Geoscience Reference
In-Depth Information
ozone hole over Antarctica and the Amazon basin in Brazil. [. . .] IGBP sees the
Earth Observation System as a very positive step towards producing the globally
consistent and reliable data needed for all nations to deal with global change' (IGBP,
2003). In its capacity to deliver essential monitoring information, Earth Observa-
tion supports a wide range of international conventions, such as the UN Framework
Convention on Climate Change and its implementation in the Kyoto protocol,
the RAMSAR convention, the UN Convention to Combat Desertifi cation and the
UN Convention on Biodiversity. This chapter aims to describe selected principles
and applications of remote sensing with a focus on land applications.
Physical Principles of Remote Sensing
Remote sensing methods are based on instruments measuring electromagnetic energy
received from a remote target (e.g., the land surface). The electromagnetic spectrum
encompasses a range of wavelengths. Here, we will describe methods utilising visible
light (wavelengths between 0.4 and 0.75
μ
m), infrared radiation (0.75 to 10
3
μ
m)
and microwaves (10
3
to 10
5
m). We distinguish active from passive remote sensing
systems. Active systems, such as Synthetic Aperture Radar (SAR), illuminate the
target by transmitting electromagnetic energy pulses and recording the return to the
sensor. Passive systems, like imaging spectrometers and multispectral scanners,
record the intensity of electromagnetic radiation originating from an independent
source, usually the sun. Dependence on solar illumination consumes less power than
active sensors but limits the imaging opportunities in the high-latitude winter season
(polar night) and at night-time.
μ
Optical/near-infrared
When observations of the land surface are required for environmental applications,
the imaging process is complicated by the atmospheric pathway that the electro-
magnetic radiation (the sunlight) has to travel before being refl ected from the target
and again on its way back. Absorption of radiation in certain parts of the electro-
magnetic spectrum through atmospheric constituents (water vapour, etc.) can change
the spectral signal. Scattering processes, such as those caused by aerosols, can alter
the direction of the electromagnetic waves, and thus, infl uence the recorded signal.
Such distortion needs to be atmospherically corrected to get the characteristic refl ec-
tance of the target (Leroy and Roujean, 1994; Los et al., 2005).
In the blue, green, red and near-infrared spectral bands, we can distinguish dif-
ferent land surface types based on their characteristic spectral refl ectance properties,
also called spectral signatures. Green plants absorb photosynthetically active radia-
tion primarily in the red spectrum, while refl ectance in the near-infrared spectrum
is almost stable, which leads to a characteristic green refl ectance from vegetation
canopies. These differences have been used to derive vegetation indices, the most
common of which is the Normalised Difference Vegetation Index (NDVI). It is based
on the refl ectance in the near-infrared (NIR) and red (RED) spectrum:
NIR
−
+
RED
NDVI
=
NIR
RED
The NDVI has been found to increase with increasing green biomass.
