CHAPTER TWELVE
A chemical-evolutionary basis
for remote sensing of tropical
forest diversity
GREGORY P. ASNER
Carnegie Institution for Science
12.1 Introduction
One of ecology?s fundamental pursuits centres on explaining controls over
the distribution of species. Interest in species distributions is also increasing
in the context of global change. There is a growing emphasis on providing
information to decision-makers on where to protect and restore ecosystems
and the species they harbour. Nowhere does this seem more obvious, or more
pressing, than in tropical forests. Land-use change continues to strike deep
into tropical forest regions, and climate change has come to the forefront in
tropical forest conservation and planning. Temperature, precipitation and
other climatic factors are shifting in some regions, or they are becoming
highly variable (Cox et al. 2004; Williams, Jackson & Kutzbach 2007), and
these changes are placing species in novel conditions that will force them to
adapt, move or die (Asner, Loarie & Heyder 2010; Loarie et al. 2009).
We currently have few tools for mapping species distributions in tropical
forests under these changing conditions, and our ability to monitor distribu-
tional changes remains limited despite their critical importance to ecology
and conservation. Today, our understanding of species distributions comes
from contrasting, and geographically limited, information sources. On the
one hand, we have incredibly dense information at small scales of up to
50 hectares or so (Condit et al. 2005; Hubbell & Foster 1986). This information
is key to understanding local-scale community patterns (Harms et al. 2001),
but sparsely distributed plots cannot resolve ecological and biogeo-
graphic processes operating at the scales of long-range dispersal, migration
or landscape evolution. On the other hand, we have longer-range, transect-
based information on species occurrence, which does not provide the compre-
hensive inventories required for ecological modelling, although it is critical
to basic conservation planning (http://fieldmuseum.org/explore/department/
ecco). Emerging satellite-based methods, especially when combined with
field observations, are beginning to fill in the geographic void with predic-
tions of community patch-matrix patterns (Higgins et al. 2011; Turner et al.
Forests and Global Change, ed. David A. Coomes, David F. R. P. Burslem and William D. Simonson.
Published by Cambridge University Press. ? British Ecological Society 2014.
2003), but current approaches do not constitute direct measurements of
species distributions or diversity levels. With mainstream technologies, it is
not possible to monitor spatially explicit change in species distributions and
biological diversity, which is likely to be non-randomover landscape, regional
and global scales.
From above, tropical forests look like monotonous green carpets, at least
until a river or forest clearing is encountered to expose water, soils and man-
made materials (Figure 12.1). A few species are conspicuous and identifiable,
like some palm (e.g Mauritia) or flowering trees (e.g. Tabebuia), but things
generally seem green and hard to differentiate. Being able to identify species,
or tomap gradients in species richness, in the canopy of tropical forests would
represent a substantial scientific advance, facilitating unprecedented forms of
conservation and resource planning. Aerial photographs have been used for
years to estimate the cover and abundance of conspicuous temperate or
plantation species (e.g. Swetnam, Allen & Betancourt 1999), and some of
that capability has been transferred to high-resolution spaceborne cameras
(Thenkabail et al. 2004). Yet standard imaging cameras have yielded little
information on species distributions or richness in tropical canopies.
A B
Figure 12.1 As in much of Amazonia, lowland forests of Madre de Dios, Peru appear as
vast expanses of green canopy, incised by water bodies and deforestation (gold mining
at A). Even a careful eye will pick out only the most basic variation in forest
composition, such as swamp areas dominated by Mauritia flexuosa palm trees (B). See
plate section for colour version.
344 GREGORY P . ASNER
Here I explore new possibilities for mapping canopy species diversity in
humid tropical forests based on both the traits expressed among canopy taxa
and the technologies needed to access information on those traits. I refer specif-
ically to the Carnegie Airborne Observatory (CAO; http://cao.ciw.edu) because I
developed it to focusmappingmeasurements on the plant traits thatmight best
express the diversity of canopies in tropical forests. I also consider the ecological
and evolutionary basis for trait-based remote sensing, with emphasis on the
chemical attributes of canopy species. The ultimate purpose of the CAO and its
biodiversity monitoring capabilities is to track compositional and/or functional
changes in ecosystems, particularly under conditions of rapid global change.
12.2 Canopy chemistry: a window into biodiversity?
Canopy chemistry may represent a window into mapping biological diversity
in tropical forests. Plants synthesise awide variety of leaf compounds utilising
numerous chemical elements and, for purposes here, it is convenient to group
them into three functional categories: (i) light capture and growth,
(ii) longevity and defence, and (iii) maintenance and metabolism (Table 12.1).
While not independent, these groups are easily recognised by plant physiol-
ogists and ecologists, and thus can be traced throughout the scientific
literature. Light capture?growth compounds include photosynthetic pig-
ments such as chlorophylls and carotenoids, nitrogen and phosphorus, as
well as the immediate products of photosynthesis such as soluble carbon
compounds (Chapin III 1991; Evans 1989; Niinemets et al. 1999). Longevity
compounds include lignin and cellulose that are structurally tough, lasting
much of a leaf?s lifetime (Melillo, Aber & Muratore 1982), while the defence
category extends this list to more dynamic polyphenols (Coley, Kursar &
Machado 1993). Maintenance?metabolism elements are those required in
small quantities to support and mediate myriad functions within the leaf, but
Table 12.1 A portfolio of chemical compounds and elements provides a multi-
dimensional signature of foliar investment in light capture and growth, longevity and
defence, and maintenance and metabolism.
