Background

The efficient and effective control of weeds, be it woody weeds, herbaceous plants or agricultural pastures, has always been the difference between a failed or viable Radiata Pine (Pinus radiata) plantation. There is no universal definition of a weed. A weed can simply be described as a plant out of place. However, more specifically “a weed is a plant which has, or has the potential to have, a detrimental effect on economic, social or conservation values” (NWS 1997). Weed management is therefore an essential and integral part of the sustainable management of natural resources (including plantation forests) and the environment, and importantly “requires an integrated, multi-disciplinary approach” (NWS 1997). Given the national importance of weed management, Commonwealth, State and Territory Ministers responsible for agriculture, forestry and the environment agreed to develop a national strategy to strengthen the cost efficiency and effectiveness of weed management in Australia. The Australian Weeds Committee (AWC) was given the lead role of drafting the National Weeds Strategy (NWS) which was published in 1997 (see footnote). Many of the fundamental principles of weed management detailed in the NWS have long been recognized by softwood planation managers in Victoria. Indeed, examples of an integrated and multi-disciplinary approach to managing weeds were introduced some ten years earlier than the release of the Strategy. Importantly, forest managers have long been required to meet the principles of the NWS in order to comply with community expectations. Herbicide use in softwood plantations has always been a controversial practice in Victoria, and high standards that largely met NWS guidelines were entrenched as far back as the mid to late 1970’s. The relevance to this article of introducing the NWS is that weed management strategies progressively identified and developed by the FCV in particular can be judged favourably against the context and principles of modern day weed management practices. It also justifies the substantial and on-going funding of weed management research by FCV.

Introduction

After an extensive screening of softwood species suitable for the development of a commercially viable industry, Radiata Pine (Pinus radiata) was identified as the most suitable species able to tolerate a wide range of rainfed site types in Victoria. Initially, new plantations were established mainly on former native forest land that had been highly disturbed by mining, including dredging, but this rapidly changed to extensive plantings on sites converted from native forest to Radiata Pine plantations. On most of these sites, the native Silver Wattle (Acacia dealbata) and eucalypt regrowth (a range of Eucalyptus species) was prolific and posed a significant threat to the survival and early growth of the plantations. Manual slashing was therefore adopted as a weed control measure. Repeated slashings were generally required to suppress the resprouting of wattle in particular. This was labour intensive and often only partially successful. As the planation estate began to expand rapidly in the 1960’s, a small research program was initiated in 1965 into chemical control of woody weeds. Until the early 1960’s, native forests were felled and broadcast burnt prior to the planting of Radiata Pine, with woody weeds controlled by slashing. But by the late 1960’s heaping of debris and windrow burning was then introduced allowing many sites to be ploughed before planting to reduce the eucalypt component of woody weed recolonization. It also temporarily controlled many herbaceous weeds. This article focuses on weed management research in Radiata Pine plantations conducted primarily by the FCV and DCFL between 1965 and 1990. Herbicide use in nurseries is beyond the scope of this article. It is important to appreciate that woody weeds in particular also present problems in the establishment phase of second and third rotations on former native forest sites cleared for softwood plantations. Equally important are the implications of a switch to former agricultural land following the cessation of clearing of native forests for softwood plantations. This shift created a new weed management challenge due to competition from both unimproved and improved pastures. By 1978 about a third of the annual planting program by FCV was being undertaken on purchased farmland. In an early study on the impacts of grass competition on Radiata Pine, Flinn (1978) conducted a study in a high rainfall zone of the Strzelecki Ranges. He found that a strip application of Vorox herbicide (see Footnote) resulted in a 3-4-fold increase in mean tree volume three years after application.

Early Research into Control of Woody Weeds

As far as the author can determine, the first formal experimental trials on controlling Silver Wattle and eucalypt regrowth during the establishment phase of Radiata Pine plantations on cleared native forest sites were conducted between 1965 and 1967. Control methods investigated included:

• Full scale trials conducted in 1965 to investigate the effectiveness of 2,4,5-T (2,4,5-trichlorophenoxyacetic acid) and Tordon to control woody weeds in new plantations (Forestry Research Report No 9 1967)
• Further work was undertaken in 1967 to study the effectiveness of herbicides (2,4,5-T and Tordon) applied monthly throughout the year (Forestry Research Report No 9 1967)
• Comparison of slashing regrowth at age two years after planting, slashing at two years after planting and then annually, and slashing 1m strips two years after planting, was also investigated in 1967 (Forestry Research Report No 8 1967). It was conducted in three Forest Districts. This trial indicated that manual slashing was still the predominant method used to control woody weeds until the mid to late 1960’s
• The first stem injection studies of regrowth using Tordon were undertaken in 1966 (Forestry Research Report No 9 1967); and
• Also reported were the results of a helicopter application of 2,4,5-T applied to ten acres of regrowth in both 2- and 3-year-old plantations in 1966 (Forestry Research Report No 9 1967)

Quantifying the Impacts of Woody Weeds on Pine Growth

It became abundantly clear from this early research that slashing woody regrowth was not cost effective and was not always successful due to the resprouting problem. However, there was little information available to assess the economic impact of uncontrolled or partially controlled regrowth. An MSc study was therefore supported by FCV to quantify the economic impacts of woody weeds in particular on the growth of Radiata Pine. This landmark study was completed in 1970 (Jack 1970). He established plots at several sites with varying levels of densities of Silver Wattle regrowth and found that there was a strong correlation between early growth rates and mortality of Radiata Pine and the basal area of the Silver Wattle regrowth. Effectively, the Silver Wattle density determined the survival and early growth rates of young plantations. Control of over-topping Silver Wattle in particular was therefore essential to avoid a failed plantation or a plantation with markedly reduced growth rates and ultimate productivity. He further found that the standing volume of Radiata Pine was related not only to stand age and site quality but also to the density of species that recolonized sites following clearing of native forest. As an example, he recorded an 80% reduction in standing volume of a mature Radiata Pine plantation growing in competition with Mountain Ash (E. regnans). The economic case for controlling woody weeds as early as possible after planting was clearly established by Jack (1970), a fundamental principle or requirement of the NWS.

