Soil Erosion and Conservation - Environmental Modelling: Finding Simplicity in Complexity - page 358

Table 22.3 Average magnitudes (absolute values) of erosivity change calculated.

Average magnitude of change
40-yr. interval 80-yr. interval

Model scenario RP RF RP RF

(%) (%) (%) (%)

HadCM3 11.8 22.5 15.9 20.9

CGCM1 HG + A1 23.4 29.1 53.4 58.3

Lafayette, IN, Temple, TX, and Corvallis, OR. Soils were
sandy clay loam, silt loam, and clay loam. Crops included
grazing pasture, corn and soybean rotation, winter wheat,
and fallow. Slopes were 3, 7, and 15%. Three scenarios of
precipitation changes were considered: (a) all precipita-
tion change occurring as number of days of rainfall, (b) all
precipitation change occurring as amount of rainfall in a
given day, and (c) half of the precipitation change occur-
ring from each source. Under these scenarios, and using
the climate generator for WEPP, changes in the number
of days of rainfall does not influence rainfall intensity,
whereas changes in the amount of rainfall on a given
day increases the duration, peak intensities, and average
intensities of rain. Levels of changes considered in each
case were approximately zero, ± 10%, and ± 20% of total
precipitation, with the same relative proportion of pre-
cipitation for the year maintained as a function of month.
Erosion rates changed much more with changes in the
amount of rainfall per precipitation event, which also
implies changes in the rainfall durations and intensities
for the events. When total precipitation in this case was
increased 10% in this case, soil loss increased an average
of 26%. Realistically, we can expect that any changes in
precipitation will come as a combination of both changes
in the number of wet days as well as in changes in the
amount and intensities of rainfall. As we discussed earlier,
historical changes in rainfall over the past century have
occurred in both of these terms (Karl et al ., 1996). For the
combined case of both changes in wet days and changes
in rainfall per day, Pruski and Nearing (2002) found that
erosion responded intermediate to the two extremes. For
a 10% increase in total precipitation, simulated erosion
increased an average of 16%.
The average results for the combined case of changes
in both number of days of precipitation and changes in
amount of rain per day from the study of Pruski and
Nearing (2002) are similar to those for the empirical
relationship proposed by Renard and Freimund (1994)

between erosivity and total annual precipitation for the
RUSLE model as discussed above. Using Renard and
Freimund's first equation for erosivity results in a 17%
change as a function of a 10% change in total annual pre-
cipitation. However, it is important to note that regardless
of this fact, obtaining the broad-scale information on
erosivity change similar to the information we obtained
from the study discussed in the previous section (Nearing,
2001) would have been extremely difficult using WEPP.
Now let's look at some of the details of the results
from the WEPP erosivity study. Greater amounts and
rates of runoff, other factors being equal, will generally
tend to cause an increase in erosion. Increased runoff
causes increased energy of surface flow, which increases
the detachment capability and the sediment transport
capacity of the flow. Interrill erosion also increases with
increased rain.
The simulation results of Pruski and Nearing (2002)
showed a general increase in soil loss with increase
in precipitation, and vice versa (Table 22.4), however,
the changes were generally not as great as for runoff
(Table 22.5). One major reason for the difference between
the sensitivity results for runoff and those for soil loss is
related to biomass production. Both runoff and soil loss
are sensitive to biomass, but soil loss is more so. Soil loss is
affected by plant canopy, which reduces the impact energy
of rainfall; by crop residues, which protect the soil from
raindrop impact and reduce rill-detachment rates and
sediment-transport capacities; and from subsurface roots
and decaying residue, which mechanically hold the soil in
place and provide a medium in which micro-organisms
can live. Thus, the increase of biomass production with
increased rainfall tends to counteract to some degree the
increased erosivity of the rain. This argument is supported
by the results of the simulations for fallow conditions in
comparison to the other treatments. The sensitivity values
for the three precipitation scenarios for fallow conditions
average 1.63 for soil loss and 1.55 for runoff. Thus fallow