GEJ & MA Geldard Pty Ltd v DN Mobbs & Ors [2010] QSC 220

[1] On 15 December 2005, a day of high temperatures, gusty winds and low humidity, extensive aerial spraying of the herbicides Metsulfuron (Ally), Grazon (Picker) and Brushwet was carried out by two fixed wing aircraft to control the regrowth of wattle trees on the cattle properties Sherwood and Wallumba in the Condamine area of South East Queensland. Grazon contains Triclopyr and Picloram. The herbicides were not mixed to the rates and ratios as set out on the labels of the products. It is not disputed that the application of chemicals was "off label" and that no permit to spray had been obtained. The application of the chemicals took over five hours.

[2] The label for Metsulfuron is in evidence.[1] The label states that aerial application is "by helicopter only" and warns the applicator to "Avoid spraying in conditions which favour temperature inversions, still conditions, or in winds likely to cause drift onto sensitive crops or fallow areas to be planted to sensitive crops". It also provides that in relation to established pastures, it should not be applied in temperatures over 30 degrees. The label also states that the chemical is "Not to be used for any purpose, or in any manner, contrary to this label unless authorised under appropriate legislation".

[3] The Geldard family through a family company (the plaintiff), grows cotton crops in the Condamine area. Two of the properties on which they grow cotton are Elgin and Noonameena. They are some 20 kilometres to the south of the fields sprayed. On 19 December 2005, four days after the spraying, yellowing and damage consistent with herbicide damage was observed on the Geldards' cotton crops on those two properties. The Geldards' cotton yield in these fields was affected by the damage to the crops. The Geldards claim that this loss of yield was caused by the spraying on 15 December 2005 on Sherwood and Wallumba and claim damages.

[4] The essential issues in this trial therefore are:

(i) Whether the spraying on 15 December 2005 on Sherwood and Wallumba affected the Geldards' cotton crops on Noonameena and Elgin; and if so

(iii) Whether the effect of the spray on the cotton crops was due to the negligence of the sixth and eighth defendants; and if so

(iv) What is the extent of the financial loss suffered if any.

[5] The first defendants (the Mobbs) are the owners of Sherwood and the second defendant (Simon Drury) was employed as the manager of that property. The third defendant (MDH Pty Ltd) is the owner of Wallumba and the fourth and fifth defendants (McDonald and Maher) were employed as managers and operators of that property. The seventh defendant (Ross York) has a business of supplying agricultural products. He supplied the herbicides that were used by the sixth defendant, Meandarra Aerial Spraying Pty Ltd (Meandarra) in the aerial spraying activities that were conducted on Sherwood and Wallumba.

[6] It is admitted that spraying occurred on Sherwood and Wallumba on 15 December 2005. There is also no dispute that the first, second, third, fourth and fifth defendants engaged Meandarra to spray the herbicides supplied by Ross York on Sherwood and Wallumba. There is no dispute that Lachlan Hill (Hill), who is the eighth defendant, flew and operated the aircraft that sprayed the herbicides. It is admitted that the sixth and eighth defendants sprayed Metsulfuron and Grazon (Picker), which is a combination of the chemicals Triclopyr and Picloram. The evidence also clearly establishes that herbicides damaged the crops at Elgin and Noonameena in the 2005/2006 season. It should also be noted at the outset that no other likely source for the herbicide damage to the crops has been established.

[7] The plaintiff's claim against the first, second, third, fourth, fifth and seventh defendants has now been settled. The plaintiff's claim continues against the sixth and eighth defendants, namely Meandarra Aerial Spraying and the pilot Lachlan Hill. For the purposes of these reasons, I will leave to one side the issue of the apportionment of liability under the Civil Liability Act 2003 (Qld) and the effect of the settlement by the plaintiff with the other defendants.

[8] The plaintiff leases three properties in the Condamine area of Queensland, namely Cullingral, Noonameena and Elgin. Two of those properties, namely Noonameena and Elgin are to the south of the properties which were sprayed, and Cullingral is to the north. The two properties to the south only were affected. Cullingral is to the north-east of Wallumba and Sherwood and its cotton crop was not affected. Wallumba, which is one of the properties on which the spraying occurred, is located 18.9 kms to the north of Elgin and 21.2 kms north of Noonameena. Sherwood, which is the other property where the spraying occurred, is located 18.5 kms to the north of Elgin and 21 kms to the north of Noonameena. The parties have agreed that a distance of some 20 kms separates those properties. The properties Toston and Coobareena, which are owned by the Keys and Ronnfeldt families respectively, were also affected in varying degrees and those properties are located between Sherwood and Wallumba on the one hand, and Noonameena and Elgin on the other. The relative location of all of those properties is shown in the following map.

[9] Ashley Geldard (Geldard) is the manager of the plaintiff company and he gave extensive evidence of the company's farming practices as well as the rainfall, crop yields and chemical use over the period 2004 to 2006. The plaintiff's three properties have historically been farmed in a cropping rotation of predominantly fallow, wheat, barley, chickpea, sorghum and cotton. Cotton is grown on all the properties. From 4 to 14 November 2005 all the fields were planted with cotton and from 17 to 28 November all fields were sprayed over the top with "Roundup Ready" herbicide. On 5 December 2005 agronomist Ed Offner noticed small areas of slight hormone damage in a couple of the fields.

[10] The properties at Elgin, Noonameena and Cullingral are divided into various fields. The relative yields as between the properties in the 2003/2004 and 2004/2005 years were similar. The plaintiff argues that because of the historical similarity, the yields on Noonameena and Elgin in 2006 should have been the same as those on Cullingral in that year. The cotton crop on Cullingral, which was to the north-east of the properties sprayed, had no yield loss in the 2005/2006 year. The plaintiff's argument is therefore that Cullingral exemplifies the yield which should have occurred in the two affected properties.