Adoption of Aerial Spraying with 2,4,5-T

Flinn and Hopmans (1977) outline the uses of 2,4,5-T in Victorian softwood plantations. They were:

• A single aerial application of 2,4,5-T (as the butyl ester) to control dense Silver Wattle regrowth in young pine plantations. The application rate was 1.1 kg per ha of the active ingredient (a.i.) in the carrier dieseline
• Basal bark spraying where the density of wattle regrowth was low and the plants small. The basal 30 cm of each stem is hand sprayed with 2,4,5-T (butyl ester) using dieseline as the carrier. For large plants, tree injection with Tordon 105 (a mixture of picloram and 2,4 5-T, both as triethylamine salts) was used. The Tordon was applied to cuts made in the stems; and
• Eucalypt regrowth was controlled by basal bark spraying with Tordon 255 (contains double the amount of picloram and 2,4,5-T than Tordon 105). For larger trees, tree injections with Tordon 105 were used.

In 1968, a large-scale aerial application of 2,4,5-T was made to a young plantation in northeast Victoria using a fixed wing aircraft (Beaver). The carrier was dieseline. Individual flight paths were marked to ensure even coverage, and Kromecote cards were strategically placed across selected spray swathes to record the droplet size and spectrum which dictates both spray drift and coverage of the target plant. The spread factor of the spray on the Kromecote medium was determined by laboratory studies (Flinn 1968) so that droplet sizes could be accurately determined. He used a state-of-the-art vibrating needle that could be calibrated to produce a range of droplet sizes. For each droplet size, a magnesium oxide coated glass slide was used to determine the precise diameter of the droplet which was also sprayed onto Kromecote cards to then determine the spread factor. Operationally, a dye was added to the spray mixture to assist in droplet measurements.

The aerial spraying was followed with basal bark spraying and stem injection of eucalypts using Tordon. Within months of the spraying, the Silver Wattle regrowth was completely defoliated and resprouting did not occur. However, a significant portion of the young pine trees developed moderate to severe foliage chlorosis and necrosis, and some apical damage. Stem deformity also varied from slight to severe. The needle scorch was attributed to the carrier (dieseline) whilst the stem deformity was due to the hormonal effect of 2,4,5-T. Plots established to monitor the young pine trees showed that they had recovered within a year of spraying.

This operation initiated a new era in the management of woody weeds, with a rapid move away from slashing to aerial application of 2,4,5-T. Over 16,000 hectares were successfully sprayed between 1968 and 1977 when its use was terminated by FCV due to the so called “2,4,5-T controversy” (Flinn and Hopmans 1977).

Flinn and Hopmans (1977) provide a detailed account of the 2,4,5-T formulations and application methods used by FCV along with the chemistry, mode of action, toxicity and degradation of applied 2,4,5-T. They note that whilst 2,4,5-T as used operationally “does not constitute a hazard to humans or environment providing application prescriptions are enforced”, the 2,4,5-T must also comply with standards presented in the relevant Pesticides Act. However, Flinn and Hopmans (1977) stress that 2,4,5-T is contaminated with 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) which is an extremely toxic compound formed during the production of 2,4,5-T. It is one of the most teratogenic compounds known to mankind. TCDD has an exceedingly low LD50, an indication of an extremely toxic compound. TCDD levels can be minimized in the production of 2,4,5-T (at a cost), but not entirely eliminated. Intense public pressure opposing the use of 2,4,5-T with its contaminant resulted in the FCV imposing a temporary suspension on the use of the herbicide in Victorian softwood plantations in 1978 (Flinn and Minko 1980). This resulted in an urgent need to identify alternative herbicides to replace 2,4,5-T. Indeed, such research was already underway with an expectation of the withdrawal of 2,4,5-T.

The loss of 2,4,5 -T was in one sense a blessing, because it stimulated research to identify an array of herbicides to efficiently and effectively control a wide spectrum of weed infestations using herbicides developed for the differing weed situations. This is a requirement in the NWS (See Fagg and Flinn 1983 for further discussion on a more strategic and integrated approach to weed management in the absence of 2,4,5-T).

Screening of Alternative Herbicides to 2,4,5-T

An intensive research program was already underway to identify replacement herbicides for 2,4,5-T. The first study on screening potential herbicides was reported by Flinn et al (1980a). This study compared three herbicides, each at two application rates, with the standard 2,4,5-T prescription (treated control). The herbicides were hexazinone, triclopyr and 3,6-dichloro-picolinic acid. They were applied (broadcast sprays) in late winter 1978 to plots established in 2-year-old Radiata Pine plantings infested by Silver Wattle and a number of eucalypt species.

They found that both hexazinone and 3,6-dichloropicolinic acid at the higher rates were of similar effectiveness to 2,4,5-T. Hexazinone also had a significant impact on the eucalypts and herbaceous weeds. Importantly, the two relatively new herbicides had minimum phytotoxicity to the young Radiata Pine trees. The authors recommended further evaluation of these two herbicides, particularly given that 2,4,5-T could be permanently withdrawn from forestry use. In 1978, an assessment of aerially applied hexazinone (3 kg per ha) was undertaken on a ten hectare site at Myrtleford where a dense population of Silver Wattle and eucalypts was competing with the 2 -year- old Radiata Pine crop plants. Plots scattered across the treated area showed that 41% of the target woody weed population was dead and a further 12-13% defoliated 11 months after spraying (Flinn et al 1980b). The authors speculated that the below average rainfall may have reduced leaching of the herbicide into the Radiata Pine root zone and reduced its potential effectiveness.

