Grains Convo

Study finds 7% of WA grainbelt at risk of wind erosion 

Summer 2021-22 hours of erosive winds, greater than 28km/hr

Project name

Identifying erosion hazard hot spots helps focus field activities 

Role of groundcover in soil conservation: WA grainbelt study 

Maintaining vegetative groundcover plays an important role in the critical function of soil and land conservation in the Western Australia (WA) grainbelt.

Insufficient ground cover is the dominant factor contributing to wind erosion.

In WA, the amount of existing ground cover is rated to assess wind erosion hazard according to national standards, with 30 percent cover being an extreme hazard, 50 per cent cover being erodible, and 50 per cent cover being regarded as the safe zone to prevent it.

A study from the Department of Primary Industries and Regional Development (DPIRD) used fractional ground cover products from Landsat satellite imagery to identify landscapes most at risk in the WA grainbelt.

This information was then used to focus on a field investigation with the West Midlands group to determine why the West Midlands landscape has the highest erosion hazard.

Imagery was processed with custom Python scripts to calculate the proportion of arable land in each ground cover class for every season over 11 years.

Individual seasons were compared with the preceding 10-year seasonal median to evaluate relative hazard.

The analysis for the summer of 2021-22 determined that 7 percent (1 million ha) of the grainbelt had inadequate ground cover to prevent erosion during a growing season that mostly experienced high rainfall.

The dry season in 2019 and 2020 decreased plant biomass production

The summer analysis determined that 16 per cent to 19 per cent of arable land had inadequate cover to prevent wind erosion during 2019 and 2020.

According to DPIRD Research Scientist, Justin Laycock, a small but worrying percentage had less than 30 per cent groundcover, which is 500,000 hectares of farmland that is highly susceptible to wind erosion.

"Add in the wind speed records, which show that most of the grainbelt receives 50 to 200 hours of erosive winds each summer, and we have a foundation for large duststorms," he said. 

"Average to above average rainfall in 2021 turned things around, with 7 per cent (1 million ha) of farmland below the 50 per cent threshold.

"However, most of this land was located in the West Midlands, and often on sandy soils that are less productive and prone to wind erosion (19% of farmland/ equates to 1 in 4 paddocks)."

Preventing carbon and other nutrient loss  

Understanding ways to reduce the occurrence of wind erosion is hoped to improve land condition, particularly in the southwest of the state.

Mr Laycock said researchers have encountered several issues throughout the study.

These include overgrazing, underperforming sandy soils, grasshopper management, and depleted seed banks due to false breaks and reduced plant numbers.

Favorable growing conditions in 2022, along with the promotion of better management, resulted in a further reduction in the percentage of farmland with inadequate cover, to 3.9 per cent of the entire Wheatbelt and 12 per cent of the West Midlands.

Mr Laycock mentioned that there is now the capacity to look at cover at the paddock scale.

“We are exploring ways to use this information to target our response and improve resource condition,” he said.

“We’re starting to integrate groundcover with soil and landscape information, as well as climatic information, to identify risks.

“Currently, we are considering ground cover as the primary hazard factor that land managers have control over."

Future research

In 2019, NASA released Landsat imagery, offering DPIRD researchers the opportunity to leverage this technology for testing data and potential outputs.

The catalog of images and seasonal trends from 1987 to the present is expected to expand, facilitating comparisons between current conditions and the past.

Some of this research includes, but is not limited to:

  • Annual field validation to ensure the accuracy and interpretability of fractional groundcover imagery across WA landscapes.
  • Targeted field surveys to investigate the soils and land management practices that may underlie consistently low groundcover.
  • Integration of groundcover data with soil and landscape information to enhance the interpretation and assessment of erosion hazards.
  • Identification of susceptible landscapes with inadequate groundcover, which could provide more comprehensive insights than groundcover alone. This data can be incorporated into a hazard matrix to calculate a multifactor erosion hazard, further developing into an erosion risk matrix to assess soil resource sustainability.


Justin Laycock  
DPIRD Research Scientist  
P: 08 9892 8407