Toward Linking Aboveground Vegetation
Properties and Soil Microbial Communities
Using Remote Sensing
Yuki Hamada, Jack A. Gilbert, Peter E. Larsen, and Madeline J. Norgaard
Abstract
Despite their vital role in terrestrial ecosystem function, the
distributions and dynamics of soil microbial communities
(
SMCs
) are poorly understood. Vegetation and soil proper-
ties are the primary factors that influence
SMCs
. This paper
discusses the potential effectiveness of remote sensing science
and technologies for mapping
SMC
biogeography by char-
acterizing surface biophysical properties (e.g., plant traits
and community composition) strongly correlated with
SMCs
.
Using remotely sensed biophysical properties to predict
SMC
distributions is extremely challenging because of the intricate
interactions between biotic and abiotic factors and between
above- and below-ground ecosystems. However, the integra-
tion of biophysical and soil remote sensing with geospatial
information about the environment holds great promise
for mapping
SMC
biogeography. Additional research needs
involve microbial taxonomic definition, soil environmental
complexity, and scaling strategies. The collaborative effort
of experts from diverse disciplines is essential to linking
terrestrial surface biosphere observations with subsurface
microbial community distributions using remote sensing.
Introduction
Microorganisms comprise a significant proportion of terrestri-
al biomass and provide diverse ecosystem services, including
decomposition of wastes, soil fertilization, and water puri-
fication (Whitman
et al.
, 1998). Their metabolism plays an
important role in all known biogeochemical cycles, and they
affect global climate (Bardgett
et al.
, 2008). Despite these con-
tributions, the function and distribution of many microbes are
poorly understood because of the significant complexity in
interactions between biotic and abiotic factors, and between
above- and below-ground ecosystems. The complexity of
the interactions is heterogeneous across scales, and cross-
scale complexity varies further across geographic locations
(Bardgett
et al.
, 2008; van der Heijden
et al.
, 2008). The
substantial difference between the scale at which microbes
act (e.g., micrometer-scale) and the scale at which climate im-
pacts are experienced (e.g., regional, continental, and global
scales) imposes additional challenges. While most microbes
were previously thought to have no spatial patterns, recent
studies have shown spatial patchiness (Ushio
et al.
2010),
restricted distributions (van der Heijden
et al.
, 2008), and
population isolation and endemism (Green and Bohannan,
2006). Because microbial community compositions can affect
the rest of the ecosystem and vice versa, knowledge about
their spatial distributions is critically important in gaining
a comprehensive understanding of terrestrial ecosystems, as
well as in other fields, such as agronomy (Ranjard
et al.
, 2010)
and climate change (Reid 2012).
If we wish to understand the relationship between vegeta-
tion properties, edaphic properties, and microbial community
structure and function,
in situ
observation or experimental
manipulation of soil systems is generally considered to be
required (Paul, 2007). While
in situ
observation of these prop-
erties is undoubtedly the “gold standard” for obtaining such
data (Larsen
et al.
, 2012a), applying these techniques to large
sample sizes and/or large areas is often constrained by cost
and physical accessibility (Graetz 1990; Hamada
et al.
, 2010;
Roughgarden
et al.
, 1991). Botkin (1986) indicated the poten-
tial utility of remote sensing for characterizing environmental
variables for studying soil microbial communities (
SMCs
) near-
ly 30 years ago, but such investigation was not possible until
recently because of limitations in remote sensing technologies
and the understanding of
SMC
structure. Remotely sensed data
provide a synoptic view of the landscape containing land
surface features and properties (Jensen, 2007). The advances
of remote sensing science and technologies, such as very high
resolution (
VHR
) imaging, hyperspectral radiometry, and Light
Detection and Ranging (lidar) systems, in conjunction with
sophisticated data processing models and algorithms, permit
more detailed and reliable characterization and mapping of
terrestrial surfaces than ever before (e.g., Dahlin
et al.
, 2012;
Laliberte
et al.
, 2011). If key terrestrial surface properties that
influence and are influenced by
SMC
dynamics are reliably
characterized and mapped at appropriate spatial and tempo-
ral scales, it will significantly contribute to the investigation
of patterns and associations between environmental factors
and microbial community properties using mathematical
modeling of ecosystem processes (e.g., Larsen
et al.
, 2012a).
This approach has been successfully applied to map distribu-
tions of microbial communities and their functions in marine
environments (Larsen
et al.
2012b).
For terrestrial environments, developing such an approach
is extremely challenging because a highly complex environ-
Yuki Hamada is with Argonne National Laboratory, 9700 S.
Cass Avenue, Bldg 240, Lemont, IL 60439, (
).
Jack A. Gilbert and Peter E. Larsen are with Argonne National
Laboratory, 9700 S. Cass Avenue, Bldg 202, Lemont, IL 60439.
Madeline J. Norgaard is with the College of Saint Benedict
and Saint John’s University, 37 College Avenue, St. Joseph,
MN 56374.
Photogrammetric Engineering & Remote Sensing
Vol. 80, No. 4, April 2014, pp. 311–321.
0099-1112/14/8004–311
© 2014 American Society for Photogrammetry
and Remote Sensing
doi: 10.14358/PERS.80.4.311
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
April 2014
311
