Modelling Soil Organic Matter Dynamics – Global Challenges - Sustainable Management of Soil Organic Matter - page 36

CO 2 increases at national and regional levels (Dumanski et al ., 1997; Smith
et al ., 1997a). If soil sinks are included under the Kyoto protocol, methods
to quantify and verify changes in soil C at local, regional and national levels
will be required. The potential for national and international trading of
carbon emission offsets, making soil carbon a commodity, will necessitate
that carbon changes be quantified to known levels of uncertainty. SOM
modelling can play a central role in such quantification systems, but it will
require a new level of integration with field measurements and other data
on environmental and management factors.
Several of the SOM models currently in use today were first developed
in the late 1970s and early 1980s (e.g. Jenkinson and Rayner, 1977; Parton
et al ., 1983). Since then, many new models have been developed (see
reviews by Jenkinson, 1990; Paustian, 1994; McGill, 1996) and older mod-
els have continued to evolve on a more or less continuous basis. Despite
their diversity, most of these models share some basic assumptions which
include the representation of SOM as multiple pools (or as a quality spec-
trum) with differing inherent decomposition rates, governed by first-order
rate constants modified by climatic and edaphic (e.g. soil physical attrib-
utes) reduction factors. Most of these models were conceived originally to
describe processes at the ecosystem or field scale. New approaches continue
to be explored, and some of the major recent trends in SOM modelling and
their application to environmental problems will be discussed below.
Collaborative Networks
A significant recent development in SOM modelling has been the increased
level of formal and informal collaborations between different modelling
groups and between modellers and experimental scientists. Whether this
has occurred as a result of the greater international collaboration spurred
by global change research, or simply as a result of increasing ease of com-
munication and the maturation of the science, is perhaps not important.
Whatever the case, such collaboration has facilitated the comparison and
evaluation of different models and approaches, such as a recent workshop
in which a number of SOM models from around the world were brought
together and evaluated using a common set of long-term experimental
data from a variety of climatic, soil and land use conditions (Powlson et al .,
1996; Smith et al ., 1997b). Such comparisons help to clearly identify the
difference in structure and assumptions of existing models (Paustian, 1994;
McGill, 1996) and their relative strengths and weaknesses for a particular
set of circumstances. These developments mirror similar recent efforts
to conduct cross-model evaluations for water quality modelling, trace gas
modelling and general circulation climate modelling, to name a few
examples.