8
Stable Isotope Tracers and Mathematical
Models in Soil Organic Matter Studies
Ronald Amundson and W. Troy Baisden
Introduction
Light stable isotopes have become widely used biogeochemical tracers in earth science and ecology
research during the last half of this century. With
some exceptions, these applications have originated
as a result of pioneering geochemical research at
the University of Chicago and the California Institute of Technology in the 1950s and 1960s (Taylor
et al. 1991). The breadth of isotopic tracer studies
in present-day ecosystem sciences is now so great
that it eludes even the most ambitious review article
or book.
In this chapter, we limit our focus to the use of
natural abundance levels of stable carbon (C) and
nitrogen (N) isotopes as tracers of soil organic matter (SOM) cycling. The use of natural abundance
levels of carbon-14 e 4 C) as a SOM tracer has been
discussed elsewhere (Trumbore 1993). Stable isotopes offer alternatives to 14C as a tracer in certain
settings: lower cost per analysis and (particularly
with N) the ability to trace different SOM processes. We define SOM as a continuum of substances ranging from fresh litter to humic substances in the mineral soil, all of which is ultimately
derived from both above- and below ground plant
material. As we discuss below, the isotopic composition of SOM reflects the balance between organic inputs and losses-losses (in the case of C)
being dominated by heterotrophic respiration of
carbon dioxide (C0 2 ), From this perspective, this
chapter is a companion piece to our recent article
on the carbon and oxygen (0) isotope composition
of soil and soil-respired CO 2 (Amundson et al.,
1998). In both, our approach is the same: the isotopic composition of both organic matter and CO2
can each be described by their own mass balance
expressions, the expressions differing slightly for
the individual isotope being considered. In both, we
review the pioneering work and modeling of other
researchers, supplementing and expanding it with
recent modeling and empirical measurements in our
laboratory. We hope to provide a clear conceptualization of the mathematics required to construct
realistic models of the behavior of C and N isotopes
in SOM.
The key to modeling is identifying relevant pools
and fluxes that must be represented mathematically.
In the next section, we discuss SOM pools and
identify the fluxes of C and N that are relevant to
understanding stable isotope dynamics in SOM.
Soil Organic Matter Pools
and Dynamics
SOM Pools
Conceptually, SOM can be considered in terms of
"boxes" or "pools." One common division in terms
of modeling has been to consider 0 horizons (if
present) and mineral horizons separately. In this
chapter, we focus on the stable isotope composition
of SOM in mineral horizons although our discussion is applicable, with modification, to 0 horizons.
We will take two approaches in our view of the
mineral soil SOM pool: (1) We will review simple
models that consider SOM as a homogenous, wellmixed pool and (2) we will review and develop
models that recognize that the characteristics of the
117
Stable Isotope Tracers and Mathematical
Models in Soil Organic Matter Studies
Ronald Amundson and W. Troy Baisden
Introduction
Light stable isotopes have become widely used biogeochemical tracers in earth science and ecology
research during the last half of this century. With
some exceptions, these applications have originated
as a result of pioneering geochemical research at
the University of Chicago and the California Institute of Technology in the 1950s and 1960s (Taylor
et al. 1991). The breadth of isotopic tracer studies
in present-day ecosystem sciences is now so great
that it eludes even the most ambitious review article
or book.
In this chapter, we limit our focus to the use of
natural abundance levels of stable carbon (C) and
nitrogen (N) isotopes as tracers of soil organic matter (SOM) cycling. The use of natural abundance
levels of carbon-14 e 4 C) as a SOM tracer has been
discussed elsewhere (Trumbore 1993). Stable isotopes offer alternatives to 14C as a tracer in certain
settings: lower cost per analysis and (particularly
with N) the ability to trace different SOM processes. We define SOM as a continuum of substances ranging from fresh litter to humic substances in the mineral soil, all of which is ultimately
derived from both above- and below ground plant
material. As we discuss below, the isotopic composition of SOM reflects the balance between organic inputs and losses-losses (in the case of C)
being dominated by heterotrophic respiration of
carbon dioxide (C0 2 ), From this perspective, this
chapter is a companion piece to our recent article
on the carbon and oxygen (0) isotope composition
of soil and soil-respired CO 2 (Amundson et al.,
1998). In both, our approach is the same: the isotopic composition of both organic matter and CO2
can each be described by their own mass balance
expressions, the expressions differing slightly for
the individual isotope being considered. In both, we
review the pioneering work and modeling of other
researchers, supplementing and expanding it with
recent modeling and empirical measurements in our
laboratory. We hope to provide a clear conceptualization of the mathematics required to construct
realistic models of the behavior of C and N isotopes
in SOM.
The key to modeling is identifying relevant pools
and fluxes that must be represented mathematically.
In the next section, we discuss SOM pools and
identify the fluxes of C and N that are relevant to
understanding stable isotope dynamics in SOM.
Soil Organic Matter Pools
and Dynamics
SOM Pools
Conceptually, SOM can be considered in terms of
"boxes" or "pools." One common division in terms
of modeling has been to consider 0 horizons (if
present) and mineral horizons separately. In this
chapter, we focus on the stable isotope composition
of SOM in mineral horizons although our discussion is applicable, with modification, to 0 horizons.
We will take two approaches in our view of the
mineral soil SOM pool: (1) We will review simple
models that consider SOM as a homogenous, wellmixed pool and (2) we will review and develop
models that recognize that the characteristics of the
117