Light capture and growth Longevity and defence Maintenance and metabolism
Anthocyanins Carbon (C) Potassium (K)
Chlorophylls Hemi-cellulose Calcium (Ca)
Carotenoids Cellulose Magnesium (Mg)
Nitrogen (N) Lignin Zinc (Zn)
Phosphorus (P) Phenols Manganese (Mn)
Water Tannins Boron (B)
Soluble carbon Iron (Fe)
REMOTE SENS ING OF TROP ICAL FOREST D IVERS ITY 345
few are considered to be highly specialised. An overarching or synthetic
structural trait is leaf mass per area (LMA); while not a chemical per se, LMA is
linked to growth traits (Wright et al. 2004; Poorter et al. 2009).
Much work has focused on the ecological significance of these foliar com-
pounds and elements. Biogeochemists tend to look at nitrogen, phosphorus
and base cations (Ca, Mg, K), as well as lignin, in mediating productivity and
decomposition (Aber & Melillo 1991; Schlesinger 1991). Physiologists often
consider photosynthetic compounds and nutrients affecting trade-offs among
growth-related processes (Peterson et al. 1999;Wright, Reich &Westoby 2001).
Here I take a broad view, thinking of the leaf chemical portfolio as expressing
strategic trade-offs among investments in light capture?growth, longevity?
defence and maintenance?metabolic needs. By doing so, we can treat the
chemical portfolios as ?signatures?. If chemical signatures are organised
phylogenetically, then this might reflect variation in biological diversity at
some taxonomic level. However, if the chemical phylogenetic signal is highly
conserved, then closely related species may have similar chemical signatures,
described broadly by Ackerly (2009) and for defence compounds by Agrawal
(2007). Or if chemical phylogenies are highly dispersed, then species may
prove to be chemically distinct from one another within a community (Kursar
et al. 2009). Independent of phylogeny, ensembles of chemical signatures ? com-
pared among taxa at the community level ? will probably describe functional
diversity among tropical forest canopies. Without such chemical information,
one might mistake many tropical forests as functionally similar or floristically
monotonous when using chemical remote-sensing methods to map them.
12.3 Factors suggesting evolution in canopy chemical traits
There are a number of issues to consider when assessing whether or not
canopy chemical traits, and particularly chemical signatures, exist among
taxa. These include intraspecific variation, inter-trait correlation and chem-
ical phylogenetic organisation. Depending upon the patterns that exist,
chemically based windows to biodiversity mapping may or may not emerge.
12.3.1 Intraspecific variation
Temperate forest ecologists sometimes observe such high intraspecific
variation that it would surprise them to find substantial chemical signature
differentiation among taxa. However, in the temperate zone, such as in boreal
or mixed deciduous forests with relatively few tree taxa (Latham & Ricklefs
1993), species are often adapted to span a wide range of climate conditions.
They span these variable conditions while negotiating nitrogen-limited sub-
strates spread over huge geographic areas (Vitousek & Howarth 1991).
Variable climate and low nitrogen availability are associated with higher
intraspecific plasticity in leaf traits (Karlsson & Nordell 1998; Niinemets,
346 GREGORY P . ASNER
Valladares & Ceulemans 2003; Reich, Walters & Ellsworth 1997). As a result,
variation in the chemistry of a temperate species might be heavily driven by
environmental factors, with adaptations centred more on increasing intra-
specific variation among relatively fewer species to negotiate environmental
gradients in space and time, thereby trumping phylogenetic expression of
chemical variation.
In humid tropical forests, growth conditions are generally good, and
although rock-derived nutrient availability may be low, especially for
phosphorus and calcium (Vitousek & Sanford Jr. 1986; Vitousek, Turner &
Kitayama 1995), production often remains high (Clark et al. 2001). High net
primary production is maintained because nitrogen flows in relative abun-
dance in many humid forests (Martinelli et al. 1999), and canopy species have
evolved to recapture other nutrients even before discarded foliage reaches the
soil surface (Vitousek 1982, 1984). However, good growth conditions are also
associated with herbivory and parasitism in humid tropical forests, and it is in
these environments where some of the strongest expressions of taxonomic
order in canopy chemistry can be found, beginning with observations of rela-
tively low intraspecific variation. For LMA, the within-species variation for
thousands of taxa averages just 16% (Asner et al. 2011b), albeit only if samples
are taken from upper, sunlit positions in the canopies.1 Observations of low
intraspecific variation of LMA at the top of tropical forest canopies aremirrored
in many leaf chemical traits (Figure 12.2). Light capture?growth traits vary
within species by an average of about 20%, depending upon the trait, in the
upper canopy (Asner et al. 2009). Longevity?defence compounds also display low
intraspecific variation, although often higher for phenols. Elements supporting
maintenance and multiple metabolic functions (i.e. Mn, B, Zn) often display the
highest levels of within-species variation (Figure 12.2).
12.3.2 Inter-trait correlation
Determining whether consistent combinations of leaf chemicals ? or
signatures ? exist also requires an understanding of potential relationships
between the contributing compounds andelements. Strong correlationbetween
traits reduces the complexity and diversity of chemical signatures, whereas
orthogonality among traits increases them.Growth factors such as chlorophylls,
nitrogen andphosphorus are, on average, about 50?60% inter-correlated (Wright
et al. 2004), and these correlations may be more strongly expressed when nitro-
gen is limiting growth, thereby down-regulating all growth-related chemicals
1 Without this control, variability in LMA and nearly all other growth compounds and
nutrients is dominated by variation in lighting and shading conditions. Sticking with the
upper canopy provides critically important control for comparative and phylogenetic analy-
sis, and it is also extremely important in the effort to measure leaf traits with remote-
sensing technology that is most sensitive to upper canopy foliage.
