Description:
Observations are presented on the natural regeneration of Eucalyptus victrix L. Johnson and K. Hill (coolibah), the dominant tree species of grassy woodlands on the floodplain of the Fortescue River, north of Newman in the Pilbara district of Western Australia. The main objectives of this study were to examine: (i) survival of newly recruited E. victrix seedlings on flooded and burnt sites, and (ii) growth and survival of established plants in a gilgaied landscape. Populations of newly-recruited seedlings following flooding or fire were monitored opportunistically at 'Marillana' and 'Ethel Creek' stations, for up to two years from tagging. A population of saplings at 'Roy Hill' station was observed over eight years. Seedling densities of E. victrix in flooded sites were 0.32 and 1.03 plants/m2 in 1995 and 1997 respectively. Locally high densities reached from 1.2 to 17.7 plants/m2 on areas of 28 and 6 m2 respectively. At the burnt site at 'Ethel Creek', 1.30 plants/m2 were observed where 58 seedlings were recorded on an area of 445 m2 a year after a natural fire. Many of these persisted well into their second year, whereas seedlings recruited in the flooded sites generally lasted less than a year (250-376 days). In contrast, persistence of a cohort of 100 established saplings <2 m tall in a gilgaied landscape at 'Roy Hill' was high. During 8 years of observations, only one sapling was lost. Those found at the edges of gilgais had attained greater heights (3.91 ± 1.14 m) than saplings located inside (3.16 ± 0.95 m) or on top (3.12 ± 0.93 m) of the gilgais. This study demonstrates that the seedling stage in E. victrix is critical, in that once newly recruited individuals reach the sapling stage (0.6-1 m), mortality rates are much reduced. Unfortunately, it was not possible during the course of this study to locate a cohort of young seedlings that did survive and so any consideration of the conditions necessary for their survival is speculative.
Description:
The paper reports soil seedbank species composition, of Eucalyptus victrix grassy woodlands, of the upper Fortescue River in the Pilbara District, Western Australia. In this study, our objectives were to investigate germinable soil seedbanks and species composition in response to three simulated seasons, using emergence. Variation in seed density from three depths was tested. Four field sites were sampled. Thirty samples were collected in late spring, after seed rain and prior to summer rainfall. From each sample spot, three soil depths (surface, 1-5, and 6-10 cm) were segregated from beneath surface areas of 100 cm(2). Samples were later incubated in a glasshouse to simulate three different seasonal conditions (autumn, winter and spring). Germinating seedlings were recorded on emergence and grown until identified. Forty-one species germinated, comprising 11 grasses (7 annuals and 4 perennials), 25 annual herbs and 5 perennial herbs. Distribution patterns of germinable seed in both the important annual grass Eragrostis japonica and the perennial Eragrostis setifolia (a preferred cattle fodder species), suggest that seedbank accumulation differs among species and between sites. In part, this may be associated with the absence of grazing. Species with most total germinable seed were E. japonica (Poaceae; 603/m(2)), and the annual herbs Calotis multicaulis (Asteraceae; 346/m(2)), and Mimulus gracilis (Scrophulariaceae; 168/m(2)). Perennial grass seed was sparse. Spring simulation gave most germination (1059), followed by autumn (892) and winter (376) sets. Greatest species diversity was produced from the spring simulation (33 species), followed by autumn (26), and winter (22). Of the total germination, 92% came from 17 species that were represented in all three simulations. Of the 1227 grass seedlings counted, most were recruited from the surface soil (735), followed by the 5 (310) and 10 (182) cm depths. Marginally more grass seedlings germinated from the spring simulation (558) than the autumn set (523). Only 11.9% of grass germinants came from the winter simulation. All grass species recruited from the soil seedbanks had a C-4 photosynthetic pathway. Except for Cenchrus ciliaris all grass species are native to Australia. Of the four sites sampled, one fenced to exclude cattle five years earlier had significantly more germination than the three unfenced sites. Seedbank sampling produced several new records for plants in the areas sampled.