Money Down The Drain
Questioning the causes, effects, and management of dryland salinity in South Australia's Upper South East
Site menu:
This update published 2nd August 2007
First published 24th January 2005
Flawed Case
Salinity and Flooding in the Upper South East
Cause of Dryland Salinity and Flooding
Management of Recharge on Non-Saline Land
Management of Recharge on Saline Land
Preface
This webpage is based largely on research and analysis selectively used by the Upper South East drainage program in the period to 2003. However, literature made available to this website's author over the past year indicates that the science of dryland salinity has been in dispute for a number of years, being split between the views of a dominant hydrologist/hydrogeologist camp (involved in the study of groundwater flow systems) and scientists and land managers with broader interests, in particular in vegetation and soil processes close to and at the land surface.
The hydrologist/hydrogeologist camp claims that dryland salinity is caused by rising groundwater levels resulting in increased evaporation and evapotranspiration of water on or at the surface of low lying land. Rising groundwater, as claimed by the hydrologists/hydrogeologists, has arisen from increased recharge caused by the loss or removal of deep-rooted perennial vegetation since European settlement. High groundwater recharge under higher ground is claimed to push up groundwater levels in low-lying areas, bringing salts previously held low in the landscape close to the land surface, where they are concentrated by evaporation and evapotranspiration.
The second group of scientists generally claims that soil structure failure and loss of vegetation on the low lying land is a result of European agricultural practices, and, combined with other factors, is the primary cause of dryland salinity.
These major differences of opinion were first formally reported in 2004 in a House of Representatives report on salinity, and subsequently in a Senate report published in 2006. The differences of opinion were also aired in May 2006 in a Channel 9 TV documentary.
The author of this webpage has been provided with numerous scientific papers and analyses that support the soils structure/vegetation arguments, which appear to be of relevance to many cases of reported dryland salinity in the Upper South East.
Over the past decade, the Upper South East Program has been driven by the hydrologist/ hydrogeologist model, with little or no consideration given to the role of soils and vegetation in dryland salinity. Indeed, recent scientific reporting for the Program indicates that the role of soil structure, including the failure of soils and sub-soils as a result of drainage, has only just started to attract the attention of the responsible government officers.
As time permits, these reports will be summarised and added in updates to this webpage.
Introduction
The program of drain construction is proceeding because it was mandated, land for the drainage network was compulsorily acquired without compensation, because Commonwealth funding was provided in response to a grossly flawed case, and because the State Government gave itself the power to impose a major levy on all landholders in the Upper South East.
Wheeler (2002) noted that as a result of complaints from landholders about the original proposed sub-schemes (two major groundwater drainage systems, and one major surface with some groundwater drainage), an independent review was conducted in 1994 by Coffey MPW (Coffey, 1994). The review concluded that major surface water drainage was preferred.
Coffey MPW argued that only surface drainage should be necessary to sustain saltland agronomy, and then only where modelling indicates that flooding could occur for a month or more with a return period of greater than 15 years. Surface drains would then be less of a threat to wetlands.
The case presented in the draft Environmental Impact Statement Supplement Report (NRC, 1994) inadequately, and in some cases incorrectly, represented Coffey MPW comments and recommendations. Selectively used statements that minimised the negative impacts of groundwater drains, eg sections 9.4, 9.4.1 and 9.4.2 and now known to be incorrect, promoted the efficacy of groundwater drains.
Wheeler (2002) notes that the scheme proposed by the State Government in 2002 did not follow Coffey MPW's recommendation.
While groundwater drains might have a role where waterlogged soils are not saline, or where saline soils are transmissive to depth and not subject to sodicity on drainage, they should not have a major role where clay features in the top 3m or so of a saline soil profile (eg the duplex soils and clay loams of the Upper South East). The long-term economic and environmental costs associated with reversing or managing chemical changes to these soils can rarely be justified by the benefits. While a drain might convey the appearance of progress in controlling the spread of salinity, it has high potential to cause even more environmental damage, harm agricultural productivity, and at the same time have only limited impact on salinity!
The Program Director reported that the "economics indicated the project based solely on drains would be marginal in terms of benefits" and that "increased water use efficiency of perennial pastures have a significant impact .... on groundwater recharge" (Johnson, 2002), but still the majority of Program funding ($45million from a total of $73.3million) was directed to drain construction. Cost-benefit analyses conducted in 1993 and 2002 also concluded that a combined pasture renovation/drainage program "barely breaks even on financing costs with little or no additional remuneration to the farm business owners" (Barber, 1993), and that "the total net benefits from improved agricultural productivity do not outweigh the costs of the scheme" (Wheeler et al, 2002)!
Salinity and Flooding in the Upper South East
State Government publicity claims that the area of saline-affected land, or land at risk of salinity (terms used interchangeably but which are significantly different in meaning) in the Upper South East has remained virtually unchanged in the period 1992 (250,840ha) (Cann et al, 1992) to 2004 (250,500ha) (Upper South East Dryland Salinity and Flood Management Program, 2003)(DWLBC, 2004a). Published information (Johnson, 2002)(DWLBC, 2004a), reporting Barnet (2000), also claims that the saline area has potential to increase without drains to a maximum of about 410,000ha.
The National Land and Water Resources Audit (Audit) initiated a comprehensive national assessment of dryland salinity employing consistent methodologies based on a groundwater hydrogeological framework (NLWRA, 2001). The Audit noted that groundwater level and trend data are recognised as fundamental requirements in evaluating the area of saline-affected land and the rate at which it was changing. The Audit projected likely increases in dryland salinity to 2050, assuming a continued rate of watertable rise and no change to water imbalance (Australian Dryland Salinity Assessment 2000, NLWRA, 2001).
The intent was to produce individual State reports on areas at risk of dryland salinity, using information on groundwater levels and trends, known incidence of salinity, soil characteristics, and topography. For those States with data on groundwater levels and trends (eg South Australia), land within 2m of the surface, or within 2 to 5m and with well demonstrated rising watertables were classed as being at high risk of dryland salinity. The Audit noted that: "This information should not be interpreted as actual areas affected since the assessments are likely to overestimate areal extent particularly in dissected (hilly) landscapes. Rather, they identify areas or regions within which dryland salinity occurs or could occur. Groundwater trend analysis at the scales used will only provide an overview. ..... These estimates also include some areas - mainly in the temperate zone - with persistent waterlogging from shallow watertables. Although salt is ubiquitous in landscapes in agricultural areas, it is acknowledged that not all of these waterlogged lands will become saline." (Australia's dryland salinity, NLWRA, 2001).
