Manly Hydraulics Laboratory

NSW Public Works

Manly Hydraulics Laboratory

Yarrahapinni Wetland Water Quality Monitoring

Peter Haskins, BSc, Grad Dip Arts, Department of Land and Water Conservation, Floodplain Management, NSW, Australia
Gus Pelosi, BE (Civil), Department of Land and Water Conservation, Sustainable Water Management, NSW, Australia
Paul Murphy, BSc, Grad Dip App Sci (Aquaculture). Manly Hydraulics Laboratory, Department of Services, Technology and Administration, NSW, Australia

Index

Abstract
1. Introduction
2. Setting
3. Instrumentation and Operation
4. Responses
5. Interpretation
6. Management Issues/Implications
7. Conclusion
8. References

Abstract

With New South Wales’ wetlands becoming ever smaller and more vulnerable, some of the most challenging tasks we face today are the protection of wetlands as natural areas and the restoration of degraded wetlands. The Yarrahapinni Wetland Rehabilitation Project involves the rehabilitation of approximately 600 hectares of degraded coastal wetlands behind floodgates. Implications for the management of the rehabilitation of the Yarrahapinni Wetland, the first estuarine floodplain project of this size to be undertaken in New South Wales, are raised in this paper.

To establish baseline water quality and water level conditions, data was collected from within the floodgated Yarrahapinni Wetland and in the Macleay River, on the Mid North Coast of New South Wales. The exercise commenced in March 1996 with the aim being to collect water quality data before and after the planned managed opening of the floodgates. Characteristic acid sulfate soil responses to critical rainfall events have been established, along with the concept of water column stratification prior to return to pre-event conditions.

1. Introduction

Yarrahapinni Wetland, formally known as Yarrahapinni Broadwater, was prior to the flood mitigation works a large estuarine wetland which encompassed a substantial portion of the lower Macleay mangroves, as well as extensive saltmarshes, seagrass beds and intertidal mudflats. Flood mitigation works were carried out to drain the wetland and create pasture land in the early 1970s. It was predicted that 1200 hectares of productive land would be protected from both flood and tidal inundation as a result of the flood mitigation works. Drainage of other wetlands within the catchment has also occurred. However, the drained Yarrahapinni Wetland has enjoyed only marginal agricultural productivity, and has suffered the impacts of acid sulfate soils and the loss of fish and crustacean nursery and habitat.

NSW Fisheries records show that since the draining of these wetlands there has been a marked decrease in fish catches from the Macleay River. Reduced fish production, water quality concerns and a desire to rehabilitate the wetland have prompted local commercial fishers, NSW Fisheries, Kempsey Shire Council and other government agencies to develop a proposal to manage the floodgates in order to re-establish this important fish habitat.

The Yarrahapinni Wetlands Reserve Trust was formed in 1996 by the NSW Government to oversee the ongoing management of the wetlands. In March 1996 Manly Hydraulics Laboratory (MHL) a business unit of the Department of Services, Technology and Administration, was commissioned to install water level and water quality monitoring (MHL in press) instruments. These were located at key sites to provide the initial baseline data and to monitor the performance of the rehabilitation project.

One of the goals of the baseline water quality monitoring has been to document the processes operating in the Yarrahapinni Wetland. A clear understanding of the processes is seen as essential to developing management strategies for the operation of the floodgates to improve water quality both in the Yarrahapinni Wetland and the receiving waters of Andersons Inlet.

2. Setting

Yarrahapinni Wetland is located in the lower Macleay floodplain south east of Yarrahapinni Peak, a prominent topographic feature midway between Eungai and Smoky Cape (Figure 1). The former broadwater had several openings which connected the system to Andersons Inlet. These breaks were closed with the construction of a continuous levee and the development of a five cell floodgate control structure to regulate flows. Upstream drainage improvements works were undertaken at the same time to allow more rapid off take of surface water. These complemented earlier drainage works carried out by landholders. The system has now experienced an overall lowering of the watertable and of the water level behind the floodgates.

