Australian Natural Resources Atlas

Natural Resource Topics

Salinity - Impacts & Costs - Western Australia

Western Australia

Location map

Dryland Salinity Impacts : Western Australia overview

The groundwater analysis was used to identify risks to infrastructure, water resources, biodiversity and agricultural land, based on the assumption that current groundwater trends continue.

Table: Key assets in areas at risk from dryland salinity in Western Australia:
Assets 2000 2020* 2050*
Agricultural land (ha) 3 552 700 4 181 700 6 490 100
Perennial vegetation (ha) 600 000 710 000 1 800 000
Important wetlands (ha) 72 500 72 500 80 000
Highways (km) 720 840 1 500
Primary roads (km) 680 745 1 165
Secondary roads (km) 1 200 1 425 2 325
Minor roads (km) 11 550 13 650 22 930
Rail (km) 1 350 1 490 2 180
Stream length (km) 1 520 1 700 2 850
Towns (number) 20 22 29
Important wetlands (number) 21 21 21

* Predictions based on current groundwater trends, and ?best guess? future land use.

Broad economic analysis of the risks to infrastructure, biodiversity and agriculture.

As part of the Audit, a broad economic analysis was carried out after identifying the impacts of shallow groundwater on infrastructure, biodiversity and agriculture. The indicative estimates of the costs of salinity on and off farm in Western Australia were prepared by Allan Herbert, Economics Group, Agriculture Western Australia.

Details and comments on the methodologies used and range testing appear in Appendix 14. Results were obtained for three impact topics (infrastructure, biodiversity, and agriculture) for three time periods - year 2000, year 2020, and year 2050. No attempt was made to quantify water resources in dollar terms. No attempt was made to cost solutions to the problem in a benefit:cost context.

What, and how much, agricultural land occurs in areas at high risk from dryland salinity?

Agricultural areas were identified as those areas not occupied by perennial vegetation. Areas were allocated to systems and total area of risk for each zone determined. Areas at high risk are shown in the table below (a full data set can be found in Appendix 13 in the Western Australian Dryland Salinity Assessment 2000 report). The data show that shallow watertables currently underlie 3.5 million hectares of agricultural land. These areas have the potential to be saline. It is predicted that this area in WA could expand to 6.5 million hectares by 2050. This figure is similar to those proposed by Ferdowsian et al. (1996) who estimated the potential extent of salinity in cleared land as 6.1 million hectares.

Table: Agricultural land at risk (ha) in Western Australia.
Soil Landscape Zones High Risk
2000		 High Risk
2020		 High Risk
2050		 Percentage
of zone allocated data		 Zone area as percentage of south west agricultural areas
211 11,929 11,929 11,929 98 1.57
212 0 0 187,368 100 1.50
213 212,779 212,779 212,779 99 1.01
221 0 0 0 70 1.22
222 125,802 125,802 125,802 25 2.55
223 0 0 0 0 2.55
224 0 0 9,763 2 2.09
225 0 0 0 75 0.97
226 32,641 32,641 54,421 36 0.59
231* 0 0 0 0 0.14
232* 0 0 0 0 0.17
233* 0 0 0 0 0.27
241 26,411 31,567 260,333 85 1.80
242 0 60,315 100,834 93 3.49
243 314,228 353,385 353,385 97 2.02
244 0 0 0 100 1.23
245 65,140 246,571 603,961 77 3.92
246 0 7,110 7,110 88 6.99
247 16,640 16,640 16,640 71 0.55
251 0 0 1,867 73 0.37
252 22,372 22,372 33,235 85 1.75
253 73,928 151,182 636,056 95 4.54
254 0 0 247,118 74 5.70
255 0 0 0 85 4.48
257 359,996 618,532 974,259 67 11.11
258 1,321,296 1,321,296 1,361,028 90 19.58
259 635,446 635,446 958,088 82 13.88
261* 0 0 0 0 0.90
271 334,127 334,127 334,127 51 2.76
272* 0 0 0 0 0.29
Total area (ha) 3,552,735 4,181,694 6,490,103
Total (% of SW region) 13 16 24

* = Insufficient data.