Barnett (9, 2000) apparently used digital terrain models of the Upper South East and data on watertable elevations and trends to predict areas at risk of dryland salinity. While noting that groundwater levels had fallen in southern South Australia in the 2 - 3 years prior to 2000, with some drier catchments experiencing falling groundwater levels since 1993 (Dryland Salinity by State - South Australia, NLWRA, 2001), and that the greenhouse effect is expected to lead to lower winter rainfall when greatest potential for recharge exists (Barnett, 2000), Barnett still projected an increase in the area at risk of salinity in the Upper South East without explanation. Barnett did not show or demonstrate that watertables were rising, or that salinity was increasing (claimed in the report by Cahalan, 2005).
Barnett confusingly listed the Upper South East area affected by secondary salinity in 2000 as 250,500ha (see eg Table 4 in Barnett (2000)), and later an identical area at risk of salinisation (see eg Table 6 in Barnett (2000)). Large areas in the Upper South East's north-west were incorrectly identified as affected or at risk of secondary salinity, although being saline for several thousand years (evidenced by gypsum and calcrete) but successfully used for agriculture.
Barnett does not appear to have considered different soil types nor fresh water stratification of the groundwater in the assessments of land at risk of salinity (18, Barnett, 2000). Both exist in the Upper South East. More porous, coarse textured soils have a lower capillary rise distance (<50cm). Groundwater tables can thus be close to the surface without leading to dryland salinity (Seelig, 2000), and are less sensitive to watertable rise as a result of recharge.
Furthermore, low salinity groundwater close to the surface should reduce the potential for dryland salinity, although it could exacerbate waterlogging and adversely affect productivity during periods of high rainfall. This will have contributed further to the area of dryland salinity being overestimated. While noting that comprehensive coverage of aerial photography would provide a moderate to high level of of confidence in the estimates, limitations were mainly related to the lack of detailed ground truthing and the complexity of dryland salinity (23, Barnett, 2000). Complexities listed include overlaps of primary (natural) and secondary (European-induced) salinity, exposed subsoils that are not necessarily associated with raised groundwater levels, and seasonal changes that can cause rapid expansion or contraction of affected areas.
The 1992 data (Cann et al, 1992) indicated that saline areas in selected parts of the Upper South East had expanded by between 10% and 32% since 1988, but Barnett (2000) subsequently noted that such increases were "most likely the result of increased awareness and better recognition of the problem, rather than the physical expansion of salinisation", although some growth might be attributed to very wet years. Subsequent studies, in particular benefit-cost analyses (Barber, 1993)(Wheeler et al, 2002) that supported development of the 1993 and 2002 cases for the drainage program, had assumed increases in saline areas based on extrapolations from the 1988 and 1992 data. The analyses assumed transitions from areas of lower salinity to higher salinity of between 4.5% to 12% per year, and an initial annual rate of increase of the total area of saline land of about 7,900ha/year (117, NRC, 1993).
Semeniuk (1993) observed that some data on hydrology had been collected over a very short period, and concluded that hydrological trends were much more complex than the single factor attributed then to increasing salinity (namely, loss of perennial vegetation). The NRC defended a general observation that the draft Environmental Impact Statement was based on short-term effects with a lengthy justification (9.2.18, NRC, 1994), which is now known to have been wrong!
A CSIRO/PISA mapping study (Furby et al, 1998) determined that 195,240ha of the Upper South East was salt-affected in 1997, a reduction in 22% from the 1992 area (or a decline of about 11,000ha/year). The estimate seems reasonable, given that watertables had peaked and started falling by the mid 1990s, that "1992 was a wet year with about double the average rainfall for June and July", that "the 1997 season was considered more average in terms of rainfall", that some areas were incorrectly being labeled as salt-affected when "the cause of [low productivity] is waterlogging not salinity, and that "salt-affected land [in the southern catchment] is being considerably over-estimated" (Furby, 1998).
It appears that after reaching a maximum in the early to mid 1990s, the Upper South East's saline area started contracting. South Australia State Government literature for more than a decade appears to have been directed more at creating concern about a fictitious growth in salinity, and justifying the case for an extensive drainage network, than presenting facts. State Government parliamentarians were provided this outdated information, which resulted in the urgent passing of the Upper South East Dryland Salinity and Flood Management Act 2002 (see eg Hansard, 2002a).
Very recent CSIRO reporting confirms that watertables in the Upper South East have been falling since 1993 as a result of lower than average rainfall (Cox, 2005), and probably because of increased planting of perennial vegetation (DWLBC, 2002). Data from observation wells in the Upper South East show this decline, and is freely available on a State Government website (PIRSA, 2005).
A review of Upper South East observation wells by Coffey MPW in 1994 (NRC, 1994a) showed that 33 under dunes indicated that watertables were rising, 46 were static, and 3 were falling (but only where groundwater withdrawal for irrigation was occurring). However, the draft Environmental Impact Statement (47, NRC, 1993), quoting Mackenzie GA et al (1992), incorrectly indicated that watertables were rising at a rate of between 0.5-1.0m/year. In the Central Catchment in 2004, 8 out of 12 observation wells on dunes indicated that watertables were falling, and 9 out of 10 observation wells on the flats showed falling trends (Cox J et al, 2005), which are potential sites of dryland salinity salinity.
More general reporting indicates that a majority of observation wells (estimated in excess of 90%) have shown falling trends across the region since 1990 (picture here from Obswell, published in Howieson, 2003). The blue arrows indicate falls, with largest arrows indicating greatest falls. Red arrows indicate rises.
The draft Environmental Impact Statement indicated that rising watertables in the period to 1992 were attributed largely to the clearance of native vegetation, but now, inconsistently, falling falls during the following decade are being attributed by Program staff to lower than average rainfall, and not to replanting of perennial pastures and native vegetation!
The justification for the regional drainage network component of the Program was to control and reverse the spread of dryland salinity, by lowering watertables (11, NRC, 1993). According to PWC (1999) reported in Hansard, "a workshop is planned for January 2000, with a panel of experts in biology, hydrology and ecology to discuss the results of monitoring to date and to determine the most appropriate management strategy for the discharge [to the Coorong]". The expert in hydrology would have been aware then that watertables were falling (many observation wells had been specially installed to monitor watertable trends in the region after 1993) and that dryland salinity was receding.
The graph shows 21-year moving rainfall averages (annual average for period ±10 years either side of each year plotted). BOM recommend using 30-year averages to detect long-term trends and remove the effects of short-term fluctuations (pers comm Clemmett, 2005). However, long-term trends are still clearly seen. Click on graph for a high resolution image.
