515
19.7 Phenology and Temporal Aggregation
Functional traits and biodiversity are normally presented as stable characteristics of
an ecosystem and are often mapped at broad scales (Butler et al. 2017; MorenoMartínez et al. 2018) as a snapshot and implemented within models as static data.
When functional traits are observed in situ, they are typically based on species-level
sampling, sometimes repeated over a phenological sequence, though most commonly near-peak biomass. By contrast, a satellite comes over at a fixed schedule and
collects data when cloud and aerosol conditions permit. Data may be selected to
capture maximum “greenness” or some other criterion (e.g., Kampe et al. 2010), or
data may be collected multiple times each year.
In many ecosystems, the species visible to a sensor in space vary (Sherry et al.
2011) over the course of a growing season. What is the best way to characterize that
system’s functional diversity? The simple sum of traits over a seasonal cycle might
double-count some species, while a peak greenness assessment might underestimate
functional diversity and neglect functionally important species specialized to early or
late season. A global imaging spectrometer with consistent repeat imaging will
enable us to both (i) address the effects of vegetation phenology on the retrieval of
foliar traits from imaging spectroscopy data and (ii) assess the phenological drivers
of variation in functional diversity at unprecedented spatial and temporal scales.
Existing studies have largely been conducted in periods of peak greenness, and neither the seasonal variability of vegetation traits nor the effects of seasonal variability
on retrieval algorithms have been addressed. These will be needed for an instrument
such as SBG, with a ~15-day return interval. Addressing this problem, like the spatial
scaling problem, requires considering the measurement in the context of ecological
theory as well as considering pragmatic reality.
19.8 Conclusion
The likely flight of a spaceborne imaging spectrometer, with global repeat coverage,
represents a transformative moment in global ecology. For the first time, a global
measurement will address not only the function of life on Earth, as do greenness and
other older measures, but at least one dimension of the diversity of life on Earth.
SBG, CHIME and precursor missions like HISUI and EnMAP will provide a unique
view of the range of plant functions and will no doubt produce new insights about
the range of function, the extent and distribution of functional diversity of plants,
and new aspects of habitat for other trophic levels. In addition, global spectroscopy
is likely to become available at a time when many other structural and functional
dimensions of biodiversity and ecosystem function are also available, reviewed in
Schimel, Schneider et al. (2019). The parallel emergence of new ecological observatories will provide a wealth of systematic ground data to aid in developing theory
and practical algorithms (Barnett et al. 2019). A number of challenges need to be
confronted in order to consistently integrate IS with ecological theory and with the
19 Prospects and Pitfalls for Spectroscopic Remote Sensing of Biodiversity…
19.7 Phenology and Temporal Aggregation
Functional traits and biodiversity are normally presented as stable characteristics of
an ecosystem and are often mapped at broad scales (Butler et al. 2017; MorenoMartínez et al. 2018) as a snapshot and implemented within models as static data.
When functional traits are observed in situ, they are typically based on species-level
sampling, sometimes repeated over a phenological sequence, though most commonly near-peak biomass. By contrast, a satellite comes over at a fixed schedule and
collects data when cloud and aerosol conditions permit. Data may be selected to
capture maximum “greenness” or some other criterion (e.g., Kampe et al. 2010), or
data may be collected multiple times each year.
In many ecosystems, the species visible to a sensor in space vary (Sherry et al.
2011) over the course of a growing season. What is the best way to characterize that
system’s functional diversity? The simple sum of traits over a seasonal cycle might
double-count some species, while a peak greenness assessment might underestimate
functional diversity and neglect functionally important species specialized to early or
late season. A global imaging spectrometer with consistent repeat imaging will
enable us to both (i) address the effects of vegetation phenology on the retrieval of
foliar traits from imaging spectroscopy data and (ii) assess the phenological drivers
of variation in functional diversity at unprecedented spatial and temporal scales.
Existing studies have largely been conducted in periods of peak greenness, and neither the seasonal variability of vegetation traits nor the effects of seasonal variability
on retrieval algorithms have been addressed. These will be needed for an instrument
such as SBG, with a ~15-day return interval. Addressing this problem, like the spatial
scaling problem, requires considering the measurement in the context of ecological
theory as well as considering pragmatic reality.
19.8 Conclusion
The likely flight of a spaceborne imaging spectrometer, with global repeat coverage,
represents a transformative moment in global ecology. For the first time, a global
measurement will address not only the function of life on Earth, as do greenness and
other older measures, but at least one dimension of the diversity of life on Earth.
SBG, CHIME and precursor missions like HISUI and EnMAP will provide a unique
view of the range of plant functions and will no doubt produce new insights about
the range of function, the extent and distribution of functional diversity of plants,
and new aspects of habitat for other trophic levels. In addition, global spectroscopy
is likely to become available at a time when many other structural and functional
dimensions of biodiversity and ecosystem function are also available, reviewed in
Schimel, Schneider et al. (2019). The parallel emergence of new ecological observatories will provide a wealth of systematic ground data to aid in developing theory
and practical algorithms (Barnett et al. 2019). A number of challenges need to be
confronted in order to consistently integrate IS with ecological theory and with the
19 Prospects and Pitfalls for Spectroscopic Remote Sensing of Biodiversity…