A project undertaken at the University of New South Wales, and supervised by Dr David Eldridge.
The reintroduction of locally extinct vertebrates into former or novel habitats is an increasingly popular goal of conservation. Many of these animals disturb the soil while foraging for food or creating habitat. This is an important process that creates patchiness and habitat for plants and other animals. One of the major challenges facing conservation agencies when they reintroduce these animals is how to determine appropriate stocking rates. In the absence of natural predators, animal numbers often increase rapidly. While some disturbance is good for the landscape, too much disturbance can lead to landscape instability and negative environmental outcomes.
Our study aimed to identify and understand how soil function changes under different intensities and frequencies of disturbance. To do this we established two discrete experiments. In the semiarid mallee near Mildura, we established an experiment with three levels of disturbance (highly disturbed by bilbies, moderate disturbance by echidnas, low disturbance by rabbits) with different frequencies of disturbance (nil, once, twice, thrice) that were grazing by cattle and sites that were long ungrazed, and tracked 15 soil functions over 4 years. In a second experiment in a mesic grassland near Sydney, we imposed a gradient in soil disturbance (0 – 60%) and herbivory (plant removal, 0-100%) to stimulate the individual and combined effects of soil disturbance and herbivory on 53 microbe and soil arthropod groups. Together these two experiments were designed to provide information on how to ensure the persistence of ecosystem functions while supporting the primary conservation goal of reintroducing soil disturbing vertebrates.
Ground-active arthropods were collected at the Penrith site using a suction device.
In the mesic grassland, we found that of 106 possible attribute x treatment (disturbance, herbivory) combinations, there were only seven significant effects in response to increasing levels of soil surface disturbance or herbivory. Modelling revealed that the greatest overall effects were due to plants (plant richness, plant biomass, litter depth), with generally positive effects of plants on surface-active arthropods, negative effects on the metazoan community, and a negative litter effect, but only on microbes. The results suggest therefore that there are unlikely to be substantial negative impacts of reintroducing soil-foraging animals on biodiversity or ecosystem functioning in the target dry woodlands, even under extremely high levels of disturbance and herbivory.
Collecting soil samples at the Penrith site
The semiarid mallee experiment produced mixed outcomes. We were unable to determine thresholds of disturbance intensity or frequency on the resilience of soil functions, with no evidence that resilience was lowest under treatments combining frequent and large disturbances, likely because these systems already exhibit naturally highly resilient. There were few effects of recovery, even up to 18 months following disturbance, but there were major effects of disturbance frequency (i.e., one, two or three disturbances). Soils were generally more resilience at ungrazed sites, but this depended on the attribute examined and the level of disturbance. For example, for small disturbances created by rabbits, any effects waned at sites beneath trees because of more rapid infilling of tree litter and more mobile soil resulting from cattle disturbance. More rapid infilling resulted in more decomposition and therefore a stronger recovery of function than in more stable systems. However, for sites that were in the open, even small disturbances maintained their low levels of function because of low levels of infilling.
Excessive disturbance by reintroducted locally-extinct animals on sandy soils in southwestern NSW.
There are two take-home messages from this research:
Reintroducing locally-extinct soil disturbing animals is unlikely to have negative effects in more subhumid systems
Recovery from excessive disturbance by reintroduced animals in more arid regions, where most reintroduction occur, is likely to be a protracted process.