|
Salinity risk across the Wimmera Plains | LC0099 |
Leah Thompson, Horsham
July, 2001 |  |
To view the Adobe Acrobat file, you will need the Adobe Acrobat reader. |  PDF 157 kb | |
This Landcare Note describes the use of risk modelling in illustrating the potential spread of salinity across the Wimmera plains and in highlighting where salinity management strategies should be concentrated. Salt affected areas are managed by establishing trees and/or salt tolerant pasture species, and by retiring the land from production.
Study area
The Wimmera Plains are defined as the area east and north of the Wimmera River, west of the Richardson River and south of Birchip. They cover approximately 507 000 hectares.
Figure 1. The Wimmera Plains
Land salinisation processes
The regional water table (Parilla sands aquifer) is greater than 20 to 25 metres below the land surface and is not causing salinity on the Wimmera plains.
Factors identified as contributing to salinity:
- Climatic contribution of 80-100 kg salt/hectare/year
- High rates of evaporation
- Poor drainage
- Localised water logging on clay soils after heavy rainfall causing the aquifer (Woorinen formation) to become saturated, creating a perched water table
- Lack of vegetation (trees and perennial grasses) to use rain water where it falls
- Red rises (red soils are mostly sodic, and contain considerable amounts of sodium)
|
Figure 2. Patchy crop growth is a symptom of localised salinity
Predicting areas at risk from salinity
Most areas currently affected by salinity are generally located at the base of red rises and low-lying areas. This is a direct result of poor soil drainage and ponding of surface water. Once the water evaporates, salt is left behind and accumulates in the soil profile. Flood plains of the Yarriambiack and Dunmunkle Creeks are not visibly salt affected, although there is potential for salinisation if current land management practices change (ie. tree clearance, impeding drainage, irrigation).
Mapping of visibly salt affected land
The first stage of determining the areas at risk from salinity was to map land already visibly salt affected. This was achieved by vehicle based mapping across the Wimmera plains. This information was then compared with independent environmental data sets to determine whether a relationship existed to the occurrence of salinity.
Environmental data sets used
- Radiometrics (Aerial geophysics survey that maps potassium, thorium and uranium isotopes present in the first few centimetres of the soil profile at resolution 1:50 000)
- Digital Elevation Model (DEM) 1:20 000 (Georeality Group, 2000)
- DEM derived data layers (ie slope, aspect, plan curvature, hill size) 1:20 000
|
Table 1. The approximate relationship of environmental layers to the mapped salinity
|
| Correlation of classified layer with salinity layer (%) |
|
| 60 |
|
| 50 |
|
| 60 |
|
| 60 |
|
| 60 |
|
| 90 |
Table 1 shows that slope had the greatest relationship with mapped salinity. This reflects the contribution of red rise recharge and run off to salinity on the Wimmera Plains. The other data layers were related to lesser extents with salinity but were collectively used to increase the accuracy of the predictive risk model.
Predictive computer modelling
The software package used was fuzzyTECH version 5.3 (INFORM GmbH/Inform Software Corporation, Germany). All environmental data layers were used in the computer model where they were compared with the salinity layer. The output identified areas at risk from salinisation.
Table 2. Computer modelling predicted percentage areas at risk
Percentage risk of salinity (%)
| Area of land at risk (ha) |
50 | 30 583 |
60 | 13 597 |
70 | 9 129 |
80 | 5 481 |
90 | 2 503 |
100 | 2 289 |
Area of mapped salinity (ha)
| 445 |
The total area identified as 100% at risk from salinity was five times greater than the area currently mapped. This result appears low with respect to the mapped area of salt affected land because,
- Only landscape attributes that are ‘fixed’ were included in the model. Hence, salinity risk mapping is a measure of landscape features rather than land use. Land use of areas identified at risk may have been modified by either drainage works or vegetation treatment.
- Variable attributes such as evaporation rates and rainfall were not included in the model. The extent of mapped salinity is affected by the climate of the year in which it was mapped, ie. dry seasons mask the problem because soils do not become waterlogged.
- The area mapped as visible salinity was much lower than the affected area due to large areas being cropped or out of view from the vehicle based surveys.
|
Recommendations
- Although salinity risk modelling has its limitations, it is still a useful tool for illustrating the potential spread of salinity and can highlight where management strategies should be concentrated.
- Areas of land identified with a salinity risk of greater than 50% can be managed by improving water use efficiency (better crop rotations, phase farming with alternate perennial pasture/cropping phases, planting red rises to trees and/or perennial pasture species.)
- Land identified with a salinity risk of 100% can be retired from production and revegetated with trees and native vegetation. Land too saline to support trees can be revegetated with salt tolerant species such as saltbush varieties, tall wheat grass (Thinopyrum ponticum) or Puccinellia (Puccinellia ciliata) and fenced off. Once ample groundcover is established, grazing should be managed to maintain adequate soil coverage.
- Areas adjacent or uphill from the saline discharge site can be revegetated with either trees or perennial pasture species, such as lucerne. Vegetation will intercept water and help reduce the severity and spread of salinity.
|
References
Georeality Group (2000). 1:80 000 soft photogrammetry digital elevation model for mineral sands exploration – Murray Basin, Digital maps only.
Hocking, M. (1999). What is causing salinity on the Wimmera Plains? Technical Report No. 53, Centre for Land Protection Research (NRE), Bendigo.
Hocking, M. (2001). Salinity risk on the Wimmera Plains Technical Report No. 72, Centre for Land Protection Research (NRE), Bendigo.
Further information
Contact the Horsham office of the Department of Natural Resources and Environment, Centre for Land Protection Research (NRE) Bendigo or the NRE Customer Service Centre on 136 186.
Acknowledgments
Compiled with the assistance of Mark Hocking, Centre for Land Protection Research (NRE) Bendigo.
This note replaces note number SC0046
The advice provided in this publication is intended as a source of information only. Always read the label before using any of the products mentioned. The State of Victoria and its employees do not guarantee that the publication is without flaw of any kind or is wholly appropriate for your particular purposes and therefore disclaims all liability for any error, loss or other consequence which may arise from you relying on any information in this publication.
|