The current and future estimates of areas at risk were based on projections of current trends, assuming no changes in land use or rainfall. The impact of changing land use was examined to determine if the potential area in 2050 could be reduced.

Economic costs to agriculture

An annual ?operating profit? was assigned to each hectare of farmed land. The values differed between zones but were assumed to be an average across all hectares within zones. High and medium risk land was scaled off against an unaffected land operating profit. The annual operating profits chosen for each zone were derived from farm business client data bases prepared recently by a bank and private consultants.

The year 2000, 2020, and 2050 predictions for salinity impacts were then used to calculate total operating profit for each zone. Year 2000 without salinity was used as a base case for comparison to calculate the additional costs of salinity in subsequent years in terms of ?lost? operating profit.

The opportunity cost of lost operating profit in year 2000 due to watertables/salinity is estimated as $80M. That is, if all the currently affected land was still able to produce normal income, farmers would have an extra $80M operating profit available to spend elsewhere in year 2000.

Up to year 2020, rising watertables/salinity will cause progressive losses in operating profit each year. The sum (present value, 7%) of these extra losses (i.e. over and above the current impact) are estimated at around $19M.

Up to year 2050, rising watertables/salinity will cause progressive losses in operating profit each year. The sum (present value, 7%) of these extra losses (i.e. over and above the current impact) are estimated at around $120M. This is significantly more than the $19M estimated after 20 years and reflects greatly increased area of land achieving the medium and high risk categories in the subsequent 30 years.

The above presents estimates of the major costs of rising watertables/salinity for rural towns, roads, railways, vegetation and agricultural land. Other likely cost items not captured in the analyses would include: water resources, lakes, wetlands, telecommunications (minor) and pipelines (minor).

All other components are at least partially captured. The vegetation component includes forests (native or commercial) and parts of wetlands and streams - although many will justifiably argue they might have higher/lower/different asset values than a protection cost.

What biological resources occur in areas of high dryland salinity risk?

Biodiversity

Impacts on Perennial Vegetation

Perennial vegetation was mapped initially fromsatellite dat interpretatin and later up dated by interpretati on of digital orthophotos. These data were analysed with the risk data to determine how much vegetation was at risk in each landscape system and zone. Vegetation in systems with insufficient data was not accounted for. Total figures can be found in Appendix 11 of the Western Australia Dryland Salinity Assessment 2000 report and show that in 2000, nearly 600,000 hectares of perennial vegetation was potentially at risk and that by 2050 this had increased to over 1.8 million hectares. There were no compiled data to indicate the likely impact on particular species. Perennial vegetation includes both remnant and planted vegetation.

Impact on Flora and Fauna

CALM have been undertaking a biological survey of the agricultural areas of WA as part of the Salinity Action Plan (1996), now revised as the State Salinity Strategy (2000). The agricultural zones cover all, or significant parts of, six of the eight biogeographic zones recognised in temperate south western Australia, (CALM 1999). The current outcomes summarised from the CALM report for flora include:

  1. An estimated vascular plant flora of some 4000 species of which over 60% are endemic to the agricultural area.
  2. 850 of these species are found only in fresh or naturally saline lowlands directly threatened by rising groundwater and salinity.
  3. Of the 4000 species, 1500 occur low in the landscape, in riverine valleys, freshwater or primary saline lands. Of these taxa, 450 are endemic to the agricultural zone and are in danger of extinction as a consequence of rising saline groundwaters.
  4. Areas affected by secondary salinisation show major declines in vascular plant biodiversity.

Outcomes for fauna studies have found a significant decline in the biodiversity of terrestrial animals. Quadrants located in areas affected by secondary salinity averaged 30% fewer species than non-salinised quadrants (see table below).

Table: Average ratio of arachnid and vertebrate species for quadrants in non-saline, primary and secondary saline areas.
Quadrant Status Average ratio of species per quadrant
Non saline 2.3 (1.3) 2.8 (0.8)
Primary salinity 1.0 (0.8) 1.0 (0.2)
Secondary salinity 1.3 (0.7) 1.9 (1.0)

(standard deviation in brackets) Source: CALM 1999

A 50% reduction in the number of waterbirds using wheatbelt wetlands due to the saline induced death of shrubs and trees was reported. Also species richness declines with salinity.

