Jane Kowald located the DAFWA holes drilled south of the Beaufort River paleochannel aquifer and supervised the drilling. A numerical model of the Beaufort River plantation and environs was created using the US Geological Survey's MODFLOW Groundwater Model (McDonald and Harbaugh, 1988). The purpose of the modeling was to predict the potential impacts of the plantation on the local groundwater system and to provide guidance for the placement of wells needed to monitor the impacts of the plantation on access to groundwater resources by neighboring landowners.
A scenario in which the maximum rate of groundwater withdrawal from the plantation was set at 300 mm/year and the maximum extraction depth was 6 m was considered to represent the most likely impact of the plantation on groundwater levels in the surface aquifer. Update the model based on the results of the proposed drilling, if required, and undertake a comparison of the model results and monitoring data. If necessary, use the calibrated model to assist in adaptive management of plantation and neighbor borehole water use.
Introduction and background
Methods
Study area
- Hydrogeology
- Groundwater pumping
- Soils
In February 1994, the Geological Survey of Western Australia drilled eleven sites in the Beaufort River area (Waterhouse et al. 1995) to specifically investigate the water resource potential of the paleochannel. These investigations discovered a similar paleochannel and the authors applied for support from the Geological Survey. Both areas were later identified as part of the same paleochannel system, which is believed to be the precursor of the modern Beaufort River (Waterhouse et al. 1995).
The mapped extent of the surficial sedimentary aquifer is shown in Figure 8 (Department of Water, 1989), the extent of the paleochannel identified by Waterhouse et al. 1995) roughly coincides with that mapping. Nine of the sites drilled by the Geological Survey in 1994 form two transects across the valley in which the plantation is located (Figure 3). It appears that groundwater in the surface aquifer may be fresher on the valley floor, but there is anecdotal evidence that this is a result of boreholes.
Conceptual groundwater model
Also shown are the plots occupied by the plantation (green shading), basal paleochannel, model domain (purple), Geological Survey bores, ridge lines (brown), and larger lakes (blue shading). The dominant soils on the valley floor are gray deep and shallow sandy duplexes and saline wet soils derived from the alluvial sediments, there are occasional aeolian lunettes. Gray deep-sandy duplexes occur on the valley sides and side slopes, they are often gravelly; sandy gravel and gray shallow loamy and sandy duplex soils are also common.
Deep sand gravel, duplex sand gravel, and shallow gravel derived from laterite or local colluvium are prominent on the upper slopes and crests. The Eocene alluvial sediments extend to the 270 m AHD, as mentioned above; Figure 8 shows both the previous mapping of their size and the 270 m contour, which is our current best estimate of their size within the model domain. Also shown are: the previously mapped extent of the paleochannel (pale red shading), the regionally mapped extent of the overlying aquifer (blue crosshatching) 270 m AHD contour within the model domain (green), current major river channels and lakes , and Albany Avenue.
Numerical groundwater model
- Model calibration
- Scenarios
The steady state model is considered to be indicative of fall conditions under current climate conditions in the Beaufort River area. The model domain includes a 10 km stretch of the Beaufort River adjacent to its eastern boundary; however, interaction between the surface aquifer and the river is not accounted for because anecdotal evidence is that the river rarely flows in summer and autumn. The actual evapotranspiration from the groundwater at any time is therefore a function of the actual depth to the water table at that time and the evaporation parameters used.
Two examples of how actual evapotranspiration from groundwater is calculated are shown in Figure 10; these examples are a subset of the evapotranspiration parameters used in the plantation scenarios (see Table 2). Reliable estimates of discharge rates were available for only three wells; SHEP01 and SHEP02 west of the plantation and KOW01 which supplies the slaughterhouse (shaded in Table 1). Seventeen (17) scenarios were conducted to assess the possible range of impacts of climate variability and change plus the plantation on the groundwater in the surface aquifer adjacent to the plantation.
Groundwater heads at the model boundaries were not changed although a reduction in rainfall would cause a reduction in groundwater levels over most of the aquifer. These scenarios may therefore underestimate the impact of the assumed rainfall reductions on groundwater levels in the area of interest. There are three main variables that influence the impact of the plantation on groundwater levels in the surface aquifer; the level of recharge reduction, the potential rate of evapotranspiration from the aquifer and the depth to which the trees are able to extract water from the aquifer.
