Journal of the Department of
Agriculture, Western Australia,
Series 4
Volume 42
Number 1 Journal of Agriculture, 2000/2001
Article 5
2001
WA soil erosion under investigation
George Richard Dr
[email protected]
Follow this and additional works at: http://researchlibrary.agric.wa.gov.au/journal_agriculture4
Part of the Environmental Monitoring Commons, Natural Resources and Conservation Commons, and the Soil Science
Commons
Recommended Citation
Richard, George Dr (2001) "WA soil erosion under investigation," Journal of the Department of Agriculture, Western Australia, Series 4: Vol.
42 : No. 1 , Article 5.
Available at: http://researchlibrary.agric.wa.gov.au/journal_agriculture4/vol42/iss1/5
This article is brought to you for free and open access by Research Library. It has been accepted for inclusion in Journal of the Department of
Agriculture, Western Australia, Series 4 by an authorized administrator of Research Library. For more information, please contact
[email protected], [email protected].
IMPORTANT DISCLAIMER
This document has been obtained from DAFWA’s research library website
(researchlibrary.agric.wa.gov.au) which hosts DAFWA’s archival research publications. Although
reasonable care was taken to make the information in the document accurate at the time it was first
published, DAFWA does not make any representations or warranties about its accuracy, reliability,
currency, completeness or suitability for any particular purpose. It may be out of date, inaccurate or
misleading or conflict with current laws, polices or practices. DAFWA has not reviewed or revised the
information before making the document available from its research library website. Before using the
information, you should carefully evaluate its accuracy, currency, completeness and relevance for your
purposes. We recommend you also search for more recent information on DAFWA’s research library
website, DAFWA’s main website (https://www.agric.wa.gov.au) and other appropriate websites and
sources.
Information in, or referred to in, documents on DAFWA’s research library website is not tailored to the
circumstances of individual farms, people or businesses, and does not constitute legal, business,
scientific, agricultural or farm management advice. We recommend before making any significant
decisions, you obtain advice from appropriate professionals who have taken into account your individual
circumstances and objectives.
The Chief Executive Officer of the Department of Agriculture and Food and the State of Western
Australia and their employees and agents (collectively and individually referred to below as DAFWA)
accept no liability whatsoever, by reason of negligence or otherwise, arising from any use or release of
information in, or referred to in, this document, or any error, inaccuracy or omission in the information.
A survey of soil erosion at 70 sites throughout Western Australia, recently completed as a part of a National
Reconnaissance Survey, has shown that erosion is occurring at an unsustainable rate. While still to be compiled into
a national publication, Dr Richard George' reports on the Western Australian results, and discusses options for
management and further investigation.
Soil loss, or sheet erosion through wind and
water, represents a significant threat to the
sustainability of land not subject to salinisation.
Accordingly, an Australia-wide assessment of
soil loss was initiated in 1990 by the National
Soil Conservation Program (now Natural
Heritage Trust). Each Australian State and
Territory used a standard sampling method to
compare the relative abundance of the
radioactive tracer, Caesium-137, with uneroded
and others sites to assess soil loss.
Samples were collected at each site and sent for
analyses to the Australian Nuclear Science and
Technology Organisation (ANSTO) laboratory
in Lucas Heights, New South Wales. Results for
Western Australia indicated that:
•
•
•
•
about 10 per cent of the sites had soil losses
in excess of 20 tonnes per hectare per year,
or approximately 15 to 20 millimetres per
decade;
about 30 per cent of the sites had soil losses
of 5 - 20 tonnes per hectare per year, or
more than 3 to 5 millimetres per decade;
about 40 per cent of the sites had soil losses
of 1 to 5 tonnes per hectare per year;
about 25 per cent of the sites had negligible
soil loss (less than 1 tonne per hectare per
year).