REMOTE SENS ING OF TROP ICAL FOREST D IVERS ITY 347
(Cordell et al. 2001). However, outside the circle of themost growth-related traits,
it can take 15 or more statistical degrees of freedom to explain the covariation
among 21 light?growth, longevity?defence andmaintenance?metabolism traits
(Asner & Martin 2011; Asner, Martin & Suhaili 2012). It is not clear whether this
occurs primarily in tropical forests, or if it is common among plants or canopy
species.Nonetheless, forhumid tropical forests, themeasurementsdemonstrate
the multidimensionality of leaf chemical signatures, and these patterns are
highly suggestive of multivariate trait evolution.
12.3.3 Taxonomic organisation
A taxonomically nested pattern for a chemical trait would suggest some
evolutionary basis to that trait, and such patterns have been uncovered in
recent studies in Hawaiian (Asner &Martin 2009), Amazonian (Asner &Martin
2011; Fyllas et al. 2009), Australian (Asner et al. 2009) and Malaysian (Asner,
Martin & Suhaili 2012) tropical forest canopies. There is often strong taxo-
nomic organisation among leaf traits inhumid tropical forest canopies, starting
with LMA: Asner et al. (2011b) found that 70% of the LMA variation among 2873
canopy individuals, collected fromawide range of humid tropical forests, could
be explained to the species level. This finding for LMA extends to foliar
Chlorophyll-a
Chlorophyll-b
Carotenoids
Water
Leaf mass per area
Nitrogen
Total carbon
Soluble carbon
Hemi-cellulose
Cellulose
Lignin
Phenols
Tannins
Phosphorus
Potassium
Calcium
Magnesium
Zinc
Manganese
Boron
Iron
0 20 40 60
CV (%)
80 100
Figure 12.2 Intraspecific variation of
foliar chemistry (expressed as a
percentage coefficient of variation, CV,
within species) is generally low in the
upper canopy of humid tropical forests.
This example showsmean and standard
deviation (box plot), as well as
minimum and maximum (whisker
lines), for 21 leaf traits collected from
sunlight positions in more than 200
species in the Peruvian Amazon.
Adapted from Asner and Martin (2011).
348 GREGORY P . ASNER
chemicals: Fyllas et al. (2009) found that up to 50% of variation in about eight
canopy foliar chemicals collected throughout Amazonia could be attributed to
phylogeny, independent of site conditions. In a detailed site-level study in the
Peruvian Amazon, Asner and Martin (2011) extended that finding to 21 traits
among 600 coexisting canopies, finding that nearly 90% of the chemical varia-
tion among canopies could be attributed to taxonomic affiliation. This pattern
has repeated itself atmany sites throughout the global humid tropics (Carnegie
Spectranomics Database; http://Spectranomics.ciw.edu).
12.4 Expressions of functional and biological diversity
in hyperspectral imagery
As described, tropical forest canopies contain foliage with emergent, predict-
able chemical properties, including relatively low intraspecific variation, low
inter-trait correlation when considering chemicals beyond the core growth-
related compounds, and demonstrable taxonomic organisation. This combi-
nation may facilitate biodiversity assessment using remotely sensed data
because these emergent chemical patterns mediate canopy spectral proper-
ties as measured with certain types of remote-sensing instruments. Many
research groups have reported taxonomic variation associated with leaf and
canopy spectral properties (e.g. Castro-Esau, Sanchez-Azofeifa & Caelli 2004;
Fe?ret & Asner 2011; Sanchez-Azofeifa et al. 2009;Williams 1991). In particular,
a rich history of study in imaging spectroscopy (also known as hyperspectral
remote sensing) has demonstrated that many of the chemicals can be meas-
ured remotely (reviews by Kokaly et al. 2009; Ustin et al. 2009). Further, Asner
et al. (2011a) recently quantified the relationship between spectral and
chemical data for thousands of tropical forest canopies. Another synthesis is
not my goal here. Instead, with the aid of CAO imagery, I present evidence for
canopy chemical diversity among tropical forest individuals, and discuss how
chemical diversity is linked to biological diversity.
The CAO includes two high-fidelity imaging spectrometers that measure
light reflected from canopies in up to 480 narrow spectral bands. By comput-
ing a principal components analysis of the CAO spectrometer data, and then
classifying the results by spectral features, I have uncovered evidence for
taxon-specific spectral signatures, such as for the well-known 50-hectare
Center for Tropical Forest Science (CTFS; http://www.ctfs.si.edu) plot on
Barro Colorado Island, Panama (Figure 12.3). The image reveals the possibility
that spectral ? and thus chemical ? diversity exists in the canopy on very short
distance scales of individual crowns. If this were not the case, then colours in
the image would appear speckled and randomly distributed. Instead, spectral
diversity is expressed crownby crown, andwhere variation does exist within a
crown, close inspection of the original imagery indicates the presence of
partial crown mortality, lianas or shadowing.
REMOTE SENS ING OF TROP ICAL FOREST D IVERS ITY 349
This crown-based pattern in tropical forest spectroscopy can be extended to
a taxon-specific interpretation such as shown for an area of lowland rain
forest on the Island of Hawaii (Figure 12.4). Combining this image with
detailed ground and laboratory analysis, we found that the crowns with
similar spectral signatures were of the same species more than 70% of the
time (Fe?ret & Asner 2011, 2013). For example, each red crown in Figure 12.4
is the invasive tree Psidium cattleianum. Field-based chemical interpretation
of the map indicates that the spectra are driven by differences in LMA,
chlorophyll, nitrogen, phenols and water content among species.
