| Author: | Irteza, Syed Muhammad |
| Title: | Forest stress analysis in Hong Kong's forested water catchments using satellite-based fluorescence |
| Advisors: | Nichol, Janet (LSGI) Shi, Wenzhong John (LSGI) |
| Degree: | Ph.D. |
| Year: | 2021 |
| Subject: | Chlorophyll -- Analysis Fluorimetry Plants -- Effect of stress on Vegetation monitoring -- China -- Hong Kong Hong Kong Polytechnic University -- Dissertations |
| Department: | Department of Land Surveying and Geo-Informatics |
| Pages: | 114 pages : color illustrations |
| Language: | English |
| Abstract: | Measurement of Sun-induced Chlorophyll Fluorescence (SIF) from a vegetation canopy, gives insight into physiological perturbations that might cause stress in vegetation. Because it can indirectly measure photochemical processes and carbon sequestration within the vegetation, SIF can be used as an early indicator of plant stress, and hence is a reliable tool to monitor overall vegetation health. Plants emit Fluorescence from Chlorophyll-a which is present in plant leaves, in the form of an electromagnetic signal, when they are exposed to light. However, SIF is only a small fraction, (approximately 1-2%) of total radiation emitted by green plant leaves. The emission spectrum of fluorescence peaks at two broadband spectral regions which are centered at 685 nm (Red) and 740 nm (Far-Red). Recently the scientific community has shown great interest in satellite-based monitoring of plant physiological and environmental changes by the detection of SIF, and several methodologies have been developed to assess the factors that contribute to plant stress. The main aim of this research is to calculate the SIF quantitatively using available hyperspectral satellite imagery as well as field-based spectrometers over Hong Kong's vegetated areas, in order to assess the vegetative health of the region and recommend the best SIF measurement methodology. In this study fluorescence is estimated from Hyperion satellite images, in the atmospheric absorption band O2A (far red wavelengths) around ~760 nm. Field surveys were conducted from May 2015 to September 2018 by using portable hyperspectral and multispectral sensors in three different structural classes of vegetation (forest, shrubland, and grassland) at Tai Mo Shan in Shing Mun Country Parks, and Kowloon reservoir. The field plan was developed while considering cloud conditions, seasonality of vegetation, and satellite overpass. Vegetation spectra were collected over different seasons at sample points dispersed in the study area. Field samples were also obtained at morning and evening times, as well as over the whole day at 20-minute intervals. The field data were used to relate ground observations with satellite observations. On-demand satellite data from NASA were also acquired to increase the number of coincident observations. High resolution Planet data were also processed monthly month, to set the field and satellite data in context of seasonal phenological changes in the study area. Two methods of fluorescence retrieval were applied to both satellite data and field data, namely the Barun Method extension FLD-M and the Fraunhofer Line Difference (FLD) method. Fluorescence results were also compared with NDVI, and analysed according to vegetation seasonality, natural vegetation structural classes, exotic plantations and five age groups of forest having median ages of 7, 20, 39, 61,70 years. Results indicate that the SIF methods are more sensitive to phenological change than NDVI. The SIF methods respond earlier and stronger, to senescence in winter and greening up in summer- which is also useful for monitoring disease in croplands and forests, as early action is needed to combat disease. Diurnal analysis from the field survey spectrometer showed that NDVI seems unresponsive to time of day (morning to evening), whereas both fluorescence methods respond to sunlight intensity - and the FLD method is more sensitive than FLD-M to the changes over the day. Both seasonal and diurnal results indicate that Fluorescence is a much better method than NDVI for measuring subtle changes in vegetation health. Although the NDVI shows differences between different structural stages of vegetation, with gradual decrease from woody vegetation to grassland, forest and exotic plantations in the study area do not show any significant difference. However, fluorescence shows very significant differences between plantations and forest, and the difference is greater in dry season. FLD-M method showed better separation between plantations and forest than the FLD method Additionally, during the end of greening season (October), the NDVI does not show significant differences between Forest, Open Forest, Shrubland and Open Shrubland, while the FLD method clearly distinguishes between the four woody structural stages of vegetation in the study area. The FLD method shows a continuous and significant increase in fluorescence emission with age up to 61 years, and the old growth forest shows minimum fluorescence. This decrease in photosynthesis for the old growth forest is much greater for the FLD retrieval than for NDVI and FLD-M. This study contributes to understanding of the dynamic functioning of vegetation and photosynthesis mechanisms. It also provides an improved operational remote sensing methodology for scientific investigation of the forest health status of Hong Kong's Country Parks. |
| Rights: | All rights reserved |
| Access: | open access |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/10986