Given the clear potential of hexazinone to control a broad spectrum of herbaceous plants, grasses and woody weeds, this herbicide was aerially applied in 1982 to a study area comprising a young plantation with the target species being eucalypts, A. verticillata and A. mearnsii near Heywood (Geary and Flinn 1985). The effectiveness of hexazinone on these two wattle species was completely unknown. The spraying (with water and a petroleum oil known as Ulvapron as the carrier-see later) was conducted in December during a drought. Plots were established over the study area and individual plants tagged to enable accurate tracking of spraying effects. Plots were assessed 16 months after spraying, with over 50% mortality of both wattles. The eucalypts were not controlled (as expected). There was no pine mortality. The results were considered “satisfactory” from an operational perspective. It may have been more effective in the absence of a drought (limiting uptake of the active ingredient), if the spraying was conducted when the target species were smaller, and a higher rate of hexazinone used.

Flinn and Minko (1981) undertook two experiments to further screen potential herbicides for woody weed control. In Experiment 1, they tested liquid and granular hexazinone over a range of rates, triclopyr, MCPA (2 formulations), 3.6-dicloropicolinic acid (2 formulations) and liquid hexazinone mixed with an emulsifiable petroleum oil. In Experiment 2, hexazinone as a liquid and as a granule were tested at three rates (broadcast) along with spot application and line drench application. The overall findings indicated that liquid hexazinone is more effective than granules on both Silver Wattle and eucalypts. They attributed this to dual uptake of the herbicide by foliage and roots. This was reinforced by the inclusion of a petroleum oil in the carrier (water) which resulted in total control of the woody weeds (see footnote). This was important given that hexazinone was more expensive than 2,4,5-T. A concern with hexazinone however was that it had a broad spectrum of activity, which is undesirable on soils with a propensity to erode when exposed.

Based on these findings, the FCV aerially sprayed around 200 ha in pine plantings in three Forest Districts using 2kg/ ha of hexazinone aerially sprayed with 60 l/ ha of water including a petroleum oil. Complete defoliation of young Silver Wattle and total leaf scorch of eucalypts was observed within four weeks of spraying at each locality. There were only minimal effects in the young Radiata Pine crop plants. This was an important step forward in identifying possible alternatives to 2,4,5-T.

A timely paper was written in 1983 that comprehensively evaluated Velpar and Lontrel (3.6-dicloropicolinic acid) for control of eucalypts and Silver Wattle in young pine plantations (Fagg and Flinn 1983). The study reports on three field experiments:

In Experiment 1, they compared the efficacy of the two herbicides applied in the presence and absence of a petroleum oil. It was found that after 20 months, Velpar plus oil killed 87% and 97% of the eucalypts and Silver Wattle populations respectively, whilst Lontrel plus oil resulted in a 20% mortality of eucalypts and 80% mortality of Silver Wattle. The addition of the petroleum oil (Ulvapron) substantially improved the efficacy of Lontrel on Silver Wattle. The authors concluded that the use of Lontrel and Velpar should be dictated largely by the species composition of the woody weed population as now required by Certification Standards and the NWS. Importantly, this experiment clearly demonstrated that viable alternatives to 2,4,5-T were now identified.

In Experiment 2, Lontrel in a spray volume comprising 33%v/v petroleum oil was aerially applied to a small plantation of pine. Spray coverage was monitored using Kromecote cards. Within 11 months of treatment, 45.4% of the Silver Wattle population that averaged 11,500 stems per hectare was killed, whilst a further 23.7% was more than 75% defoliated. This finding supported results from experiment 1 in terms of Silver Wattle control. They concluded that the result was operationally satisfactory, though the treatment cost was significantly higher than a routine 2,4,5-T treatment. Nonetheless, the experiment again confirmed Lontrel as a replacement herbicide for 2,4,5-T.

The 3^rd Experiment focused on the control of eucalypt coppice using the liquid and granular formulations of Velpar, with the quantity of herbicide applied to the soil being dictated by the height of coppice clumps which varied from 30 to 300 cm. All treatments resulted in a mortality in excess of 90% within 6 months of application. The treated control was stem injected with picloram /2,4,5-T and or slashing/grubbing. Once again, the experiment conclusively showed that soil-applied Velpar will effectively control eucalypt coppice as a substitute for the treatment containing 2.4.5-T. The best treatment was 1g liquid active ingredient of Velpar per meter height of the clump. The authors noted however that caution was needed in extrapolating results from the experiment because edaphic and climatic factors along with eucalypt species composition may influence the level of control.

The three experiments undertaken by Fagg and Flinn (1983) clearly demonstrated that control of Silver Wattle through the application of Lontrel plus a petroleum oil in the spray mixture at a rate of around 2kg/ha of active ingredient will satisfactorily control dense Silver Wattle regrowth in young pine plantations without any significant long-term damage to the crop plants. This treatment is a viable alternative to the previous widespread use of 2,4,5-T, albeit at a significant increase in the cost of the treatment. The experiments also identified Velpar as a very effective treatment for eucalypt regrowth, both when aerially or ground applied as dictated by the extent of eucalypt regrowth. The study was also the first record of the evaluation of the effects of an added petroleum oil (Ulvapron) on the efficacy of Lontrel (see Footnote). Experiment 1 clearly showed that the addition of Ulvapron substantially increased the efficacy of Lontrel in controlling Silver Wattle, thereby resulting in a significant cost saving. It is noteworthy that the cost of the petroleum oil is only a fraction of that of Lontrel.

Control of Pine Wildings

Radiata Pine wildings are encountered in two situations. Firstly, dense natural regeneration of Radiata Pine often develops when a seed-bearing stand is killed by wildfire. With few exceptions, this regeneration is not managed to produce the next crop for a variety of reasons including sacrificing productivity gains from on-going tree breeding programs. In 1979, a wildfire killed around 300 ha of planation in southwest Victoria, resulting in a dense cover of pine wildings with a mean density of around 250,000 plants per hectare (Flinn et al 1982). Plots were established to assess the effectiveness of four herbicides, each applied at two rates. The treatments were two formulations of Dicamba, one formulation of Trichloroacetate, and Paraquat plus Diquat. Total control was achieved with two of the herbicides applied in spring, those being Dicamba as the dimethylamine salt and Paraquat plus Diquat. An autumn spraying did not provide adequate control. The plots were split into sub plots to examine the minimum safe period before Radiata Pine seedlings could be safely planted following the herbicide applications. It was found that plots treated with herbicide in the first spring after the fire can be safely planted in the following winter. This study complemented the work of Minko and Stephens (1986) who investigated the effectiveness of a low application rate of Krenite to control unwanted natural pine regeneration along with competing vegetation in a pine plantation killed by wildfire near Bright. The treatment was partially successful by reducing the number of pine seedlings from around 15,000 to 4,000 stems/ha.