Finally, evidence suggesting a direct link between spectral and biological
diversity in tropical forests is emerging in new comparative analyses of CAO
data taken over different forests. For example, amonotypic stand of Eucalyptus
grandis maintains a nearly constant spectroscopy (Figure 12.5A). Note the
expression of low intraspecific variation by near constant colour; where
there is variation, it is caused by shadows among crowns. In contrast, a
high-diversity Amazonian forest shows a kaleidoscope of spectral variation
(Figure 12.5B). Here, it is again observed that the results are expressed at the
crown level. Areas of near-constant canopy spectroscopy in the left portion of
the image (shown in purple) turn out to be near-monotypic stands of Mauritia
flexuosa (see also Figure 12.1). That is, even in high-diversity landscapes, one
A
B
Figure 12.3 This CAO image of the 50-hectare Center for Tropical Forest Science (CTFS)
plot on Barro Colorado Island, Panama, shows (A) a natural colour composite as seen by
the naked eye, and (B) spectroscopic analysis of three dominant chemical components
throughout the canopy. In this case, the image was interpreted in terms of chlorophyll
(reds), leafmass per area (greens) andwater content (blues). Note that the expression of
chemical variation occurs primarily at the scale of individual crowns. White bar
indicates approx. 30 m of distance. See plate section for colour version.
350 GREGORY P . ASNER
AB
C
Figure 12.4 This three-dimensional CAO image of a Hawaiian tropical forest indicates
that canopy chemical traits are differentially expressed among species, including
(A) Psidium cattleianum, (B)Mangifera indica and (C) Trema orientalis. Species-level chemical
signatures are created by low intraspecific variation, low inter-trait correlation and
phylogenetic organisation in chemical traits. See plate section for colour version.
A
B
Figure 12.5 CAO imagery taken over
(A) Eucalyptus grandis plantation and (B) high-
diversity lowland Amazon forest indicates
the role of chemical diversity in expressing
biological diversity. Chemicals expressed
here include chlorophyll, water and phenols.
Notice the near-constant chemistry (colour)
in the plantation canopy and the crown-scale
variation in chemical signatures in the
Amazon case. The near-constant purple
canopies in the left portion of the Amazon
image are small monotypic stands of Mauritia
flexuosa as seen in Figure 12.1. White bar
indicates 100 m of distance. See plate section
for colour version.
REMOTE SENS ING OF TROP ICAL FOREST D IVERS ITY 351
can find cold spots of spectral diversity expressing local areas of lower canopy
biodiversity in the CAO imagery.
Although for years we have known that spectral properties are directly
controlled by the foliar chemical properties in forest vegetation, only now
are we getting a top-down look at the degree to which these patterns occur
and how they may be related to tropical forest biodiversity. These are the first
quantitative glimpses, and the degree to which species dictate the chemistry
and spectroscopy of forest canopies is still far from understood. Much more
study is needed to advance the role of this new observational capability in
scientific, conservation and forest management contexts.
12.5 Ecological and evolutionary causes for chemical signatures
What ecological and evolutionary processes give rise to individual canopies
and species having unique spectral and chemical signatures within a commun-
ity? Some new information, combined with long-standing theories, provides
important clues.
In humid tropical forests, geologic and topographic variation is often asso-
ciated with changes in plant species composition (Gentry 1988; Higgins et al.
2011; Phillips et al. 2003), and in some regions such as Amazonia, floristic
changes are connected to differences in soil cations, phosphorus availability
and texture (ter Steege et al. 2006; Tuomisto et al. 1995, 2003). Floristic vari-
ation also occurs with geographically varying climate regimes. Critically, the
observed compositional turnover is often paralleled by changes in average
(community-level) chemical concentrations in canopy foliage (Asner & Martin
2011; Fyllas et al. 2009). That is, we see both a change in floristics and a change in
the average chemical make-up of the canopies when crossing soil- and climate-
mediated thresholds and boundaries. On either side of these thresholds, con-
vergence in the canopy chemical signatures among coexisting species can be
viewed as a syndrome, and we know that soil-mediated chemical syndromes
occur among growth-related traits, especially for leaf nitrogen and phosphorus.
For example, high-fertility soils harbour species with higher nutrient concen-
trations in their foliage; opposite patterns are often found in species found in
low-nutrient conditions (Townsend et al. 2007; Townsend, Asner & Cleveland
2008). We also see community-scale shifts in foliar trait syndromes related to
longevity?defence and maintenance?metabolism (e.g. phenols and carbon frac-
tions), although these patterns can be more subtle (Asner & Martin 2011).
Moreover, plant species tend to display suppressed levels of phenotypic plasti-
city in foliar longevity-and-defence compounds in resource-poor conditions
(Stamp 2003). It seems that abiotic resources are key factors mediating plant
evolutionary strategies, setting spatially explicit templates for the types of
defence that may evolve (Kursar & Coley 2003). Furthermore, although growth
(e.g. nitrogen) and defence (e.g. lignin) traits are often viewed as correlated in
352 GREGORY P . ASNER
temperate forest ecosystems (Melillo, Aber &Muratore 1982), we canfind a clear
disconnect between them in tropical canopies (Asner & Martin 2011; Fyllas et al.
2009). This provides a critical insight: it suggests that evolution in highly com-
petitive, high-production environments, such as humid tropical forests, leads to
highly diverse, uncorrelated foliar chemical strategies. Why?