The rainfall trend appears to be a feature of the Upper South East, which has experienced a decline in rainfall of between 10 - 20mm/decade since 1950 (Figure 45, p28, BOM, 2005b), amounting to a reduction of about 15% at Lucindale and Kingston over 50 years. See the Bureau of Meteorology trend maps website for latest information.
Heneker (2006) reported that a 10% decrease in rainfall may result in a 30-40% decrease in runoff, and also noted a 12% decrease in locations south and east of the Upper South East region.
Longer term rainfall trends in the region, deduced by Professor Patrick De Deckker, ANU, from analysis of sediments in Mount Gambier's Blue Lake (reported on the ABC TV's Catalyst program), indicates that the recent observed downward trend is part of a decline that started about 1,000 years ago. The picture shows deduced Blue Lake depth, which is assumed to be proportional to annual rainfall (adapted from Catalyst)
Analysis of monthly rainfall records (BOM, 2005a) shows that current dry season (Nov- Apr) annual rainfall across the Upper South East appears to be similar for all centres studied, averaging around 140mm ±20mm since 1990, after following a generally downward trend from 210mm ±20mm in 1970, with a second peak of about 240mm ±40mm further back in 1945.
The period from May - August is significant, because this is when rainfall usually exceeds evaporation and transpiration, and significant recharge can occur where the watertable is below the root zone, ie on dunes and ranges. Since 1945, long-term (21-year moving) May - August averages have each varied in a band of less than 40mm for all centres studied. Keith and Bordertown current averages are about 230mm, Marcollat (Padthaway) and Naracoorte about 275mm, and Kingston and Lucindale a little over 310mm. Click on graph for a high resolution image.
Current short-term (5-year moving average) May - August averages for all centres are equal to, or a little higher than, their long-term averages. All centres had short-term averages nearly 50mm higher in 1990, which then fell to minima about 50mm lower in 1999! The graph shows a clear cyclic variation in rainfall with a period of about 9 years. Click on graph for a high resolution image.
Current rainfall during the period of greatest recharge potential (May to August) is thus similar to the long-term average. While 2006 has been a year of well below average rainfall, such variability is not uncommon. Most of the decline in annual rainfall averages has occurred during the dryer months, between November to April. The period May to August is also when prolonged periods of waterlogging are most likely to occur on flats where watertables are close to the surface or where the permeability of soils is poor.
Annual rainfall in the Upper South East has followed a downward trend since the mid-1900s. In 2030 it is projected to be 5% below the 1990 average, and in 2070 12.5% below the 1990 average (Whetton, 2001). Most models predict that where average rainfall decreases there will be more dry spells, and also, in general, a possible increase in extreme daily rainfall events. Where rainfall decreases significantly, models predict a reduction in extreme rainfall (Whetton, 2001)! Predicted annual rainfall averages for 2030 range from about 440mm for Keith to about 540mm for Lucindale, and in 2070 about 400mm for Keith and about 500mm for Lucindale.
Empirically, each 25mm reduction in rainfall results in a reduction in potential stocking rates by one dry sheep equivalent per hectare (dse/ha). In today's prices, a reduction of one dse/ha is equivalent to a lowering of gross margins by about $35/ha.
Assertions in the draft Environmental Impact Statement (107, NRC, 1993) that "there is no doubt that the hydrological imbalance driving dryland salinity has been caused by excessive native vegetation clearance and the loss of lucerne from the dunes", and that the area affected by dryland salinity would continue to expand at a rate determined by data collected between 1988 and 1992, to a maximum area determined from soil mapping, was incorrect. It appears that higher than average rainfall towards the end of the 1980s played a key role, which destroys the central argument used to justify a regional drainage network. Conversely, forecast lower than average rainfall will continue to have a strong influence on lowering watertables for the foreseeable future, which would have been reinforced by an active revegetation and pasture renovation program.
The requirement for a regional groundwater drainage network has thus been removed, which, based on data available already in 1999, should have been concluded well before the current stage of drain construction. It was clear in 1999 that watertables, a key indicator of dryland salinity, had been trending downwards since 1993 (see data at PIRSA, 2005). Contrary to Australian Dryland Salinity Assessment 2000 recommendations (Summary, NLWRA, 2001), it is incomprehensible that neither of these key drivers of the Program were not re-validated in 2002 when the Program was reviewed, prior to the Upper South East Act being passed and Stage 3 approval given by the PWC (2003). It is even more incomprehensible that National Action Plan for Salinity and Water Quality managers (NAP, 2003) should approve a further $38.3 million of tax-payer's money in 2003 to support Stage 3 completion with the following targets for the Upper South East Program:
-
Resource condition targets (land salinity):
-
Area of saline land to be reduced by 30% within the drained area of the Upper South East by 2010 (22% achieved in 1998, so at least 30% probably achieved in 2000, without drains!!).
-
Predicted rate of increase in salinity across all other salinity affected areas in the Upper South East reduced by 50% by 2020 (probably achieved in 1993, without drains!!).
-
-
Management action targets (land salinity):
-
Determine the salt and water balance under perennial salt-land pasture in the Upper South East system by 2008.
-
Determine the salt and water balance under established production systems of the shallow groundwater system of the Lower South East by 2008.
-
650km of drains completed by 2010.
-
Deep rooted perennial vegetation planted over 85,000 ha by 2007 (according to McEwan et al (2002), over 88,000 ha (considered to be a "gross underestimate) of dryland lucerne had been planted prior to Stage 3 drain construction approval, and over 131,000 ha of native vegetation was recorded!!).
-
National Action Plan for Salinity and Water Quality funding for the current stage of the Upper South East Program was thus justified in 2003 with objectives only of controlling and reversing trends in dryland salinity, which had been in natural decline since the early to mid 1990s!
Temperatures in the region are expected to rise by an average of 0.75ºC by 2030, and by 2.5ºC by 2070. Each 1ºC increase in global average temperatures is expected to result in the moisture balance in the Upper South East decreasing by an estimated -90mm, arising from higher evapotranspiration and lower rainfall (the current balance is about -650mm).
The combination of lower than average rainfall and higher temperatures, and ongoing planting of perennial vegetation, can only mean that dryland salinity in the Upper South East will have continued to decline, without drains!