[11] Geldard explained that historically, the plaintiff had farmed those properties in the various cropping rotations outlined above, in a 17 metre strip cropping management system. In May 2005 however, all of the cotton country was converted to a broad acre management system. This system involves the planting of two rows of cotton, one metre apart with a total of five metres "skip" in-between alternative rows of cotton.

[12] Geldard outlined that this configuration was to better utilise available soil stored moisture and in-crop rainfall. Whilst there was a difference in rainfall, Geldard explained that the yields very closely reflected what they would expect to see from the rainfall and the known amounts of soil moisture stored in the soil prior to the crop.

[13] Geldard indicated that the cotton crops on these properties are not irrigated and dry farming methods are used. Moisture retention is therefore a vital component of the methods used. He indicated that the soil in all the fields was similar. Soil moisture comparisons were also carried out, which indicated that the crops at Noonameena and Elgin had root damage and were therefore not able to utilise the moisture that was available in the soil compared to the Cullingral crop, which had utilised the available moisture.

[14] It is uncontentious that cotton is a crop which is susceptible to herbicide damage. Geldard gave evidence that in December 2005 and early January 2006 significant yellowing was noted in the cotton in various fields, namely the north-west corner of Noonameena two and Elgin one and three. Furthermore, he stated that the crops at Noonameena and Elgin were showing more signs of stress than those at Cullingral, despite the fact that those properties had the same history and rainfall. By February 2006 phytotoxicity symptoms were observed on Noonameena and Elgin. Those symptoms included yellowing, discolouration, deformity and crinkling sections on the cotton plants. Geldard also gave evidence of the areas of pasture and tree-lines that he observed to have been damaged. He considered that the damage was particularly noticeable on one windward face of the trees in an area to the south and south-west of Wallumba and Sherwood. Damage to crops and trees is clearly observable in an extensive series of photographs which have been tendered as exhibits.[2]

[15] Miranda Ronnfeldt gave evidence of the damage she observed to some of the wilga trees on the property Coobareena which she runs with her husband. The property is located directly to the north of Elgin and one property north of Noonameena. Paul Keys who owns the property Toston, which is directly to the south of Sherwood and Wallumba, gave evidence of the damage to trees on his property which extended to his southern border. He also stated that the eighth defendant had visited his property and agreed that damage had occurred and that this was unacceptable.

[16] On 3 January 2006 two cotton leaf samples were taken from the affected cotton and sent for testing. The testing showed the plants were nutritionally healthy but affected by herbicides. The analysis of cotton plant tissue from Elgin 3 tested positive for Metsulfuron. All levels were below the level of detection in the Elgin 3A sample.

[17] Wilga trees are also susceptible to herbicides and three samples of wilga leaves were collected from trees at intervals between the properties Sherwood, Wallumba, Noonameena and Elgin on 10 February 2006. All three samples had detectable levels of Metsulfuron. Picloram was below the level of detection in all samples. A number of those plant samples were positive for the presence of Metsulfuron and Triclopyr. There was a positive test for Triclopyr in the wilga sample at 4.6 kilometres (sample 2W) with the level of Metsulfuron at 0.06 mg/kg. Wilga samples at 16 (sample 7W) and 19 kilometres (sample 8W) were also positive for Triclopyr but they were below the level of quantification. There was a gradient declining of Metsulfuron concentration to 0.01 at the 16 and 19 kilometre samples. All wilga leaf samples contained the herbicide 2,4-D which was not sprayed by the defendants on 15 December.

[18] A soil sample was collected from Elgin on 10 February 2006 which contained Metsulfuron above the level of detection of 0.002mg/kg.

[19] The plaintiff initially sued in both negligence and nuisance but has abandoned the claim in nuisance. It is clear that since the High Court decision in Burnie Port Authority v General Jones Pty Ltd[3] issues, such as those raised in this case, "should be dealt with according to the ordinary principles of negligence".[4] The plaintiff alleges that the defendants breached the duty of care and that they have suffered loss as the result of negligence. As previously indicated, all the defendants other than the sixth and eighth defendants have settled the action.

[20] The plaintiff, therefore, claims that the sixth defendant, Meandarra Aerial Spraying and Hill, owed a duty of care to the plaintiff to:

(a) take all such reasonable steps as were necessary to avoid there being any risk of loss or damage to the plaintiff's properties and/or crops; and

(b) take all reasonable steps to ensure that there was no risk of any loss being occasioned or damage to the plaintiff's properties and/or crops; and

(c) to take all reasonable steps as were required to prevent the herbicide being sprayed or emitted onto the plaintiff's properties.

[21] The plaintiff alleges that the sixth defendant, Meandarra Aerial Spraying, was negligent in that it:

(a) used herbicide sprays where they had not been mixed to the correct ratio and rate;

(b) sprayed and used herbicides when it was inappropriate to be used for aerial spraying;

(c) sprayed when the weather conditions were inappropriate;

(d) used a herbicide or herbicides which should not have been used for aerial spraying; and

(e) aerial sprayed in such a manner as to cause damage to the plaintiff's property when such damage could have been avoided by the use of reasonable care.

[22] In relation to the negligence of the eighth defendant, the plaintiff claims that the pilot Lachlan Hill:

(a) used herbicide sprays when they had not been mixed to the correct ratio;

(b) sprayed and used herbicides when it was inappropriate to be used for aerial spraying;

(c) sprayed when the weather conditions were inappropriate;

(d) used herbicides which should not have been used for aerial spraying;

(e) aerial sprayed in such a manner as to cause damage to the plaintiff's property when such damage could have been avoided by the use of reasonable care;

(f) failed to advise or give proper advice and make reasonable and proper inquiries or investigations concerning the conditions under which aerial spraying should be conducted;

(g) sprayed with off-label concentrations for Metsulfuron and Grazon in excess of the appropriate maximum concentrations; and

(h) failed to give proper advice concerning the use of those herbicides.

