Soil Formation - Soil Carbon: Science, Management and Policy for Multiple Benefits - page 93

and Bliss, 2002) show that over a timespan
that is miniscule compared to many geo-
logic processes, agriculture has transformed
the landscape. As a consequence, Amund-
son et al . (2003) estimate that in the USA,
mollisols (which correlate with grassland
vegetation) have lost 28% of their undis-
turbed area to land-use change - primarily
agriculture. Pimentel (2006) and Montgom-
ery (2007) estimate that erosion rates from
agricultural fields average 1-2 orders of
magnitude greater than rates of soil produc-
tion, with potentially negative consequences
for human food security and environmental
quality. As described above, these elevated
erosional rates have consequences for local
soil function because soil erosion is associ-
ated with the destruction of soil aggregates,
leading to organic C loss as previously pro-
tected C is exposed to weathering (Lal, 2003).
Cumulatively, loss of organic C from agricul-
tural and pasture lands may even have influ-
enced the planetary carbon balance (e.g. Lal,
2003; Janzen, 2004; van Oost et al ., 2007;
Chapter 20, this volume).
Humans have also had a significant
role in altering soil organic matter at scales
from local to planetary (Chapter 3, this volume).
The authors of Chapter 3, this volume, pro-
pose the existence of a soil carbon transi-
tion curve. The curve consists of a rapid
decline of the initial endowment of soil
carbon due to human clearing of natural
vegetation for agricultural land use and
management practices such as conventional
tillage, followed by a phase of greatly di-
minished soil fertility resulting from deple-
tion of this initial stock, and finally by
recovery of soil organic content once agri-
cultural practices improve. Lal and Follett
(2009) present generalized trends of relative
soil carbon loss as a function of time since
conversion from native to managed land
status. This loss is rapid in the initial dec-
ade or so after conversion. Over periods of
several decades, carbon loss (driven in part
by enhanced erosion) may amount to up to
80%. The absolute amounts vary with the
setting. For example, carbon loss may amount
to 50-70 Mg ha - 1 in tropical rainforests,
30-35 Mg ha - 1 in prairies and as high as
200-220 Mg ha - 1 in peatlands.

Conversely, recovery in soil organic
carbon has taken place in many parts of the
globe owing to practices such as no-till agri-
culture. Follett et al . (2009) reported that
modern farming practices across a signifi-
cant portion of the USA led to a mass of car-
bon in the top 100 cm that averaged 78% of
that in paired native sites. This topic is ex-
plored much more comprehensively in
Chapter 3, this volume.

The Importance of Critical Zone
Observatories

The science of Earth's critical zone com-
bined with environmental observatory re-
search methodology provides a powerful
framework to advance understanding of soil
functions and soil threats. Critical Zone
Observatories (CZOs) are field research fa-
cilities with intensely focused multidiscip-
linary research (cf. CZO Special Issue, Vadose
Zone Journal , Volume 10, 2011). The experi-
mental design and methodology is strongly
hypothesis driven, with a focus on nested
scales of observation from molecular-to-
grain, profile-plot, catchment and basin
scales. There is a strong emphasis on pro-
cess understanding and integration of the
multiple scales of observation with inter-
pretation through mathematical modelling
and computational simulation.
The vertical integration of process under-
standing, linking the aboveground vegetation
with soil functions and deeper regolith and
aquifer processes, is particularly powerful in
assessing the chain of impact that results from
environmental change. Aboveground changes
in climate driven by global warming, or land-
use change driven by demographic develop-
ments, set off a chain of impacts that are
transmitted in many cases through the result-
ing changes in soil functions. Defining and
quantifying this chain of impact allows as-
sessment of the potential consequences of
change, and the benefits of different interven-
tion options. CZOs provide the essential data
sets and process models to enable these as-
sessments, and their link to important social
and economic drivers of change.