The second situation is where pine wildings are not controlled when pine logging slash remains unburnt after clearfelling. On infertile sites in particular, the practice of slash burning ceased in order to conserve nutrients, organic matter and soil moisture, thereby preventing a decline in productivity of subsequent rotations (Flinn et al 1979a). This sometimes leads to a problem with advanced regeneration in multi-thinned stands and germination post clearfelling. Between 1982 and 1984, the following alternative methods for dealing with the problem were investigated by Geary et al (1987) using a large series of plots:

• Pre clearfelling (3 treatments) Low intensity prescribed fire, rotary slashing plus low intensity fire, and delimbing flail.
• Post clearfelling (8 treatments) Single drum chopper-roller, Duplex chopper-roller, rotary slasher (RS), RS plus hand-cleaning, RS plus brushcutters, RS plus flame thrower, RS plus delimbing flail, and RS plus six formulations of amitrole.

It was found that a non-herbicide treatment comprising slashing wildings prior to clearfelling followed by a low intensity burn was the most cost effective and operationally feasible treatment of those tested. This treatment, however, generally requires some hand cleaning to eliminate surviving wildings and new germination following clearfelling. Amitrole was only moderately effective.

Evaluation of Herbicides for Stem Injection of Woody Weeds

Restrictions on the use of 2,4,5-T stimulated urgent research to identify alternatives to stem injection with Tordon which contains 2,4,5-T and picloram. Manual stem injection using a “Tordon spear” is particularly suited to sparse populations of eucalypt and wattle regrowth and residual advanced eucalypt regrowth following aerial application of 2,4,5-T or any other broadscale herbicide that does not control eucalypts when treating dense Silver Wattle regrowth. An advantage of this technique is that chemical usage in the environment is minimized, a fundamental NWS principle.

Minko and Flinn (1982) compared the efficacy of the following five herbicides:

• Velpar L (Hexazinone)
• Banvel 200 (Dicamba, amine formulation)
• Garlon M/3724 (Triclopyr, amine formulation)
• Garlon 480 Herbicide (Triclopyr, ester formulation)
• Thistle Killem (MCPA, amine formulation)

These herbicides were applied by stem injections in either spring or winter to plots comprising twenty adjacent stems of eucalypts and Silver Wattle of around 7m in height. A Tordon spear was used to dispense 1ml of each herbicide at 7 cm intervals around the base of each stem. Three concentrations of each herbicide were tested on a total of 5040 tagged stems. Tordon was not included as a treated control.

Results showed that injection in spring of all five undiluted herbicides resulted in high mortality of the target species. Similar results were found for diluted Velpar L, Garlon M/3724 and Garlon 480 Herbicide when injected in spring. The authors concluded that all five herbicides used in the study could be used operationally to effectively control Silver Wattle and/or eucalypts within the size range investigated if attention is given to season of injection and concentration of the product. They also noted that the choice of herbicide is likely to be dictated more by availability of products, season of injection, and toxicology considerations. The study met the objectives of identifying alternative herbicides to Tordon for stem injection of advanced competing woody weeds.

In 1984, Tordon Timber Control Herbicide and Roundup were registered in Victoria for control of unwanted eucalypts by stem injection. This prompted a study by Fagg et al (1990) that compared the efficacy of these two herbicides together with Velpar for control of eucalypt regrowth by stem injection. It was a far more comprehensive study than that of Minko and Flinn (1982). Fagg et al (1990) tested the three herbicides on a range of eucalypt species and sizes. They showed that all eucalypt species on the study area could be effectively controlled using Roundup or Tordon Timber Control Herbicide at the then recommended label rate and Velpar at half the label rate. Testing was conducted in spring, autumn and winter on upper (and harsher) and lower slopes. An important finding was that eucalypt regrowth on upper slopes is more difficult to control than that on lower slopes. Season of application on upper slopes was important, where spring treatments should be avoided.

Herbaceous Weed Control

Herbaceous weeds comprising a very wide range of native perennial plants are common to most pine plantations established on cleared native forest sites. Their density varies significantly in response to soil and climatic factors. Whilst such weed populations usually do not result in significant early pine mortality, their competition for soil water, light and nutrients is known to impact on early growth rates of the crop plant, with up to two years growth being irretrievably lost on sandy soils in low rainfall zones. A number of studies have been conducted on the control of herbaceous weeds. Minko (1974) appears to have prepared the first formal report on herbaceous weed control using herbicides during the establishment phase of pine plantations on cleared native forest sites. He applied Vorox AA (amitrole plus atrazine) plus Weedazol TL (amitrole) to control a very wide range of mainly herbaceous weeds that recolonized his experimental site. He found that the strip treatment of these two active ingredients on moist soil conditions improved survival of the pine outplants and increased their early growth. Minko (1974) tabulated the long list of species that recolonized his study area, something that is often overlooked by weed scientists. It allowed results to be extrapolated to other sites with similar floristics.

Minko (1978) conducted a further study on the control of herbaceous weeds at a study area near Myrtleford. He applied Roundup before and after planting pine seedlings on a former native forest site. Roundup killed most seedlings in the post planting treatment. But a pre-pant application four weeks before planting effectively controlled regenerating vegetation for a considerable period. This study was unique in that the author identified the 64 native species that recolonized the study area, a painstaking task that few herbicide studies have undertaken. As such, it was a landmark study that demonstrated the methodology of weed research on herbaceous species (excluding woody weeds) invading cleared sites. Poa species were found to be the main competing plants, and Roundup effectively controlled them during the establishment phase.

