The ecology and management of lumbricid earthworms in the Midlands of Tasmania

Earthworms have the ability to impart substantial physical, chemical and biological effects on the soil through their burrowing and casting activities. Earthworm communities are distributed throughout Tasmania's agricultural soils, however in the low rainfall region of the Midlands of Tasmania ( < 600 mm p.a.), earthworm density and species diversity is low. This may be restricting potential gains in pasture production in the region. The purpose of this study was to examine the ecology and management of earthworms in the Midlands, and their effect on pasture production during 1991-1993. The earthworm Aporrectodea longa was also introduced at two sites in the Midlands to determine its effects on pasture production. This knowledge can then be utilised to increase earthworm activity and hence, pasture productivity in the Midlands. Earthworm population dynamics were studied at 14 sites in the Midlands during 1992-1993. Earthworm number and biomass was recorded monthly and found to be significantly correlated with soil moisture; maximum earthworm activity in the surface soil was evident during the wetter months of winter and early spring, followed by an aestivation in the surface and sub-soils during the drier summer months. The two most abundant earthworm species found in the Midlands were Aporrectodea caliginosa (maximum of 174.8/m2 or 55.06 g/m2) and A. trapezoides (86/m2 or 52.03 g/m2), with Octolasion cyaneum (22/m2 or 13.0 g/m2) and Lumbricus rubellus (31/m2 or 13.58 g/m2) combined constituting < 15 % of the total population. However, the behaviour of A. caliginosa to rainfall and soil texture was contrary to that of A. trapezoides in this study. A. caliginosa was particularly dependent upon rainfall in the Midlands: population density, cocoon production and adult development of A. caliginosa was reduced under low rainfall. The number and biomass of A. caliginosa also tended to be lower on the sandy soil in this study. In contrast, the density and biomass of A. trapezoides was unaffected by rainfall between 425-600 mm p.a.; cocoon production and adult development continued unabated at low rainfall. Density and biomass of A. trapezoides was similar on all soil types. The depth of earthworm aestivation was examined during the summers of 1992- 1994 from the same sites used to examine earthworm population dynamics. Aestivation behaviour of earthworms was similar in each year. Most individuals were in aestivation at a depth of 150-200 mm, regardless of species, soil moisture or texture. Smaller aestivating individuals were located nearer the soil surface, shown by an increase in mean mass of aestivating individuals with depth. There was a high mortality of up to 60 % for juvenile, and 63 % for adult earthworms associated with summer aestivation in 1993 in the Midlands. Cocoons did not survive during the summers of 1992 or 1994, but were recovered in 1993, possibly due to the influence of rainfall during late winter and early spring. The influence of ivermectin on earthworm growth and cocoon production was investigated. The growth and cocoon production of four pasture earthworm species provided with dung from sheep treated with ivermectin were not significantly impeded over a five week laboratory study. Pastures at Oatlands and Perth were chosen in the Midlands to determine the effects of lime (L), nitrogen (N), organic matter (0) and fertilisers (F), and the introduction of A. longa, on earthworm numbers and pasture growth. Treatment application and introduction of A. longa produced contrasting results at Oatlands and Perth. A. longa increased pasture production at Perth within seven months by up to 17 %, but had no effect at Oatlands. Application of F and L at Oatlands increased pasture growth, whilst 0 and L initially decreased pasture growth at both sites. Numbers of A. trapezoides and A. longa were increased at Oatlands by L and 0, whilst F increased numbers of A. trapezoides and L. rubellus. The increase in earthworm numbers in response to treatments appears to be an indirect response to greater amounts of high N food reserves in the soil. In contrast, the initially low population density of earthworms at Perth were further reduced by N and F. The differences in response at Perth and Oatlands to treatment application and A. longa introduction is discussed in terms of differences in climate, soil type and pasture composition between sites.