The environmental factors supporting elevated productivity in tropical
forests (abundant nitrogen, light, water) are also correlated with pest and
pathogen pressure (Ayres 1993; Mattson 1980), and the diversity of herbi-
vores, parasites, viruses and other leaf-attacking lifeforms may well have led
to the pronounced radiation of host strategies designed to combat, accommo-
date and outlast these threats in order to increase survival and fitness. This
builds directly on the seminal work of Janzen (1970) and Connell (1970), who
argued for the existence of pest- and pathogen-mediated density dependence
in plant seedling establishment and recruitment. They suggested that, by
host-specific co-evolution, herbivores and other leaf-threatening lifeforms
maintain very low densities of conspecific plants locally. Janzen also proposed
that tropical plants found in low-nutrient environments allocate heavily to
chemical defences to protect foliage, whereas plants found on higher-nutrient
soils invest more in chemicals yielding higher growth rates. Key work by
Coley (1983) also suggests that light-gap species maintain chemical traits
that are different from those found in shaded understorey environments. So
while soil nutrient status fosters the community-scale foliar chemical syn-
dromes described earlier, host-specific growth and defence strategies create
unique chemical signatures among plant species coexisting within a com-
munity. Add to this the possibility for highly competitive plant-to-plant inter-
actions and evolved syndromes for light and nutrients (Coley 1993; Coley &
Barone 1996; Fine, Mesones & Coley 2004), and a multi-scale kaleidoscope of
chemical traits emerges, as we observe in CAO imagery of humid tropical
forests.
How diverse are chemical signatures among coexisting canopy species?
Does each have a unique combination of chemical traits? At some level, the
answer is yes, but intraspecific variation ? as low as it can be at times ? still
leaves a buffer of chemical variability for each species, making it hard to
answer these questions. Species-specificity is also dependent upon the num-
ber of traits considered within a chemical signature: combining traits results
in increased differentiation among species, but does this have ecological or
evolutionary meaning? The CAO imagery again provides some clues. In the
Hawaiian image example (Figure 12.4), the forest is dominated by a relatively
recent compilation of exotic species introduced from forests of Oceania, Asia,
Central and South America (Mascaro et al. 2008; Zimmerman et al. 2007). The
forest thus harbours species with chemical traits established in far-flung com-
munities evolved under diverse environmental conditions and biogeographic
REMOTE SENS ING OF TROP ICAL FOREST D IVERS ITY 353
histories. Perhaps as a result, each canopy species is very distinct and separa-
ble in the imagery. In the Peruvian Amazon image, the canopy comprises
species evolved in similar, late-Holocene climate conditions (Williams,
Jackson & Kutzbach 2007). There was thus a shared environment for chemical
trait evolution, but the species also come from differing pre-Holocene biogeo-
graphic origins (Hoorn et al. 2010). Chemical diversity is high and many
species are chemically distinct, yet the imagery suggests greater overlap
among coexisting canopies (i.e. a shared syndrome) than we find in the
Hawaiian forest canopy occupied by species from very diverse evolutionary
and biogeographic backgrounds.
These tantalising hints suggest that canopy biodiversity maps based on
chemical plant traits could unlock doors to understanding forest function
and community assembly, and how species have evolved under contrasting
environmental regimes. Far more work is needed to measure the spectral
patterns and to interpret them chemically and in terms of biological diversity.
Yet this early evidence suggests a novel pathway forward to unravelling the
secrets of the ecology and evolution of tropical forest communities.
12.6 Biodiversity mapping and global change
There is a pressing need to take biodiversity mapping and monitoring beyond
the plot or transect scale. The velocity of climate and land-use changes is now
likely to exceed the migration potential of many species and functional
groups of species (Asner, Loarie & Heyder 2010; Loarie et al. 2009), and this
process has already begun to rearrange the composition of our biosphere
(Parmesan 2006; Wake, Hadly & Ackerly 2009). In short, our biosphere is in
a state of non-equilibrium change (Biggs, Carpenter & Brock 2009; Schimel,
Asner & Moorcroft 2013). Without leaps forward in the area of biodiversity
monitoring, changes in biospheric composition will go unaccounted, and
thus will have unknown impact on prognostic studies of the future global
environment. New technologies such as imaging spectroscopy are key to
making sure biodiversity change is quantified, but technology represents
only part of the solution going forward. New and innovative approaches are
needed to take advantage of the developing technologies. I propose that the
best approaches will be seated in an understanding of the underlying evolu-
tionary drivers of trait-based diversification among species. More generally,
mapping compositional and functional changes at the regional level will
require effort among experts from multiple fields including engineering,
chemistry and evolutionary biology. The required interdisciplinary links
still need forging, yet the effort is central to advancing biodiversity monitor-
ing for conservation planning, ecosystem management and resource policy
development in this century of rapid global change.
354 GREGORY P . ASNER
Acknowledgements
I thank the many colleagues who have influenced my thinking on the topics
presented here. I thank Roberta Martin, Alan Townsend and two anonymous
reviewers for comments on the manuscript, and David Burslem and David
Coomes for organising the writers and for editing. The Carnegie Airborne
Observatory is made possible by the Andrew Mellon Foundation, Avatar
Alliance Foundation, Gordon and Betty Moore Foundation, Grantham
Foundation for the Protection of the Environment, John D. and Catherine T.
MacArthur Foundation, Margaret A. Cargill Foundation, Mary Anne Nyburg
Baker and G. Leonard Baker Jr, W.M. Keck Foundation and William R. Hearst III.
References
Aber, J. D. & Melillo, J.M. (1991) Terrestrial
Ecosystems. Philadephia: Saunders College
Publishing.
Ackerly, D. (2009) Conservatism and
diversification of plant functional traits:
Evolutionary rates versus phylogenetic
signal. Proceedings of the National Academy of
Sciences USA, 106, 19699?19706.
