University of Twente Student Theses

Login
This website will be unavailable due to maintenance December 1st between 8:00 and 12:00 CET.

Estimating Mangrove Foliar Nitrogen using Spectroscopy and Remote Sensing

Walalite, Tanapipat (2010) Estimating Mangrove Foliar Nitrogen using Spectroscopy and Remote Sensing.

[img] PDF
291kB
Abstract:Numbers of spectroscopic remote sensing methods have been developed for estimation of foliar biochemical at canopy level, but the successes of the estimation are restricted with site and ecosystem specifications. The research aimed to explore the potential of spectral reflectance (laboratory and airborne data) to estimate mangrove foliar biochemical (nitrogen and chlorophyll) at leaves and canopy level. Continuum removal and first-difference transformation methods were applied to reflectance spectra. Stepwise multiple linear regression and cross-validation analysis were employed to find relationships between the transformed spectra and mangrove foliar biochemical. For nitrogen concentration, the best predictive model obtained from first-difference transformation method applied to laboratory spectra (R2cv = 0.58 and RMSEcv = 0.21%dry matter, n=60) and to hyperspectral HyMap data (R2cv = 0.46 and RMSEcv = 0.27%dry matter, n=36). For chlorophyll content (used SPAD values as an indicator), although the best models achieved very high R2cv (0.80 at leaves level, 0.83 at canopy level), but the models included too many number of wavebands (10). The wavebands selected in the regression model to estimated nitrogen concentration at leaves and canopy levels are related to known absorption features of chlorophyll and protein. High humidity and the forest physical structure such as wet soil, water and roots on the forest floor may have strong influence to mangrove canopy reflectance which would explain the slightly lower accuracy than the results from previous studies conducted on other ecosystems with HyMap data. The result models demonstrated a potential to estimate mangrove foliar biochemical. Although the accuracy of the models are slightly lower than the results from other previous studies, but the results are encouraging to applied another techniques to improve the accuracy of the estimation model for mangrove ecosystem. Keywords: Spectroscopy, Hyperspectral, Nitrogen, SPAD, Mangrove
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:https://purl.utwente.nl/essays/90726
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page