427
16.1.1 Why a Chapter on Scale?
A key thesis of this chapter is that much of the uncertainty in the RS of biodiversity
arises from the lack of attention to sampling scale, which affects both traditional
biodiversity metrics and our ability to detect biodiversity remotely. In most cases,
the sampling scales of typical satellite or airborne RS methods do not match those
of our biological definitions or biodiversity measurements on the ground, confounding our interpretation of biodiversity from RS. In part because of scale mismatches,
the interpretation of remotely sensed data from one time or place often cannot be
applied to another, and we lack a universal, operational approach to RS of biodiversity. For RS of biodiversity to be meaningful, a careful consideration of scale is
essential.
The purpose of this chapter is to address this need, with the goal of contributing
to the design of an effective, operational global biodiversity monitoring system. Our
primary focus is on optical diversity (a.k.a. “spectral diversity”) using passive optical RS in the visible to shortwave-infrared (VIS-SWIR) range (400–2500 nm)
because this approach allows species and functional diversity assessment. However,
we acknowledge that other methods, including lidar (Asner et al. 2012; Lausch
et al., Chap. 13, this volume), can also make important contributions to our understanding of biodiversity. Our key examples involve optical studies of terrestrial vegetation primarily at the level of alpha and beta diversity (Whittaker 1972), with the
underlying assumption that vegetation diversity may be related to the diversity of
other trophic levels via surrogacy (Magurran 2004) or to belowground diversity via
biogeochemical cycling (Madritch et al. 2014; Madritch et al., Chap. 8). Similarly,
in aquatic environments optical diversity (often expressed as “ocean color”) can
reveal dynamic structure related to the distribution of phytoplankton (Moses et al.
2016; Muller-Karger et al. 2018) and benthic organisms (Goodman and Ustin 2007),
and scaling principles discussed here may apply in these cases. While a comprehensive assessment of all aspects of biodiversity in all environments is beyond the
scope of this review, our hope is that the principles discussed here with a primary
focus on terrestrial vegetation will enable progress toward an operational global
biodiversity monitoring system involving RS.
16.1.2 What Is Optical (Spectral) Diversity?
Optical diversity can be defined and measured in many ways. It is often based on
spectral reflectance of leaves and canopies, in which case the term “spectral diversity” is often applied. One definition is based on the number of different kinds of
reflectance spectra (“spectral types,” “spectral species,” or “spectral signatures,”
Fig. 16.1) present in a given area, a direct analogy to the biological concept of species diversity (Féret and Asner 2014). The categorical spectral type concept presumes distinct and stable spectral patterns exist for a given species. However, this is
16 Consideration of Scale in Remote Sensing of Biodiversity
16.1.1 Why a Chapter on Scale?
A key thesis of this chapter is that much of the uncertainty in the RS of biodiversity
arises from the lack of attention to sampling scale, which affects both traditional
biodiversity metrics and our ability to detect biodiversity remotely. In most cases,
the sampling scales of typical satellite or airborne RS methods do not match those
of our biological definitions or biodiversity measurements on the ground, confounding our interpretation of biodiversity from RS. In part because of scale mismatches,
the interpretation of remotely sensed data from one time or place often cannot be
applied to another, and we lack a universal, operational approach to RS of biodiversity. For RS of biodiversity to be meaningful, a careful consideration of scale is
essential.
The purpose of this chapter is to address this need, with the goal of contributing
to the design of an effective, operational global biodiversity monitoring system. Our
primary focus is on optical diversity (a.k.a. “spectral diversity”) using passive optical RS in the visible to shortwave-infrared (VIS-SWIR) range (400–2500 nm)
because this approach allows species and functional diversity assessment. However,
we acknowledge that other methods, including lidar (Asner et al. 2012; Lausch
et al., Chap. 13, this volume), can also make important contributions to our understanding of biodiversity. Our key examples involve optical studies of terrestrial vegetation primarily at the level of alpha and beta diversity (Whittaker 1972), with the
underlying assumption that vegetation diversity may be related to the diversity of
other trophic levels via surrogacy (Magurran 2004) or to belowground diversity via
biogeochemical cycling (Madritch et al. 2014; Madritch et al., Chap. 8). Similarly,
in aquatic environments optical diversity (often expressed as “ocean color”) can
reveal dynamic structure related to the distribution of phytoplankton (Moses et al.
2016; Muller-Karger et al. 2018) and benthic organisms (Goodman and Ustin 2007),
and scaling principles discussed here may apply in these cases. While a comprehensive assessment of all aspects of biodiversity in all environments is beyond the
scope of this review, our hope is that the principles discussed here with a primary
focus on terrestrial vegetation will enable progress toward an operational global
biodiversity monitoring system involving RS.
16.1.2 What Is Optical (Spectral) Diversity?
Optical diversity can be defined and measured in many ways. It is often based on
spectral reflectance of leaves and canopies, in which case the term “spectral diversity” is often applied. One definition is based on the number of different kinds of
reflectance spectra (“spectral types,” “spectral species,” or “spectral signatures,”
Fig. 16.1) present in a given area, a direct analogy to the biological concept of species diversity (Féret and Asner 2014). The categorical spectral type concept presumes distinct and stable spectral patterns exist for a given species. However, this is
16 Consideration of Scale in Remote Sensing of Biodiversity