145
The robust and portable field spectroradiometers developed in previous decades
have evolved from the non-imaging spectrometers currently used in the laboratory.
Fibre optic bundles providing different FOV angles has become widely used.
Depending on the application considered, manufacturers offer two kind of spectroradiometer: (1) small and light devices that are designed to work only in the VNIR,
with levels of SNR around 250:1; (2) less small and light devices that work across
the entire solar spectrum, with refrigerated SWIR detectors and an SNR of around
1000:1. Both instrument types have a VNIR’s full FWHM of nearly 3 nm but, for
SWIR instruments, the FWHM is nearly 10 nm.
Methods in field spectroscopy are well described by several authors (Salisbury
1998; Goetz 2012). The basic recommendations for better spectra acquisition are to
measure under cloudless sky, at high sun zenith angles and with the same illumination and atmospheric conditions for the panel and target. The most widely used
acquisition methodology to obtain near-ground reflectance is the single-beam,
where the same instrument is used to measure both the target and the reference
panel spectral radiance. In this case, Spectralon
®
(Labsphere, North Sutton, NH,
USA) has been established as the standard material for panels, due to its high
degree of perfect diffuse and lambertian response. Measurements with field spectroradiometers are often hand-held, usually with the sensor head mounted on a pole
or yoke to keep it away from the operator’s body. In this regard, we should mention
the novel carrier- lift system MUFSPEM@MED (Mobile Unit for Field SPEctral
Measurements at the MEDiterranean, Manakos et al. 2010), which has the capacity
for automated signature acquisition.
Spectral libraries are collections of spectra that characterize the reflectance or
emissivity spectral response of Earth’s surfaces and materials. Characterizing plant
species is always challenging due to variations in vegetation elements, growing
states, and phenology (Asner 1998). A large number of studies have focused on the
acquisition of plant spectra in the field over different vegetation formations, including shrublands (Manevski et al. 2011), marshlands (Zomer et al. 2009), forest
(Somers and Asner 2012) and sub-aquatic environments (Fyfe and Dekker 2001).
Nevertheless, there is no standard protocol for the acquisition of a plants reflectance
spectral response (Pfitzner et al. 2010), so a universally applied methodology for the
collection of field spectra is needed (Manakos et al. 2010).
Spectral Unmixing
When working with mono-specific plant communities, the estimation of plant species cover by remote sensing is better accomplished by sub-pixel algorithms. In
remote sensing imagery, the total signal integrated in a pixel is a function of:
• The optic properties of the components inside the Ground Instantaneous Field Of
View (GIFOV), and the relationships between them. The components contribution can follow a linear mixture or a non-linear mixture model (Keshava and
Mustard 2002);
Sub-pixel Mapping of Doñana Shrubland Species
The robust and portable field spectroradiometers developed in previous decades
have evolved from the non-imaging spectrometers currently used in the laboratory.
Fibre optic bundles providing different FOV angles has become widely used.
Depending on the application considered, manufacturers offer two kind of spectroradiometer: (1) small and light devices that are designed to work only in the VNIR,
with levels of SNR around 250:1; (2) less small and light devices that work across
the entire solar spectrum, with refrigerated SWIR detectors and an SNR of around
1000:1. Both instrument types have a VNIR’s full FWHM of nearly 3 nm but, for
SWIR instruments, the FWHM is nearly 10 nm.
Methods in field spectroscopy are well described by several authors (Salisbury
1998; Goetz 2012). The basic recommendations for better spectra acquisition are to
measure under cloudless sky, at high sun zenith angles and with the same illumination and atmospheric conditions for the panel and target. The most widely used
acquisition methodology to obtain near-ground reflectance is the single-beam,
where the same instrument is used to measure both the target and the reference
panel spectral radiance. In this case, Spectralon
®
(Labsphere, North Sutton, NH,
USA) has been established as the standard material for panels, due to its high
degree of perfect diffuse and lambertian response. Measurements with field spectroradiometers are often hand-held, usually with the sensor head mounted on a pole
or yoke to keep it away from the operator’s body. In this regard, we should mention
the novel carrier- lift system MUFSPEM@MED (Mobile Unit for Field SPEctral
Measurements at the MEDiterranean, Manakos et al. 2010), which has the capacity
for automated signature acquisition.
Spectral libraries are collections of spectra that characterize the reflectance or
emissivity spectral response of Earth’s surfaces and materials. Characterizing plant
species is always challenging due to variations in vegetation elements, growing
states, and phenology (Asner 1998). A large number of studies have focused on the
acquisition of plant spectra in the field over different vegetation formations, including shrublands (Manevski et al. 2011), marshlands (Zomer et al. 2009), forest
(Somers and Asner 2012) and sub-aquatic environments (Fyfe and Dekker 2001).
Nevertheless, there is no standard protocol for the acquisition of a plants reflectance
spectral response (Pfitzner et al. 2010), so a universally applied methodology for the
collection of field spectra is needed (Manakos et al. 2010).
Spectral Unmixing
When working with mono-specific plant communities, the estimation of plant species cover by remote sensing is better accomplished by sub-pixel algorithms. In
remote sensing imagery, the total signal integrated in a pixel is a function of:
• The optic properties of the components inside the Ground Instantaneous Field Of
View (GIFOV), and the relationships between them. The components contribution can follow a linear mixture or a non-linear mixture model (Keshava and
Mustard 2002);
Sub-pixel Mapping of Doñana Shrubland Species