[23] As a result of the defendants' negligence, the plaintiff claims it has suffered loss in the sum of $461,237.23, which is essentially made up of the production loss of the affected cotton crops on Noonameena and Elgin in 2006.

[24] In order to show the defendants were negligent, the plaintiff must establish that a duty of care was owed to it and the determination of that question involves a consideration of the proximity of the relationship and whether the injury caused to the plaintiff was reasonably foreseeable.

[25] The issue of foreseeability was discussed by Mason J in The Council of the Shire of Wyong v Shirt:[5]

"Notwithstanding this Australian support for a narrower version of the foreseeability doctrine as applied to breach of duty, this Court would be well advised to accept that the law upon the point was correctly stated and applied by the Judicial Committee in The 'Wagon Mound' (No. 2.). I say this not only because The 'Wagon Mound' (No.2) was a unanimous decision given on appeal from the Supreme Court of New South Wales, but also because there are sound reasons for accepting it as a correct statement of the law.

In essence its correctness depends upon a recognition of the general proposition that foreseeability of the risk of injury and the likelihood of that risk occurring are two different things. I am of course referring to foreseeability in the context of breach of duty, the concept of foreseeability in connexion with the existence of the duty of care involving a more generalized enquiry.

A risk of injury which is quite unlikely to occur, such as that which happened in Bolton v. Stone [1951] UKHL 2; (1951) AC 850, may nevertheless be plainly foreseeable. Consequently, when we speak of a risk of injury as being 'foreseeable' we are not making any statement as to the probability or improbability of its occurrence, save that we are implicitly asserting that the risk is not one that is far-fetched or fanciful. Although it is true to say that in many cases the greater the degree of probability of the occurrence of the risk the more readily it will be perceived to be a risk, it certainly does not follow that a risk which is unlikely to occur is not foreseeable.

In deciding whether there has been a breach of the duty of care the tribunal of fact must first ask itself whether a reasonable man in the defendant's position would have foreseen that his conduct involved a risk of injury to the plaintiff or to a class of persons including the plaintiff. If the answer be in the affirmative, it is then for the tribunal of fact to determine what a reasonable man would do by way of response to the risk. The perception of the reasonable man's response calls for a consideration of the magnitude of the risk and the degree of the probability of its occurrence, along with the expense, difficulty and inconvenience of taking alleviating action and any other conflicting responsibilities which the defendant may have. It is only when these matters are balanced out that the tribunal of fact can confidently assert what is the standard of response to be ascribed to the reasonable man placed in the defendant's position.

The considerations to which I have referred indicate that a risk of injury which is remote in the sense that it is extremely unlikely to occur may nevertheless constitute a foreseeable risk. A risk which is not far-fetched or fanciful is real and therefore foreseeable. But, as we have seen, the existence of a foreseeable risk of injury does not in itself dispose of the question of breach of duty. The magnitude of the risk and its degree of probability remain to be considered with other relevant factors."

[26] The relevant guiding principles of law are set out by Balkin and Davis in Law of Torts 3rd Edition as follows:

"In Wyong Shire Council v Shirt Mason J held that, in order to determine what a reasonable person would do by way of response to a given risk, the trier of fact would need to consider 'the magnitude of the risk and the degree of the probability of its occurrence, along with the expense, difficulty and inconvenience of taking alleviating action and any other conflicting responsibilities the defendant may have'. The standard of care required of the reasonable man or woman placed at the time of the accident in the shoes of the defendant is an objective one and only by balancing these factors can it confidently be asserted that this standard has been met. The need for establishing this balance ostensibly remains even in situations which would formerly have attracted the rule in Rylands v Fletcher. In such circumstances, however, the High Court has now ruled that a reasonable prudent person would exercise 'a high degree of care' which, depending on the magnitude of the danger, might well involve 'a degree of diligence so stringent as to amount practically to a guarantee of safety'. Principles along the same lines as those stated by Mason J have now been included in legislation in some of the Australian jurisdictions, applicable generally to causes of action arising other than from the use of a motor vehicle or in the workplace. It is assumed that, when interpreting this legislation, the courts will be guided by the principles already developed by the common law over the previous century or more."

[27] The Civil Liability Act 2003 has been enacted in Queensland and the relevant principles as examined by Mason J are dealt with in ss 9, 10, 11 and 12 of that Act. Once a duty is established, it must then be shown that the duty which was owed to the plaintiff was breached and the determination of that issue also involves a consideration of whether the event which gave rise to the injury was reasonably foreseeable and whether the defendants failed to do what a reasonable person would have done in the circumstances. The final issue is that the plaintiff must establish that the breach caused the injury complained of.

[28] It is clear that the plaintiff has the burden of proving each element of its claim[6] and that this burden of proving causation always remains with the plaintiff.[7] The plaintiff must also prove the nature and extent of its claimed loss.

[29] The defendants clearly have a duty of care which includes taking precautions to avoid spray drift which would cause damage to non-target crops and plants. The defendants' duty is to exercise the care and caution that a reasonable applicator should use in the same or similar circumstances. The duty of an aerial applicator of herbicides was discussed in Bonic v Fieldair (Deniliquin) Pty Ltd & Ors[8] where Davies AJ held:

"The nature of the chemicals being sprayed was such that all persons responsible for the operation, that is the four defendants, had a non delegable duty to ensure that properties in the vicinity of the Rendell land were not damaged by the spraying."

[30] Therefore, the plaintiff must essentially establish that the sixth and eighth defendants by spraying the chemicals in the way they did breached their duty of care and that, because of that breach, the sixth and eighth defendants caused the chemicals to drift onto their property in a sufficient quantity to cause the damage alleged.

[31] In the present case, the parties submit that the real issue in contention is this question of causation. I agree that this is the central issue.

[32] A number of experts were retained by the respective parties and they produced a number of reports which are exhibits in these proceedings.[9] These experts included:

Peter Tremain, an agricultural consultant from Wellington, New South Wales, who has been involved in the investigations of chemical drift in the past.