Agrawal, A. A. (2007) Macroevolution of plant
defense strategies. Trends in Ecology &
Evolution, 22, 103?109.
Asner, G., Martin, R. & Suhaili, A. (2012) Sources
of canopy chemical and spectral diversity in
lowland Bornean forest. Ecosystems, 15,
504?517.
Asner, G. P., Loarie, S. R. & Heyder, U. (2010)
Combined effects of climate and land-
use change on the future of humid
tropical forests. Conservation Letters, 3,
395?403.
Asner, G. P. & Martin, R. E. (2009) Airborne
spectranomics: Mapping canopy chemical
and taxonomic diversity in tropical forests.
Frontiers in Ecology and the Environment, 7,
269?276.
Asner, G. P. & Martin, R. E. (2011) Canopy
phylogenetic, chemical and spectral
assembly in a lowland Amazonian
forest. New Phytologist, 189,
999?1012.
Asner, G. P., Martin, R. E., Ford, A. J.,
Metcalfe, D. J. & Liddell, M. J. (2009) Leaf
chemical and spectral diversity of
Australian tropical forests. Ecological
Applications, 19, 236?253.
Asner, G. P., Martin, R. E., Knapp, D. E. et al.
(2011a) Spectroscopy of canopy chemicals in
humid tropical forests. Remote Sensing of
Environment, 115, 3587?3598.
Asner, G. P., Martin, R. E., Tupayachi, R. et al.
(2011b) Taxonomy and remote sensing of
leaf mass per area (LMA) in humid tropical
forests. Ecological Applications,
21, 85?98.
Ayres, M. P. (1993) Plant defense, herbivory, and
climate change. In Biotic Interactions and
Global Change (eds. P.M. Kareiva,
J. G. Kingsolver & R. B. Huey), pp. 75?94.
Sunderland, MA: Sinauer Associates.
Biggs, R., Carpenter, S. R. & Brock, W. A. (2009)
Turning back from the brink: Detecting an
impending regime shift in time to avert it.
Proceedings of the National Academy of Sciences
USA, 106, 826?831.
Castro-Esau, K. L., Sanchez-Azofeifa, G. A. &
Caelli, T. (2004) Discrimination of lianas and
trees with leaf-level hyperspectral data.
Remote Sensing of Environment, 90, 353?372.
Chapin III, F. S. (1991) Integrated responses of
plants to stress. BioScience, 41, 29?36.
Clark, D.A., Brown, S., Kicklighter, D.W. et al. (2001)
Net primary production in tropical forests: An
evaluation and synthesis of existing field data.
Ecological Applications, 11, 371?384.
Coley, P. D. (1983) Herbivory and defensive
characteristics of tree species in a lowland
REMOTE SENS ING OF TROP ICAL FOREST D IVERS ITY 355
tropical forest. Ecological Monographs, 53,
209?234.
Coley, P. D. (1993) Gap size and plant defenses.
Trends in Ecology and Evolution, 8, 1?2.
Coley, P. D. & Barone, J. A. (1996) Herbivory and
plant defenses in tropical forests. Annual
Review of Ecology and Systematics, 27,
305?335.
Coley, P. D., Kursar, T. A. & Machado, J.-L. (1993)
Colonization of tropical rain forest leaves by
epiphylls: effects of site and host-plant leaf
lifetime. Ecology, 74, 619?623.
Condit, R., Ashton, P., Balslev, H. et al. (2005)
Tropical tree alpha-diversity: Results from a
worldwide network of large plots. Science,
55, 565?582.
Connell, J. H. (1970) On the role of natural
enemies in preventing competitive
exclusion in some marine animals and in
rain forest trees. In Dynamics of Population
(eds. P. J. Den Boer & G. R. Gradwell).
Wageningen: Pudoc.
Cordell, S., Goldstein, G., Meinzer, F. C. &
Vitousek, P.M. (2001) Regulation of leaf life-
span and nutrient-use efficiency of
Metrosideros polymorpha trees at two extremes
of a long chronosequence in Hawaii.
Oecologia (Berlin), 127, 198?206.
Cox, P.M., Betts, R. A., Collins, M. et al. (2004)
Amazonian forest dieback under climate-
carbon cycle projections for the 21st
century. Theoretical and Applied Climatology,
78, 137?156.
Evans, J. R. (1989) Photosynthesis and nitrogen
relationships in leaves of C3 plants.
Oecologia, 78, 9?19.
Fe?ret, J.-B. & Asner, G. P. (2011) Spectroscopic
classification of tropical forest species using
radiative transfermodeling. Remote Sensing of
Environment, 115, 2415?2422.
Fe?ret, J.-B. & Asner, G. P. (2013) Tree species
discrimination in tropical forests using
airborne imaging spectroscopy. IEEE
Transactions on Geoscience and Remote Sensing,
51, 73?84.
Fine, P. V. A., Mesones, I. & Coley, P. D. (2004)
Herbivores promote habitat specialization
by trees in Amazonian forests. Science, 305,
663?665.
Fyllas, N., Patin?o, S., Baker, T. et al. (2009)
Basin-wide variations in foliar properties
of Amazonian forest: phylogeny, soils
and climate. Biogeosciences, 6,
2677?2708.
Gentry, A. H. (1988) Changes in plant community
diversity and floristic composition on
environmental and geographical gradients.
Annals of the Missouri Botanical Garden, 75,
1?34.
Harms, K. E., Condit, R., Hubbell, S. P. &
Foster, R. B. (2001) Habitat associations of
trees and shrubs in a 50-ha neotropical
forest plot. Journal of Ecology, 89, 947?959.