Threatened Ecosystems (TEC), as defined by English and Blyth (1999), are also being examined by CALM. The work to date has identified 14 occurrences of TECs in the south western region (see table below). Not all systems where TECs have been identified, have sufficient data to determine risk. For some systems, (e.g. 257Bv) shallow groundwater may be an imminent threat to these communities. Factors other than shallow groundwater and salinity may be impacting some TECs.

Threatened Ecosystems and risk of shallow watertables.
Soil-Landscape System* TEC Number Occurrences 2000 2020 2050
221Ea VU 1 - - -
221Ga CR 1 L L L
224Ir PD 1 M M H
241St CR 1 L L L
257Bv EN 1 H H H
257Wg LR 1 - - -
258Kb CR 1 H H H
259Co CR 1 M M M
259Sh CR 1 M M H
271Ko VU 1 - - -
271Mw VU 2 - - -
271Ng PD 1 H H H

* = Define by the Natural Resource Assessment Group Agriculture WA.
CR = Critically endangered, EN = Endangered, VU = Vulnerable,
PD = Presumed destroyed, LR = Low risk.
L = Low risk, M = Moderate risk, H = High risk
Source: CALM Hamilton-Brown pers. comm.

Natural Diversity Recovery Catchments

Natural Diversity Recovery Catchments are catchments identified by the State Salinity Strategy (2000) as catchments with biological significance. Three catchments have been identified to date: Lake Muir, Lake Toolibin and Lake Warden. Regional risk analysis for these catchments has shown that Lake Muir has the worst prognosis with approximately 58% of the catchment area currently impacted by shallow watertables, increasing to 83% by 2050 (table below).

Table: Areas of Recovery Catchments at high risk of shallow watertables.
Catchment Name Area of Catchment (ha) Area (ha) at Risk in 2000 Area (ha) at Risk in 2020 Area (ha) at Risk in 2050
Lake Muir 59,700 34,458 34,458 49,824
Lake Toolibin 48,661 9,058 9,058 9,058
Lake Warden 191,442 0 0 95,106
Percentage at Risk Percentage at Risk Percentage at Risk
Lake Muir 58 58 83
Lake Toolibin 19 19 19
Lake Warden 0 0 50

Impact on Wetlands

Environment Australia supplied a list of Important wetlands (as defined by the Australian Nature Conservation Agency, 1996). Wetlands were identified that could be subject to hydrologic change due to the development of shallow watertables. Of the 54 wetlands located within the agricultural areas of Western Australia, 47 were fully or partially located in systems that could be allocated a risk category. The final analysis used 35 wetlands with data that covered more than 75% of the wetlands total area.

The wetlands at high risk vary in size from less than one hectare to thousands of hectares.

Both Lake Muir and Toolbin Lake were allocated a high risk. The catchments surrounding Lake Muir and Toolibin Lake have been nominated as Recovery Catchments in WA (Salinity Action Plan 1996). Whilst many wetlands currently have or will have a high watertable, the dynamics of the wetlands are likely to be impacted by increasing inundation and salinity. CALM (1999) documented that the death of many trees and shrubs in wheatbelt wetlands due to salinity has caused a 50% decrease in the number of water bird species using those wetlands. The data show that there are many important wetlands at high risk which are currently not receiving a similar level of investigation and management as the Recovery Catchments.

Economic costs to vegetation

A ?protection? cost based on the capital and operating costs of a pumping Strategy was allocated as a proxy value to a hectare of vegetation. While the capital costs were held constant, differential operating costs were applied for each risk profile. Conversion to an annuity provided an estimate of the annual cost of protection and allowed a series of cost flows to be constructed. The present values of these cost flows then provided a basis for estimating the $ impact of increasingly affected areas of vegetation.

All vegetation in the same risk category was therefore valued equally - ?medium risk? value was $126/ha/yr and ?high risk? value was $209/ha/yr. There was no consideration of differential biodiversity, habitat, or rare/endangered species values. The same pumping Strategy proxy value was used throughout - that is, regardless of landform, geological structure, soil types, etc.

