Monitoring vegetation dynamics represents a fundamental practice to evaluate the response of the vegetation to environmental changes. Optical proximal sensed data allow the monitoring of the temporal and spatial variability of vegetation properties under natural conditions. Vegetation optical properties can be used to indirectly derive information about the phenological and/or physiological status of the plants. Optical sensors commonly used in the field can be divided into broadband sensors (usually multispectral, e.g. RGB cameras) and narrowband sensors (both multispectral and hyperspectral devices, e.g. spectroradiometers). The potential applications of these two categories of sensors differ. While broadband sensors have been applied in the last years to track the phenological development of the vegetation, spectroradiometers have been shown to be suitable also for the characterization of plant physiological status. In fact hyperspectral systems provide a more detailed optical characterization of the analysed targets, nevertheless the systems have to be accurately characterized in terms of spectral and radiometric performances in order to obtain repeatable and comparable. The first part of the research has been addressed to the determination of the sources of uncertainty of the optical measurement systems. Several of the most common optical devices available on the market have been characterized and compared. A particular attention has been paid to instrumental differences in the optical components which could affect the radiance measurements. Moreover the performances of spectroradiometers have been evaluated both indoors and outdoors in order to evaluate the impact of sensor characteristics on the estimation of parameters commonly used in vegetation studies. The analysis has been initially focused on the effect of different cosine receptors on the estimation of biochemical and biophysical properties of the vegetation, such as leaf area index and chlorophyll content. In a second step the analysis referred to the impact of instrumental characteristics (mainly spectral resolution and signal to noise ratio) on the estimation of the absolute value of sun-induced chlorophyll fluorescence. The results obtained suggest that instrument components affect the measurements and according to the required parameter estimations accuracy some instruments are more suitable rather than others. The second part of the study has been focused on the use of proximal sensors to monitor the dynamics of terrestrial vegetation. For this purpose both broadband and narrowband sensors have been separately considered. The first analysis referred to the possibility of using broadband imaging systems to investigate the temporal and spatial phenological dynamics of an alpine ecosystem. The phenological analysis has been conducted using a 3-year time series of digital RGB images collected in a grassland site. This spatiotemporal analysis provided interesting insights into the role of plant species composition on phenology in complex ecosystems, such as alpine grasslands. This study case indicates the potential of using RGB digital cameras as a tool for long term phenological monitoring, allowing the spatial characterization of the investigated ecosystem. The possibility of using hyperspectral narrowband sensors to detect vegetation physiological changes was also evaluated. In particular, this analysis focused on the use of sun-induced chlorophyll fluorescence for the early detection of vegetation stress. The study was conducted during a controlled experiment designed to modify the functional status of actual photosynthesis. The results indicate that fluorescence is immediately affected by physiological changes as a demonstration that such estimates can be used to track physiological traits better than traditional remote sensing techniques based on optical broadband vegetation indices.

(2015). Optical proximal sensing for vegetation monitoring. (Tesi di dottorato, Università degli Studi di Milano-Bicocca, 2015).

Optical proximal sensing for vegetation monitoring

JULITTA, TOMMASO
2015

Abstract

Monitoring vegetation dynamics represents a fundamental practice to evaluate the response of the vegetation to environmental changes. Optical proximal sensed data allow the monitoring of the temporal and spatial variability of vegetation properties under natural conditions. Vegetation optical properties can be used to indirectly derive information about the phenological and/or physiological status of the plants. Optical sensors commonly used in the field can be divided into broadband sensors (usually multispectral, e.g. RGB cameras) and narrowband sensors (both multispectral and hyperspectral devices, e.g. spectroradiometers). The potential applications of these two categories of sensors differ. While broadband sensors have been applied in the last years to track the phenological development of the vegetation, spectroradiometers have been shown to be suitable also for the characterization of plant physiological status. In fact hyperspectral systems provide a more detailed optical characterization of the analysed targets, nevertheless the systems have to be accurately characterized in terms of spectral and radiometric performances in order to obtain repeatable and comparable. The first part of the research has been addressed to the determination of the sources of uncertainty of the optical measurement systems. Several of the most common optical devices available on the market have been characterized and compared. A particular attention has been paid to instrumental differences in the optical components which could affect the radiance measurements. Moreover the performances of spectroradiometers have been evaluated both indoors and outdoors in order to evaluate the impact of sensor characteristics on the estimation of parameters commonly used in vegetation studies. The analysis has been initially focused on the effect of different cosine receptors on the estimation of biochemical and biophysical properties of the vegetation, such as leaf area index and chlorophyll content. In a second step the analysis referred to the impact of instrumental characteristics (mainly spectral resolution and signal to noise ratio) on the estimation of the absolute value of sun-induced chlorophyll fluorescence. The results obtained suggest that instrument components affect the measurements and according to the required parameter estimations accuracy some instruments are more suitable rather than others. The second part of the study has been focused on the use of proximal sensors to monitor the dynamics of terrestrial vegetation. For this purpose both broadband and narrowband sensors have been separately considered. The first analysis referred to the possibility of using broadband imaging systems to investigate the temporal and spatial phenological dynamics of an alpine ecosystem. The phenological analysis has been conducted using a 3-year time series of digital RGB images collected in a grassland site. This spatiotemporal analysis provided interesting insights into the role of plant species composition on phenology in complex ecosystems, such as alpine grasslands. This study case indicates the potential of using RGB digital cameras as a tool for long term phenological monitoring, allowing the spatial characterization of the investigated ecosystem. The possibility of using hyperspectral narrowband sensors to detect vegetation physiological changes was also evaluated. In particular, this analysis focused on the use of sun-induced chlorophyll fluorescence for the early detection of vegetation stress. The study was conducted during a controlled experiment designed to modify the functional status of actual photosynthesis. The results indicate that fluorescence is immediately affected by physiological changes as a demonstration that such estimates can be used to track physiological traits better than traditional remote sensing techniques based on optical broadband vegetation indices.
COLOMBO, ROBERTO
Proximal sensing, Field spectroscopy, Sun-induced chlorophyll fluorescence, Land surface phenology
GEO/10 - GEOFISICA DELLA TERRA SOLIDA
English
24-feb-2015
Scuola di dottorato di Scienze
SCIENZE AMBIENTALI - 09R
27
2013/2014
open
(2015). Optical proximal sensing for vegetation monitoring. (Tesi di dottorato, Università degli Studi di Milano-Bicocca, 2015).
File in questo prodotto:
File Dimensione Formato  
phd_unimib_724794.pdf

accesso aperto

Descrizione: Tesi dottorato
Tipologia di allegato: Doctoral thesis
Dimensione 3.5 MB
Formato Adobe PDF
3.5 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/70505
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
Social impact