Fagg et al (1988) reported on a study in southwest Victoria that investigated spot spraying Velpar and amitrole/atrazine to first (1R) and second (2R) rotation pine plantations on sandy soils. The weed spectra on the experimental sites were typical of those on 1R and 2R sites in the locality. Hexazinone was applied at 3 rates whilst only a1part amitrole plus 4 parts atrazine was used for this treatment. The timing of application (soon after planting and around six months post planting) were trialed. After three years, spot spraying in September after a July planting on both sites resulted in the largest and very substantial growth response. They note that spot spraying is significantly cheaper than broadcast or strip spraying and has a particular application on 2R sites where slash is retained.

Hopmans and Tomkins (1992) studied the interaction between fertilizer additions and weed control in pine plantations, an important consideration in pine plantation management. They undertook the study at Neerim East comprising a factorial design with 3 rates of superphosphate (nil, 180 g/tree, and 700 kg/ha) and 4 levels of Velpar (nil, spot, broadcast granules and broadcast liquid). At three years of age, survival of outplants was less than desirable, particularly on unfertilized trees treated with Velpar. Circumstantial evidence indicated that this was due mainly to being browsed by animals. The weed spectrum comprised a mixture of herbaceous and woody weeds. The Neerim East sites were known to be severely P deficient, and they found that the spot application of P followed by a broadcast application at age five years increased basal area around ten-fold at age ten years. The combination of the two fertilizer additions plus complete weed control resulted in a near twenty- fold increase in basal area over the same period. This study demonstrated that herbicide addition should be avoided on sites where nutrient availability is the most limiting factor in the establishment phase of softwood plantations. Limiting growth factors including nutrient deficiencies must therefore be addressed before the application of weed control measures. This requirement is consistent with NWS guidelines.

Flinn et al (1979b) measured the above-ground biomass (herbaceous and woody weeds) at four sites at Neerim East. They found that this ranged from 3,720 to 6,570 kg/ha three years after planting. Variable responses to applied P at their four sites was attributed to differential competition for soil moisture across the sites. Based on the work of Jack (1970), a weed biomass of over 6,500 kg/ha during the establishment phase would have impacted significantly on early growth.

Minko (1978) undertook an important study into the longer-term impacts of applying herbicides on native plant diversity and density. This is an important ecological factor of herbicide use in young plantations on former native forest sites. He studied the impact of Vorox AA and found that by the end of the second year after herbicide application, plant diversity varied little between control and treated plots, though the mass of recolonizing plants was significantly less three years after treatment. He concluded that Vorox AA at the routine operational rate had no long-term effect on plant species balance.

Control of Native and Improved Pastures

This article has primarily focused on control of woody, and to a lesser extent, herbaceous weed populations up to around 1990. The control of native and introduced grasses, including improved pastures on cleared land purchased for establishment of pine plantations following cessation of clearing native forests for pine plantations, was the subject of a significant research program. This was conducted in close collaboration with the Woods and Forests Department of South Australia. The formation of VPC and subsequent privatization (HVP) of the Victorian Government’s pine estate, along with the formation of CFTT (a business unit in DCFL) meant that the reports on many studies commissioned by VPC and HVP were commercial-in-confidence.

Stream Water Monitoring and Spray Drift

Apart from measures taken to minimize spray drift, including the addition of a petroleum oil to reduce droplet evaporation and herbicide efficacy when water is used as the carrier, there are two other measures that have been investigated, Firstly, where the option is available, the use of a granular rather than a liquid formulation is an obvious measure that can be taken if the efficacy of the granular formulation is as effective as the liquid alternative. This consideration is particularly relevant to the use of Velpar. The two formulations of Velpar have been intensively evaluated in the studies report above, including the experiments of Fagg and Flinn (1983). Important considerations in choosing a formulation include the efficacy of the granular and liquid formulations (especially where an adjuvant is used where water is the carrier) along with topography, and soil and climatic factors (e.g. leaching of the active ingredient into the root zones of the target weeds).

Another option is the use of a tractor mounted rope-wick applicator. This alternative application which effectively eliminates spray drift was investigated by Fagg et al (1987) for control of Silver Wattle using Lontrel and Roundup with and without the petroleum oil (Ulvapron). The evaluation was conducted on a study area of mild topography (5-14 degrees) where Silver Wattle populations varied from 2,500 to 4,000 stems /ha. Lontrel at the time of the study was the recommended treatment for pure Silver Wattle competition. Fagg et al (1987) investigated eleven treatments replicated three times as follows:

• Lontrel/water and Lontrel/water/Ulvapron using 1 and 2 passes
• Roundup/water and Roundup/water/Ulvapron at 1 and 2 passes with a thickener added to the Roundup to limit high flow rates; and
• Water only

An average plot area of 1350 square meters was used. Three transects were established on each plot before treatment, with 15 wattles covering the full range of Silver Wattle stems being tagged on each transect. The results were very encouraging. Using the proportion of tagged Silver Wattle stems in the 76-100% dead or defoliated eight months post treatment as a guide, Lontrel resulted in significantly better control than Roundup for either one or two passes. Roundup resulted in minor tip damage to pines while Lontrel as expected did not cause any damage to Radiata Pine. The authors concluded that the rope-wick treatment with Lontrel at a relatively low rate was effective and durable, and the total cost of the rope-wick application followed by manual stem injection on areas unsuited to the treatment was comparable with the aerial application of Lontrel for medium density Silver Wattle populations. Improvements such as front-mounting rather than rear-mounting rope-wick, an increase in rope angle, and the use of a large 4-wheel drive tractor were suggested improvements that would most likely lead to even better outcomes. From a spray drift viewpoint, the rope-wick applicator offers a cost-effective alternative to aerial spraying for control of relatively low-density wattle populations on sites with mild topography free of substantial ground debris. Furthermore, the treatment should be considered where spray drift needs to be completely avoided or in small areas that do not justify aerial treatment.