Acid sulfate soils of the Borirgalla Soil Landscape (Eddy in press) have been identified in the immediate catchment of the Yarrahapinni Wetland. These reflect an episode of deposition of estuarine sediments which on exposure to aerial oxidation have the potential to become extremely acidic. Similar sediments are widespread on the NSW coastal floodplains and the areas with the most severe acid concerns are now being recognised as ‘hot spots’ meriting urgent remedial attention.

Figure 1: Location of study site.

Stratification of the waters in Yarrahapinni Wetland has been observed, with a very acidic and fresh layer of crystal clear water at the surface. This is underlain, with a sharp interface, by a cloudy, less acidic, brown layer marked by the presence of flocculated iron hydroxide complexes. This lower layer may grade progressively to a nearly black colour, with an increasing conductivity and a higher temperature than the surface layer.

3. Instrumentation and Operation

Five sites were selected for the collection of water level and water quality parameters (Figure 1).

Double Island - water quality measuring water level, pH, temperature, conductivity and dissolved oxygen.
Middle Island - water quality measuring water level, pH, temperature, conductivity and dissolved oxygen.
South West Rocks - water level and pH.
South West Rocks Creek - water level only.
Cockle Island - water level only.

Middle Island and Double Island sites were chosen respectively for their potential to best represent overall water quality conditions within the Yarrahapinni Wetland and in the receiving waters below the control structure. The already operating water level recorder at South West Rocks was upgraded with the addition of a pH sensor to provide a further measure of water quality downstream. The monitoring function of the South West Rocks Creek and Cockle Island water level recorders was discontinued when it became evident that the sufficient data for the baseline exercise was being generated at the other three locations.

Water quality monitoring at the sites began in mid March 1996 and was discontinued in February 1999, when it was considered adequate data had been collected to establish baseline and event processes. Monitoring should be resumed shortly before modification of the current floodgate management operation.

Rainfall data was originally collected at Upstream Euroka gauge. This was found to be unrepresentative of the events in the Yarrahapinni Wetland catchment, due in part to the influence of Yarrahapinni Peak. Retrospective data was acquired from the Bureau of Meteorology (BoM) station at Smoky Cape Lighthouse and compared with data supplied from a local landholder’s records. This data gave a very positive correlation with both observed water quality responses and the Smoky Cape data. The Smoky Cape data has been used as the control for rainfall events in the Yarrahapinni Wetland.

Water quality (and level) data capture has been highly effective with more than 95% availability of all data streams from all sources. Accuracy has been high with the proviso that dissolved oxygen records frequently show progressive drift due to occlusion of the cell by biofilms, which is considered normal owing to the 6-weekly deployment schedule.

All water quality probes were fixed level automatic instruments logging each hour, and storing the data internally for the deployment period. At the end of each deployment, the instrument was returned to MHL to be downloaded, cleaned and re-calibrated. A fresh instrument was deployed at each changeover. Vertical profiles were recorded with the new instrument prior to deployment.

Following the unauthorised removal of 4 of the 5 floodgates on the system in May 1998, which resulted in significant saline inundation within Yarrahapinni Wetland and deep penetration up the drainage system, a local investigation including further vertical and transverse profiling was carried out.

4. Responses

The baseline water quality monitoring has extended over a period just short of 3 years. During this time, 4 clear response signatures have been identified as patterns that can be correlated with particular conditions or events. A single aberrant signature was also recognised, with the unauthorised removal of the floodgates.

‘Normal’

The downstream recorders located at Double Island and South West Rocks produced ‘normal’ lower estuarine responses for the greater part of the time. The dominant control feature being the tidal movement, the records demonstrate increasing pH and (conductivity and dissolved oxygen) with increasing tides and corresponding decreases on falling tides.

‘Jerky’

In contrast, the Middle Island recorder (inside the Yarrahapinni Wetland) reveals an extremely ‘jerky’ water level pattern for the greater part of the time. The ‘jerks’ appear as sharp breaks, characterised in the water level record, synchronous with tidal lows observed in the mainstream (Double Island and South West Rocks records). These represent attenuation of the mainstream tidal cycle and, as they persist even through extended dry spells when in catchment flow is minimal, are considered to be the expression of sustained entry of tidal waters punctuated by outflows when critical low tide levels are reached.