Higgins, M., Ruokolainen, K., Tuomisto, H. et al.
(2011) Geological control of floristic
composition in Amazonian forests. Journal of
Biogeography, 38, 2136?2149.
Hoorn, C., Wesselingh, F. P., ter Steege, H. et al.
(2010) Amazonia through time: Andean
uplift, climate change, landscape evolution,
and biodiversity. Science, 330, 927?931.
Hubbell, S. P. & Foster, R. B. (1986) Biology,
chance, and history and the structure of
tropical rainforest tree communities. In
Community Ecology (eds. J. Diamond &
T. J. Case), pp. 314?329. New York: Harper
and Row.
Janzen, D.H. (1970) Herbivores and the number
of tree species in tropical forests. American
Naturalist, 104, 501?528.
Karlsson, P. S. & Nordell, K. O. (1998) Intraspecific
variation in nitrogen status and
photosynthetic capacity within mountain
birch populations. Ecography, 11, 293?297.
Kokaly, R. F., Asner, G. P., Ollinger, S. V.,
Martin, M. E. & Wessman, C. A. (2009)
Characterizing canopy biochemistry from
imaging spectroscopy and its application to
ecosystem studies. Remote Sensing of
Environment, 113, S78?S91.
Kursar, T. A. & Coley, P. D. (2003) Convergence in
defense syndromes of young leaves in
tropical rainforests. Biochemical Systematics
and Ecology, 31, 929?949.
356 GREGORY P . ASNER
Kursar, T. A., Dexter, K. G., Lokvam, J. et al. (2009)
The evolution of antiherbivore defenses and
their contribution to species coexistence in
the tropical tree genus Inga. Proceedings of the
National Academy of Sciences USA, 106,
18073?18078.
Latham, R. E. & Ricklefs, R. E. (1993) Global
patterns of tree species richness in moist
forests: energy-diversity theory does not
account for variation in species richness.
Oikos, 67, 325?333.
Loarie, S. R., Duffy, P. B., Hamilton, H. et al. (2009)
The velocity of climate change. Nature, 462,
1052?1057.
Martinelli, L. A., Piccolo, M. C., Townsend, A. R.
et al. (1999) Nitrogen stable isotopic
composition: tropical vs. temperate forests.
Biogeochemistry, 46, 45?65.
Mascaro, J., Becklund, K. K., Hughes, R. F. &
Schnitzer, S.A. (2008) Limited native plant
regeneration in novel, exotic-dominated
forests on Hawai?i. Forest Ecology and
Management, 256, 593?606.
Mattson, W. J. (1980) Herbivory in relation to
plant nitrogen content. Annual Review of
Ecology and Systematics, 11, 119?161.
Melillo, J. M., Aber, J. D. & Muratore, J. F. (1982)
Nitrogen and lignin control of hardwood
leaf litter decomposition dynamics. Ecology,
63, 621?626.
Niinemets, U., Tenhunen, J. D., Canta, N. R. et al.
(1999) Interactive effects of nitrogen and
phosphorus on the acclimation potential of
foliage photosynthetic properties of cork
oak, Quercus suber, to elevated atmospheric
CO2 concentrations. Global Change Biology, 5,
455?470.
Niinemets, U., Valladares, F. & Ceulemans, R.
(2003) Leaf-level phenotypic variability and
plasticity of invasive Rhododendron
ponticum and non-invasive Ilex aquifolium co-
occurring at two contrasting European sites.
Plant, Cell and Environment, 26, 941?956.
Parmesan, C. (2006) Ecological and evolutionary
responses to recent climate change. Annual
Review of Ecology, Evolution, and Systematics, 37,
637?669.
Peterson, A. G., Ball, J. T., Luo, Y. et al. (1999) The
photosynthesis-leaf nitrogen relationship at
ambient and elevated atmospheric carbon
dioxide: A meta-analysis. Global Change
Biology, 5, 331?346.
Phillips, O. L., Vargas, P. N., Lorenzo, A. et al.
(2003) Habitat association among
Amazonian tree species: a landscape-scale
approach. Journal of Ecology, 91, 757?775.
Poorter, H., Niinemets, U., Poorter, L.,
Wright, I. J. & Villar, R. (2009) Causes and
consequences of variation in leaf mass per
area (LMA): a meta-analysis. New Phytologist,
182, 565?588.
Reich, P. B.,Walters, M. B. & Ellsworth, D. S. (1997)
From tropics to tundra: Global convergence
in plant functioning. Proceedings of the
National Academy of Sciences USA, 94,
13730?13734.
Sanchez-Azofeifa, G. A., Castro, K., Wright, S. J.
et al. (2009) Differences in leaf traits, leaf
internal structure, and spectral reflectance
between two communities of lianas and
trees: Implications for remote sensing in
tropical environments. Remote Sensing of
Environment, 113, 2076?2088.
Schimel, D. S., Asner, G. P., & Moorcroft, P. R.
(2013) Observing changing ecological
diversity in the Anthropocene. Frontiers in
Ecology and the Environment, 11, 129?137.
Schlesinger, W.H. (1991) Biogeochemistry, An
Analysis of Global Change. San Diego, CA:
Academic Press.
Stamp, N. (2003) Out of the quagmire of plant
defense hypotheses. The Quarterly Review of
Biology, 78, 23?55.
Swetnam, T.W., Allen, C. D. & Betancourt, J. L.
(1999) Applied historical ecology: Using the
past to manage for the future. Ecological
Applications, 9, 1189?1206.
ter Steege, H., Pitman, N. C. A., Phillips, O. L. et al.
