热带流域河流-含水层相互作用的区域尺度完全耦合模拟

Regional-scale, fully coupled modelling of

stream–aquifer interaction in a tropical catchment

Adrian D. Werner

a,

*

, Mark R. Gallagher

b

, Scott W. Weeks

a

a

Natural Resource Sciences, Department of Natural Resources and Mines, Queensland Government,

120 Meiers Road, Indooroopilly, Brisbane, Qld 4068, Australia

b

the code MODHMS, calibrated to near-stream observations of watertable behaviour and multi-

ple components of gauged stream flow. This model has been tested using other methods of esti-

mation, including stream depletion analysis and radon isotope tracer sampling. The coarseness

of spatial discretisation, which is required for practical reasons of computational efficiency,

limits the model’s capacity to simulate small-scale processes (e.g., near-stream groundwater

pumping, bank storage effects), and alternative approaches are required to complement the

model’s range of applicability.

mates from three different hydrograph separation techniques, which were found to be unable

to reflect the dynamics of Sandy Creek stream–aquifer interactions. The model was also used

created separately from river management models; or

highly simplified representations of either the surface water

or groundwater domains are adopted. These single-domain

models are often unsuitable for basin-wide applications to

low-flow hydrology issues (Halford and Mayer, 2000) such

Coupling of groundwater and surface water models for

regional-scale application is not straightforward and only a

few well-tested examples are available (e.g., Said et al.,

2005). Model developers need to overcome practical issues

of differences in scale, temporal and spatial discretisation,

and flow and head variability between surface and subsur-

face flow systems and their respective mathematical repre-

sentations. Codes also need to be adequately accurate,

efficient and robust to effectively manage the large compu-

tational burden associated with the dual domain simulation

BRANCH (Swain and Wexler, 1996), MODHMS (Panday and

Huyakorn, 2004; Hydrogeologic Inc., 2003) and DAFLOW

(Jobson and Harbaugh, 1999).

In this study, a MODHMS model of Sandy Creek and the

Pioneer Valley aquifers has been developed. The Pioneer

Valley is the subject of an intensive water resources plan-

ning and management investigation, which aims to establish

sustainable water resource operation practices that main-

tain riparian and hypothetic biodiversity and ensure that

water supply reliabilities are protected (NR&M, 2003).

the Sandy Creek area can impact on water supply reliabili-

ties in both systems. Sandy Creek is chosen to be the main

focus of this study, rather than other waterways in the Pio-

neer Valley (e.g., the Pioneer River, Bakers Creek) that are

generally regulated through supplemented supply and/or in-

stream weirs.

The primary objectives of the study are: (1) to outline an

efficient and effective methodology for regional-scale

stream–aquifer interaction model development and calibra-

tion; (2) to evaluate the limitations of the MODHMS model by

(Fig. 1) where the tropical climate is characterised by high

average rainfall (1551 mm/year) and a distinct summer

wet season (December–March) (Werner, 2004). Sugarcane

cultivation is the predominant land-use and irrigators use

both groundwater and surface water in accordance with

seasonal rainfall patterns. The vast majority of high-yielding

bores are situated in alluvial and fluvial sediments deposited

by the Pioneer River, Sandy Creek and Bakers Creek and

their predecessors, although fractured rock aquifers are

also accessed (Murphy et al., 2005).

ral Resources and Mines (NR&M) maintains a stream flow

gauging station in the lower reach of Sandy Creek and a

broad coverage of observation bores to monitor the sys-

tem. The locations of the closest 12 observation bores to

Figure 1 Locality map of the Pioneer Valley.

498 A.D. Werner et al.

Sandy Creek and the stream gauging station are shown in

Fig. 1.

The Sandy Creek catchment area is 420 km

2

die, 2004) and 326 km

and the catchment size. High-flow events in Sandy Creek are

predominantly contained within the flow channel, and the

system does not contribute significantly to flood plain re-

charge through riverbank overtopping.

Riparian habitats along the Sandy Creek estuary are of

ecological significance (Cook et al., 2004; Brodie, 2004).

Preserving the biodiversity of these habitats which are sub-

ject to considerable anthropogenic and climatic stresses is a

priority for natural resource managers (NR&M, 2003).

Other features of MODHMS are fully described in Hydrog-

Regional-scale, fully coupled modelling of stream–aquifer interaction in a tropical catchment 499