Contamination of stream water flowing from herbicide-treated catchments soon became a sensitive issue following the first broadscale aerial application of 2,4,5-T. Whilst the FCV developed stringent specifications to protect riparian vegetation and to minimize spray drift through application being confined to calm conditions and using spray systems that delivered a droplet spectrum void of very small droplets (McKimm 1973). He demonstrated that significant spray drift occurred using conventional agricultural spray systems, even under relatively calm conditions. McKimm (1973) tested solid stream nozzles which were superior in terms of droplet spectra, droplet density and the level of spray drift. A balance must, however, be met between large droplets to minimize spray drift and smaller droplets to ensure satisfactory coverage and increased herbicide efficacy.

Whilst all possible precautions were taken during aerial spraying of 2,4,5-T, community concern about stream contamination became intense. This stimulated a number of stream water sampling programs conducted before, during and well after spraying. Prior to this sampling program, some background checks were made to determine 2,4,5-T sorption by the two most common containers used for stream water sampling to monitor chemical contaminants (but not herbicides). The most common containers were glass and low density polyethylene (LDPE) plastic. Hopmans (1975) undertook laboratory studies which showed that the sorption of the butyl ester formulation of 2,4,5-T may reach as much as 90% of the initial amount of herbicide within a 1-hour contact period. In contrast, sorption by Pyrex glass appeared to be relatively small and independent of contact time. The work clearly demonstrated that polyethylene bottles are not suitable for the monitoring of 2.4,5-T residue levels in stream water. The study also demonstrated the extreme lengths taken to ensure that the community could have confidence in the results of herbicide residue studies, the first of which was conducted in 1977 at Scrubby Creek, Carboor (McKimm and Hopmans 1977). This monitoring program was in response to an EPA request following concern about contamination of Scrubby Creek and dams on adjacent farmland properties. The treated area was 162 ha which was dissected by gullies which only flowed after heavy rain. Five strategic sampling points were identified, three in the adjacent landholder’s dams, one at a culvert which consistently carried water from the sprayed area during the monitoring period, and one in Scrubby Creek upstream from the first domestic takeoff point. Sampling commenced the day before spraying and continued for six days after the herbicide treatment. Samples were collected twice daily except for the Scrubby Creek sampling point where a composite sample was collected hourly. Samples were analysed in the laboratory using a technique with a detection limit of 0.3 ug/l of the active ingredient. Some samples were also provided to a government laboratory for independent analysis.

The only samples where 2,4,5-T was detected (4.2 ug/l) were from those collected from the culvert pipe draining the main gully of the sprayed area. This concentration was well below the upper limit of 20 ug/l for 2,4,5-T in drinking water set by the National Health and Medical Research Council (NHMRC). This result demonstrated that the prescriptions adopted during the spraying were effective in minimizing 2,4,5-T residues in stream water to concentrations well below NHMRC guidelines.

McKimm and Hopmans (1978) used a similar but more intensive sampling regime for a monitoring study at Narbethong. A stream immediately below a 108-ha sprayed area was sampled twice daily for five days before spraying, and at 12 minute intervals for a period of 13 days during and following the 2,4,5-T spraying. The herbicide was only detected in samples collected on the 7th and 8th days post-spraying during a period of heavy rain. The maximum concentration found was 10 ug/l, still comfortably below the limit set by the NHMRC for potable water.

As detailed above, following the withdrawal of 2,4,5 -T from use in Victorian plantations, new herbicide treatments were identified to treat woody weeds in young plantations. One of these herbicides was Velpar. Whilst Velpar is considered to have a relatively low toxicity to humans, fish and birds (Fagg et al 1982) and has a relatively favorable half-life depending on soil type, a study was required to establish the levels of Velpar in stream water following its aerial application to young pine plantations under operational conditions. Fagg et al (1982) undertook such a study at Yarram. A sampling program similar to that used in previous 2,4,5-T stream water residue programs was adopted. Samples were analysed by the Institute of Drug Technology Ltd to determine the concentrations of hexazinone and 4 of its metabolites in the stream water samples. Neither hexazinone nor 4 of its metabolites were detected in any of the samples at detection limits of 0.05 to 0.07 ppm for hexazinone and metabolite A, and metabolites B, D and E respectively. This compares with the NHMRC maximum concentration of 0.6 ppm in drinking water. The authors noted that given 91.8 mm of rainfall was measured during the sampling period, it was unlikely that the herbicide would have been detected due to its rapid uptake by competing vegetation and its adsorption by soil colloids.

A follow up study was undertaken in 1983 by Leitch and Flinn (1983) to monitor Velpar levels in stream water at the Cropper Creek Hydrologic experimental area (Bren et al 1979). Velpar was applied to this steeply sloping 46.4 ha catchment with residues being studied over a 9-week period using automatic samplers at intervals of 0.25-2.0 hours throughout the nine weeks of the sampling, with the more intensive sampling being employed immediately following spraying and during substantial storm events. A total of 69 representative samples were analysed, with 4 ug/l active ingredient being detected in six of these samples. The authors attributed the low residue levels to the spraying by helicopter being undertaken in calm conditions, the spray system which delivered relatively large droplets and a vegetation riparian reserve along the entire length of the stream. Soil type and its moisture status at the time of spraying were other factors influencing this favourable result.

The final steam water herbicide residue study was conducted by Fagg and Leitch (1983) in the Ryan’s Creek catchment in northeast Victoria where Lontrel was applied to control Silver Wattle. This catchment provides water for the City of Benalla. As for the Cropper Creek study, automatic sampling was employed together with some manual sampling at downstream sampling points over a 19-day period. There were seven substantial rainfall events totaling 143 mm during the sampling period. A total of 103 individual samples and 24 composite samples were analysed for Lontrel residues. The highest concentration detected was 0.015 mg/l active ingredient compared with the maximum recommended level of 1 mg/l active ingredient (NHMRC). As in previous studies, the low residue levels were attributed to several factors including an unsprayed and undisturbed strip of streamside native vegetation, the seemingly low mobility of Lontrel in the soil, and minimal spray drift due to a combination of calm spraying conditions, large droplets and the use of an anti-evaporant oil (Ulvapron) in the spray mix.

