Monitoring mice in Australia

Author: Steve Henry (CSIRO Health & Biosecurity, Canberra) and Peter Brown (CSIRO Agriculture & Food, Canberra) | Date: 13 Feb 2018

Take home messages

  • Mouse numbers are currently moderate across most regions of southern New South Wales (NSW), South Australia (SA) and northwest Victoria (VIC). There is potential for economic damage at sowing in 2018.
  • Current efforts to monitor mice are not sufficient to detect variations in mouse abundance between and within cropping regions. Growers need to stay informed about potential increases in mouse numbers from the mouse monitoring updates that the project publishes at the end of each monitoring session, known as mouse alert.
  • Growers should conduct their own monitoring to ensure they know what is happening in their own paddocks in the lead up to sowing each autumn. Growers should follow the recommendations outlined in the GRDC GrowNotes™ and the Tips and Tactics: Better Mouse Management page.
  • Broad-scale application of zinc phosphide bait is the only rodenticide available to growers to control mice in their paddocks. Timely application of mouse bait at the prescribed rate is paramount for reducing the impact that mice have on crops at sowing. Strategic use of bait is more effective than frequent use of bait.

Background

‘Surveillance and forecasts for mouse outbreaks in Australian cropping systems’ is a GRDC investment to monitor and model mouse populations across the grainbelt of Australia. The project started in October 2012 as a collaboration between Landcare Research (New Zealand), CSIRO and the Invasive Animals Cooperative Research Centre.

The aim of the project is to monitor mouse populations across the grain growing regions of Australia and develop predictive models to forecast mouse outbreaks. A key element of the project is to publicise the results of the monitoring and predictions to growers and industry through GRDC and other communication networks to enhance awareness of increases in mouse activity.

Mouse populations are monitored in typical grain farming systems in Western Australia (WA), South Australia (SA), Victoria (VIC), New South Wales (NSW) and Queensland (QLD) at three key times each year, coinciding with important crop stages (e.g. at sowing of winter crops) and critical times in the build-up of mouse populations (e.g. commencement of breeding in spring). The monitoring is used to collect information about the population size, breeding status and overall activity of mice. This information is used in predictive models to determine the probability of changes in mouse abundance. These models were developed at long-term monitoring sites in the northern Adelaide plains in SA, the northwest Mallee in VIC and the Central Darling Downs in southern QLD.

Mouse monitoring

The monitoring of mouse populations occurs at three levels of intensity on 110 transects across 11 sites (Figure 1):

  1. Benchmark sites in the Adelaide Plains (SA), Northwest VIC, and the Darling Downs (QLD), where long-term trapping has been conducted for more than 20 years and where forecast models have been developed. Live trapping data is collected at three key times per year and the data is used in the models to predict the likelihood of outbreaks for those regions.
  2. Quantitative rapid-assessment sites in Geraldton and Ravensthorpe (WA), Horsham and Walpeup (VIC), Riverina, Central West and Moree (NSW), Mallala and Yorke Peninsula (SA) and the Darling Downs and Goondiwindi (QLD) where there are two types of monitoring — mouse chew cards set out overnight (10 chew cards at 10m spacing along 100m survey lines), and active burrow counts along 4m x 100m survey lines. Monitoring is conducted three times a year.
  3. Qualitative monitoring networks in all the areas with rapid-assessment sites where key growers and agronomists are contacted to collect information about mouse activity in the region, as well as any reports of the use of rodenticides.

Map of mouse monitoring site locations is Australia, all listed in the previous dot points

Figure 1. Location of monitoring sites across western, southern and eastern Australia.

Issues with monitoring

Current models are performing well and mice are being monitored at a large number of sites across the grain belt, but only a snapshot of what is happening with mouse populations is being gained.  There is considerable variability in mouse activity between regions, farms or between paddocks on individual farms that cannot be captured by the monitoring regime. In an effort to capture this variability, a mobile phone application, MouseAlert (www.mousealert.org.au) was developed with the aim of having growers and agronomists supplement the monitoring program with data about mouse abundance on their farms (Figure 2).

Unfortunately, use of the App has been low and the data collected has been insufficient to use in predictive models.  However, the App still provides growers with the opportunity to enter data and view observations of other growers about the level of mouse activity in their district.

Unfortunately, use of the App has been low and the data collected has been insufficient to use in predictive models.

Figure 2. Records of mouse observations in MouseAlert web site/phone app since 2014 (containing > 500 records).

Monitoring outcomes

Over the five years that the monitoring project has been running, mouse numbers have fluctuated at all of the monitoring sites — on one occasion in QLD, mouse numbers were significant and damage was recorded on the Darling Downs.

In the spring of 2016, based on the trapping data at Walpeup and Mallala, the models predicted a high likelihood of an outbreak in autumn 2017 (Figure 3). Through the summer, numbers of mice in southern NSW, central and western VIC and most cropping regions of SA continued to rise and as a result, growers had to undertake significant baiting programs to reduce damage from mice at sowing in 2017. Despite warnings about the potential for significant increases in mouse numbers, many growers were caught unprepared. This was probably the result of high stubble loads after an exceptional 2016 harvest masked the signs of mouse activity.In some cases where growers anticipated high numbers, baiting was ineffective due to high feed load in the system.

Mice continued to be a problem throughout the 2017 crop. Monitoring early in the spring showed little or no sign of activity associated with active burrow counts or crop damage, but a significant level of activity was recorded on the chew cards. Adjusted trap success in north western VIC was significantly higher than expected for the spring trapping, indicating that breeding had started early.

Later in the spring, significant amounts of damage were recorded in many of the developing crops. Anecdotal reports of damage to all types of crops continued to be reported right up to harvest and reports of higher than expected numbers of mice through the harvest were not uncommon. Severe weather events during the 2017 crop resulted in significant crop losses in some areas due to dropped grain or frost damage, resulting in a greater than normal supply of food for mice.

Graph of mouse populations, description of these trends in preceding paragraphs.

Figure 3. Current mouse population abundance at benchmark sites in VIC and SA compared to outbreaks in the past.

Mouse control issues

More data is needed to make accurate predictions about changes in mouse abundance across cropping regions. One way to achieve this would be to develop a remote monitoring system that could detect changes in mouse activity on a broad scale.

The current approach to bait application is to spread bait on a broad scale across entire paddocks. Our understanding of mouse ecology and behaviour is based on work undertaken in conventional cropping systems. Understanding mouse ecology in zero and no-till cropping systems could lead to more strategic application of bait, potentially reducing the quantity of bait spread or increasing the effectiveness of bait by targeting high activity zones in paddocks.

Future development of new rodenticides for mouse control is still some time away and the development of novel biocontrol techniques has potential, but is still in the very early stages of development. In the interim, there is a need to find ways to use the tools that are available to control mice more effectively.

Acknowledgements

The research undertaken as part of this project, IAC00002, is made possible by the significant contributions of growers through both trial cooperation and the support of the GRDC. The authors would like to thank them for their continued support.

Contact details

Steve Henry
CSIRO Health & Biosecurity
GPO Box 1700
Canberra ACT 2601
(02) 62464088
Steve.Henry@csiro.au

@mousealert

GRDC Project code: IAC00002