The highest rates of cumulative soil loss
occurred where the following land uses were
prevalent:
•
grazed pastoral areas in the Kimberley and
West Pilbara-Gascoyne regions;
J O U R N A L OF A G R I C U L T U R E
•
sloping gravelly soils in the South West
woolbelt (500 to 700 millimetres rainfall);
Intense summer storms
wheatbelt slopes used for crop - pasture
rotations;
removing nutrient rich
can cause significant
erosion and damage,
•
sediment and organic
matter. Erosion rates in
•
high rainfall sites used for intensive
horticulture (row cropping).
the woolbelt can be
very high when
associated with
Lowest soil losses were associated with
permanent pasture land on low slopes, provided
they had not been subject to wind erosion, and
many pasture-based systems in the woolbelt and
wheatbelt regions also had lower soil loss rates.
In addition, pastoral areas in the Kalgoorlie and
Eucla areas appeared less eroded, although
there were too few samples to conclude why
soil loss rates were lower in these particular
areas.
cultivation.
1 - Senior Research Scientist,
the Department of
Agriculture, Bunbury
15
Vol 42 2000/01
Table 1. Locations of Western Australian sites sampled for ,37Cs.
Locations
Land use
Kimberley
Rangelands
Annual rainfall
(mm/year)
Number
of sites
500-700
9
West Pilbara-Gascoyne
Rangelands
4
<300
Southern rangelands
Rangelands
6
<300
Wheatbelt
Crop pasture
27
300-500
Woolbelt
Pasture
7
500-700
South West
Pasture - horticulti re
17
700-1400
Table 2. Summary of net soil loss rates in relation to land use.
Net soil loss
(t/ha/yr)
Pasture &
cropping
sites
Horticulture
(potato) sites
Annual
pasture
sites
Rangeland
sites
%
30-49.9
2
0
0
0
2.9
20-29.9
1
0
0
4
7.1
10-19.9
6
1
1
4
17.1
5-9.9
3
1
0
3
10
1-4.9
16
3
3
5
38.6
<1
7
1
6
3
24.3
%
50
8.6
14.3
27.1
100
Methodology
The Caesium technique was considered the best
indicator of soil loss. Caesium-137 (137Cs) is a
radioisotope that was released during
atmospheric nuclear weapons tests in the
mid-1950s, and the fallout from these tests has
been used as a tracer of soil movement.
The 137Cs isotope, which has a half-life of 30.5
years, is strongly fixed to the soil and is lost
when soils are eroded. By comparing the
amount of isotope in an agricultural soil with
nearby areas that had not been disturbed since
the 1950s, it was possible to estimate how
much soil had been lost in the past 50 years.
Of the 70 sites surveyed, 50 per cent had been used for annual
pasture-crop rotations, more than 25 per cent were rangelands
located in the West Pilbara, Gascoyne, Kimberley and the State's
southern regions, and the remainder were annual pastures or
under potato-vegetable-cropping rotations (see Table 1).
The usefulness of the 137Cs technique is often
debated as it is strongly influenced by site
history and land use. However, the level of
erosion estimated using this technique appeared
robust in the many areas sampled, so long as
care was taken during the survey and when
interpreting site-specific results.
For example, soil loss rates were dependent on
the relative loss between sites chosen as
reference sites, the sampling techniques used,
16
and any recent history of soil disturbance (eg.
date of clearing and cultivation).
McFarlane et al (1992), Loughran et al (1993),
and Harper and Gilkes (1994) had also
conducted previous studies of soil erosion using
the 137Cs isotope in Western Australia. All
reports from these surveys indicated that the
137
Cs technique was appropriate as a tool to
assess variations in soil loss, so long as the field
methods and interpretations were supported by
sound field evidence. In particular, the
estimates of erosion are dependent on the
accuracy of the reference value. At some sites,
especially those cleared after 1955, this
assumption may not always be valid.
For this most recent survey, the method proved
useful in providing a long term or cumulative
measure of soil loss, but could not determine
when the losses had occurred. It was therefore
possible that some sites were heavily eroded,
but had subsequently been stabilised.