The original Environmental Impact Statement (NRC, 1993) indicated that the northern catchment had reached a hydraulic equilibrium, and that recharge reduction is likely to reduce salinisation due to local flow cell effects. A change in land use over half of the area (approximately 20,000ha) would reduce recharge by about 50%. (Rural Solutions, 2002)
Specifically, a salinity risk assessment for the Upper South East indicates that around a third of one landholder's grazing property (840ha) in the most salt affected area of the Upper South East is "highly to extremely saline land" and "non-productive", yet the majority of this land had been successfully sown to the very productive grazing pasture, puccinellia. The existence of gypsum on this land (evident over an area of at least 200ha) also indicates that salinity is naturally occurring, and pre-dates European settlement by nearly 20,000 years.
A further third of the land was classified as "already too saline for many field crops", but was successfully sown to lucerne and veldt grass, and watertables had been falling since the mid 1990s.
Analysis of Upper South East salinity land categories in 1992 (Cann (1992) correlated with Anon (2000)) identified:
-
256,050ha not affected and unlikely to be affected (Cat1),
-
175,050ha not currently affected but at risk (Cat2),
-
109,540ha showing low levels of salinity, and on which lucerne and some clovers will not grow (Cat 3),
-
75,350ha showing medium levels of salinity, and vegetation dominated by salt barley grass (Cat 4), and
-
65,950ha showing severe levels of salinity (some natural), with sparse sea barley grass, and land dominated by samphire and bare scalds (Cat 5).
Cat 1 and 2 land is suitable for conventional perennial pastures, and despite the above formal descriptions, at least 80% of the land on one property identified as Cat 4 grows lucerne and veldt grass (photograph below). Land identified as Cat 5 supports productive puccinellia and tall wheat grass pastures.
How wrong are the risk assessments for other land in the Upper South East?
 Newly Sown Lucerne on "Saline" Flat. Winter waterlogged Puccinellia patch (arrowed) will reduce in size as groundwater is lowered under lucerne. |
 Cattle Grazing Freshly Sown Lucerne on a "Saline" Flat |
In a survey of 98 Upper South East landholders conducted in 2002, only 8% considered they had a salinity problem (Truscott, 2002)!
Key problems associated with sowing and establishing lucerne in the Upper South East, namely non-wetting sands and aphids, have been overcome with specially-modified direct-seeding combines, and by the breeding of aphid-resistant lucerne species. Productivity on saline land can now be as good as, and in many cases better than, on non-saline land (Strugnall, 2004)(McFarlane, 2005) .
The Causes of Dryland Salinity and Flooding in the Upper South East
The Upper South East covers an area of about 680,000ha, with average rainfall ranging from 450mm in the north to 550mm in the south. Most rainfall occurs in the cooler months, when evapotranspiration is lowest, and hence recharge of the groundwater is potentially highest. Potential evapotranspiration averages 1,100mm. In the southern half of the Upper South East, a series of remnant NNW-aligned sand dunes are separated by interdunal flats, which are often vast and generally have sandy clay and calcareous soils. In the northern half of the Upper South East, the dunes are jumbled. O'Driscoll (1960) provides excellent descriptions of the soils and hydrology of the Upper South East.
The flats are 1 - 10km wide and slope gently down from east to west (1:2,000 to 1:5,000), and from south to north (1:5,000 to 1:10,000). This results in periods of seasonal inundation, in particular on the western sides of the flats where swamps and wetlands are often located. The main areas affected by dryland salinity are reported to be the flats, which are formed from an ancient marine sedimentary basin. The prime agricultural land is generally on the eastern side of the flats. Native vegetation now covers about 20% of the region (McEwan et al, 2002). Before drains were constructed, up to half of the South East would be inundated in wet years (Goyder, reported in Hansard (2002), and Turner and Carter (1989)).
The principal cause of dryland salinity in the Upper South East was reported to be rising groundwater, resulting from large-scale clearance of native vegetation in the 1950s and 1960s, followed in the 1970s by the demise of vast areas of lucerne killed by drought, aphids, wingless grasshoppers, over-grazing, especially by sheep, and then a massive flood event in 1981. The underlying saline watertable rose rapidly, resulting in the development of areas of dryland salinity (155, SENRCC, 2003)]. The net effect after clearance and the loss of lucerne was an increase in rainfall reaching the watertable under dunes, from less than 0.1% to 10 - 15%. Increased water pressure under the dunes then caused saline groundwater under the flats to rise (Walker et al, 1997).
Surface soils were initially of low salinity due to winter leaching. Highly productive pastures could be established and maintained, even in areas of high sub-soil salinity and shallow groundwater levels. The spread of agricultural development caused a more frequent and extensive downstream transfer of water along the watercourses, resulting in inundation and waterlogging. Wet winters prior to 1993 accelerated these processes, and increased grazing pressures on higher ground. In turn, this and flooding on the flats resulted in further losses of perennial pastures and an increase in recharge. Loss of pastures on the flats resulted in increased evaporation with consequent salinisation. (5, NRC, 1993)
Less well reported is the role of stop banks (including weirs) in causing salinity and flooding. These have been constructed to halt surface flows along natural watercourses, and prevent inundation of down-stream areas converted to agriculture. A side effect though has been to hold back up-stream water, which has inundated low lying areas and locally raised watertables. Lack of fresh water flushing flows resulted in salts building up at the surface and in the root zones of native vegetation, killing them. Stop banks need only be very small to have a major damaging effect on native vegetation. The lower right photograph is of an area of dead native vegetation caused by saline sub-soils. Excessive recharge raised saline groundwater levels and a 300mm high stop bank constructed by a down-stream farmer to prevent water inundating his paddocks, also prevented fresh water flushing salts out of the system.
 High recharge land - products of poor surface cover and Upper South East weather. Less water from rain is used by vegetation, resulting in more entering the groundwater, causing the watertable to rise, which in turn causes dryland salinity. |
 Dual cause of death of native vegetation. High recharge, and a 300mm high stop bank. |
When the watertable is close to the surface (few centimetres to 0.5 m for coarse structured soils such as sandy loams, and 2 - 3 m and more for finer grained clayey soils), capillary rise causes water to flow to the surface, where it evaporates leaving dissolved salts behind. The dynamics are such that salt accumulates in the top-soils over summer, which are leached back to the groundwater in winter. The nature and extent of vegetation cover can reduce evaporative loss by lowering soil temperatures, locally increasing humidity, reducing wind velocity, and, if salt-tolerant, using soil moisture and thus controlling the amount reaching the surface. Vegetation also causes salts to become concentrated close to roots, as fresh water is extracted from groundwater.
Net groundwater discharge (when discharge exceeds recharge) causes salts to accumulate within 2m of the surface (Walker et al, 1997). Groundwater salinity ranges from less than 1,500mg/L in the east and local areas in the south of the Upper South East, to greater than 12,000mg/L in the north (Fig 3.3, SENRCC, 2003).
