1. Canopy nitrogen: a remote sensing and modelling approach
- Author
-
Yasmina Loozen, Jong, S.M. de, Wassen, M.J., Karssenberg, D.J., Rebel, K.T., and University Utrecht
- Subjects
canopy nitrogen ,vegetation index ,remote sensing ,satellite ,modelling ,forest ,grass ,N:P ratio ,random forests ,P ratio ,random forests [canopy nitrogen ,N] - Abstract
In the last decades, CO2 emissions from fossil fuel burning and land use change have caused an increase in the CO2 concentration of the atmosphere. This human induced rise in atmospheric CO2 concentration is extremely likely to be causing climate change, which poses a threat to human communities worldwide. The increase in atmospheric CO2 influences the terrestrial biosphere. Over the last decade, the terrestrial biosphere acted as a carbon sink, absorbing an estimated 29% of the yearly anthropogenic CO2 emissions. The behavior of the terrestrial C sink with future CO2 emission is, however, uncertain. Nitrogen (N) is an essential and limiting nutrient for vegetation growth. Due to the resulting close links between the carbon (C) and N cycles, the N cycle influences the biosphere response to the rise in atmospheric CO2. How the N cycle influences the C cycle in terrestrial biosphere is studied with global vegetation models. Data on the N cycle are needed for the models. However, data on the N cycle are lacking at global scale. Remote sensing methods could provide insights on the spatial pattern of canopy N, defined as the foliar N scaled to the whole vegetation canopy, at large scale. In this context, this thesis explores the possibility to estimate canopy N concentration across scales and vegetation types. This is done under different conditions using vegetation indices (VIs) computed from remotely sensed spectral reflectance. In the first study (Exploring the use of vegetation indices to sense canopy nitrogen to phosphorous ratio in grasses), the reflectance spectra of a grass species, Holcus lanatus L., was measured under controlled conditions in a laboratory experiment. Several VIs were subsequently computed and related to the grass canopy N concentration using linear regressions. In the second study (Remote sensing of canopy nitrogen at regional scale in Mediterranean forests using the spaceborne MERIS Terrestrial Chlorophyll Index), canopy N was estimated in a Mediterranean forest ecosystem at regional scale in Catalonia, Spain, using the MERIS Terrestrial Chlorophyll Index (MTCI) obtained from the MERIS sensor abord ESA-Envisat satellite. In the third study (Mapping canopy nitrogen in European forests using remote sensing and environmental variables with the random forests method), canopy N was estimated in European forests at continental scale using either only satellite-based variables or both VIs and environmental variables as predictors with the random forests method algorithm. In the last study (Comparison of simulated foliage nitrogen by the O-CN and LPJ-GUESS vegetation models with a canopy nitrogen map based on forest sampling), the obtained canopy N map in European forests was compared with foliage N simulated by two global vegetation models, O-CN and LPJ-GUESS, in Europe.
- Published
- 2022