143
and conservation of natural protected areas have been given the responsibility to
establish long term monitoring programs for terrestrial ecosystems that allow the
inference of trends and rates of change (Vaughan et al. 2001). However, these monitoring programs should be informed by periodic and repeatable measurements
(Schmeller 2008). Standard methodologies allow better comparison with other
maps of natural areas. In the work by Oakley et al. (2003), guidelines for monitoring
protocols are outlined, with these highlighting the importance of the collection,
management, analysis and reporting of the data.
In this study, we present an approach for establishing a protocol for mapping the
distribution of plant species based on airborne imaging spectroscopy. Robust and
widely used methodologies for hyperspectral airborne images and field spectroscopy acquisition, processing and analyzing are selected to generate spatially-explicit
maps of plant species distribution. The aim behind this work is to facilitate programs to monitor ecological communities based on mapping derived from imaging
spectroscopy. These maps will help to interpret shifts in species composition in
response to environmental changes induced by climate and land use change and
other anthropogenic impacts.
We present a practical case study to demonstrate the usefulness of the program
and protocols. In the ecosystem of stabilized sand dunes of Doñana National Park
in south west Spain, the shrub communities are a very important habitat for fauna.
In-depth knowledge of the spatial distribution of the shrub species is also essential
for managing shrubland habitats (Cobo et al. 2002).
Background of Plant Species Mapping Activities Using
Imaging Spectroscopy
Before describing the procedure proposed for mapping plant species in this work,
we should explain some aspects of the key techniques that underpin the protocols:
the characteristics of the airborne imaging spectroscopy, the basis of field spectroscopy data, and the spectral unmixing algorithms applied to the hyperspectral data.
Airborne Imaging Spectroscopy
Airborne remote sensing is characterised by its flexibility in imagery acquisition
conditions and continuous maintenance and calibration of the sensors installed.
These characteristics offer great advantages for the acquisition of seasonal and diurnal
processes (e.g., drought and fire impacts) and can determine the viability of specific
research applications. There are several aerial platforms for imaging spectroscopy
(e.g., balloons and helicopters) and more recently the Remotely Piloted Air Systems
(RPAS), formerly Unmanned Aerial Systems –UAS- (Hruska et al. 2012). However,
non-pressurised aircraft are the most widely used platform due to the better
Sub-pixel Mapping of Doñana Shrubland Species
and conservation of natural protected areas have been given the responsibility to
establish long term monitoring programs for terrestrial ecosystems that allow the
inference of trends and rates of change (Vaughan et al. 2001). However, these monitoring programs should be informed by periodic and repeatable measurements
(Schmeller 2008). Standard methodologies allow better comparison with other
maps of natural areas. In the work by Oakley et al. (2003), guidelines for monitoring
protocols are outlined, with these highlighting the importance of the collection,
management, analysis and reporting of the data.
In this study, we present an approach for establishing a protocol for mapping the
distribution of plant species based on airborne imaging spectroscopy. Robust and
widely used methodologies for hyperspectral airborne images and field spectroscopy acquisition, processing and analyzing are selected to generate spatially-explicit
maps of plant species distribution. The aim behind this work is to facilitate programs to monitor ecological communities based on mapping derived from imaging
spectroscopy. These maps will help to interpret shifts in species composition in
response to environmental changes induced by climate and land use change and
other anthropogenic impacts.
We present a practical case study to demonstrate the usefulness of the program
and protocols. In the ecosystem of stabilized sand dunes of Doñana National Park
in south west Spain, the shrub communities are a very important habitat for fauna.
In-depth knowledge of the spatial distribution of the shrub species is also essential
for managing shrubland habitats (Cobo et al. 2002).
Background of Plant Species Mapping Activities Using
Imaging Spectroscopy
Before describing the procedure proposed for mapping plant species in this work,
we should explain some aspects of the key techniques that underpin the protocols:
the characteristics of the airborne imaging spectroscopy, the basis of field spectroscopy data, and the spectral unmixing algorithms applied to the hyperspectral data.
Airborne Imaging Spectroscopy
Airborne remote sensing is characterised by its flexibility in imagery acquisition
conditions and continuous maintenance and calibration of the sensors installed.
These characteristics offer great advantages for the acquisition of seasonal and diurnal
processes (e.g., drought and fire impacts) and can determine the viability of specific
research applications. There are several aerial platforms for imaging spectroscopy
(e.g., balloons and helicopters) and more recently the Remotely Piloted Air Systems
(RPAS), formerly Unmanned Aerial Systems –UAS- (Hruska et al. 2012). However,
non-pressurised aircraft are the most widely used platform due to the better
Sub-pixel Mapping of Doñana Shrubland Species