The existence of gypsum (and calcrete) over a relatively large area in the Upper South East indicates that dryland salinity has existed there for at least 20,000 years (Keeling et al, 2001). Although still saline, this area has been successfully used for grazing after sowing with salt-tolerant pastures. Assertions that dryland salinity in the Upper South East has arisen from human activity (eg Johnson, 2002) was thus wrong.
Groundwater in the region is associated with an unconfined aquifer that flows generally from east to west, and which originates across the State border in western Victoria (SENRCC, 2003; Understanding salinity management, NLWRA (2001)). Perturbations in the watertable occur locally due to local recharge and discharge (Fig 14.1, page 210, SENRCC, 2003; Walker et al, 1997). The aquifer thickness in the Upper South East varies from approximately 19m to less than 30m (Fennell at al, 1992). The groundwater gradient is greatest to the east of the Upper South East (1:1,000, resulting in faster groundwater flow rates), is moderate to low in the central region (1:2,500 to 1:5,000, with slowest flows in the north), increases towards the coast (1:1,500 ) (Fig 3.4, page 45, SENRCC, 2003), then slows again at the coast because of a thinning of the aquifer (Walker et al, 1997).
Groundwater flow rates are proportional to the product of the potentiometric gradient and soil transmissivity. Groundwater flow rates vary from 0.5 m/year (Walker et al, 1997) to 100 m/year (page 46, SENRCC, 2003). The slow movement means that the groundwater is frequently incapable of removing all water entering by recharge, so surplus is accommodated by the watertable rising. Salinity also increases gradually as groundwater moves across a flat due to evaporative concentration (Walker et al, 1997), being greatest generally where gradients are smallest and the watertable is closest to the surface.
Recharge
Recharge to the unconfined aquifer occurs by a number of different processes (page 46, SENRCC, 2003):
-
Lateral groundwater recharge, from areas of higher potentiometric head, generally from the east, and locally from recharge areas.
-
Diffuse recharge, usually the dominant source of replenishment of the unconfined aquifer, which is infiltration of rainfall through the soil into the aquifer. The magnitude of recharge is controlled principally by rainfall amount, soil type, depth to watertable, and type of vegetative cover.
-
Direct recharge, through naturally occurring sink holes.
-
Point recharge, through localised infiltration of surface water from numerous swamps in the region.
-
Upward leakage, from the deeper confined aquifer.
Along the eastern margin of interdunal flats, the aquifer can also receive recharge from groundwater outflow from the adjacent dune range, evidenced by the occurrence of springs (Mackenzie et al, 1992).
The interdunal flats, where significant discharge occurs during the summer months, are also significant recharge areas in winter. Recharge on the dunes under native vegetation and lucerne is a few mm/year, and under annual pastures has been measured at 50 - 75mm/year. On the flats recharge can be as high as 400mm/year, presumably in the absence of any vegetation (SENRCC, 2003), and under annual pastures is up to about 100mm/year (Mackenzie et al, 1992). The magnitude of recharge on the flats makes the management of dryland salinity with solely plant-based options practically impossible, because plants can only use a portion of the water during the wet season, when temperature limits growth.
However, these figures contradict research conducted in the Upper South East in the mid 1950s by CSIRO (reported in O'Driscoll (1960)) during particularly wet years, believed to be 1955 and 1956. The CSIRO research found that rainfall did not percolate lower than 5m below sparse vegetation found on some dunes.
In 1992, groundwater levels in a "large part" of the Upper South East were reported to be rising at up to 0.5-1.0m/10-years, particularly under the "more elevated dune ranges and the topographically higher area to the east of the Keith-Naracoorte road". However, this observation was not evident in the 24 hydrographs presented in (Mackenzie, 1992), which showed a mix of rising, falling and generally static trends. Some hydrographs (4 total) showed a distinct rise in watertables during the 4-year period to 1992 (about 0.5m rise), a period of rising rainfall which peaked in 1992. Other hydrographs (4 total) showed a general rising trend (up to 2m/10-years) over a period of 10-15 years to 1992, but where the watertable was 10-20m below the soil's surface! One hydrograph showed a rising trend (about 1m/10-years, and was about 2.5m below the soil's surface in 1992) to the west of the Keith irrigation area, but 13km to the east a hydrograph showed a falling trend (about 1m/10-years to be about 8.5m below the soil's surface in 1992)! The remaining 14 hydrographs showed an approximately equal mix of stable, slightly rising and slightly falling trends.
Modelling of groundwater drainage (Armstrong et al, 1992) used transmissivity data collected at a number of locations in the Upper South East. The test data recorded soil characteristics to depths of between 5m to 8m, and showed up to 7 distinct soil layers (Fennell et al, 1992), although modelling assumed homogeneous soils. In 2005, the Program's drain construction manager acknowledged that transmissivity data were incorrect, because of the methodology used, and the time at which measurements had been made (in autumn when watertables were at their lowest).
Furthermore, transmissivity data of water through soils incorrectly assumed that soil characteristics do not change as saline water is flushed with fresher water. In fact, transmissive soils can rapidly become poorly transmissive, measured in cm/year (pers comm Merry, 2004), as sodicity symptoms appear (Fitzpatrick et al, 2003)(Strugnall, 2004a), and result in poor drainage and increased waterlogging!
Discharge
Currently, the generalised depth to the watertable is in the range 0 - 2m over about 50% of south Upper South East, and 2 - 5m and deeper over more than 70% of the north. A small area about 20km north-west of Keith in the north has a generalised watertable in the range 0 - 2 m. (Fig 14.1, SENRCC, 2003) The main discharge process is the evaporative loss of groundwater in areas where the watertable is within about two metres of the land surface (Mackenzie, 1992).
Groundwater salinity trends to higher values in a northerly direction (Fig 3.3, page 43, SENRCC, 2003), as does the area of land identified as being highly saline (Fig 12.2, page 160, SENRCC, 2003). The higher groundwater salinities in the north-west of the region are likely to be a result of lower vertical recharge rates (from rainfall or wetlands), higher temperatures and hence evapotranspiration (see BOM, 2005), reduced aquifer flushing in an area where the aquifer thickness is relatively small, poorly developed surface flows, and possibly historic surface water discharged to the area carrying dissolved salts (arising from north-west moving surface water flows along interdunal flats) (page 44, SENRCC, 2003). This does not satisfactorily explain all high to extreme dryland salinity in the north of the region where the groundwater is 2-20m below the surface, unless waterlogging (resulting in death of vegetation) and sodicity (dense, infertile, impermeable soils) are being misidentified as salinity.
