Exotic pine plantations are a major landuse within the coastal lowlands of southeast Queensland, extending from close to the shoreline to the hinterland ranges. These plantations are within a sub-tropical climatic zone, and in most years, the summers are appreciably wetter than the winters. This terrain, in general, has been highly weathered and the soils are poor in nutrients. Environmental factors such as the climate, topography and weathering profile (including soil) are found to be important controls on vadose zone hydrology, which, in turn, has a great impact on tree growth and consequently on the design of management practices. This research project takes a holistic approach to investigate the influence of these environmental factors at different scales, and is designed to fulfil the following objectives:
(1) To build a spatial model of forest productivity for the entire Tuan Toolara State Forest (TTSF), southeast Queensland, by analysing the spatial patterns of many environmental variables that may have controls on soil water distribution.
(2) To determine how some of these environmental factors are responsible for the development of water-logging and soil salinisation by examining in detail an area of low site index that is severely affected by these two processes.
(3) To develop a model to assess the risks of water-logging spatially and temporally.
A multiple regression model was constructed to predict the forest productivity (measured by the value of site index, the average dominant tree height at 25 years of age). The independent variables were derived from a digital elevation model (elevation, slope, curvature, hillshade, flow accumulation and distance to streams), γ-ray spectrometry (potassium, thorium and uranium), and interpolated rainfall. The model explained up to 60% of the variance in the site indices and produced predictive maps of site index for two species: P. elliottii Engelm. and Queensland hybrid, a P. elliottii × P. caribaea Morelet hybrid. The model also identified the lowest site index area at the northern Tuan State Forest (NTSF), likely due to a greater risk of water-logging and salinisation.
The NTSF area is of low relief and, therefore, the focus has been on the vertical controls of deep weathering profile. The methodology included setting up a network of groundwater bores screened at different depths within the weathering profile, characterising the profile (mineralogy, EC, and pH) and the groundwaters within it (water levels, physico-chemical parameters, major and minor ions). It is found that water-logging is caused by perched groundwater formed on top of the ferricrete or mottled saprolite after prolonged rainfall. Localised salinisation is related to the discharge of brackish groundwater occurring within the mottled saprolite. The deep aquifer within the coarse saprolite is fresh and not responsible for salinisation, a situation that differs from many other settings in Australia.
The ability of using the Soil and Water Assessment Tool (SWAT) computer model to simulate soil water balance and to assess the risks of water-logging was tested in a selected catchment in the TTSF. The model successfully simulated stream flow at 2 weirs for a period of 6 years; the achieved R2 were 0.752 and 0.858, respectively. Long-term simulation for a 30-year period showed that there are pronounced seasonal patterns in rainfall and evapotranspiration as well as in soil water. For mature plantation with slopes of 3-15%, the mean annual duration of water-logging ranged from 161 days in the humus podzols, to 110 days in the gleyed podzolic, and to 90 days in the yellow podzolics.
The outcomes of this research suggest that forest management can be strongly supported by understanding the impacts of these environmental factors (e.g. climate, topography and weathering profile) on vadose zone hydrological processes; the selection of optimum approach will depend on the research objective or purpose. The models and analytical tools that were developed or tested here have the potential to be successfully applied elsewhere if the input data are available.
| Identifer | oai:union.ndltd.org:ADTP/265783 |
| Date | January 2008 |
| Creators | Wang, Qing |
| Publisher | Queensland University of Technology |
| Source Sets | Australiasian Digital Theses Program |
| Detected Language | English |
Page generated in 0.0017 seconds