314
to analyze seasonal trajectories. Increasingly, a greater number and diversity of
spectral bands (e.g., coastal, red edge and short wave infrared) are being provided
by other spaceborne VHR sensors, with the Worldview-2 and 3 sensors being notable examples.
Alongside the release of satellite data archives, algorithms and software for processing dense time series of remote sensing data has also been developed, with these
including those that identify breakpoints and trends associated with change events
or processes (e.g., the TIMESAT program for analyzing time-series of satellite sensor data (Jönsson and Eklundh 2004) or Breaks For Additive Season and Trend or
BFAST (Verbesselt et al. 2010)). Others use past data to establish typical trends,
deviations from which can indicate a change in land cover (e.g., Zhu et al. 2012).
The benefit of these algorithms is that they can be used to better describe and target
analysis of change, which allows greater responsiveness to and management of
adverse events and processes.
Airborne observations have proved to be highly beneficial, if not essential, in the
interpretation of data acquired by spaceborne sensors and derived products and as
standalone sources of information, and key amongst these has been airborne
LiDAR. Whilst discrete return LiDAR have been available for over two decades,
many LiDAR today are operating using full waveform technology (Lim et al. 2003)
and there is a drive towards the use of multi-spectral LiDAR (Morsdorf et al. 2009).
These sensors now have the potential to provide highly detailed information on the
three-dimensional structure of vegetation but also facilitate the discrimination of
plant species as well as component materials (leaves and wood) and the generation
of Digital Surface Models (DSMs). Time-series of these data are also being exploited
to better understand the dynamics of plant communities, with notable successes
being in savanna woodlands (Cho et al. 2012). Digital Terrain Models (DTMs)
generated from these data also provide new opportunities to understand ecological
processes, including loss of sediments through erosion and water flows.
Drone technology has also revolutionized the capacity to obtain very high (centimeter) resolution information that supports the interpretation, calibration and validation of products from spaceborne but also airborne sensors (Colomina and Molina
2014). The data from these sensors fill a significant gap in terms of scaling groundbased measurements to larger areas, with these including distributions of plant species (through classification of RGB or multi-spectral data) or canopy height maps
from point clouds derived through stereo imaging (photogrammetry). Nowadays,
drones (either with fixed or rotatory wings) are available at affordable prices, and
can support a range of sensor types (e.g., RGB and multispectral cameras). As the
weight of hyperspectral and LiDAR sensors decreases, these are also being deployed
on drones allowing, for example, better discrimination of plant species and interrogation of land covers in multiple dimensions. Accordingly, many applications
addressed by spaceborne or airborne images can also exploit drones, with a major
advantage being the possibility to obtain images à la carte. Other new applications
beyond pixel mapping are emerging in relation to fauna detection, surveying and
censoring (e.g. counting nests in bird colonies; Díaz-Delgado’s et al. (2017 - this
book)).
R. Lucas et al.
to analyze seasonal trajectories. Increasingly, a greater number and diversity of
spectral bands (e.g., coastal, red edge and short wave infrared) are being provided
by other spaceborne VHR sensors, with the Worldview-2 and 3 sensors being notable examples.
Alongside the release of satellite data archives, algorithms and software for processing dense time series of remote sensing data has also been developed, with these
including those that identify breakpoints and trends associated with change events
or processes (e.g., the TIMESAT program for analyzing time-series of satellite sensor data (Jönsson and Eklundh 2004) or Breaks For Additive Season and Trend or
BFAST (Verbesselt et al. 2010)). Others use past data to establish typical trends,
deviations from which can indicate a change in land cover (e.g., Zhu et al. 2012).
The benefit of these algorithms is that they can be used to better describe and target
analysis of change, which allows greater responsiveness to and management of
adverse events and processes.
Airborne observations have proved to be highly beneficial, if not essential, in the
interpretation of data acquired by spaceborne sensors and derived products and as
standalone sources of information, and key amongst these has been airborne
LiDAR. Whilst discrete return LiDAR have been available for over two decades,
many LiDAR today are operating using full waveform technology (Lim et al. 2003)
and there is a drive towards the use of multi-spectral LiDAR (Morsdorf et al. 2009).
These sensors now have the potential to provide highly detailed information on the
three-dimensional structure of vegetation but also facilitate the discrimination of
plant species as well as component materials (leaves and wood) and the generation
of Digital Surface Models (DSMs). Time-series of these data are also being exploited
to better understand the dynamics of plant communities, with notable successes
being in savanna woodlands (Cho et al. 2012). Digital Terrain Models (DTMs)
generated from these data also provide new opportunities to understand ecological
processes, including loss of sediments through erosion and water flows.
Drone technology has also revolutionized the capacity to obtain very high (centimeter) resolution information that supports the interpretation, calibration and validation of products from spaceborne but also airborne sensors (Colomina and Molina
2014). The data from these sensors fill a significant gap in terms of scaling groundbased measurements to larger areas, with these including distributions of plant species (through classification of RGB or multi-spectral data) or canopy height maps
from point clouds derived through stereo imaging (photogrammetry). Nowadays,
drones (either with fixed or rotatory wings) are available at affordable prices, and
can support a range of sensor types (e.g., RGB and multispectral cameras). As the
weight of hyperspectral and LiDAR sensors decreases, these are also being deployed
on drones allowing, for example, better discrimination of plant species and interrogation of land covers in multiple dimensions. Accordingly, many applications
addressed by spaceborne or airborne images can also exploit drones, with a major
advantage being the possibility to obtain images à la carte. Other new applications
beyond pixel mapping are emerging in relation to fauna detection, surveying and
censoring (e.g. counting nests in bird colonies; Díaz-Delgado’s et al. (2017 - this
book)).
R. Lucas et al.