Dr Chris Preston, an Associate Professor at the University of Adelaide. He is an expert in weed management and has a PhD in Plant Biochemistry and has been involved in herbicide resistant weed management and research into how herbicides work in plants, how herbicides behave in the environment, how plants respond to herbicides and how plants become resistant to herbicides.

Ken Bullen, an agricultural loss assessor with a degree in agricultural science who has been conducting assessments of agriculture losses since 1982. He has worked with the University of Sydney and Agri-search, which is a research and agricultural chemical company. He researches agricultural chemicals, in particular herbicides, including Metsulfuron.

Andrew Somervaille, an agricultural consultant. He has expertise in the area of field development of herbicides in broad acre crops and row crops. His particular interest is in spray application as it relates to effectiveness of pesticides and the reduction of spray drift.

Bill Gordon, an agricultural consultant holds degrees in the area of plant protection, specialising in pesticide application technology. He also has expertise in aerial spraying, pesticide application and drift management.

Ed Offner, an agronomist who was employed by the plaintiff to monitor the cotton crop during the relevant period.

[33] Given the number of reports and the extent of the evidence, I will refer in detail to the evidence which I consider to be significant and my reasons for reliance on it. The defendants made a significant number of objections[10] to the expert reports of Bullen, Gordon and Somervaille. I have taken those objections into account and, without specifically ruling on each and every one of those numerous objections, the extent to which I have specifically relied on a aspect of a report which was contentious I have set that out in these reasons. I consider that many of the objections were, in fact, overtaken by the joint report.

[34] Offner gave evidence of his observations, in his role as the plaintiff's on farm agronomist, of the development of symptoms at Elgin and Noonameena. He gave evidence of his observation of hormone damage prior to the spray event on 5 December 2005. This early damage consisted of crinkling symptoms and he considered it could have been consistent with 2,4-D damage. He also gave evidence of his observation of the subsequent yellowing symptoms on 19 December which he stated was quite different to the earlier symptoms. These subsequent symptoms were not consistent with hormone damage. He considered the symptoms were "unusual"[11] and were indicative of a further problem.

[35] He also stated that the yellowing was not due to inversion because there was not a lot of yellowing in the drainage lines. He stated:

"An inversion is when there's a different layer of air temperature and instead of something being blown on by the wind it actually shifts slowly like a mist and it usually lays in lower areas of the paddocks. So if it's been caused by an inversion you can usually spot them in drainage lines and other areas like that."

[36] Offner also reported that after the 26 December 2005 inspection, the crop behaviour at Noonameena and Elgin began to differ to that at Cullingral. The crops at Noonameena and Elgin developed a shorter node length.

[37] Offner stated that by January 2006 the plaintiff had called in other experts to investigate the yellowing and he "stepped out of the picture in terms of reporting".[12]

[38] Initially Tremain indicated that he did not consider that the chemicals could have travelled 20 kms and his view was that even at 800 metres there would be insufficient quantities of chemicals to cause damage. He also considered that some of the damage to cotton attributed to Metsulfuron could, in fact, be due to the chemical 2,4-D, which is also an auxin type herbicide. Neither was Mr Tremain convinced that there was a clear chemical trail of tree damage from the spray site to the cotton fields and he also considered that he would expect a gradient of damage down wind. He agreed that components of Metsulfuron, Triclopyr and 2,4-D were found in the samples taken from wilga trees, but there was no evidence of Grazon found in the sample. He also agreed that there was no Grazon or 2,4-D in the cotton residue samples.

[39] In my view, there is clear evidence that the trees on the Keys' property were damaged by the spray drift and this is accepted by the sixth and eighth defendants. The Keys' property is some six kilometres from Sherwood and Wallumba. Accordingly, I consider that many of the assumptions in his report must be incorrect given the clear evidence of damage.

[40] Furthermore, I do not consider that Tremain has the same level of expertise as either Somervaille or Bullen in relation to herbicides, their application or the damage caused. He indicated that he had no specific experience of the combination of Brushwet and Metsulfuron or done any assessments before in the Condamine region of Queensland. I therefore place more reliance on Bullen's report in regard to chemical damage.

[41] Dr Preston's evidence largely related to the assessment of the results of the samples taken and whether they were consistent with damage from the application of the herbicides on 15 December 2005.

[42] Dr Preston indicated that Metsulfuron is a sulfonylurea herbicide which produces symptoms that include yellow to red or purple colouration of new foliage, which gradually progresses through the plant. Leaf curl may also be apparent and new leaves can fail to properly expand. The plants will also be stunted, with shortened internodes. He agreed that the symptoms observed four days after the application on 15 December was yellowing of the cotton leaves, which is clearly consistent with Metsulfuron damage. He indicated, however, that symptoms which are similar to Metsulfuron symptoms can be caused by deficiencies of iron, zinc or nitrogen, or could be due to water logging, drought stress and glyphosate use.

[43] Dr Preston's evidence was that, assuming the herbicides were evenly mixed in the spray, he would have expected 2.5 times as much Triclopyr and 0.83 times as much Picloram deposited in any area. He stated that whilst it is clear that Metsulfuron was present in the samples, he would have expected the presence of the other chemicals, namely Picloram and Triclopyr, as well. He also indicated that the symptoms from those chemicals would appear more quickly than the symptoms of Metsulfuron. Accordingly, he would have expected those symptoms to have appeared before the Metsulfuron symptoms.

[44] Dr Preston's report indicated that once inside a plant a herbicide may be detoxified through metabolism and that Metsulfuron is "poorly" metabolised in susceptible plants, but that Triclopyr and Picloram are "slowly" metabolised. He stated that those factors will influence the amount of herbicide detected in plant samples. He stated that there is no information available on the specific fate of those herbicides on cotton and wilga plants.