According to Walker et al (1997), the amount of water lost by evapotranspiration is hardly discernible in the seasonal fluctuation of the watertable. The regional influence on groundwater discharge is small, amounting to about 1% of that lost over summer, and 10% of net discharge. This means that local land management can have an impact, because discharge is not unduly affected by recharge outside of the region.
The relatively slow lateral movement of groundwater means that long distance influences on the watertable arising from areas of high recharge are attenuated. Pasture renovation and revegetation of recharge areas could thus have a local impact on watertables, which from experience does work. One could thus also conclude that increased local discharge, and consequently dryland salinity, has arisen because of excessive local recharge and/or poor surface cover, eg caused by local farm management practices, such as cropping, reliance on annual pastures, and over-grazing.
Long-term net discharge determines salt accumulation, whereas short-term recharge and discharge processes affect groundwater fluctuations (Walker et al, 1997). Groundwater discharge by capillary rise is a very sensitive function of the depth to the groundwater, so a small lowering of the watertable can have major effect on the amount of salt accumulating on the soil's surface. Surface cover reduces evaporation and salt build up by shading the surface from the sun, and slowing wind at the surface, allowing more humid conditions to be established. Periodic leaching by rainfall into the sub-soils minimises the impact of salts on shallower rooted plants. Thus, strategically revegetated recharge areas could cause a rapid reduction in the amount of salt existing in the root zones of plants, but groundwater would still remain saline at depth.
While revegetation might seem an ineffective option for managing dryland salinity, it should be remembered that the total salt load in the Upper South East has remained virtually unchanged since European settlement - only its distribution has changed. Because agricultural productivity is inextricably linked to soil water of low salinity, a primary objective of any management option should be to remove salt from the root zone. Two possibilities exist - either cause the salt to move lower in the soil profile (by lowering watertables), or remove it (using drains). Alternatively, one could exploit the available soil water using salt-tolerant vegetation. In practice, the optimum solution is likely to be a combination of saltland agronomy, recharge management and drains.
Surface Water
Despite a fairly extensive drainage scheme in the Lower South East, waterlogging still remains a significant problem because of high watertables. Over 59% of the whole of the South East is susceptible to waterlogging, and 15% is wetland (page 176/177 and Fig 14.1, SENRCC, 2003). Surface water originates mainly from winter rainfall, when rainfall exceeds evapotranspiration and watertables rise. In other cases, waterlogging occurs when rainfall exceeds infiltration rates, mainly because of hard pans and sealed surfaces in sub-soils. In general, fresh water will lie on the top of saline groundwater, unless a saline watertable has been pushed up as a result of lateral flows from under adjacent dunes and ranges applying an upward pressure on the flats.
Surface water moves slowly across the flats of the Upper South East towards the coast (in the west) until it is directed northwards along the eastern side of a range. The ultimate discharge points for peak surface water flows used to be the terminal wetlands in, and south of, the Messent Conservation Park and the Coorong at Salt Creek (page 39, SENRCC, 2003).
Why is Salinity Bad?
Soils are defined as saline when they contain sufficient soluble salts to adversely affect the growth of crops and pastures. The predominant soluble salt is sodium chloride, but can also include appreciable quantities of sodium bicarbonate, sodium bicarbonate, and chlorides and sulfates of calcium and magnesium. Less soluble salts such as calcium sulfate (in its hydrated form known as gypsum), and calcium carbonate (lime) can contribute to salinity if present in quantity. The highly soluble salts sodium carbonate and sodium bicarbonate are capable of alkaline hydrolysis (they react with water to produce hydroxyl ions), which increases soil pH to 9.5 and higher. High concentrations of sodium salts also cause sodicity, which arises when sodium ions become attached to clay and organic matter particles. Salts originate from different sources, eg rain (typically in the range 20-60 parts per million), left behind from when the land was under the sea, and dissolution of salts in the soil profile as rain leaches through.
Salinity is a problem for plants for two main reasons. Firstly, the plant must expend more energy extracting water from saline soils. Dissolved salts create a force (osmotic pressure) which opposes the movement of water from a saline soil environment to the relatively fresher environment within the roots. As the concentration of dissolved salts increase, so does the force opposing the extraction of soil water by the roots. Thus, a symptom of salinity is often wilting, arising from the inability of a plant to take up sufficient water. The dissolved salts are also sources of ions that are toxic to plants if present in large numbers, and can cause high pH which limits the uptake of critical nutrients.
See (Abrol et al, 1988)(Warrence et al, 2005) and (Seelig, 2005) for more information on salinity.
Salinity Management
Visitors to the Upper South East will not be surprised that significant recharge of groundwater has occurred. Large areas of cleared land are covered with weedy annual grasses, and some areas are completely devoid of any vegetation. One can also see paddocks renovated with lucerne thrive well on one side of the road (high profitability, low recharge), with only samphire and silver grass on the other (low profitability, high recharge).
While the potential exists to raise profitability, and significantly reduce recharge in the Upper South East through pasture renovation, many farmers either do not have the confidence, money, or will, to make the necessary improvements. Upper South East soils are fragile and degrade rapidly if abused. However, they do respond well to pasture improvement and fertiliser application, but this requires a reinvestment strategy not followed by many farmers.
If the salt could be re-distributed so that it is out of the root zone of pastures and crops, then farms could return to conventional techniques. Theoretically this is possible using vegetation, since evapotranspiration in the region is nearly twice that of rainfall.
An extensive recharge management program could thus have an immediate localised impact on dryland salinity. However, large quantities of recharge accumulated under dunes might take some time to dissipate due to low lateral flow rates, although this seems to have been dissipating naturally over the past decade. A modest lowering of the watertable though should have a major impact on salt accumulation, and open up possibilities to use less salt-tolerant but more productive, high water-use pastures.
Declining rainfall, which is likely to continue for the foreseeable future, is aiding the process of lowering watertables, but this also slows down leaching of salts from the top soils into the groundwater.
As salt is leached further down the soil profile, the effects of sodicity (discussed later) lower the transmissivity of underlying clays, reducing its ability to leach salts further down the soil profile. It is therefore important to initiate soil remediation to counter the effects of salt-induced sodicity as soon as the watertable declines, and before the soil's structure starts to fail.