It can be concluded from the above studies that aerial spraying of a range of herbicides can be safely undertaken and avoid significant levels of herbicide residues in stream water providing common sense prescriptions are strictly adhered to, including:

• First and foremost, consistent with the principles of the NWS, undertake a thorough and objective assessment of the weed problem and prescribe the most appropriate strategy that takes account of all relevant factors including any environmental issues
• Spraying under calm conditions
• Using spray systems that avoid very small droplets
• Maintaining a significant strip of undisturbed native forest along the entire lengths of both annual and perennial streams
• Using an anti-evaporant to reduce evaporation where water is the carrier
• Having a clear understanding of the mobility of the active ingredients over a range of soil types
• Using Kromecote cards to instantaneously monitor any spray drift; and
• Consideration of topographic conditions and hydrologic issues, especially on soils with a propensity to erode under heavy to intense rainfall events

A Move Towards Integrated Weed Management

The NWS emphasizes the need to take an integrated weed management (IWM) and multi-disciplinary approach. This means that routine aerial applications of 2,4,5-T were inappropriate in the absence of a quantifiable analysis of the target weed population in terms of density, species composition, patchiness of the weed problem and other relevant factors. The withdrawal of 2,4,5-T provided an opportunity to rapidly move to an IWM system. FCV in fact adopted this strategy in advance of the NWS, and in advance of Certification requirements to use the principles of IWM to minimize use of chemicals and to minimize environmental impacts including stream water contamination.

Fagg and Flinn (1984) provide a case study of the application of the IWM approach for a plantation at Morgans Creek in northeast Victoria. Each weed problem needs to be carefully evaluated before prescribing control techniques. This evaluation process must give consideration to management objectives, topography, labour availability and costs, herbicide costs, climate, soil type, environmental constraints and the size, extent, density and species composition of the weed population. This approach results in an overall cost-effective weed control program that minimizes environmental impacts. It generally results in two or more techniques being used in a plantation block. In the Morgans Creek case study, the following six contrasting treatments were required to control woody and herbaceous weeds:

• Ploughing pre-planting to partly control wattle and eucalypts
• Spot spraying with amitrole and atrazine to control herbaceous weeds
• Aerial spraying with Lontrel to control dense Silver Wattle
• Aerial spraying with Velpar to control dense eucalypt seedlings
• Soil application with Velpar or stem injection with Roundup to control individual clumps of eucalypt coppice; and
• Nil treatment

This IWM approach was adopted by the FCV and used routinely across the State from the mid 1980’s. It pre-dated the NWS which recommended the use of IWM in all situations where weed control was prescribed in agriculture, forestry and other sectors.

Conclusions

Effective and efficient strategies have been progressively developed over the past five decades to manage the contrasting range of weed problems encountered in Victorian pine plantations. Such strategies are of fundamental importance in the overall management of a successful softwood plantation program in Victoria and elsewhere. Indeed, weed management is a critically important and significant cost component of a high proportion of softwood plantations in Victoria. Many plantations across the State would simply fail in the absence of such strategies. The work of Jack (1970), who quantified the loss of growth due to woody weeds on plantation productivity, brought into focus the need to control weed populations. Failure to manage the contrasting weed populations encountered across the softwood estate can lead to failed or unproductive plantations.

In the early 1970’s there was a rapid move away from manual slashing of woody weeds (which was costly when repeated treatments were required due to resprouting). The introduction of 2,4,5-T to Victorian plantation management was a milestone development. This herbicide was found to be exceeding effective in controlling dense Silver Wattle competition. This herbicide in mixture with other herbicides was also used to treat eucalypt regrowth by stem injection. Over 16,000 ha were treated by aerial application of 2,4,5-T in the years that followed. The use of this herbicide however was a controversial issue due to a contaminant in the formulation that is exceedingly toxic. Community opposition to the use of this herbicide in softwood plantations in Victoria led to its voluntary withdrawal by FCV of its use in plantation forestry. Researchers responded to the issue well before its withdrawal by evaluating alternative herbicides. A suite of options was rapidly developed that enabled a more integrated and strategic approach to weed management in recognition that blanket applications of 2,4,5-T were no longer appropriate.

A new era of weed management followed. With more options identified by R&D for the full suite of diverse weed populations in terms of species, size and density along with soil type, topography and climate (especially the frequency of intense storm events) which enabled every plantation to be critically evaluated, and an integrated weed management strategy implemented for the individual plantations. This approach pre-dated this basic requirement as detailed in the NWS. It also met the requirements of Certification Standards including the AFS and FSC. In many respects therefore, the loss of 2,4,5-T was a blessing. Any reader interested in the dioxin controversy is encouraged to read the paper by Flinn and Hopmans (1977). Unlike the 2,4,5-T/Tordon era, researchers identified an arsenal of herbicides that could be effectively and efficiently used on the full spectrum of weed problems across the pine estate. The choice of a herbicide regime for each plantation also considered advice from soil scientists and hydrologists in particular.

Whilst the community appeared much more comfortable with the new IWM approach, controversy continued about the aerial herbicide treatments in particular. FCV accordingly developed more stringent specifications for aerial applications and undertook several intensive herbicide residue studies of stream water flowing from treated plantations. Residue levels in all instances for the three herbicides studied were well below standards for potable water. Minimisation of spray drift was given special attention using a combination of measures including retained and unsprayed vegetation strips along entire lengths of streams, use of spray systems that emitted medium to large droplets, the use of a petroleum oil in the spray mix to reduce droplet evaporation, and most importantly strictly meeting the requirement of calm conditions during spraying. Use of granular formulations where appropriate, the use of Ulvapron to reduce the evaporation of droplets and to increase herbicide efficacy offers the opportunity to reduce rates of herbicide application and hence reduce the level of chemical use in plantation management. Further development of rope-wick applications for small plantations on favourable slopes should be further studied, especially as Certification moves to smaller holdings where aerial application may not be appropriate.

References

All Forest Research Reports listed below are available in the Site Library. Other publications may also be available.