[45] Dr Preston's report also indicated that the limit of detection (LOD) for the three herbicides was such that the detection and quantification limits were five times lower for Metsulfuron than Picloram or Triclopyr. Whilst Dr Preston agreed that the wilga samples were consistent with the application of the chemicals sprayed on 15 December, he did not consider the presence of 2,4-D in the wilga sample was consistent with the spray event. Ultimately, however, Dr Preston agreed that the 2,4-D issue was probably a "red herring" and that it actually did not play a huge role.[13]

[46] Dr Preston also stated that he considered that the development of yellowing symptoms just four days after the application was exceptionally fast and not consistent with the known behaviour of Metsulfuron. He also considered that the amount of Metsulfuron which needed to be applied to produce a 50 per cent reduction in production was in the order of one to two grams.

[47] Dr Preston also stated that other possible factors contributing to the damage to the cotton was the application of Metsulfuron prior to the planting of the cotton or the application of glyphosate to the cotton. He also reported that Metsulfuron is a soil persistent herbicide which persists longer in alkaline soils and that the persistence of Metsulfuron can be exacerbated by dry weather. He noted drier than normal weather conditions in 2005 and the high pH levels in the soils and concluded that "Metsulfuron-methyl residues present in the soil from previous applications of the herbicide Ally are very likely contributor (sic) to symptoms in the cotton crops given the very high soil PH in these fields".

[48] In relation to the possible distance of spray drift, Dr Preston stated that the Australian Pesticides and Veterinary Medicines Authority (APVMA) had developed a model comparing wind speeds of eight to 20 km/h and spray qualities of coarse, medium and fine. He stated that based on those scenarios, the worst case scenario (fine spray at 20km/h) would result in a four per cent drift of the herbicide at 800 metres. A medium spray would result in drift of one per cent of the herbicide over the same distance and coarse spray would result in a drift of 0.5 per cent. He considered that long distance drift is more often associated with still wind conditions where temperature inversions occur than it is with windy conditions.

[49] Dr Preston's report indicated that whilst the weather conditions may have been appropriate for spraying from 6.40 am to 1.00 pm, he considered that the data from the weather stations indicated that spraying would have been marginal from about noon, for a maximum period of about 1.5 hours.

[50] Dr Preston considered that from the temperature records there was no evidence that inversion conditions were present. In his 5 May 2009 report, Dr Preston ultimately concluded, "It is my opinion that, given the weather conditions at the time of application and with reference to the APVMA scenario modelling, spray drift of the extent claimed over 20 km is very unlikely".

[51] Dr Somervaille explained the herbicide Metsulfuron in the following terms:

"Metsulfuron-methyl is a herbicide that inhibits the biosynthesis of the amino acids and valine and isoleucine required for the production of plant proteins. The effect of this is to stop or inhibit cell division. At sub-lethal levels this would affect both shoots and roots of affected plants. A secondary effect of Metsulfuron therefore would be to limit access of plants to water and nutrients that otherwise would be available for plant growth and in particular production of fruiting bodies (square and bolls).

Where plant growth is affected by depletion of soil water and high temperatures, this would be axecerbated [sic] where plants had been inhibited in their production of a fully functioning root system. This effect would be in addition to the direct effect of the herbicide on production of biomass and therefore production of cotton."

[52] Dr Somervaille acknowledged that instances of injury to crops by Metsulfuron are not common generally because of the low rates of the product that are used. He considered, however, that in the present case there were a number of significant factors operating. These factors included the high rate of Metsulfuron used, the height of release and the prevailing winds at the time of spraying. He gave evidence that, in his view, the effects of less than expected crop growth on the properties Elgin and Noonameena was attributable to the effects of drift of low levels of Metsulfuron and Grazon. He bases this conclusion on a number of factors. First, the appearance of yellowing and leaf distortion on the cotton within a few days of the aerial application of Metsulfuron upwind of the cotton farms; second, the persistence of the symptoms and the progressive increase in their severity in the weeks following the initial observation. He considered that the plants exhibited distorted growth, consistent with that caused by a sub-lethal dose of herbicides such at Metsulfuron, whilst others appeared to be affected by auxin like herbicides such as Picloram, which is a component of Grazon.

[53] He did not consider, however, that the initial leaf twisting and crinkling was typical of that expected from an application of 2,4-D. He also stated that the assay of Metsulfuron showed that it was in vegetation three weeks following the drift event, including samples of cotton and trees. Dr Somervaille also stated that there was a known sensitivity of cotton to low levels of the herbicides Metsulfuron, Picloram and Triclopyr.

[54] Dr Somervaille stated in his report of 10 February 2010 that his inspection revealed that the height at which 2,4-D symptoms appeared in the crop was at around node 20. Accordingly, as this would likely have been above the position of the top harvest boll, he concluded that the 2,4-D effects observed would not have significantly affected the yield potential of the crop. Furthermore, he stated that the effects of 2,4-D at a moderate to severe level would normally be associated with fruit loss on branches where foliar effects were noted, which would result in an increase in internode length and plant height. However, there was no such loss of fruit or increase in plant height noted in the crop. The effect of this evidence is that in his view 2,4-D damage was irrelevant to crop loss.

[55] Dr Somervaille also discounted the possibility of soil residual herbicides affecting the crop. He noted that soil residual herbicides were applied in January 2005 on Noonameena 5 and Elgin 1 in alternate 17 metre strips. He also noted that part of Noonameena 1 and 2 was also treated with Metsulfuron in November 2004. However, he observed that the reduction in crop performance occurred in areas not actually previously treated with Metsulfuron

[56] Gordon's report involved the use of computer generated modelling to predict whether it was possible for the chemicals to have travelled more than 20 km from the site of application. Gordon's evidence was that the distance that spray droplets may move and deposit at some point down wind is able to be predicted using computer based models such as Agdisp, which is a model initially developed by NASA and now used by government regulators, including Australia. The model takes into account factors such as aircraft type, manner of operation, droplet spectrum, environmental conditions, as well as the impact of the terrain and the nature of the sprayed material.