The following results are based on 10% of the affected areas being protected.

The annual cost of protection of 10% of the year 2000 affected areas is estimated as $63M.

If 10% of the increased areas affected in year 2020 are protected, the annual cost is estimated as $64M.

If 10% of the increased areas affected in year 2050 are protected, the annual cost is estimated as $78M. The total sum (present value, 7%) of the extra costs of protection over the 50 years is estimated as $15M.

What water resources occur in areas of high dryland salinity risk?

Water Resources

The risk to water resources was considered in terms of the length of the major stream (at 1:250,000) in each system. Risk was attributed and total lengths in each risk category summed for each zone. The data showed the biggest increases in stream length likely to be affected by shallow watertables occurred for Zones 212, 241, 242, 252, 253, 257, 242 and 242. In all of these zones the stream length at high risk doubled between 2000 and 2050. (Full figures can be found in Appendix 9 of the Western Australia Dryland Salinity Assessment 2000 report.)

The Water and Rivers Commission (WRC 1999) provided water quality data for monitored streams. Monitoring sites were mapped and risk values attributed. Monitoring sites which had non potable water resources (water quality >500 mS/m), were not considered. An assessment was made on those sites with potable water (<500 mS/m) as increasing salinity could degrade potential water resources.

Whilst the majority of wheatbelt streams/rivers do not contain potable water, there are fresh waterways in high rainfall and coastal regions.

A total of 83 stream monitoring sites were identified of which 44 had potable water. Of these, 16 were located in systems with insufficient data to allocate risk. For the remaining 28, only four were in a high risk category in 2000. In 2050, eight were located within high risk systems (Table below). The most significant of these waterways in the high risk category are the Denmark River (AWRC Basin 603), Preston River (AWRC Basin 611), the Harvey River (AWRC Basin 613) and the Ellen Brook (AWC Basin 616). These basins were all larger than 500km2.

The data shows that shallow watertables are not always associated with salinity. The Preston River and Ellen Brook are examples where the allocated risk is high, i.e. these areas have shallow watertables in year 2000, but have high quality surface water.

Table: Risk of shallow watertables in systems occupied by surface water monitoring stations.
River Basin Station Name EC mS/m Risk 2000 Risk 2020 Risk 2050
603 603136

Denmark River

Mt Lindesay

 130 M M H
607 607220

Warren River

Barker Rd Crossing

 170 M M M
608 608151

Donnelly River

Strickland

 41 M M M
611 611004

Preston River

Preston Bridge

 58 H H H
611 611111

Thomson Brook

Woodperry Homestead

 56 M M M
612 612014

Bingham River

Palmer

 55 M M M
612 612022

Brunswick River

Sandalwood

 29 M M M
612 612006

Collie River

Mt Lennard

 185 M M M
613 613052

Harvey River

Clifton Park

 53 M M H
616 616189

Ellen Brook

Railway Parade

 105 H H H

Source: WRC (1999).

Water Resource Recovery Catchments are catchments specified for additional research and management to address salinity impacting on a water resource. The Collie, Helena, Kent and Warren catchments have been nominated. Currently the Warren is the only catchment with any areas falling into the high risk category. In the Warren this is a relatively small area (7%). All resource catchments, except the Collie, have potential to have extensive areas at risk of shallow watertables by 2050 (see table below). Full figures can be found in Appendix 10 in the Western Australia Dryland Salinity Assessment 2000 report.

The impact of shallow watertables and salinity on biodiversity is not easily illustrated. However, a number of data sets were examined to give an indication of the likely impacts on flora and fauna throughout the state. The data sets used include areas of perennial vegetation, lists of wetlands (Environment Australia 1999), and threatened ecosystems and biodiversity work currently being undertaken by the Department of Conservation and Land Management (CALM).

