Wang, Q., Cox, M. E., Hammond, A. P. and Preda, M.

School of Natural Resource Sciences, Queensland University of Technology, GPO Box 2434, Brisbane, Queensland 4001, Australia

Abstract:Plantation forests are a major landuse within the coastal plains of southeastern Queensland. In low-lying areas such as the north Tuan State Forest, one of the biggest issues for forest management is about wet-season water-logging. Here we report on a study focused on a perched water table formed at the top of a deep weathering profile.

Deep weathering profiles typically consist of horizons including soil, (ferricrete), mottled zone, pallid zone and bedrock. The soil mainly consists of quartz sand, which overlies ferricrete or clayey mottled zone at shallow depths. Hydraulic conductivity drops quickly from the soil to ferricrete or mottled zone. Because of the flat topography, water drainage is impeded and under high rainfall, perched groundwater forms above the mottled saprolite and water-logging occurs consequently.

During the last wet season, water-logging started after the highest rainfall in December and finished as the wet season ended. The highest water level was observed either in December or January depending on site drainage conditions. Water-logging risk map, which was derived from subtraction of interpolated highest-recorded water table from DEM, shows that the closer to waterways, the higher the risk is. This supports the preference of maintaining native vegetation in these buffered zones rather than economic plantations.

Keyword:  water-logging, weathering profile, perched groundwater, forest hydrology, north Tuan State Forest, Australia

Introduction

Water-logging can be a major problem with some forms of landuse, and in southeast Queensland, significant proportions of the state-owned plantation forests are affected by seasonal water-logging (Bubb, 2002). The presence of excess water in the root zone of plants can result in poor gas exchange and anaerobic conditions; the impact on tree growth depends on the duration of the saturation conditions and the proportion of the potential root zone affected (Moore and McFarlane, 1998). Water-logging also has an impact on forest management, such as harvesting regimes.

Water-logging occurs when an area receives more water than is being discharged from it, resulting in a rise in the water table that may reach the soil surface (Pearce and Bohl, 2004). For these conditions to occur, a sufficient supply of water and mechanisms by which water is retained within the root zone are important (Williamson, 1998). Three main types of water-logging summarised in McFarlane and Williamson (2002), are: 1) ephemeral perched water tables in texture contrast (duplex) soils; 2) areas with very low gradients usually with fine-textured soils and poor soil structure; and 3) water-logging caused by discharge from permanent groundwater systems where hydraulic head is at or above the groundwater surface.

The first type of water-logging associated with duplex soils is the most reported as shown by studies in Western and Southern Australia (McFarlane and Cox, 1992; Cox and McFarlane, 1995; Hatton et al., 2002; Silberstein et al., 2002). Duplex soils are characterised by a sharp reduction in hydraulic conductivity from the A to B horizon due to a rapid increase in texture (Isbell, 2002; Northcote, 1979), which often impedes the drainage of water at the horizon boundary. More broadly, in any type of soil or weathering profile, a layer that presents a permeability contrast over a short vertical distance can form a restrictive layer (Tolmie and Silburn, 2006). This condition may include a permeability contrast associated with shallow soil profile on rock or permanently cemented layers, or increased clay content at shallow depths within a weathering profile(Schoknecht, 2002).

The north Tuan State Forest is a section of the biggest pine plantation in Queensland. Early hydrogeological research in the area was mainly directed to assessing deep groundwater aquifers within the bedrock below a deep weathered sandstone profile (Laycock, 1969; Laycock, 1975; QDNR&M, 2005). The causes of water-logging and relation to the weathering profile are, however, not well understood. This paper, presents preliminary findings about the causes and risk of water-logging, and the relation to the physical properties of the profile.

Settings

The study area is located around 300 km north from the capital city, Brisbane (Figure 1), and includes all or part of the Melaleuca, Boonooroo and Boronia logging areas with an area of around 100 km². The region has a subtropical climate typical of southeast Queensland. The Maryborough weather station, around 5 km to the northwest, has recorded the annual rainfall of approximately 1155 mm. The wettest season is during summer from December to February, when summer cyclonic conditions are common. The winter is comparatively dry.