Further analyses and more thorough sampling
will be required before the 137Cs technique can
be used as a quantitative or predictive tool.
Sites sampled
Over 70 sites were sampled throughout Western
Australia. At each site, between 4 and 20 cores
(average of 10) were taken on a hillslope
(catena), taking note of soil conditions, land use
and site history.
Sample cores were extracted from a depth of 20
centimetres (maximum 40 centimetres), and
were sieved and weighed before being sent for
analysis to the Australian Nuclear Science and
Technology Organisation in New South Wales.
Net soil loss was calculated by comparing the
field value to samples taken from a reference
site and other local sites. The field value was
also compared to soil loss rates derived for both
cultivated and pasture land uses using predictive
equations from elsewhere in Australia (see
McFarlane et al, 1992, 2000).
Discussion
Impact
The impact of soil loss is dependent on land
use, timing of the loss, soil characteristics (eg.
depth and nutrition) and soil formation rates.
In eastern Australia, soil loss rates of less than
0.5 tonnes per hectare per year have been
regarded as equivalent to 'natural rates' of
erosion in pasture cropping systems. Rates of
1.2 tonnes per hectare per year are considered
J O U R N A L OF A G R I C U L T U R E
Vol 42 2000/01
acceptable in areas where there are higher rates
of soil formation, such as might be expected in
the wetter and northern rangelands. The
United States Department of Agriculture
considers 5 tonnes per hectare per year as a
tolerable soil loss rate.
In higher rainfall areas
the combination of
steep slopes, intensive
gazing and heavy
rainfall at the break of
the season creates a
high risk of erosion.
Soils in this area are
In the South West of Western Australia, where
soil formation is likely to be the lowest in
Australia, perhaps in the order of one metre
every million years, soil loss at almost any rate
is unlikely to be sustainable.
With results from this study indicating loss
rates of typically more than 5 tonnes per
hectare per year or approximately 3 to 5
millimetres per decade, and extreme rates of
erosion equivalent to a loss of 35 to 50
millimetres per decade, mismanagement of
erosion can have a serious effect on agriculture.
With such low soil formation rates and the
extremely low nutrition status typical of soils in
Western Australia, erosion can lead to the loss
of any inputs of agriculture used over the last
50 years in building up these soils.
In addition, soil erosion has offsite impacts.
For example, eroding soils can carry nutrients
to streams and cause eutrophication, while the
sediment deposition from storms may result in
direct damage to infrastructure and the
environment.
In other areas, soil loss may impact in different
ways. For example, sediment loads into Lake
Argyle from rangelands have been recorded at
over 24 Megatonnes per year, and potentially
threaten to shorten the design life of Lake
Argyle. In other areas, soil loss may remove
usually well covered by
winter and longer term
rates of erosion can be
minimised.
finer textured soils, such as clays and silts,
which may lead to the loss of soil nutrients and
the lowering of potential yields, or during wind
events, result in direct crop damage due to
sandblasting.
Management
Management of soil loss depends on land use
and site conditions. In cropping areas the use
of grade banks, contour systems, no-tillage, and
stubble retention is likely to result in lowered
rates of soil loss. Pastured areas, when mixed
with cropping, are especially vulnerable to
erosion in dry years and careful grazing
management is required.
Results
There were 23 sites sampled within the pasture-cropping rotation and rangelands sites that had net soil losses between
5 and 49.9 tonnes per hectare per year, and there were two potato-vegetable-pasture rotation sites and one annual
pasture site with the same order of erosion.
In the soil loss category of 1 to 4.9 tonnes per hectare per year, there were between 3 and 16 occurrences in each land use
group, the greatest number being under a pasture-cropping rotation. Approximately 25 per cent of sites had soil losses
of less thanl tonne per hectare per year.
Two sites had net soil losses in the highest category (30 to 49.9 tonnes per hectare per year), both with pasture-cropping
rotations. Little ,37Cs remained on either transect.
