|
Project properties |
|
| Title | Automtic quality checks for eddy-covariance fluxes. At KNMI the eddy-covariance fluxes were hand-checked until now. Now they want to automated that, without loosing the high quality of there flux data |
| Group | Meteorology and Air Quality Group |
| Project type | internship |
| Credits | 24 |
| Supervisor(s) | Fred Bosveld (KNMI) |
| Examiner(s) | Prof. dr. AAM Holtslag |
| Contact info | Fred Bosveld (KNMI) |
| Begin date | 2018/01/01 |
| End date | 2019/02/01 |
| Description | To devise and implement procedures to detect (and if needed correct) errors in eddy-covariance data and derived fluxes.
The eddy-covariance method is by now the reference method to determine surface fluxes of heat, water vapour, momentum and CO2. This implies that many end-users make use of those fluxes under the assumption that they give a correct representation of reality. Whereas the idea behind the eddy-covariance method is simple, the reality is that many things can go wrong: instruments that do not function, or function improperly, datalogging equipment that adds information (noise, spikes...), meteorological conditions may disturb the measurements. Until today, the eddy-covariance fluxes as measured at Cabauw by KNMI where quality-checked by hand. While this is probably the best way to do it, it is also the most time consuming (=expensive) one. The objective of this research is to investigate which quality checks can be done automatically, and how they should be implemented. This can be based in part on what other researchers have done already. Is it possible to attain the same quality of fluxes using automated quality control as with manual checking? |
| Used skills | |
| Requirements | Atmosphere Land Interactions + Field Experiment Meteorology and Air Quality, experience with FORTRAN or another programming language (and willingness to learn to use FORTRAN) |