The elevation is largely less than 30 m above sea level, gently decreasing towards the marine coastline in the east. Most of the area has a slope less than 1% (Figure 2). High slopes commonly correspond to a rapid change to topographic high points, e.g. the southwest corner, or the poorly developed surface drainage system. Several small tidal creeks flow to the east to discharge at the coast of Great Sandy Strait. To the west, streams flow to the Mary River.

The geology of the region was described by Ellis (1968). Basement rocks are mudstone, shale, siltstone and sandstone of the Maryborough Formation. The upper 60 m of the Maryborough Formation is silicified and forms a north-northwest elongate outcrop across the central area. These Mesozoic age sediments were unconformably overlain by the Elliott Formation of Tertiary age. The exposed Elliott Formation was lateritised since the Miocene and this covers most of the land surface as a deep lateritic weathering profile (Figure 2).

Methods

Groundwater bores network

A network of 33 groundwater bores was established and sited to evenly cover the study area (Figure 2).Twenty-nine are shallow bores, at an average 2 m deep and slotted within sandy soils at the top. The other four went deeper, with the deepest one, C2d, finished at 13 m, and slotted within a sand-gravel aquifer.

Soil/regolith description

Soil/regolith samples from the drillholes were collected at 0.5 m intervals and any changes observed. For each collected sample, a generalised soil/regolith description

Figure 1. Location of north Tuan State Forest in SE Queensland.

was completed including the grain size, colour (matrix and mottles), and presence of iron nodules.

X-Ray Diffraction analysis

Selected samples from the drillhole C2d were analysed using X-Ray Diffraction (XRD) for mineralogy. A standard procedure for preparing and analysing randomly-orientated powder samples was followed(Jenkins and Snyder, 1996). Ten percent corundum was added for each sample to identify potential amorphous contents.

Water level measurement

The bores were developed using standard methods (Brassington, 1998). Over the wet season from December 2005 to February 2006, field measurements of water levels were taken once a month.

Results and Discussion

Cuttings from the drillhole C2d, as well as old geological logging of groundwater bores from Laycock (1969), depicted a typical deep weathering profile, which includes the following distinctive horizons: soil, (ferricrete), mottled zone, pallid zone and fresh bedrock (Figure 3).

This common form of zoned deep weathering profile was first formalised by Walther (1915). They are widespread in the inter-tropical belt between latitudes 35ºN and 35ºS, such as in Australia, Africa, India and South Africa (Anand and Paine, 2002). Different authors have preferred terminology for describing the profiles (Britt et al., 2001), but they are somewhat equivalent such as replacing the pallid zone with saprolite.

The soil at the top is variable, but is commonly yellowish coloured fine to medium sand. The depth is largely less than 2 m, and shallower when it is close to the bedrock outcrop. Near the coast, the soil is usually covered by a layer of aeolian sands. Pisolithic iron nodules may be present in the soil.

The mottled zone and pallid zone are both clay-rich. The difference is that the pallid zone is depleted in iron whilst the mottled zone is enriched in iron. The mottled zone contains patches of iron oxides in the form of vermiform mottles or pisolithic segregations in a matrix of pallid zone saprolite; it may merge with the ferricrete (if present).

At site C2d, the mineralogy of regolith samples representing different weathering zones further characterised this in-situ weathering profile (Figure 4). The soil consists mainly of quartz sand. Within the mottled zone, around one third of the composition is kaolinite; iron oxides are mainly goethite and hematite. Separated iron nodules have similar mineral contents as the mottled zone, but less kaolinite and more iron oxides. Contrary to the mottled zone, no goethite and hematite are present in the pallid zone, more than 80% of which are kaolinite. Weathering is less intense near the base of the profile (gravely sandy clay), where smectite, rather than kaolinite, constitutes the major part of the clay minerals, and primary minerals such as feldspar are well preserved.

