. The British Geological Survey has a long-standing interest in these questions. We use modern statistical approaches to planning sampling, we use advanced statistical methods to analyse our archived data and we undertake research on the fundamental underlying statistical problems. We undertake research for industry and government bodies that involves the design and application of efficient sampling schemes. We also contribute to training in these areas through our involvement in research council doctoral training, notably in the STARS Doctoral Training Centre for soil science. Our staff have published in this area, both research papers and a textbook. This puts us in a unique position to offer a training course to PhD students and early-career researchers who want a better understanding of how to undertake sampling and associated data analysis. The course will draw on both our practical experience and our active research interests in the cutting edge of this branch of science. It will also benefit from our experience of undertaking training in this area in the UK and overseas (Germany, Australia, central Africa). A particular strength of this proposed course is the involvement of the Environment Agency (EA). EA staff along with the course leader, will develop case studies for the course based on their experience in environmental regulation. As a culmination of the course participants will develop sampling strategies for these real-world problems, and will analyses associated data. They will then present their proposals and findings to the partners, and receive valuable feedback. The EA partner will also speak to the course participants, explaining their perspectives on environmental sampling. This will complement the theoretical elements of the course, and give the participants a unique insight into how statistical principles enable the environmental scientist to provide information which is reliable, but also collected as efficiently as possible. Participants in this course will be exposed to the cutting edge of statistical science, but also to real-world examples of its application and the interests of leading public bodies. They will be trained in the analysis of data using the widely-used, and free R platform, and provided with code which they can take away and use after the course. This will give them tools to undertake their own research more effectively, and insight into how end users of environmental research view the sampling problem. This will impact both on their current research, and on their ability to contribute as scientists in their future careers.
- NERC Reference:
- NE/N018176/1
- Grant Stage:
- Completed
- Scheme:
- Doctoral Training
- Grant Status:
- Closed
- Programme:
- Advanced Training
This training grant award has a total value of £16,850
FDAB - Financial Details (Award breakdown by headings)
| Total - Other Costs |
|---|
| £16,850 |
If you need further help, please read the user guide.