Details of Award
NERC Reference : NE/M010171/1
Consistent monitoring of the Earth land surface from EO
Training Grant Award
- Lead Supervisor:
- Professor PE Lewis, University College London, Geography
- Grant held at:
- University College London, Geography
- Science Area:
- Terrestrial
- Overall Classification:
- Terrestrial
- ENRIs:
- Biodiversity
- Environmental Risks and Hazards
- Global Change
- Natural Resource Management
- Pollution and Waste
- Science Topics:
- Biogeochemical Cycles
- Earth Surface Processes
- Ecosystem Scale Processes
- Land - Atmosphere Interactions
- Abstract:
- Aim. Develop and test new approaches for monitoring vegetation disturbance (fire, pest etc.) from Earth Observation (EO). We see such EO monitoring in this context as mainly one of observational opportunity (limited by cloud cover and orbital/sensor sampling) and so seek to maximise this using data from a range of sensors and scales within a data assimilation (DA) framework. The work will test hypotheses and models relating observations from different sensors and develop methods and tools to ingest optical EO to update land surface state and detect and attribute sudden changes. The improvements offered by such an integrated system will be quantified by comparison with independent measures. Context. Disturbance of vegetation involves the influence of external factors (such as fire or pest) on the state of the vegetation. It has a large impact on ecosystem functioning and the transport of carbon etc (e.g. fire emissions to the atmosphere). It is only poorly understood and monitored but will likely play an increasing role in the Earth System under climate change. EO data provide the only realistic route to routine monitoring of such phenomena, with measurements in different parts of the EM spectrum providing complementary information for monitoring. Current EO methods rely on either a direct observation of the disturbance event (e.g. active fire detection) or evaluating evidence that a sudden change has taken place in the land surface (e.g. from optical EO before and after the disturbance event). Exisiting optical approaches use data from a single EO sensor type to facilitate any comparison of pre- and post-event spectral characteristics over a given spatial support. Some approaches attempt post hoc merging of 'detections' from different sensors, but these have not made good use of the information available. Method. Sensor inter-operability needs to deal with angular, spectral and spatial variations in the observations, and studies on methods to reconcile data over these domains will form the main focus of the work. The main methods are: (i): investigate physically-based and empirical evidence for combining data at different angles and wavebands, including analysis of the vast quantities of optical data collected this century (MODIS/MERIS) to give a DA system for sensors with similar spatial characteristics for change detection on the combined data; (ii) examine spatial constraints such as regularisation to the spatial scaling of EO data, exploiting the vast amounts of existing data (e.g. MODIS/SPOT/Landsat), giving a multi-resolution state estimation DA system, able to ingest data from sensors with different spatial resolutions (such as Sentinel-2 and -3 data); (iii) combine (i) and (ii) into a generic multi-scale optical EO DA system that can detect sudden changes in the land surface. The Bayesian framework developed will allow the impact of other sources of information (e.g. active fire detections) to be incorporated as constraints. The tool will be tested using several case studies and the ability to monitor and accurately map disturbance compared with independent data sources (e.g. fire radiative energy from active fire detections) over a range of biomes. The empirical analyses will be the first of their kind exploiting these large data records, and should also provide information to test the application of various constraints to the land surface state estimation.
- NERC Reference:
- NE/M010171/1
- Grant Stage:
- Completed
- Scheme:
- DTG - directed
- Grant Status:
- Closed
- Programme:
- Industrial CASE
This training grant award has a total value of £93,122
FDAB - Financial Details (Award breakdown by headings)
| Total - Fees | Total - Student Stipend | Total - RTSG |
|---|---|---|
| £16,587 | £65,538 | £11,000 |
If you need further help, please read the user guide.