Multiple ecological processes underpin the eruptive dynamics of small mammals: House mice in a semi-arid agricultural environment

Abstract

Mouse plagues are a regular feature of grain-growing regions, particularly in southern and eastern Australia, yet it is not clear what role various ecological processes play in the eruptive dynamics generating these outbreaks.This research was designed to assess the impact of adding food, water, and cover in all combinations on breeding performance, abundance, and survival of mouse populations on a typical cereal growing farm in northwestern Victoria.Supplementary food, water, and cover were applied in a 2 × 2 × 2 factorial design to 240 m sections of internal fence lines between wheat or barley crops and stubble/pasture fields over an 11-month period to assess the impact on mouse populations.We confirmed that mice were eating the additional food and were accessing the water provided. We did not generate an outbreak of mice, but there were some significant effects from the experimental treatments. Additional food increased population size twofold and improved apparent survival. Both water and cover improved breeding performance. Food and cover increased apparent survival.Our findings confirm that access to food, water, and cover are necessary for outbreaks, but are not sufficient. There remain additional factors that are important in generating mouse plagues, particularly in a climatically variable agricultural environment.

Keywords: abundance; breeding; food addition; mouse plague; survival.

Figure 1

Feral house mice (Mus musculus domesticus) can reach very high densities during outbreaks and cause significant damage to grain crops in Australia

Figure 2

Layout of study site between Walpeup and Toritta, northwestern Victoria. There were 16 experimental sites established, each 240 m in length. There were two replicates of each treatment (untreated control, food, water, cover, food + water, food + cover, water + cover, and food + water+cover), which were assigned randomly. Sites were established along internal fence lines between a wheat crop (shaded) and pasture for sheep grazing (nonshaded)

Figure 3

Schematic representation of a 240 m experimental site showing the approximate location of the three trap lines, the food and water containers and the wire netting (three sections of 40 m) along a fence line between a wheat crop and a pasture crop used for sheep grazing. Three additional water stations (20 L drums) were provided (not shown). For sites without supplementary food or water, empty food and water containers were provided. On sites without supplementary cover, no wire netting (cover) was provided

Figure 4

Rainfall records from the Mallee Research Station spanning the experiment. The number above the column shows the percentage of rainfall records for the month from the last 91 years that fall below the value observed during the study

Figure 5

Mouse population changes throughout the experiment on the 16 plots (both replicates shown for each treatment). Population estimate based on the mark‐recapture Jackknife estimator. Error bars are 95% confidence intervals

Figure 6

Model‐averaged estimates (±SE) of population size in response to treatments in autumn (April) 2005, based on the top 5 models (combined Akaike weight 0.85) shown in Table 2, that is, the 5th ranked null model is included in the estimates

Figure 7

Model‐averaged estimates (±SE) of population size in response to treatments in midwinter (June) 2005, based on the top four models (combined Akaike weight 0.87) shown in Table 2. Treatment combinations are supplementary food and water (FW), food and added cover (FC): water and cover (WC), and food, water, and cover (FWC)

Figure 8

Model‐averaged estimates of apparent survival from the top 10 models in Table 3. The water treatment made no biologically significant difference to apparent survival rates, so it has been removed from the figure to aid clarity. For example, the model‐averaged predictions for the survival rates under the food and cover and water treatment were essentially the same as those for the food and cover treatment