[57] Gordon stated that both Preston and Tremain initially used model outputs generated by a different model known as APVMA as an indication of the likely amounts of product which would move. Gordon stated that the APVMA Model uses a release height of not more than three metres. He considered that this was not the relevant release height.[14]

"In the aerial operations conducted by Meandarra Aerial Services where application was made over mature trees of more than 6 m in height, and in many instances at height well over 10 m. Hence the distance the herbicides could be expected to move downwind would be much greater than those predicted for a release height of 3 metres. The likely level of disposition at various points downwind from the point of application would also be expected to be greater when the release height is higher."

[58] Gordon outlined in his report the various parameters he had used to conduct his modelling and the report set out in full the results of all this modelling. He concluded:[15]

"... there is sufficient evidence, based on Agdisp predictions, to suggest that it is likely that chemical from the Application made by Meandarra Aerial services on December 15, 2005 could have travelled 20 km or more from the site of application."

[59] Ken Bullen was retained by the plaintiff on 2 February 2006, about six weeks after the spray event, to investigate the crop symptoms. He explains that the basis of his investigation includes physical observations, assessments and residue analysis. He also conducted a modelling of cotton agronomic parameters and yield potential. He has prepared a number of reports including a Quantum Report dated 20 March 2007, Liability Report dated 10 July 2007 and a Supplementary Report dated 30 July 2009.

[60] All of Bullen's reports are extensive and involve a detailed analysis of data which relate to a number of issues including the development of symptoms, field and crop details, crop yields and residue sampling. Those reports include 27 maps which show the location of the properties, rain gauges, sample sites and plant assessment locations. There are also six graphs setting out comparison yields over a three year period, as well as comparisons of rainfall and in ground moisture. His reports also exhibit an extensive number of contemporaneous photographs of the crop damage.

[61] The report of 10 July 2007 contains a convenient summary of the factual matters which relate to the chemicals. It is clear that all of the chemicals applied on 15 December 2005 were applied "outside of legal label registrations".[16]

[62] In relation to Metsulfuron, therefore, Bullen states that it was applied at a significantly higher than label rate on the basis that:

• The maximum registered label rate for Metsulfuron application by fixed wing aircraft in Qld is 7.0 grams per hectare for cereal crops.

• Aerial application of Metsulfuron for brush control is not registered in Queensland. The label states that it is not to be applied in temperatures above 30 degrees Celsius.

• Based on the current label, Metsulfuron was applied in excess of 14 times the label rate on 15 December 2005.

[63] In relation to Grazon (Picker), which contains Triclopyr and Picloram, Bullen states that it was applied at below total water rate because:

• The minimum registered water volume for fixed wing aircraft in Queensland is 200 litres per hectare.

• The label states that the potential for damage is reduced by avoiding unsuitable spraying conditions and using spray pressure and nozzles to minimize production of small droplets.

[64] In relation to Brushwet, Bullen states that it was applied at significantly higher than label rate because:

• The label rate for brush control is up to 200 mls per 100 litres for aerial spraying.

• The 15 December 2005 application was 10 times the label rate.

[65] Bullen concluded in relation to the chemicals as follows:

"6.12.26 We conclude from our investigations of the products applied by Meandarra Aerial to the properties Sherwood and Wallumba on 15 December 2005, that Metsulfuron 600WG, Picker herbicide and Brushwet, were all applied outside of label registrations.

6.12.27 Metsulfuron 600WG was applied at significantly higher than label rate, Picker herbicide was applied at significantly below total water rate and Brushwet was applied at significantly higher than label rates.

6.12.28 The combination of all non label applications of these products would have potentially contributed to the extent, distance and quantity of spray drift of herbicide...."

[66] Bullen outlined the history of the plaintiff's previous crops and rotations and the development and pattern of the yellowing symptoms on the 2005 crop, which was planted between 4 November and 13 November. He stated that crops were at the four to eight node stage of growth at the date of the spraydrift event and were, therefore, at a very susceptible stage.

[67] Bullen also indicated that in February and March 2006 he carried out four separate crop inspections and noted significant crop phytotoxicity at Noonameena and Elgin. He indicated that this was manifested by "general chlorisis and distortion of new growth, cupping of leaves especially leaf margins and in the worst damage cases, necrosis of some leaf margins. Additionally, shortening of plant internode length resulted in plant stunting."[17]

[68] Bullen considered that the phytotoxicity symptoms were predominantly those associated with "Sulfon-Urea products of which Metsulfuron belongs to this group. Additionally, these symptoms indicated the presence of Auxin type products such as those containing Picloram and Triclopyr."

[69] Bullen considers that "Product applications of higher rates of Metsulfuron, lower rates for Picker herbicide and higher rates of Brushwet surfactant all exacerbated drift potential through droplet size, herbicide concentration in droplets and increased length of survival of droplets due to level of surfactant."[18]

[70] Bullen also considered that the soil analysis confirmed the ability of Noonameena and Elgin cotton fields to utilise soil moisture was significantly reduced which he considered indicated reduced root growth.

[71] Bullen also considered that the crop records, nutrition, soil moisture, tissue analysis and seasonal conditions all indicated that the yields at Noonameena and Elgin should have been similar to Cullingral. I specifically note the defendant's objections to the use of the water data and Bullen's conclusions as to yield and have not specifically relied on that aspect of the report. I have however relied to a considerable extent on Bullen's views as to yield in formulating my own conclusions.

[72] Clearly there is a significant difference of opinion between the experts. None of the experts, other than Bullen and Somervaille, had the advantage of seeing the fields at the time. All the other experts had to base their reports on photos and, indeed, data which were contained in the reports of Somervaille and Bullen.

