182
8.4 Case Studies
8.4.1 Remote Sensing of Belowground Processes via Canopy
Chemistry Measurements
Plants act as aboveground signals for belowground systems. As such, RS of plant
spectra can provide information about belowground systems. Plant spectra can provide a wealth of biological information important to plant physiology and community and ecosystem processes across multiple spatial scales (Cavender-Bares et al.
2017). Some researchers have used direct spectral measurements (e.g., NIRS) for
direct measurements of soil characteristics (reviewed by Stenberg et al. 2010;
Bellon-Maurel and McBratney 2011; Soriano-Disla et al. 2014), and there are
limited examples of remotely sensed spectroscopic measurements of soils (reviewed
by Ustin et al. 2004; Cecillon et al. 2009). Here we focus on remotely sensed spectral measurements of plant communities as a surrogate for belowground processes.
The optical surrogacy hypothesis (sensu Gamon 2008) argues that plant spectra can
serve as a surrogate for important belowground processes.
Direct spectral measurements have been used to assess belowground processes for
decades. For instance, direct NIRS of leaf litter can be used to predict decomposition
rates in a variety of systems (Gillon et al. 1993; Gillon et al. 1999; Shepherd et al.
2005; Fortunel et al. 2009; Parsons et al. 2011). RS of canopy traits to predict belowground processes is becoming increasingly useful. Spectroscopic measurement of
δ
15
N is of particular interest for ecosystem processes (Serbin et al. 2014) because
stable N isotopes can provide important information regarding ecosystem N cycling
(Robinson 2001; Hobbie and Hobbie 2006). RS of forest disturbance (e.g., fire severity) and subsequent belowground processes is relatively common (e.g., Holden et al.
2016). Sabetta et al. (2006) used hyperspectral imaging to predict leaf litter decomposition across four forest communities. Fisher et al. (2016) were able to distinguish
between arbuscular and ectomycorrhizal tree-mycorrhizal associations using spectral
information gleaned from Landsat data. While the above examples focus on remotely
sensed spectral information, remotely sensed forest structural information developed
from lidar data can also provide information about belowground systems, as Thers
et al. (2017) were able to use remotely sensed lidar data to estimate belowground
fungal diversity. The growing number of examples that employ remotely sensed data
to provide information about belowground systems points to the potential of plant
spectra to be used as surrogates for ecosystem processes.
8.4.2 Forest Systems: Aspen Clones Example
An example of optical surrogacy in practice is illustrated by work completed in
trembling aspen (Populus tremuloides) systems across the Western and Midwestern
USA. Trembling aspen is the most widespread native tree species in North America
(Mitton and Grant 1996) and is an ecologically important foundation species across
M. Madritch et al.
8.4 Case Studies
8.4.1 Remote Sensing of Belowground Processes via Canopy
Chemistry Measurements
Plants act as aboveground signals for belowground systems. As such, RS of plant
spectra can provide information about belowground systems. Plant spectra can provide a wealth of biological information important to plant physiology and community and ecosystem processes across multiple spatial scales (Cavender-Bares et al.
2017). Some researchers have used direct spectral measurements (e.g., NIRS) for
direct measurements of soil characteristics (reviewed by Stenberg et al. 2010;
Bellon-Maurel and McBratney 2011; Soriano-Disla et al. 2014), and there are
limited examples of remotely sensed spectroscopic measurements of soils (reviewed
by Ustin et al. 2004; Cecillon et al. 2009). Here we focus on remotely sensed spectral measurements of plant communities as a surrogate for belowground processes.
The optical surrogacy hypothesis (sensu Gamon 2008) argues that plant spectra can
serve as a surrogate for important belowground processes.
Direct spectral measurements have been used to assess belowground processes for
decades. For instance, direct NIRS of leaf litter can be used to predict decomposition
rates in a variety of systems (Gillon et al. 1993; Gillon et al. 1999; Shepherd et al.
2005; Fortunel et al. 2009; Parsons et al. 2011). RS of canopy traits to predict belowground processes is becoming increasingly useful. Spectroscopic measurement of
δ
15
N is of particular interest for ecosystem processes (Serbin et al. 2014) because
stable N isotopes can provide important information regarding ecosystem N cycling
(Robinson 2001; Hobbie and Hobbie 2006). RS of forest disturbance (e.g., fire severity) and subsequent belowground processes is relatively common (e.g., Holden et al.
2016). Sabetta et al. (2006) used hyperspectral imaging to predict leaf litter decomposition across four forest communities. Fisher et al. (2016) were able to distinguish
between arbuscular and ectomycorrhizal tree-mycorrhizal associations using spectral
information gleaned from Landsat data. While the above examples focus on remotely
sensed spectral information, remotely sensed forest structural information developed
from lidar data can also provide information about belowground systems, as Thers
et al. (2017) were able to use remotely sensed lidar data to estimate belowground
fungal diversity. The growing number of examples that employ remotely sensed data
to provide information about belowground systems points to the potential of plant
spectra to be used as surrogates for ecosystem processes.
8.4.2 Forest Systems: Aspen Clones Example
An example of optical surrogacy in practice is illustrated by work completed in
trembling aspen (Populus tremuloides) systems across the Western and Midwestern
USA. Trembling aspen is the most widespread native tree species in North America
(Mitton and Grant 1996) and is an ecologically important foundation species across
M. Madritch et al.