|
Project properties |
|
| Title | Predicting functional traits (leaf and branch traits) using spectrometry in the Atlantic forest
|
| Group | Forest Ecology and Forest Management Group |
| Project type | thesis |
| Credits | 24-36 |
| Supervisor(s) | Rens Brouwer |
| Examiner(s) | Prof. dr. Frans Bongers |
| Contact info | Rens.brouwer@wur.nl
|
| Begin date | 2022/03/01 |
| End date | |
| Description | MSc thesis / MSc internship /
The linking of individual functional traits to ecosystem processes used extensively in ecology, however the measurement of individual trait values requires extensive measurements and is time consuming. Recently, advances have been made in the up-scaling of trait mapping through the use of spectrometry. Spectrometry is the recording of light properties after irradiation of an object or substance. If we apply this on tree leaves, the technique allows us for trait data to be inferred, since the reflectance, transmittance, and absorbance of light depend on the size, density, and shape of plant tissues and the content of chemical components. We have hyperspectral data set for over 200 tree species, and for 6 leaves per species. These species are growing in natural regeneration, restoration plantations and old growth forest in the Atlantic Forest region of Brazil. For the same species we have a data set of 7 commonly measured functional traits (e.g. SLA, LDMC, Wood density etc.). During this thesis you will help collect more trait and reflectance data in Brazil and use machine learning (PSLR) to predict leaf traits from reflectance data. This thesis will be part of the NewFor project: https://www.wur.nl/en/project/Understanding-restored-forests-for-benefiting-people-and-nature-in-the-Atlantic-Forest.htm Topics (Choose appropriate topic(s) from list): Biodiversity and functional diversity/ Forest restoration and succession Region(s) (choose): the Netherlands/ / America's/ Climate(s) (choose): Tropical zone |
| Used skills | Statistical skills (R) |
| Requirements | Standard for MSc thesis/internship:
FEM-30306 Forest Ecology and Forest Management and REG-31806 Ecological Methods I Recommended: GRS-20306 Remote Sensing for basic understanding of image spectroscopy |