[73] Ultimately I have placed considerable reliance on Bullen's various reports as he was the expert who actually went out into the field and conducted property and crop assessments on 6 February 2006, seven weeks after the spray event. In short, he saw the damaged crops in situ. Furthermore, Bullen has particular experience with the chemical Metsulfuron given he was the project manager for a number of studies conducted by the chemical company which developed it. He also has post graduate qualifications in cotton production and vast experience in investigating spray drift claims. I note that all of the reports of the experts relied, to a significant extent, on Bullen's report dated 10 July 2007. I also consider that Bullen's reports contain the most thorough analysis of the issues, including a very detailed assessment of the views of the other experts.

[74] Somervaille, whose expertise is in pesticide and crop performance, also conducted his assessments just two months after Bullen on 4 April 2006.

[75] Whilst I noted Dr Preston's undoubted expertise with weed management, he had no experience of dry land cotton. Neither had he any particular experience with the combination of the chemicals at the rates that they were actually applied. Neither did he have any qualifications in relation to the aerial application of herbicides. Whilst Dr Preston placed considerable weight on the absence of Picloram and Triclopyr in the samples, he nonetheless agreed that the crops and plants showed symptoms of damage from a herbicide of that nature. He also indicated that there were numerous factors which could influence the residue of chemicals found in a plant. I consider that the absence of Picloram and Triclopyr is not significant due to those factors. I also consider that the presence of 2, 4-D is not relevant.

[76] I have also relied heavily on the findings of the joint report of August 2009,[19] as well as the evidence given during the concurrent evidence session on 11 February 2010.

[77] The issue is really whether the combination of the applications of these products would have contributed to the quantity of the herbicide which was deposited, the distance it travelled and the extent of the damage that was caused to the Noonameena and Elgin cotton crops in 2005/2006. This, therefore, raises the contentious issue of spray drift and whether the chemicals could have reached the plaintiff's properties 20 kms away in sufficient quantities to cause the damage to the extent alleged.

[78] On 24 July 2009, orders were made at a directions hearing that the parties' experts meet and identify the matters on which they agreed and the matters on which they disagreed. On 28 and 30 July 2009 the experts met and their report signed at various dates in August 2009 is Exhibit 27.

[79] Whilst there were still significant areas of disagreement at the time that report was signed, the experts agreed as follows:

(1) That a spray event occurred on 15 December 2005 on the properties Sherwood and Wallumba.

(2) That the chemicals applied by Meandarra Aerial Services were as follows:

(3) That Brushwet would not have a direct impact on the cotton but may enhance the uptake of chemicals such as Metsulfuron and Grazon into cotton plants.

(4) That both Metsulfuron (Group B) and Grazon (Group I) would, if applied in sufficient quantities, adversely impact cotton.

(5) That the first reference of symptoms that were attributed to herbicide was made on 5 December 2005 when the plaintiff's agronomist noticed 'Hormone Damage in small areas of a couple of paddocks only'

(6) That symptoms which were later attributed to damage that is the subject of this litigation were first detected on 19 December 2005 by the plaintiff.

(1) That there is sufficient evidence to indicate that herbicide/herbicides damaged the cotton crops at Elgin and Noonameena in the 2005/06 season.

(2) The yellowing and stunting symptoms observed are typical of an exposure to a Group B herbicide of which Metsulfuron is in this group.

(3) That Auxin (Group I) like symptoms were also prevalent in the cotton crop of which the herbicide Grazon (Picker) is in this group.

(4) That the symptoms observed on 19 December 2005 if due to exposure to Metsulfuron would have been primarily caused by leaf uptake.

(5) That the symptoms of yellowing are consistent with exposure to Metsulfuron.

(6) Symptoms consistent with exposure to a member of the Group I herbicides (Auxin type symptoms) were evident in photos of 3.1.06.

Three of the five experts agreed that other than the aerial application at Sherwood and Wullumba, the claimed extensive area of damage to the Elgin and Noonameena cotton fields could have only have come from a ground rig application spraying if a very large area was spraying under still or inversion conditions. Tremain and Preston disagreed with this.

Assuming that Elgin 3 was not sprayed with a SU in previous seasons, residual levels of SU would not have been the cause of the symptoms observed in Elgin 3.

Contamination of spray equipment highly unlikely to have caused damage."

[80] Three of the five experts consider that other than the aerial application at Sherwood and Wallumba, the extensive area of damage to the Elgin and Noonameena cotton fields could only have come from a ground rig application if there was a large area sprayed under still or inversion conditions. However, it is generally agreed between the experts that there were no inversion conditions during the relevant period.

[81] Against that factual background of matters which have been agreed upon, it is clear that there is no consensus in relation to the other issues. The first issue, therefore, which needs to be determined by this Court, is whether the herbicides sprayed on Sherwood and Wallumba on 15 December 2005 actually reached the cotton crops on Elgin and Noonameena. Whilst Gordon considers that the herbicide spray did travel 20 km, both Preston and Tremain consider it "unlikely".

Did the spraying event on 15 December 2005 reach the cotton on Noonameena and Elgin?

[82] In order to answer this question, it should be remembered that all the experts agree that two of the herbicides sprayed on 15 December 2005, namely Metsulfuron and Grazon (which contains Picloram and Triclopyr), if applied in sufficient quantities can affect cotton and that the other chemical, Brushwet, can enhance the uptake of those chemicals.

[83] It is also generally agreed that cotton and wilga are susceptible to these herbicides. It is also clear that the cotton on Noonameena and Elgin was damaged by herbicides and that the symptoms were consistent with Metsulfuron and Grazon exposure.

[84] The plaintiff essentially argues that because of the quantities of chemicals used, the weather conditions and the height of application of those chemicals, the conditions were such that there was a drift event and the spray reached the cotton crops. The plaintiff alleges there was a pattern and gradient of damage in a south-westerly direction towards their properties. The plaintiff argues that this conclusion is supported by the crop phytotoxicity symptoms, the residue analysis of cotton and other vegetation, as well as spray drift modelling, using wind vectors.