The ‘jerky’ water level character becomes modified at non-periodic intervals and exhibits a more ‘tidal’ sinusoidal wave form. This wave form typically has a sudden commencement point and relates closely to an overall increase in water level inside Yarrahapinni Wetland. This wave form progressively returns to the former ‘jerky’ mode as the water level falls.

Highly correlative responses in pH and conductivity values have been observed with each ‘jerk’. Characteristically the fall is sharper than the rise, however this is not necessarily evident in all cases due to the hourly nature of the data capture occasionally resulting in a failure to pick up the lowest value.

Temperature and dissolved oxygen records exhibit anticipated patterns associated with normal diurnal changes. These are modified to a degree by local changes, particularly extended cloud cover which tends to depress dissolved oxygen values. Some modifications are also experienced due to tidal leakage. Limited instances of apparent supersaturation of dissolved oxygen have been noted and are considered to be linked to algal blooms, an interpretation that has been confirmed in one instance when the occurrence was observed during a changeover.

‘Critical’

Rainfall events above a certain size (as determined by the Smoky Cape data) are correlated with abrupt departures from the ‘jerky’ water level pattern, accompanied by an overall rise inside the wetland. These have been identified as ‘critical’ event patterns and are typified by a rainfall event of 75 mm of rainfall in a 24-hour period. Characteristically, pH and conductivity values fall sharply within the 24-hour period following the cessation of rainfall, and may reach alarmingly low levels for biota conditioned to a near neutral, mid-saline environment.

A typical example of this type of response can be seen for the rainfall event of 28 July 1996 (Figure 2). Dissolved oxygen values may actually rise on the day of such an event, and fall back to their former levels. For smaller events, dissolved oxygen and temperature commonly fall, probably due to the immediate depression of photosynthesis by cloud cover.

Over the period of data collection, there have been 6 ‘critical’ rainfall events which occurred after significant dry spells. These resulted in the sharp characteristic falls in pH and conductivity, and shared a common feature in that more than 75 mm of rain fell in the 24-hour period proceeding the slump.

Figure 2. Yarrahapinni Water Quality, July 1996.

‘Sub-critical’

There was a single instance of a smaller event (58.4 mm over a 48-hour period) which caused a sharp pH and conductivity fall. However, the degree of fall was not as significant with pH values stopping above 4.5, whereas the other ‘critical’ rainfall events saw pH values fall below 4.

Where larger rainfall events occurred within 14 days or less of a ‘critical’ rainfall event as defined above, the depression of pH and conductivity varied from minor to negligible. A typical example (Figure 3) occurs in late November 1998, when a ‘critical’ event on 19 November was followed by a further heavy rainfall over 25-26 November, with no significant pH response.

Figure 3. Yarrahapinni Water Quality, November 1998.

‘Aberrant’

A very significant ‘aberrant’ water quality response occurred in mid May 1998 (Figure 4), with the unauthorised removal of the floodgates. Near full tidal performance was recorded inside the Yarrahapinni Wetland at the Middle Island water quality site
Widespread upstream inundation and drain penetration by saline water from the estuary occurred as a result of entry through the open floodgates at the control structure. A period of 6 days passed before the floodgates were restored, after which time pH and conductivity values exhibited a strong downward trend. On-site investigation by MHL and the Department of Land and Water Conservation demonstrated the progressive development of a stratified water column within the Yarrahapinni Wetland, with highly acidic, relatively fresh releases on low tides. The stratification appeared to peak some 4 weeks later, but remained evident in water quality responses until a ‘sub-critical’ rainfall event was superimposed on the system.

Figure 4. Yarrahapinni Water Quality, May 1998.

5. Interpretation

The lowering of the water table in the Yarrahapinni Wetland has resulted in local accelerated oxidation of the Borirgalla Soils. Frequent pulses of highly acidic waters leading from these soils have been identified, typically after a significant rainfall event in the catchment following a extended dry spell. These acidic pulses are characterised by low pH values (increased acidity) and very reduced conductivity, along with various visual expressions in the form of turbidity (‘black water flows’), blooms (flocculation of iron and aluminium) and crystal clear flows.