Table: Areas of Water Resource Recovery Catchments at high risk of shallow watertables.
Catchment Name Area (ha) of Catchment Area (ha) at Risk in 2000 Area (ha) at Risk in 2020 Area (ha) at Risk in 2050 % Coverage of catchment
Collie 278,550 0 3 431 100
Helena 147,780 0 0 122,535 83
Kent 240,517 0 0 144,025 61
Warren 413,293 29110 29110 108,213 72
Catchment Name Percentage at Risk 2000 Percentage at Risk 2020 Percentage at Risk 2050 % Coverage of catchment
Collie 0 0 0 100
Helena 0 0 83 83
Kent 0 0 60 61
Warren 7 7 26 72

What infrastructure occurs in areas of high dryland salinity risk?

Infrastructure was examined in terms of roads (highways, primary, secondary and minor roads), rail and major towns. The length (in kilometres) of each rail and road type were measured in each system, a risk allocated and then total lengths for each risk category determined. The totals for the High risk category are shown in the table below.

These figures were compared to other studies. The Main Roads Western Australia (McRobert et al. 1997) estimated that in 1997, 500km of main roads were affected by salinity and that this was likely to double over the next 20 years.

A second study, by the SS2020 implementation project, NLWRA was also reviewed. This project looked at the length of road networks affected by salinity in the 2.3 million hectares mapped by the Land Monitor project (an NHT and WA State Government project to map and monitor the extent of salinity through satellite imagery). Land Monitor found a total length of 3333 km of road at risk from salinity. If the area is scaled up to cover the entire south west, (an area of 26.6 million hectares) the figures are comparable and of the same magnitude.

Table: Total lengths (km) of road and rail at high risk of shallow watertables and potential salinity for 2000, 2020 and 2050.
Infrastructure 2000 2020 2050
Highways 721 841 1506
Primary Roads 680 746 1166
Secondary Roads 1196 1427 2326
Minor Roads 11552 13852 22936
Rail 1359 1488 2182

Road types are designated according to the body controlling maintenance. National highways and primary roads are serviced by Main Roads Western Australia, secondary and minor roads are maintained by local government.

The risk of loss of infrastructure for towns was allocated according to the system in which the town was located. Many rural towns are located near or within close proximity to rail lines. Most rail lines are located in low parts of the landscape. It is therefore likely that a high proportion of rural towns will be subject to salinity.

Only major towns were considered. The full list of towns and allocated risk can be found in Appendix 8 in the Western Australia Dryland Salinity Assessment 2000 Report. The table below lists towns that are currently being investigated for groundwater and salinity issues as part of the Rural Towns Program funded by the WA State Salinity Strategy. These towns were all nominated as high risk by this Audit.

Table: List of towns at which have, or will have a shallow watertable, and therefore high risk by 2050. Towns currently being investigated by the Rural Towns Program are noted.
Town Risk 2000 Risk 2020 Risk 2050 Rural Towns Program
Harvey H H H
Three Springs M M H
Gnowangerup M M H
Jerramungup H H H
Cranbrook H H H *
Boyup Brook H H H
Darkan M H H
Boddington M M H
Walpole H H H
Mt. Barker M M H
Northam M M H
Moora H H H *
Katanning M M H *
Brookton H H H *
Carnamah H H H *
Coorow H H H
Calingiri M H H
Wagin H H H *
Williams H H H
Beacon H H H *
Kellerberrin H H H *
Koorda H H H *
Merredin H H H *
Mukinbudin H H H *
Narembeen H H H *
Kondinin H H H
Perenjori H H H *

* denotes towns currently being investigated for groundwater and salinity under the Rural Towns Program, West Australian State Salinity Strategy. H = High Risk, M = Moderate Risk.

Economic costs to infrastructure

A draft consultant?s report (Dames and Moore - URS, June 2000) on the cost of salinity for the town of Merredin provided the base assumptions for both whole towns and individual components of infrastructure costs. The town was ?zoned? according to land use (e.g. residential, commercial, industrial, vacant land, recreation, civic buildings, roads, railways, etc) and individual repair/maintenance costs attached to each - according to when watertables reached to within 1.5 metres and 0.5 metres of the surface. Current and predicted watertable profiles were applied across the townsite such that repair and maintenance costs and asset write-offs were progressively brought into a cost flow budget at designated times.