Management of Recharge on Non-Saline Land
Recharge studies (Fig 4.2, Walker et al, 1997) have shown that a soil's clay content can cause a significant reduction in potential recharge where the watertable is not shallow. A nearly ten-fold decrease in recharge results from increasing clay content in the top 2m of soil from zero to 10%. Clay also reduces the water repellence of non-wetting soils, and increases inherent soil fertility. However, 10% clay in 2m weighs the equivalent of about 3,000T/ha, which is far in excess of that which can reasonably be applied by clay-spreading.
While it might be possible to reduce recharge on dunes by revegetating with very high water use perennials (eg lucerne <5mm/year), it is more difficult to do on flats that are saline. During the '70s, about 45% of Upper South East pastures were sown to lucerne, which, according to recent State Government literature, provided “an effective means of reducing groundwater recharge”. Despite the availability of Natural heritage Trust grants for recharge projects, the State Government claimed in 2003 that only about 40,000ha (7%) of the Upper South East was sown to lucerne (Upper South East Dryland Salinity and Flood Management Program, 2003). The clear but incorrect message from the State Government was that because of the poor uptake of lucerne, drains were needed to take away excess groundwater.
However, the figure referred to 1993 figures (Johnson, 2002), and grossly understated the current area under lucerne. In fact, according to a 1999 survey (reported in DWLBC, 2002), of 186,461ha of the Upper South East, 72,327ha (39%) was identified as established to perennial pastures (the majority lucerne, but also including phalaris, cocksfoot, kikuyu, primrose and veldt grass). Furthermore, in 2002, 88,326ha of lucerne had been planted according to records, which was also considered to "grossly underestimate", possibly by as much as five-fold, the actual area planted (McEwan et al, 2002). This might provide another reason why watertables have been going down in the Upper South East over the past decade!
A minority of land in the Upper South East is sown to annual crops, which are inefficient users of available rainfall, and contribute significantly to groundwater recharge. Croppers in Western Australia use lucerne as a means of controlling rising groundwater on their land, by rotating with cereal crops in 3 and 4 year cycles. Lucerne also has the added benefit of fixing nitrogen for use by following crops. Alternatively, changed cropping practices that eliminate summer fallow significantly reduce recharge.
Management of Recharge on Saline Land
Any recharge that occurs on the flat can only be removed on the flat, and the best time to do this is during periods of highest rainfall, when salt concentrations are at their lowest. However, as water is removed from the soil's profile, salts become more concentrated reducing a plant's ability to extract water.
A useful combination of salt-tolerant pastures could be puccinellia, tall wheat grass, and possibly tall fescue, which between them have a long growing season, spanning from late autumn to the end of the following summer. Promotion of winter and spring growth through use of nitrogen-based fertilisers should also increase water use during the high winter recharge period, and thus impact more positively on the watertable. According to State Government reporting (McFarlane, 2005)(Upper South East Dryland Salinity and Flood Management Program, 2003), improved fertiliser and grazing management of salt-tolerant perennial pastures such as puccinellia and tall wheat grass, can increase stocking rates to up to double district averages.
On low to moderately salinised land with moderate waterlogging, options include tall wheat grass, tall fescue and balansa clover. As waterlogging increases in severity, puccinellia and salt water couch are more adapted. Pastures on saline flats have the added benefit of reducing evaporation and the build up of salts on the soil’s surface that occurs during dry, warm and windy weather. Saltbush and bluebush can also have a significant effect on groundwater levels, as well as providing palatable feed for stock. Saltbush is more tolerant of waterlogging(SGSL, 2000).
Puccinellia does not appear to use any groundwater, whereas the deeper rooted tall wheat grass uses groundwater late in the summer when the surface is dry (Walker et al, 1997). Puccinellia is not drought tolerant, and is likely to become stressed and possibly die, if the top-soil becomes too dry following drainage. Poor soil structure arising from sodicity, which in turn affects infiltration and potential waterlogging, should be treated before attempting a concerted program of saline soil recharge management.
At least 60% of salt-affected land in the Upper South East had been sown to salt-tolerant pastures (DWLBC, 2002).
Role of Drains
The obvious purpose of a drain is to remove water not required in one area and transport it to another. In the Upper South East, new drains are trapezoidal in cross section, with a base width typically of the order of 2 - 3 m, and sloping banks (batter) of relatively shallow gradient of about 2.5:1. Spoil from construction is placed on a spoil heap, which runs parallel to the drain about 5m from the drain edge. The area between the drain edge and spoil is known as a berm.
The theory of drain construction is well known, and there are many text books on the subject (eg Karvonen, 2002). Groundwater flows into a drain through the sides and base, at a rate which depends on the drain design and aquifer characteristics, eg aquifer thickness (typically 20 - 30m in the Upper South East) and hydraulic conductivity. Hydraulic conductivity can be estimated by analysis of the aquifer sub-soils and sediments, using borehole slug tests, or by pump tests.
Pump tests have been used to estimate hydraulic conductivity in the Upper South East (Fennell et al, 1992). The pump tests involved drilling bore holes to 8m deep. Soil and lithology reports indicate layers of different sediments, ranging from clay of low transmissivity, to sand of high transmissivity. Hydraulic conductivity is estimated using information on aquifer thickness, flow rates into the bore, and depth and extent of drawdown.
Estimates of hydraulic conductivity clearly represent an average that incorporates the different sediments in the aquifer. Notwithstanding this, the Program's drain construction manager recently acknowledged that the 1992 measurements were flawed, because they were conducted during a period of low watertables, when they were at a level similar to or lower than the base of the proposed deep drains. Transmissivity was thus over-estimated, which resulted in the performance of the drains also being over-estimated. A trial conducted in the Upper South East in 1999 indicated that drain performance had been over-estimated by five-fold. The author understands that the erroneous transmissivity data was still being used in studies in late 2005.
Many saline soils (particularly clayey soil) become sodic on drainage, resulting in an impermeable layer forming on the upper side of clayey sub-soils (see later section on sodicity). Rainfall is then prevented from entering the groundwater, and waterlogging is likely to increase over time. Furthermore, iron oxides accumulate in drain walls, where capillary waters rise and are oxidised, causing soil pores to clog (Fitzpatrick et al, 2004). Groundwater drain effectiveness will thus degrade over time. More recently, Kemp (pers comm, 2005) reported that dispersing clay from soil heaps filtered down through the soil to form a barrier to the movement of water to a trial drain.
To date, modelling of drain effectiveness has assumed homogeneous aquifer characteristics, although it is known that soils are layered, and tend to vary from limestone to clay from east to west across interdunal flats. Furthermore, modelling has not considered the impact of declining rainfall on watertable trends, nor the effect of revegetating dunes and ranges. The NRC recommended that further economic assessments be conducted, and that sensitivity analyses of variable factors should be reviewed (xiii, NRC, 1994c), neither of which were done.