The Kimberley (13.5 tonnes per hectare per year) and West Pilbara-Gascoyne regions (14.5 tonnes per hectare per year)
had the highest average net soil losses. Land uses in the southern rangelands typically had low rates of net soil loss
(1 tonne per hectare per year). The horticulture-pasture belt (4.6 tonnes per hectare per year) had relatively high erosion
(particularly associated with potato cropping), which was consistent with other independent studies carried out in the same
area.
The average soil loss in the crop-pasture region (wheatbelt) was 6.6 tonnes per hectare per year, while in the woolbelt
(pasture-crop), it was 9.8 tonnes per hectare per year. Many of the sites in this region were known to have experienced
serious sheet erosion, caused by heavy grazing and summer storms.
Analysis of the impact of removing all sites cleared after 1955 (21 sites) from the database did not significantly alter the
magnitude of soil loss, or change the relative erosion ranking of the regional land uses.
J O U R N A L OF A G R I C U L T U R E
17
Vol 42 2000/01
Acknowledgements
The author thanks Department of Agriculture
colleagues Russell Speed and Dan Carter for
dieir significant contribution to the survey,
including report editing.
The author is also grateful to the landholders of
Western Australia and colleagues in the
Department of Agriculture for their help,
including Ian Watson, Paul Findlater, Geoff
Elliott, John Simons, Jan-Paul van Moort, Rob
Hetherington, Peter Tille, Don Bennett and
Geoff Russell.
This survey has not been a definitive study of the erosion status
of Western Australian soils, although the indications from the
range of erosion rates encountered show there is cause for
concern in the management of our soil resources.
Erosion from upland
areas accumulates in
lower slopes that are
also prone to salinity.
Saline areas are
unstable without plant
cover and in turn allow
sediment to be moved
to streamlines and river
pools.
Mr Greg Elliot (ANSTO) managed the sample
analysis process. Dr Bob Loughran
(Department of Geography, University of
Newcastle, New South Wales) and Dr Don
McFarlane (now with Water and Rivers
Commission, Perth), managed the project and
were senior authors of the final report - A
National Reconnaissance Survey of Soil Erosion
of Australia, Western Australia.
Assessing soil erosion in
the State's South West.
References
In areas such as the Swan Coastal Plain, wind
breaks and crop residue management may result
in lower erosion rates, while in the rangelands,
livestock and feral animal management may
produce similar outcomes.
Conclusions
For a truer indication of the State's erosion
status, the survey should be expanded to take
into account the natural variation of land use,
landscape and erosion rates. This was not
possible with the resources available at the time
of the survey. Also, analyses of erosion,
independent of the 137Cs technique, should be
employed before firm conclusions are drawn.
The results of the National Reconnaissance
Survey indicate that land management requires
continued attention, both in terms of research
and extension of methods that are currently
available for erosion control.
18
J O U R N A L OF
Harper, R.J. and Gilkes, R.J. (1994).
Evaluation of the ™Cs technique for estimating wind erosion losses for some sandy
Western Australian soils. Australian Journal
of Soil Research, 32: 1369-1387.
Loughran, R.J., McFarlane, D.J., Campbell,
B.L. and Shepherd, R. (1993). The distribution of caesium-137 in rangeland soils at
three sites in Western Australia. Rangeland
Journal, 15 (1): 24-38.
McFarlane, D.J., Loughran, R.J. and Campbell,
B.L. (1992). Soil erosion of agricultural
land in Western Australia estimated by
caesium-137. Australian Journal of Soil
Research 29: 533-546.
McFarlane, D.J., George, R.J., Loughran, R.J.,
Elliott, G.L., Ryder, A.T., Bennett, D. and
Tille, P.J. (2000). A National
Reconnaissance Survey of Soil Erosion of
Australia, Western Australia, ISBN
0725909919, University of Newcastle,
NSW, 2308, 264.
AGRICULTURE
Vol 42 2000/01