Wayne Houston

Replacement of littoral native vegetation with the ponded pasture grass Hymenachne amplexicaulis: effects on plant, macroinvertebrate and fish biodiversity of backwaters in the Fitzroy River, Central Queensland, Australia

Changes in plant and macroinvertebrate communities were found following replacement of extensive zones of floating-attached/submergent native vegetation within Fitzroy River backwaters by the major environmental weed Hymenachne amplexicaulis (Poaceae). Impacts of H. amplexicaulis on native littoral flora and fauna (macroinvertebrates and fish) were assessed by comparing three sites previously supporting native vegetation and now invaded by H. amplexicaulis with nearby stands of native backwater vegetation. Plant biomass of Hymenachne plant beds was 30-fold greater than native plant beds, whereas plant species diversity (richness) was significantly less. Macroinvertebrate communities of Hymenachne beds were significantly lower in abundance of insect orders Ephemeroptera, Hemiptera and Odonata, while Coleoptera were more abundant in Hymenachne beds. Non-metric multidimensional scaling ordination of macroinvertebrate family abundance and composition data showed that Hymenachne plant beds had a different assemblage to that in native plant beds. In common with other studies of weed invasions, an increased abundance of some vertebrate fauna was observed (in this case an introduced fish species Xiphophorus maculatus comprised 75% of fish captured in Hymenachne beds compared with 0% in native plant beds). Change in vegetation structure was implicated as an important factor influencing macroinvertebrate and fish faunal composition, and with potential to impact on waterbird habitat values of wetlands.

Severe hail damage to mangroves at Port Curtis, Australia

A hailstorm in October 1994 was found to have moderately or severely impacted on 5.3% of the mangrove forests in Port Curtis. All mangrove species showed evidence of hail damage, including the three most common species (Rhizophora stylosa, Ceriops tagal and Avicennia marina). Physical effects of hail damage included stripping of leaves from plants, holes punched through leaves, bruising to bark, divots removed from bark, branch and plant death. Species‐specific differences in vulnerability to the effects of hail were observed with C.tagal experiencing relatively higher mortality rates than the other two common mangrove species. A delayed pattern of mortality was observed in two species _ C.tagal and A.marina. Alterations to mangrove community structure included: (1) reductions in stem density, stem diameter and basal area, (2) reductions in canopy cover (based on a photographic index of foliage projective cover) and (3) changes in relative abundance of species in mangrove zones. Recovery was observed in some stands but others had not recovered to pre‐hail levels of canopy cover two years after the hailstorm. Recovery had occurred by regeneration of fresh leaves but no recruitment of young plants had been observed during the study. The forests in the impact area were dominated by either Ceriops tagal or Rhizophora stylosa with Avicennia marinaas a subdominant in places. C.tagal dominated forests within the impact area were relatively more severely affected (41.8% in the severe category) than R.stylosa dominated communities (only 17.4% in the severe category). This indicated that C.tagal dominated forests were more vulnerable to the effects of hail damage than R.stylosa dominated forests. In addition, hail‐impacted C.tagal dominated forests represented a relatively high percentage of the area of C.tagal dominated forests in Port Curtis (44.3%). This percentage was much higher than hail‐impacted R.stylosa dominated forests in Port Curtis (2.7%). These two factors – relatively severe impact on C.tagal communities and a relatively higher percentage affected within the Port Curtis area – illustrate that hailstorms, as a form of natural disturbance, are an important influence on the forest ecology of mangrove ecosystems in this region.

Termite (Isoptera) diversity of riparian forests, adjacent woodlands and cleared pastures in tropical eastern Australia