Discovered recently are vertical sand columns which exist throughout the Upper South East in clayey sub-soils. These are exposed during drain construction (see photographs below), and comprise columns measuring from a few to several cm across. They provide a low conductivity drainage path for fresh water to drain to the the groundwater, bypassing the saline clays. They also appear to allow the upward flow of groundwater, when the head of the underlying groundwater is greater than the exposed end of the column. The existence of green-grey clay around the columns indicates that they have been continuously discharging since this section of drain was constructed 16 months before the photograph was taken, and suggest that groundwater flow into the base and sides of the drain is via a less transmissive route.
Thus, fresh water will also preferentially flow down the columns to any lowered watertables, resulting in salts remaining behind in the clayey sub-solis. Indeed, the action of fresh water between the sand-clay interface is likely to result in the clay dispersing and sealing the column, and locking soluble salts in the clayey sub-soils.
 Several sand columns visible (drain wall is ~2m high). Tops of columns identified by moist green-grey surface. |
 Section of a vertical sand column, surrounded by moist green-grey clay. |
The above photographs were taken at the end of winter 2005, which is the end of the Upper South East's peak rainfall period. This section of drain at the down-stream end of a 50km+ drain was designed to take peak flows up to about 260ML/day, but has barely reached levels any greater than shown. The water is flowing at about 7ML/day, and declining as the Upper South East heads towards the drier months.
The drains in the northern catchment of the Upper South East scheme are thus over-designed, and assumed a run-off factor of 1mm/day (multiply by catchment area to give daily flow rate). In a region where average annual rainfall is only 450mm, it is unlikely that a run-off of 1mm/day would ever have occurred, even if the catchment had been concreted over! Run-off figures of 0.1 - 0.3mm/day have been suggested by hydrologists as being more appropriate, which would have reduced capital construction costs for smaller drains by at least 50%, and reduced ongoing maintenance and operating costs. Studies have reported recharge figures of 40mm/year (Cox et al, 2005a) to 100mm/year (Mackenzie et al, 1992) on the flats, which equate well with the suggested run-off figures.
Drains only treat the symptoms of dryland salinity, and not the cause! Drains alone will not improve productivity, which will only occur following a program of revegetation. Drain construction is thus only one investment in a number that must be made to increase productivity on saline soils.
Water is also a valuable resource that logically should only be allowed to drain uncontrolled from the landscape as a last resort. While drains might help to relieve inundation when it occurs, during drought conditions water remaining in the soil becomes critical. Predictions of declining rainfall in the Upper South East make it more important to make efficient use of the limited rainfall, rather than draining it away to become unavailable for agricultural production. There is expected to be a 10% decrease in summer and autumn productivity 1km either side of a 2m drain (Schrale et al, 1987).
Drains also remove water from, potentially, the most fertile land in the Upper South East. While a revegetation program could have stopped recharge over half of the Upper South East, with a significant increase in net profits, less than 4% of the region will benefit from drainage, with marginal, if any, increase in economic returns. While some adverse environmental effects arising from drainage are common to more effective recharge management, a sounder economic base would have rendered their remediation easier to achieve by farmers.
Soil salinity profiles will vary according to distance from a drain. Close to a drain, perhaps up to 100m in duplex soils, non-sodic top- and sub-soils will be leached of salts over a period of several years, when they could then be sown to traditional deep-rooted and more productive perennial pastures. However, measurements close to a 3m deep drain in an area of duplex soils showed that salinity levels in the top-soils (0 - 30cm) were in fact similar two years after construction at distances up to 2km away, and surprisingly were increasingly higher (and excessively saline) closer to the drain in the clayey-sub-soils (Fitzpatrick et al, 2004)! High salinity levels in sub-soils restrict options to salt-tolerant perennial pastures. However, the productive puccinellia, which requires moist top-soils, is dying (Strugnall, 2004a).
Once healthy puccinellia dying up to 150m from a drain
Clayey sub-soils are expected to remain saline for years, possibly forever, because saline groundwater remains within capillary rise distance. The poorer transmissivity of clayey soils also means that salts leach very slowly. Shallower drains in the up-stream sections of the Northern Catchment of the Upper South East are not expected to have a significant effect on soil salinity for many years, if ever, unless groundwater levels are lowered to below about 2m.
Between 100m and 500m from a drain, sub-soils are likely to remain saline forever, and so only salt-tolerant perennial pastures should be sown. While non-saline top-soils could support traditional annual pastures (recommended by Upper South East program staff!), trials in the Upper South East have shown them to have very low productivity on duplex soils, with yields less than 3% of phalaris and tall fescue (Strugnall, 2004a). Beyond 500m, top-soils are likely to remain saline, restricting pasture choices to highly salt-tolerant species, such as puccinellia and tall wheat grass.
Such variation in soil types leads to significant management challenges, in addition to challenges created by the existence of a drain with limited crossing points.
The draft Upper South East Dryland Salinity and Flood Management Plan was published in 1993 for comment. Of 188 public and government submissions returned, 29% supported the Plan, 36% had concerns with the Plan, and 35% had serious concerns with or opposed the Plan (NRC, 1994b). Most concerns were associated with the need for a groundwater drainage network, which had not been adequately justified.
An independent review of the Plan (Semeniuk, 1993) concluded that analysis of hydrological trends and the justification for drains were unconvincing (now known to be valid criticism). A second review, including of the independent review, concluded that many aspects of hydrological data and modelling were deficient, and that "only surface drainage is necessary to sustain salt land agronomy" (Coffey MPW in Appendix 3, NRC, 1994a)! These conclusions did nothing to change the Program's preferred option of major groundwater drains in the southern and central catchments and major surface water drains in the northern catchment.
In 2003, Program staff sought "written acceptance of the proposed [drainage] works" in the northern catchment from affected landholders, but "lack of landowner support was highlighted by the fact that only 17 landowner agreement letters were returned from 35 sent, primarily because the depth of drain was perceived to be detrimental to agricultural production in the upper sections" (Upper South East Program Board, 2003). Landholders in the lower sections were not consulted on drain depth. After revising the drain depth in the upper sections, construction went ahead without seeking landowner acceptance!
The current and vociferous adverse reaction from landholders to the drainage scheme should thus not have come as a surprise to Program staff. Very recent State Government statements that opposition to the drainage scheme (in particular groundwater drains) has only recently become apparent (Cahalan et al, 2005), and other statements that there was unanimous or broad support for it, are clearly wrong.
|
|