The above scenarios therefore cover the likely range of these variables to provide an assessment of the level of impact of the plantation in the absence of more detailed data. In each of these scenarios it was assumed that the plantation trees have unlimited access to groundwater, up to the indicated depth and that there are no impeding layers preventing root growth or hydraulic connection with the rest of the surface aquifer; neither of these is likely to be true across the plantation. The semi-elliptical buffer on the southern boundary of the plantation has been raised above and beyond that shown in Figure 11 in the more recent, approved plantation design.
Simulated groundwater impacts immediately south of the plantation will therefore be slightly overestimated from those expected with a larger exclusion zone.
Results
- Calibration—Base case
- Reduced rainfall scenarios
- Plantation impact scenarios
- Zero Recharge under planted area
- Low groundwater use scenarios
- High groundwater use scenarios
- Combined reduced recharge -plantation groundwater use scenarios . 30
- Summary of predicted impacts
No similar explanation exists for the systematic underprediction of heads (−0.08 to −1.0 m) south of the plantation. Areas predicted to have groundwater within half a meter of the ground surface are shown in Figure 13. The results of the two reduced rainfall scenarios are shown in Figure 14 and Figure 15 in terms of changes in groundwater level in the surface aquifer compared to the base case .
In the 10 percent recharge reduction scenario (R10), the level of impact over most of the affected area is less than 0.5 m. The maximum head reduction in the surface aquifer is less than 1.5 m and the impact over most of the area is between 0.1 m and 1 m. The predicted impact of the low groundwater use scenarios is shown in Figure 17 to Figure 21.
Note that the predicted actual evapotranspiration is only a fraction of the assumed maximum rate (300 mm/year) because this rate is only applicable where groundwater is at the soil surface (Figure 10). This influence extends several hundred meters east and west of the plantation, but is most noticeable south of the planted plots. Pale blue - groundwater within 0.5 m of the ground surface, orange and red - groundwater rises above the ground.
The predicted effects of the high groundwater use scenarios are shown in Figure 23 to Figure 26. When pumping has stopped, the model predicts lower abstractions near the well than before. Light blue - groundwater within 0.5 m of the soil surface, orange and red - groundwater above the ground.
The range of predicted lateral impacts of the plantation on groundwater levels is at the upper end of observed values for south-west Western Australia. The predicted impact of the plantation on water levels is within the range of observations by George et al. In scenario P500_10, the water table northeast of the plantation dropped by a sufficient amount for the model to automatically stop pumping at bore KOW01, which supplies the abattoir on Leggoe Road.
Discussion and recommendations
Groundwater monitoring
In addition, as new plantations complicate the interpretation of the impact of the existing system, no new plantations should be established in areas predicted to experience a significant impact from the current plantation without reference to the Management Committee.
Further groundwater modelling
Proposal for Monitoring and Review Program
FPC to manage the establishment of the committee (STF Reference Committee and/or Council) and operational plan, potentially using the services of a consultant at the initial stage to ensure independence. Committee membership and roles and responsibilities will need to be decided up front. Activities to be undertaken (before #b-#g below) may then include, (i) details of benefit measures, (ii) access to land and water, (iii) monitoring responsibilities and related methodology, ( iv) membership - chairman of the commission, (v) communication, (vi) data exchange and.
Determination of current conditions in and near the edge of the plantation as they reflect the structure and function of aquifers in the area; the extent and structure of the Beaufort palaeochannel and relation to local hill aquifers. Report as base case (for use in submission) .. c) Analyze areas known to be 'at risk' given reduced recharge and/or water tables through site-specific assessment (drilling transects, pumping tests on existing wells, etc.). Establish an agreed program of well monitoring, which will likely include monitoring current withdrawal rates from pumped wells (including wells on Fysh and Sattler properties, those supplying the abattoir and other production wells).
Build a simple aquifer model and run scenarios that reflect the likely .. consequences of existing and/or proposed extraction and the effect of plantation. This also enables better targeting of monitoring and prediction of long-term effects, both benefits and potential yield implications. e) Design and establish a monitoring system to assess changes over time for each observed scenario. You can do this later. f) Define measurable "trigger conditions" and agreed upon "responses" for each.
Previous advice from the Ministry of Water in relation to access to water in unlicensed areas is that protection of water availability to existing users (prior use) is the main criterion used in the allocation of water resources. For each risk and area, it is proposed that the trigger conditions will be based on the current availability of water for users. In determining the cause of any change in supply, information such as drawdown in monitoring wells relative to the previous period and relative to wells outside the area affected by the plantation will be used to attribute any effects.
This will require knowledge of current water consumption (who, when, how). g) Assess the feasibility of each response and develop a management plan.
Bore details