Based on experience with basement rock aquifers in Africa, Thomas (1994) summarised typical curves of hydraulic properties for deep weathering profiles (Figure 5). He pointed out that the relative permeability is high at the surface in residual materials from which many fine-grained components have been winnowed or eluviated. It decreases rapidly in mottled and pallid zones where clay-sized materials have not been removed or have been added by illuviation. The fact that porosity is relatively high in these zones while permeability is low, suggests the clay minerals maintain the porosity but block fluids passing through the material. At the base of the weathering profile, however, the permeability increases again and falls off in the fresh basement.

Figure 2. DEM, slope and surface geology with layout of bores.

Figure 3. Schematic of typical weathering profile for the area.

Figure 4. Mineralogy of selected soil/regolith samples from site C2d.

Figure 5. Typical curves for hydraulic properties (after Thomas, 1994).           

The sharp reduction of hydraulic conductivity from the sandy soil to ferricrete or clayey mottled zone forms a restrictive layer which impedes water drainage and under high rainfall, perches groundwater above the layer. Rainfall records were collected from December 2005 to February 2006 (Figure 6). After the highest rainfall in mid-December, water-logging conditions started and ended as the rainfall decreased in February. Depending on the site, during the whole season, the highest water tables were either recorded in December or in January. For example, if a site is well drained, the perched water table can be reduced quickly. The water table at poorly-drained site, however, may continue for a longer time or even rise if there is additional rainfall.

The interpolated highest-recorded water tables smoothly followed the spatial trend of surface elevation (Figure 7A). The aim for using the highest water level instead of contemporary data was to better simulate the potential of overall saturation throughout the period. Subtraction of the highest water table from the DEM produces the risk map of water-logging (Figure 7B). Water-logging conditions are defined as a raised water table to within 50 cm of the ground surface. Where the water table has never reached within this depth, the site is considered not prone to water-logging. Based on the risk map, the highest risk occurs along the buffered zone of alluvial channels, as they are relative low points in topography and locations that accumulate water. In most of the places, native vegetation has been preserved.

Figure 6. Three times of water-level measurements and the rainfall records from the Bureau of Meteorology.

Figure 7. Maps showing (A) interpolated highest-recorded water table during the wet season and (B) depths to the highest recorded water table which reflects the water-logging risk and influence of the drainage system.

Conclusion

The deep weathering profile developed in the study area has vertically distinctive zones including soil, (ferricrete), mottled zone, pallid zone and bedrock. The soil at the top is shallow and mostly quartz sand, while the mottled zone consists of one third of kaolinte, which substantively reduced the permeability. Two major causes of the water-logging are flat topography and a restrictive layer at shallow depths where the hydraulic conductivity drops from

the sandy soils to ferricrete or clayey mottled saprolite zones.

Rainfall is the primary recharge source for the shallow perched groundwater. Water-logging occurred following the largest rainfall in December, but the highest water table recorded varied from December to January depending on the drainage conditions of the specific site. Water-logging ended with the wet season. The risk of water-logging is largely related to the topography; where it is close to the waterways, there is greater risk.

Acknowledgements

The drilling program was funded by Queensland University of Technology and the project was conducted within the properties of the Queensland Department of Primary Industries/Forestry.

References

[1] Anand, R.R. and Paine, M., 2002. Regolith geology of the Yilgarn Craton, Western Australia: Implications for exploration. Australian Journal of Earth Sciences, 49: 3-49.

[2] Brassington, R., 1998. Field Hydrogeology, 2nd Edition. John Wiley, Chichester, New York.

[3] Britt, A.F., Smith, R.E. and Gray, D.J., 2001. Element mobilities and the Australian regolith - a mineral exploration perspective. Marine Freshwater Research, 52: 25-39.

[4] Bubb, K.A., Croton, J. T., 2002. Effects on catchment water balance from the management of Pinus plantations on the coastal lowlands of south-east Queensland, Australia. Hydrological Processes, 16: 105-117.