Such acid flow events can and do produce conditions contributing to fish kills. Large volumes of acidic water can become impounded behind the floodgates and may persist for extended periods. As a result, regular discharges of acidic water to Andersons Inlet can occur on falling tides when the mainstream level drops below that of water in the Yarrahapinni Wetland, causing the floodgates to open.

Recovery of water quality values to dry weather conditions in the Yarrahapinni Wetland after a significant rainfall event can be quite slow, in some cases taking weeks rather than days, to return to pre-event conditions (Figure 5). Several factors are involved in this process, including the stratification described above, the limiting outflow potential of the existing floodgates, the upstream leakage that occurs at the floodgates and significant passage of mainstream water through the porous levee back into the system on higher tides. Field observation has confirmed marked temporary stratification of the water column in Yarrahapinni Wetland after the unauthorised opening. The mechanism for acidification is considered to be the leaching of acid material from the inundated acid sulfate soils of the Borirgalla Soil Landscape. This occurred only when the floodgates returned to normal operation and the watertable began unloading back to Yarrahapinni Wetland. The water quality records show the pH falling below the pre-opening level of 5, to a minimum of 3 - despite only very minor rainfall occurring in the period - before returning to the pre-opening condition.

Figure 5. Yarrahapinni Water Quality, March 1998.

Subsequent outflows are seen to reflect the earlier ‘jerky’ water pattern, with sharp, short bursts of more acidic and fresher water leaving the system at the low tide minima.

Plotting of these bursts shows an extremely high degree of correlation between the level of response in the water quality parameters (pH and conductivity) and the elevation difference in the level of waters either side of the floodgates. Characteristically, if this elevation difference is less than 200 mm, there will be little or no change in the pattern of pH and conductivity values. Conversely the greater the elevation difference, the more marked is the degree of change.

The pH and conductivity scales broadly reflect the concentration factors of hydrogen ions and total dissolved solids. However, the pH is expressed as a log scale while conductivity is a linear scale. This tends to conceal the real magnitude of pH change as values drop towards the lower end of the scale.

From observations made during and after the unauthorised opening of the floodgates, and of the response to rainfall events of varying intensities, it appears that within the lower reaches of Yarrahapinni Wetland, mainly the pool area of the former broadwater, a very significant volume of acidic water may be developed and held for extended periods.

Over a very short period, usually only a matter of days, this becomes stratified with the more acidic and less conductive waters at the top. Subsequent influx of more saline and less acidic water, from both bypasses at the floodgates and through the levee, acts to break down this stratification as the overall level of water in Yarrahapinni Wetland responds to external tidal levels.

6. Management Issues/Implications

Although the floodgates were installed some 25 years ago, there is still the development of a significant acid pool as a result of critical or near critical rainfall events in the catchment, from surface accessions leaching the soil profile. Unless the management of the floodgates is improved and/or the levee is removed or modified, the acid water discharges will continue and the wetland will continue to deteriorate.

The apparent rapid stratification of the acid pool leads in turn to concerns for sustained and heightened acidic discharges from the Yarrahapinni Wetland and their impacts on the downstream environment.

7. Conclusion

Water quality monitoring over an extended time range, covering spot events, seasonal change and longer term trends, has been practical and valuable in terms of baseline data. Interpretation of that data has outlined the water processes operating in the system and raised management issues that must be addressed in terms of water quality. In particular, the exercise has highlighted the sensitivity of the system and its potential, if the floodgates are opened and closed with a short period of time, to generate large volumes of poor quality water.

Sound and careful management of the rehabilitation project can ensure minimal impacts. These impacts potentially are far outweighed by the long term environmental and commercial benefits to be gained by improving water quality and reinstatement of the Yarrahapinni Wetland role as a healthy fish and crustacean nursery and habitat.
The rehabilitation project will provide a valuable demonstration site for other future wetland rehabilitation projects.

8. References

Eddy, M.W. (in press). Soil Landscapes of the Macksville-Nambucca 1:100 000 Map Sheet (Map and report). NSW Department of Land and Water Conservation: Sydney.

Manly Hydraulics Laboratory (in press). Yarrahapinni Water Quality Monitoring. MHL Report No.986. NSW Department of Services, Technology and Administration:Sydney.