The Fitzroy River (Central Queensland) has the second largest ocean‐flowing catchment in Australia and retains relatively extensive remnant woodlands. However, clearing of the more fertile alluvial plains is proportionally much greater. Typically, forested habitat on the alluvial plain is restricted to narrow riparian corridors and small remnants on the adjacent terraces. Riparian forests are known to be important repositories of biodiversity in cleared and uncleared landscapes. However, little is known about the invertebrates of these habitats. Termites are ecosystem engineers promoting soil health and organic decomposition, and keystone taxa by virtue of mound and nest construction, and the hollowing of living trees. A comparison of the termite fauna of riparian forests and vegetation of the adjacent terraces (either remnant woodlands or cleared for grass‐dominated pastures) was made at five locations in the Fitzroy River basin. Terrace pastures had fewer species than terrace woodlands while termite encounters within the riparian forests were lowest. Termite assemblages also differed between the three vegetation types. Some species showed zonation patterns. These patterns may relate to ecosystem drivers such as frequency of flooding in relation to life history parameters (e.g. arboreal or ground nesting). Termite species distribution also appeared to be coupled to resource availability such as presence of live and dead wood and percentage cover of litter, grass and dung. To retain natural levels of termite diversity and functionality, it is recommended that land managers retain both riparian forests and adjacent woodlands in a matrix with cleared pastures.

Mitochondrial phylogeography of the critically endangered Capricorn yellow chat (Epthianura crocea macgregori)

Abstract. The critically endangered Capricorn yellow chat (CYC) is endemic to coastal central Queensland on marine plains where it occurs in three areas, numbering <300 birds. Recent industrial expansion in the region has increased the threat to the CYC. To assist management of the subspecies, a phylogeographical evaluation of the CYC using mitochondrial DNA was undertaken. We found no genetic diversity within, nor genetic divergence between, the two areas at the northern and southern extremes of their current distribution, and only slight morphological differences. These findings suggest that the two groups of CYC represent daughter populations of an ancestral population that was affected by a genetic bottleneck in the recent past. Implications for conservation of the subspecies could be increased vulnerability to environmental change. A preliminary evaluation of the divergence between the CYC and its nearest subspecies, the widespread inland yellow chat, indicate a time to the most recent common ancestor of 215 000 years or less. This timespan overlaps two periods of glacial aridity during which xeric habitats used by yellow chats for breeding, such as semiarid and arid swamps, may have expanded, allowing colonisation of the coastal marine plains. CYCs may represent a relictual population from a previously more xeric era that has subsequently become isolated as the region became wetter following glacial maxima.

Grazing and tree ‘clearing’ alter grass-associated invertebrate assemblages in an Australian tropical grassy woodland

To evaluate the response of invertebrates to ‘clearing’ and grazing pressure impacts, a previously grazed but uncleared grassy woodland in central Queensland was manipulated to provide four grazing pressures (destocked, low, moderate and high) and two tree treatments (with trees, i.e. untreated, and ‘cleared’, i.e. trees and saplings poisoned with herbicides), with two replicates of each, making 16 plots in total. Monitoring was carried out in 1998, approximately four years post-establishment of the treatments. Two types of samples were taken: pitfall for ground-active fauna and suction for grass-associated fauna. Overall, 23 orders of invertebrates were sampled by pitfalls and 22 by suction. Significant effects of grazing on invertebrate assemblages were detected by both methods, but no effects were detected from ‘clearing’. There was a gradation in the invertebrate assemblages from low to high grazing pressure, the invertebrate assemblages in the paddocks with the highest grazing differing most from those in the destocked and low-grazing-pressure paddocks. Notwithstanding the lack of effect of ‘clearing’ at the assemblage level, ground-active invertebrates and some grass-associated invertebrates increased in abundance following ‘clearing’, possibly reflecting an increase in the quality of the resource base. However, ground-active invertebrates and grass-associated invertebrates showed contrasting responses to grazing pressure, the former increasing, possibly reflecting changes in trapability due to the more open vegetation structure at higher grazing pressures. The abundance of grass-associated invertebrates declined by 50–80% with increased grazing – although with complex changes in assemblage structure. Despite those declines, the basic trophic pyramid remained, and, along with that, the potential for recovery of invertebrate assemblages and associated ecosystem services with reduction in grazing intensity. With 80% of Queensland grazed, the reduction in invertebrate abundance has implications for the viability of insectivores, particularly mobile fauna such as birds, at a landscape scale. It is recommended that the utility of using suction samples as a basis for assessing ecosystem functional health be investigated and that grazing pressure be reduced to increase invertebrate assemblages of rangeland pastures and to improve sustainability.