[5] Cox, J.W. and McFarlane, D.J., 1995. The causes of water logging in shallow soils and their drainage in southwestern Australia. Journal of Hydrology, 167(175-194).

[6] Ellis, P.L., 1968. Geology of the Maryborough 1:250,000 Sheet area, Geological Survey of Queensland Report, 26.

[7] Hatton, T.J., Bartle, G. A., Silberstein, R. P., Salama, R.B., Hodgson, G., Ward, P. R., Lambert, P. and Williamson, D.R., 2002. Predicting and controlling water logging and groundwater flow in sloping duplex soils in western Australia. Agricultural Water Management, 53: 57-81.

[8] Isbell, R.F., 2002. The Australian Soil Classification, Revised Edition. CSIRO Publishing, Collingwood, Victoria.

[9] Jenkins, R. and Snyder, R.L., 1996. Introduction to X-ray Powder Diffractometry, Chapter 9 (Specimen preparation). Chemical Analysis, 138. Wiley, 231-259 pp.

[10] Laycock, J.W., 1969. Mary Valley Groundwater Investigations - Hydrogeological report on the area between Tiaro and Pialba, Department of Mines, Geological Survey of Queensland.

[11] Laycock, J.W., 1975. Mary Valley Groundwater Investigations - Drilling and testing sand-gravel aquifer - Boonooroo area, Department of Mines, Geological Survey of Queensland.

[12] McFarlane, D.J. and Cox, J.W., 1992. Management of excess water in duplex soils. Australian Journal of Experimental Agriculture, 32: 857-864.

[13] McFarlane, D.J. and Williamson, D.R., 2002. An overview of water logging and salinity in southwestern Australia as related to the 'Ucarro' experimental catchment. Agricultural Water Management, 53: 5-29.

[14] Moore, G.A. and McFarlane, D.J., 1998. Water logging. In: G.A. Moore (Editor), Soil Guide - A Handbook for Understanding and Managing Agricultural Soils. Agriculture Western Australia, South Perth, pp. 94-108.

[15] Northcote, K.H., 1979. A Factual Key for the Recognition of Australian Soils. Rellim Tech. Pubs, Glenside, South Australia.

[16] Pearce, M.W. and Bohl, J.P., 2004. Response to wet season rainfall of an alluvial aquifer system in the lower Herbert floodplain, Queensland, Australia. Australian Geographical studies, 42: 34-45.

[17] QDNR&M, 2005. Groundwater Database - Bore Card Report. Queensland Department of Natural Resources & Mines.

[18] Schoknecht, N., 2002. Soil groups of Western Australia: a simple guide to the main soils of Western Australia, Edition 3, Resource Management Technical Report 246, Natural Resources Assessment Group, Department of Agriculture, Western Australia, Perth.

[19] Silberstein, R.P., Bartle, G. A., Salama, R. B., Hatton, T.J., Reggiani, P., Hodgson, G., Williamson, D.R. and Lambert, P., 2002. Mechanisms and control of water logging and groundwater flow in the 'Ucarro' sub-catchment. Agricultural Water Management, 53: 227-257.

[20] Thomas, M.F., 1994. Geomorphology in the Tropics: a study of weathering and denudation in low latitudes. John Wiley, New York.

[21] Tolmie, P. and Silburn, M., 2006. Soil limitations to water entry - understanding restrictive soil layers, Fact sheets, The State of Queensland (Department of Natural Resources, Mines and Water).

[22] Walther, J., 1915. Laterit in Westaustraliean. Z. dt. Geol.Ges., 67B: 113-140.

[23] Williamson, D.R., 1998. Land degradation processes and water quality effects, water logging and salinity. In: J. Williams, R.A. Hook and H.L. Gascoigne (Editors), Farming Action - Catchment Reaction, The effect of Dry-land Farming on the National Environment. CSIRO Publishing, Collingwood, Victoria, pp. 162-190.