Bren, L J, Flinn, D W, Hopmans, P and Leitch, C J (1979). The hydrology of small forested catchments in North-eastern Victoria.1. Establishment of the Cropper Creek Project. For. Comm. Vic., Bull. No 27, 48pp.

Fagg, P C and Flinn, D W (1983). Aspects of woody weed control in Radiata Pine plantations in Victoria. Forest Research Report 213, 22pp.

Fagg, P C and Flinn, D W (1984). Review of weed control practices in radiata pine plantations in Australia. Forest Research Report 268, 15pp.

Fagg, P C and Leitch, C J (1983). Residues of 3,6-DCPA in streamwater following aerial application to a radiata pine plantation in Ryan’s Creek catchment, north-eastern Victoria. Forest Research Report 238, 16pp.

Fagg, P C, James, J M, Flinn, D W and Mainey, P (1982). Residues of hexazinone and four of its metabolites in streamwater after aerial spraying of a Pinus radiata plantation near Yarram. Forest Research Report 193, 9pp.

Fagg, P C, Cameron, C J and Flinn, D W (1987). Silver wattle control with Lontrel and Roundup applied with a tractor-mounted rope-wick applicator. Forest Research Report 323, 15pp.

Fagg, P C, Flinn, D W and Hepworth, G (1988). Spot-spraying of hexazinone and amitrole/atrazine in the establishment of first and second rotations of Pinus radiata in south-western Victoria. Plant Protection Quarterly 3:74-77.

Fagg, P C, Hepworth, G, Smith, W and Cameron, G J (1990). Evaluation of Roundup, Tordon TCH and Velpar L herbicides for the control of eucalypt regrowth in softwood plantations by stem injection. Forest Research Report 342, 17pp.

Flinn, D W (1968). Determination of Spread Factor. Student project, School of Forestry, University of Melbourne, 21pp plus appendices (unpublished).

Flinn, D W (1978). Early growth response of Radiata Pine to grass control and superphosphate in the Strzelecki Ranges. Forest Research Report 104, 5pp.

Flinn, D W and Hopmans, P (1977). Pesticides in the forest environment. 1. Use of 2,4,5-T for control of woody weeds in Pinus radiata plantations in Victoria. For. Tech. Paper No 26, 5-10. (also in Forest Research Report 105).

Flinn, D W and Minko, G (1981). Advances in control of woody weeds in Radiata Pine plantations in Victoria. Forest Research Report 184, 7 pp.

Flinn, D W, Minko, G and Barrand, L (1980a). Screening of herbicides for woody weed control in Pinus radiata plantations. Forest Research Report 155, 9 pp.

Flinn, D W, Hopmans, P, Farrell, P W and James, J M (1979a). Nutrient loss from the burning of Pinus radiata logging residue. Aust. For. Res. 9: 17-23.

Flinn, D W, Hopmans, P, Moller, I and Tregonning, K (1979b). Response of radiata pine in the Neerim East Plantation to fertilisers containing N and P applied at planting. Forest Research Report 133, 11 pp.

Flinn, D W, Minko, G and Barrand, L (1980a). Screening of herbicides for woody weed control in Pinus radiata plantations. Forest Research Report 155, 9 pp. (also in For. Tech. Paper No 28, 12-17).

Flinn, D W, Minko, G and Stokes, K G (1980b). Aerial application of “Velpar L” for woody weed control in established pine. Aust. For. Grower 3:55.

Flinn, D W, Minko, G and Barrand, L (1982). Chemical control of young, natural, Radiata Pine regeneration. Forest Research Report 199, 7pp. (also in Aust. Weeds 2 (2): 53-55).

Geary, P W and Flinn, D W (1985). Control of Acacia verticillate and other woody weeds in a young Pinus radiata plantation by an aerial application of hexazinone. Forest Research Report 283, 10pp.

Geary, P W, Pollett, S J, Thomson, D S and Flinn, D W (1987). Control of natural regeneration of Radiata Pine during second rotation establishment using slash retention techniques. Forest Research Report No 325, 29pp.

Hopmans, P (1975). The sorption of 2,4,5-T by LDPE plastic containers. Forest Research Report 71, 5pp.

Hopmans, P and Tomkins, I B (1992). Response of radiata pine to fertiliser and chemical weed control at Neerim East. Forest Research Report 351, 17pp.

Jack, J B (1970). The weed problem in Victorian plantations of Pinus radiata D. Don. MSc Thesis, Melb. University, 216pp

Leitch, C J and Flinn, D W (1983). Residues of hexazinone in streamwater after aerial application to an experimental catchment planted with Radiata Pine. Forest Research Report 216, 9pp.

McKimm, R J (1973). An evaluation of two low-volume spray systems for aerial use. Forest Research Report 35, 11 pp.

McKimm, R J and Hopmans, P (1977). 2,4,5-T concentrations in Scrubby Creek, Carboor following aerial spraying of a young Pinus radiata plantation. Forest Research Report 93, 9pp.

McKimm, R J and Hopmans, P (1978). 2,4,5-T residues in stream water following aerial spraying for woody weed control in the Narbethong plantation. Forest Research Report 112, 11pp.

Minko, G (1974). Atrazine and Amitrole for the control of vegetation competing with outplants of Pinus radiata at Myrtleford. Forest Research Report 42, 13pp.

Minko, G (1978). Glyphosate for the control of competing vegetation in Pinus radiata plantations at Myrtleford. Forest Research Report 103, 15pp.

Minko, G and Flinn, D W (1982). Evaluation of herbicides for stem injection of woody weeds in young Radiata Pine plantations. Forest Research Report 188, 10pp. (also in Aust. For. 44, 260-266).

Minko, G and Stephens, L A (1986). Krenite for control of natural regeneration of Radiata Pine and competing vegetation following wildfire. Forest Research Report 312, 11pp.

NWS (1997). Agric. and Resource Management Council of Aust. and N Z, Aust. and NZ Env. and Conserv. Council, Forestry Ministers. 1997. The National Weeds Strategy: A strategic approach to weed problems of national significance. 52pp. ISBN 0 642 21401 8.