(2006) Continental-scale patterns of canopy
tree composition and function across
Amazonia. Nature, 443, 444?447.
Thenkabail, P. S., Enclona, E. A., Ashton, M. S.,
Legg, C. & De Dieu, M. J. (2004) Hyperion,
IKONOS, ALI, and ETM plus sensors in the
REMOTE SENS ING OF TROP ICAL FOREST D IVERS ITY 357
study of African rainforests. Remote Sensing of
Environment, 90, 23?43.
Townsend, A. R., Asner, G. P. & Cleveland, C. C.
(2008) The biogeochemical heterogeneity of
tropical forests. Trends in Ecology and the
Environment, 23, 424?431.
Townsend, A. R., Cleveland, C. C., Asner, G. P. &
Bustamante, M.M. C. (2007) Controls over
foliar N:P ratios in tropical rain forests.
Ecology, 88, 107?118.
Tuomisto, H., Poulsen, A. D., Ruokolainen, K. et al.
(2003) Linking floristic patterns with soil
heterogeneity and satellite imagery in
Ecuadorian Amazonia. Ecological Applications,
13, 352?371.
Tuomisto, H., Ruokolainen, K., Kalliola, R. et al.
(1995) Dissecting Amazonian biodiversity.
Science, 269, 63?66.
Turner, W., Spector, S., Gardiner, N. et al. (2003)
Remote sensing for biodiversity and
conservation. Trends in Ecology and Evolution,
18, 306?314.
Ustin, S. L., Gitelson, A. A., Jacquemoud, S. et al.
(2009) Retrieval of foliar information
about plant pigment systems from high
resolution spectroscopy. Remote
Sensing of Environment, 113, S67?S77.
Vitousek, P.M. (1982) Nutrient cycling and
nutrient use efficiency. American Naturalist,
119, 553?572.
Vitousek, P.M. (1984) Litterfall, nutrient cycling,
and nutrient limitation in tropical forests.
Ecology, 65, 285?298.
Vitousek, P.M. & Howarth, R.W. (1991) Nitrogen
limitation on land and in the sea ? How can
it occur? Biogeochemistry, 13, 87?115.
Vitousek, P.M. & Sanford Jr, R. L. (1986) Nutrient
cycling in moist tropical forest. Annual
Review of Ecology and Systematics, 17, 137?167.
Vitousek, P.M., Turner, D. R. & Kitayama, K.
(1995) Foliar nutrients during long-term
soil development in Hawaiian montane
rain forest. Ecology, 76, 712?720.
Wake, D. B., Hadly, E. A. & Ackerly, D. D. (2009)
Biogeography, changing climates and
niche evolution. Proceedings of the
National Academy of Sciences USA, 106,
19631?19637.
Williams, D. L. (1991) A comparison of
spectral reflectance properties at
the needle, branch, and canopy level for
selected conifer species. Remote
Sensing of Environment, 35, 79?93.
Williams, J.W., Jackson, S. T. & Kutzbach, J. E.
(2007) Projected distributions of novel and
disappearing climates by 2100 AD.
Proceedings of the National Academy of Sciences
USA, 104, 5738?5742.
Wright, I. J., Reich, P. B. & Westoby, M. (2001)
Strategy shifts in leaf physiology, structure
and nutrient content between species of
high- and low-rainfall and high- and low-
nutrient habitats. Functional Ecology, 15,
423?434.
Wright, I. J., Reich, P. B., Westoby, M. et al. (2004)
The worldwide leaf economics spectrum.
Nature, 428, 821?827.
Zimmerman, N., Hughes, R. F., Cordell, S. et al.
(2007) Patterns of primary succession of
native and introduced plants in lowlandwet
forests in Eastern Hawaii. Biotropica,
doi:10.1111/j.1744?7429.2007.00371.x.
358 GREGORY P . ASNER
A B
Figure 12.1 As in much of Amazonia, lowland forests of Madre de Dios, Peru appear as
vast expanses of green canopy, incised by water bodies and deforestation (gold mining
at A). Even a careful eye will pick out only the most basic variation in forest
composition, such as swamp areas dominated by Mauritia flexuosa palm trees (B).
AB
C
Figure 12.4 This three-dimensional CAO image of a Hawaiian tropical forest indicates
that canopy chemical traits are differentially expressed among species, including
(A) Psidium cattleianum, (B)Mangifera indica and (C) Trema orientalis. Species-level chemical
signatures are created by low intraspecific variation, low inter-trait correlation and
phylogenetic organisation in chemical traits.
A
B
Figure 12.3 This CAO image of the 50-hectare Center for Tropical Forest Science (CTFS)
plot on Barro Colorado Island, Panama, shows (A) a natural colour composite as seen
by the naked eye, and (B) spectroscopic analysis of three dominant chemical
components throughout the canopy. In this case, the imagewas interpreted in terms of
chlorophyll (reds), leaf mass per area (greens) and water content (blues). Note that the
expression of chemical variation occurs primarily at the scale of individual crowns.
White bar indicates approx. 30 m of distance.
AB
Figure 12.5 CAO imagery taken over
(A) Eucalyptus grandis plantation and
(B) high-diversity lowland Amazon forest
indicates the role of chemical diversity in
expressing biological diversity. Chemicals
expressed here include chlorophyll, water
and phenols. Notice the near-constant
chemistry (colour) in the plantation canopy
and the crown-scale variation in chemical
signatures in the Amazon case. The near-
constant purple canopies in the left portion
of the Amazon image are small monotypic
stands of Mauritia flexuosa as seen in
Figure 12.1. White bar indicates 100 m of
distance.