Rural towns

The total cost of salinity for the town of Merredin (high risk category) was used to scale off all other towns in zones with a high or medium risk - based on respective town populations. Changes in town risk profiles between year 2000, year 2020, and year 2050 allowed calculation of the extra costs over the 20 and 50 year periods. Medium risk towns were assumed to attract 25% of the costs (compared to a high risk town) - after scaling for population.

The present value of costs for 60 affected towns under the year 2000 risk profile was estimated as near $68M. This is today?s value of the sum of all future expected costs of repair and maintenance and asset write-off over a 50 year period.

The present value of costs for each of the years 2000 and 2020 risk profiles were converted to an annuity as an estimate of what each town would need to spend each year to combat rising watertables/salinity. The difference in annual cost flows provides an estimate of the extra costs with progression of the salinity problem. Over the first 20 year period, the total additional cost is $0.8M. This is a relatively small increment reflecting the relatively small increase in risk for the first 20 year period. The implication is that timing of implementation of major control works is not critical. Dealing with the current problem is important but there is time to properly plan control strategies to address the more major increases in salinity risk in the following 30 years. (This is a sweeping generalisation and not necessarily the case for individual towns).

The present value of costs for year 2050 risk profile was also converted to an annuity for comparison with the year 2000 profile. There is greatly increased impact on rural towns in the 20 to 50 year period and the total additional cost (i.e. over and above the current impact) is estimated at $22.5M. This is a large increase over the $0.8M estimated for the first 20 years and reflects the greatly increased risk in the 30 years after year 2020. Control strategies designed for the year 2020 situation will be inadequate and need to have increased capacity to address the increasing watertable rises in the subsequent period - notwithstanding that earlier intervention might slow the process.

Roads

An approximate annual cost of repair per kilometre was calculated for each type of road and risk profile. Comparison of normal maintenance cost with the increased cost gave an estimate of the extra costs due to watertables/salinity. Length of road affected at the three time periods was used to frame a series of cost flows over the 20 and 50 year periods of analysis. Costs for road repair and maintenance within towns are already captured in the total rural towns costs above.

The annual cost in year 2000 of extra repairs due to watertables/salinity is estimated as $505M. That is, annual road expenditure is $505M more than it would be if no roads were affected. This appears to be an excessively large number and is probably caused by the difficulty in discerning between ?normal? costs and costs associated with watertable impacts - and the large difference in estimated repair costs between the two situations.

Rising watertables will increase the lengths of roads affected with a total extra repair cost of $91M. This is today?s value of the sum of all future extra costs of repair and maintenance over a 20 year period.

Today?s value of the sum of all future extra costs of repair and maintenance over a 50 year period is estimated at around $288M. The large increase over the $91M for the first 20 year period is not only because an additional 30 years of costs has been brought to account. There is also greatly increased lengths of roads affected in this latter period.

Railways

An approximate annual cost of repair per kilometre was calculated for railways and for each risk profile. Comparison of normal maintenance cost with the increased cost gave an estimate of the extra costs due to watertables/salinity. Lengths of railway affected at the three time periods was used to frame a series of cost flows over the 20 and 50 year periods of analysis.

The annual cost in year 2000 of extra repairs due to watertables/salinity is estimated as $11M. That is, annual railway expenditure is $11M more than it would be if railways were not affected. As for roads, this appears to be an excessively large number and is probably caused by the difficulty in discerning between ?normal? costs and costs associated with watertable impacts - and the large difference in estimated repair costs between the two situations.

Rising watertables will increase the lengths of railway affected with a total extra repair cost of near $2M. This is today?s value of the sum of all future extra costs of repair and maintenance over a 20 year period.

Today?s value of the sum of all future extra costs of repair and maintenance over a 50 year period is estimated at around $7M.

Both the 20 year and 50 year additional costs are relatively small numbers compared to other infrastructure components. While the assumed repair/maintenance cost differences need more detailed investigation, it also reflects the low density of railway lines in rural areas - compared with roads.

Further information

Western Australian Dryland Salinity Assessment 2000

Link to Map maker to make a map using this information.

Before you download

Most publications are downloadable as PDF files. Adobe Acrobat Reader  is required to view PDF files.

If you are unable to access a publication, please contact us to organise a suitable alternative format.

Key

   Links to an another web site
   Opens a pop-up window