EFFECTS OF SUMMER IRRIGATION AND TRAMPLING IN DAIRY PASTURES ON SOIL PHYSICAL-PROPERTIES AND EARTHWORM NUMBER AND SPECIES COMPOSITION

In 1989 a replicated split-plot trial on a Krasnozem soil was establis hed at Elliott Research Station (ERS) in the north-west of Tasmania, a s well as 14 on-farm trials in newly irrigated pastures on 3 different soil types (Alluvial, Podzolic, Krasnozem) in the dairy districts of Scottsdale, Smithton, and Deloraine. There were 3 main treatments at E RS: irrigated before grazing, irrigated after grazing, and grazed and not irrigated. Part of each main plot was fenced to prevent trampling but still allowed grazing. Effects of summer irrigation and trampling by dairy cows were examined for pasture production, and soil chemical and structural properties. Summer irrigation at ERS and on-farm trials has led to a decline in soil structure indicated by slower ponded wat er infiltration rates on irrigated plots compared with the dryland plo ts. The decline in ponded water infiltration rates suggests a reductio n in macroporosity, especially in the soil surface. However, other ind icators for soil structural change in the top 100 mm - percentage wate r-stable aggregates (>2.5 mm) and bulk density - revealed no significa nt variation between the irrigated and dryland paddocks. There were, h owever, higher water infiltration rates and lower bulk densities in th e untrampled areas than the trampled ar eas at ERS. Pasture production at ERS was about 50% more with irrigation in each of the 2 years of t he study. Data collected at ERS in autumn and spring on the numbers of Aporrectodea caliginosa (Savigny) and Lumbricus rubellus (Hoffmeister ) earthworms showed that they respond quite differently to irrigation. After 2 irrigation seasons, A. caliginosa numbers in irrigated plots dropped by over 50%, whereas in the dryland plots densities of this ea rthworm have remained around 390 earthworms/m(2). In contrast the dens ities of L. rubellus at ERS rose under irrigation practices, especiall y in the autumn-winter sampling period. Therefore, with the advent of summer irrigation at ERS, there was a shift in earthworm composition f rom a fauna dominated by A. caliginosa to a fauna with an increasing p roportion of L. rubellus and a decreasing number of A. caliginosa. The typical dairy pasture in the on-farm trials recorded 2 main species, A. caliginosa (70%) and L. rubellus (30%). Total earthworm densities w ere highest in the north-west (Smithton) region of the State (293-351 earthworms/m(2)) regardless of soil type, and the lowest densities wer e recorded in the Alluvial soils of Deloraine (96 earthworms/m(2)). Th e north-west area also had the most diverse earthworm fauna, with 5 sp ecies recorded in one site: A. caliginosa, A. longa, Allolobophora chl orotica, L. rubellus, and O. cyaneum. Summer irrigation effects after 2 seasons on earthworm composition and abundance on dairy farms caused no significant change in A. caliginosa numbers, but there was a 45% i ncrease in the numbers of L. rubellus in irrigated treatments. L. rube llus was considerably more active over summer in irrigated paddocks (2 5 earthworms/m(2)) than in non-irrigated paddocks (7 earthworms/m(2)). In contrast the number of A. caliginosa recorded in dryland paddocks was not statistically different to the irrigated paddocks, but the A. caliginosa in dryland paddocks were mostly inactive 8-20 mm from the s oil surface.