Soil and Hillslope (Eco)Hydrology - Environmental Modelling: Finding Simplicity in Complexity - page 168

between bulk density and earthworm-community

to the surface may be of the order of 20-40 t ha − 1 y − 1 .
To illustrate how striking the effect of earthworms can
be, Lavelle et al . (2006) cite examples where 'invasion' of
soils by worms has led to rapid changes in the physical
and biochemical properties of those soils. One of the
examples is Chauvel et al . (1999) who studied soil com-
paction in Amazonian pastures near Manaus in N Brazil
created after deforestation. Chauvel et al . (1999) suggest
that the earthworm Pontoscolex corethrurus Miller - an
exotic species that rapidly colonizes such pastures - could
cause as much or even more compaction than machin-
ery (during tree removal) or cattle (after tree removal).
They attribute this effect to the churning and near total
dispersion of soil particles in the earthworms' guts, such
that the casts produced by the worms form a higher
density soil with a lower macroporosity than that which
was ingested. Unfortunately, Chauvel et al . (1999) do not
provide details of their methods or of the replication of
treatments in their experiments, so it is difficult to judge
the reliability of their findings. It is noteworthy that other
studies, such as Zund et al . (1997) have suggested the
exact opposite of Chauvel et al . (1999) - that the same
species of earthworm can cause decreases in bulk density
and increases in macroposity. Zund et al . (1997) per-
formed controlled experiments on replicated treatments,
and the changes they found in soil properties - which
included reductions in soil bulk density of 5-17% in
the upper 10 cm of the soil - were after an experimental
period of 12 weeks. However, their study was based on
initially homogenized soil held in laboratory pots and it
is unclear how this affected their results and how much it
might explain the difference in their results from those of
Chauevel et al . (1999). Because Chauvel et al . (1999) do
not provide details on their methods and experimental
replication, it is impossible to compare the two studies
and explain the apparently diametrically opposite find-
ings. Regardless of which is the more reliable study, it
is clear that earthworms can cause substantial and rapid
changes in soil properties even when that soil is (in a
sense) artificial.
Other soil invertebrates may also have pronounced
effects on soil hydrological processes. For example,
in a review of the effects on Australian soils of the
Aphaenogaster genus of ants, Richards (2009) notes
that ants may be responsible for more bioturbation
than earthworms - mounding rates from ants in the
genus may exceed 5 t ha − 1 y − 1 - and that their nests
(which have funnel-like openings at the soil surface, with
tunnels and galleries extending to depths of up to 2 m)
can control the partitioning of water between the soil
surface (overland flow) and the subsurface (infiltration).

structure.
It is interesting that Margerie et al . (2001) found a
mismatch between surface pattern (in the vegetation)
and subsurface pattern. Such a mismatch emphasizes the
danger of using surface pattern to infer subsurface pro-
cesses and also confirms the cautionary tale of the dryland
study of Mueller et al . (2007) mentioned in Section 10.2
that patterns of a range of soil hydrological and chemical
properties do not necessarily coincide and may not mirror
vegetation patterns. The subsurface patterns modelled by
Mueller et al . (2007) were the result of water-flow and
chemical-transfer processes but some pattern mismatches
may indicate an ecological memory effect, and such an
effect is discussed in Section 10.4.
The role of soil invertebrates, such as earthworms,
as 'bio-engineers' has been put into a wider context
by Lavelle et al . (2006) who make a convincing case
for soils being CAS. They suggest that soils have the
attributes of CAS as identified in Section 10.1 and that
soil invertebrates play a key role in soil formation and the
maintenance of biophysical structures across a range of
scales. In particular, they note:

Soil invertebrates are key mediators of soil function for
the diversity of ecosystem engineering processes in which
they partake. The comminution and incorporation of
litter into soil, the building and maintenance of structural
porosity and aggregation in soils through burrowing,
casting and nesting activities, the control of microbial
communities and activities, plant protection against some
pests and diseases, acceleration of plant successions are
among the many effects they have on other organisms
through their activities ... In so doing, they develop
multiple interactions with other organisms, at different
scales and across the whole range of chemical, physical
and biological processes that sustain the provision of soil
ecosystem services. (p. S6)

Taking just the effect of earthworms on soil physi-
cal properties, it is worth citing Darwin (1881) again.
Although individually seemingly insignificant, an earth-
worm, through its burrowing, may radically change the
infiltrability of a soil over a scale of a few tens of square
centimetres. At the scale of a square metre, variations in
earthworm density can cause variations in soil physical
properties. Moving up two or three orders of magnitude
in linear scale (i.e. from 10 0 mto10 2 -10 3 m), numbers
of earthworms may exceed 100 000 ha − 1 (p. 158; here
Darwin cites the work of Hansen but does not provide the
full details of the paper written by Hansen; see also Poier
and Richter, 1992) and the amount of material brought