The NERC Research Experience Placement scheme
Research Experience Placements (REPs) are paid Summer internships for Home/EU undergraduates from any university who are studying Mathematics, Physics, Chemistry, Engineering, Computing, or other quantitative disciplines, who wish to gain experience of research in the Environmental Sciences.
Successful undergraduate applicants are paid a stipend of at least £200 per week for a placement of 8-10 weeks during the Universities' summer recess. Up to £500 funding is also available to supervisors to use for expenses incurred by the research project. Successful candidates will later be encouraged to apply for a NERC-funded PhD in the Environmental Sciences.
Applications are open for EnvEast's 2018 REP scheme.
Candidates may apply for up to two EnvEast REP projects (see below) each year. To apply, candidates should send a C.V., a statement of up to 200 words stating which project you wish to apply for and why, and the name of a member of staff who can vouch for your academic abilities (e.g. your academic adviser/tutor).
The deadline for applications is Tuesday 15 May 2018.
Any enquiries about the EnvEast REP scheme should be sent to David/Alison at firstname.lastname@example.org.
2018 EnvEast Research Experience Placement projects:
2. Choose up to two projects from the list below.
3. Apply by sending a CV (to include an academic referee) and covering letter to email@example.com, by 27 May 2016. Separate covering letters should be sent for each project applied for.
One candidate for each project will be selected by supervisors to be put forward for a short interview by an independent panel at UEA in Norwich, on 10 June 2016. Telephone/Skype interviews will be available. At least four top ranked candidates will then be offered placements (possibly more if funding is available).- See more at: http://webcache.googleusercontent.com/search?q=cache:safIQf0c30cJ:www.enveast.ac.uk/research-experience-placements/information-for-applicants+&cd=2&hl=en&ct=clnk&gl=uk&client=firefox-b-ab#sthash.ZutLtDL3.dpuf
Deriving reflectance measures from hyperspectral UAV imagery using a modelling approach
Lead supervisor: Dr France Gerard, Earth Observation, Climate System group.
Location of the Internship: Centre for Ecology and Hydrology, Crowmarsh Gifford, Wallingford, Oxfordshire.
Duration (start date TBC and by arrangement with the supervisors): 10 weeks
The objective is to develop an approach that delivers reflectance from hyperspectral drone data that does not require the use of reference sheets.
Background of the project
Optical remote sensing collects and analyses the solar radiance (electromagnetic radiation) that is reflected from the Earth’s surface to evaluate the status of the land surface or its vegetation. Reflected radiances (i.e. radiance) vary with solar illumination conditions (i.e. irradiance) and the optimal method to account for changes in irradiance during data acquisition, is by calculating and using the relative proportion of irradiance that is reflected (i.e. reflectance = radiance/irradiance).
UAVs (unmanned aerial vehicle or drone) are used to collect hyperspectral reflected radiances to provide information on vegetation status and health. To allow for changing illumination conditions during the UAV acquisitions, the resulting image frames are typically normalized relative to one another. However this approach does not produce reflectance values, required for multi-temporal studies. A conversion to reflectance can be achieved through a simple empirical calibration, using reference reflectance sheets deployed in situ, but this approach does not allow for changes in illumination conditions that occur during an UAV campaign.
In July 2017 we flew an UAV, carrying a hyperspectral camera, across a woodland canopy to evaluate if hyperspectral UAV imaging could be used to detect diseased oak trees. The UAV camera, a Rikola, collects radiance for ~ 30 bands across 500nm to 900nm. It also continuously collects relative irradiance data integrated across 500nm to 900nm. The idea is to calculate solar irradiance, as it is changing during the UAV campaign, by combining the relative irradiance data collected by the Rikola with a look-up-table of modelled (using a radiative transfer model) solar irradiance spectra. Leaf level reflectance spectra obtained from sampled oak trees show a clear but subtle difference between diseased and healthy trees. By deriving canopy level reflectance from the UAV acquisitions, you will enable the comparison between leaf and canopy level data.
What will you do?
- Carry out a short literature review and online search for the most appropriate radiative transfer model (e.g. SMART)
- Develop a strategy to adapt simulated irradiance spectra (generally developed for clear sky conditions) to represent cloudy conditions following the example of Nann and Riordan 1991 or others.
- Produce a look-up table of (1) irradiance spectra simulated for cloudy conditions and (2) integrated irradiances (500nm-900nm) derived from these simulated irradiance spectra.
- Develop a code using e.g. R, C, or python that will for each Rikola image pixel:
- Compare the simulated integrated irradiances with the Rikola measured integrated irradiance and select the simulated irradiance spectrum for which the integrated irradiance shows the best match.
- Calculate a reflectance spectrum using the Rikola radiance observation and the selected simulated irradiance (i.e. reflectancewaveband = radiancewaveband/irradiancewaveband)
- Test and run the code using the available UAV imagery acquired in early August 2017
- Participate in a planned UAV campaign in early August 2018 to collect further imagery of the sampled trees.
- Implement your code using the UAV imagery acquired in early August 2018.
- Finally, you will compare canopy reflectance spectra with leaf reflectance spectra.
What will you learn?
In CEH, research groups are multi-disciplinary. Also, in CEH research is very much a team effort. By being part of such a group you will be exposed to a variety of research, have access to a wide range of expertise and experience a team based research organisation. You will be exposed to UAV remote sensing, gain specialised understanding on how UAV imagery is being collected and pre-processed, and learn how electromagnetic radiation interacts with vegetation. The project is designed to allow you to investigate, take initiative and develop problem solving skills. You will acquire skills in programming, in particular how to manipulate multi-dimensional binary data. To help increase your presentational skills you will also be given the opportunity to present your work at group meetings.
We are looking for a student who is studying for a physics degree and who has a good mathematical background. An aptitude and interest in coding is essential.
Enquiries about the project are welcome: Dr France Gerard, firstname.lastname@example.org; 01491 692383.
Please check your eligibility before applying. You can apply for up to TWO projects from those advertised on this website by sending a C.V., a statement of up to 200 words stating which project you wish to apply for and why, and the name of a member of staff who can vouch for your academic abilities (e.g. your academic adviser/tutor).
Applications should be sent to email@example.com by Tuesday 15 May, 2018.