[85] The sixth and eighth defendants initially argued that even though the relevant chemicals can cause the symptoms observed and even though those symptoms were observed in the cotton, there was no "mechanism" which could have moved those chemicals a distance of 20 kms from the area sprayed. The defendants' case altered during the course of the trial, with the defendants ultimately submitting that, even if there was some spray drift, not enough chemicals drifted from the spray site to the cotton fields to cause the damage claimed.

[86] The defendants argue that they did not breach their duty of care as they checked the wind, they ceased operations for two hours when the wind increased and they sprayed at acceptable levels of 1.5 to two metres above foliage. The evidence of both pilots is, therefore, crucial to an understanding as to what occurred on the day of spraying and I will refer in some detail to the evidence they gave in relation to some important factual issues.

[87] The aerial spraying was conducted on 15 December 2005 by Lachlan Hill, the eighth defendant, and Michael Baker, who was a pilot employed by the sixth defendant. Spraying initially commenced about 6.40 am and finished at 8.00 am, and then resumed between 9.50 am and 1.30 pm. The interval of almost two hours occurred as conditions were not considered appropriate for spraying by the pilots.

[88] Both pilots gave evidence and indicated that they were aware that the plaintiff grew cotton on its properties and they were aware of the location of those properties. Another crop dusting pilot Gale who operated in the Condamine area was called by the plaintiff to give evidence about flying conditions on 15 December 2005. However, given the defendants' objections to his evidence, I have not taken this evidence into account.

[89] Baker's evidence was that this was a difficult spraying exercise for a number of reasons, which included the terrain on the properties, the nature of the work they were undertaking and the weather conditions on the day. He also stated that he was not aware that they were spraying "off label" products. He indicated that he subsequently became aware that the spraying had caused damage on the neighbouring property and had seen photos of damaged trees.[20] Baker's evidence was in the following terms:

"Because you were aware that with an off-label mix and a mix such as this that it can increase the risk factor of damaged crops, can't it?-- Yes.

But you were given no advice whatsoever beforehand; you were just told to do the job?-- Yes.

There's trees of six metres in height, thereabouts?-- Yeah.

So it's an open canopy, undulating, with some very tall trees, I suggest?-- Yeah, some tall trees, yeah.

Have you been shown photographs of the damage to the trees? - I've seen them myself when I was driving in there.

And, in fact, you would have had to clear those trees, I suggest, by at least a couple of metres?-- Yes.[21]

All right. Now, you agreed with me before lunch that the spray should not have drifted on to Paul Keys's property, should it?-- No.

Shouldn't have gone that far, should it?-- Well, no, it shouldn't have.

That - that was unexpected, wasn't it?-- Yes.[22]

And it's unacceptable to - for it to drift further than the target area, being the properties at Sherwood and Wallumba; isn't that the case?-- Yes.

It wasn't supposed to go further than Sherwood and Wallumba?-- No, and I didn't think it was going further.

No, and if it has, it means something's gone wrong; would you agree?-- Well, yes.

Windier than you thought, or the temperature higher, or the humidity lower perhaps, would you agree with that, that those factors can affect spray drift?-- Yes.[23]

Okay. Why was it a particularly hard job?-- Because I've never - never had to spray trees and - that's what the job was, and probably - well, I don't like to kill trees. Like, I'm, you know, but that's just a part of my job, and it was a delicate job and that's - that's what I'm in the job to do what the best of my ability and that's what I thought I was doing that my - the best of my ability and that's why I got up in the morning, up to have a look at it to try and sort everything out, to get everything right.[24]

You weren't? Well, could you please explain how it is you've got trees there that are burnt at the top, burnt throughout?-- Well, it was probably a - I've come and you've got vortexes on the side - end of your wingtips and I was - I might have went too close and the vortex have come out and swirled, like you see on a Jumbo jet, and they've come up, but I have not gone and I have not sprayed over trees.

But, Mr Baker, even if that explanation is true, that shouldn't happen, should it?-- I - I obviously have just misjudged it and with the - it's the - it's the precision of - of agricultural flying, it is a very fine line of - of - to become fully, like, aware of everything.

Okay. That's something that should not have occurred, you've acknowledged, and was unintended; is that correct?-- That was unintended.

Like spray drifting on to the Keys' property and the Ronnfeldt's property?-- If it was, it was unintended.

Yeah, and it should not have occurred?-- I didn't want it to occur, no.[25]"

[90] Lachlan Hill also gave evidence in relation to the chemicals used as well as the conditions on the day. His evidence was as follows:

"And you don't disagree with that, do you? That even for the operation you were carrying out it was an off label product? - I'll accept it was off label.

Yes. And, in fact, the use of such chemicals here, metsulfuron with the grazon, was such that you're not to use it with a fixed winged aircraft?-- That's correct.[26]

Well, you never saw a permit, did you?-- No.[27]

Well, did you see, though, that you as a pilot still have a duty to see that if everything's appropriate?-- I'd accept that.

But you just didn't bother?-- You could say that, I suppose.[28]

And, in fact, you're aware here that in fact you also had in the mix that you were spraying brushwet?-- That's correct.

And you were aware that that was designed to enhance the uptake of the herbicides that it was being sprayed with, or mixed with? You're aware of that?-- I am. That's the - that's the - sorry.

And you were aware, I suggest, that that was - that is the brushwet at a level 10 times the maximum label rate?-- Yes.

And you didn't think that any of these circumstances were a bit unusual and if ever there was a case to follow the recommendations on the label this was such a case?-- Quite obviously, as I stated earlier, is why the job was rejected initially.[29]

MR HOWE: I suggest to you - I put it to you that you should not spray, particularly a chemical mix such as this, where the wind speed is in excess of 15 kilometres per hour?-- We weren't spraying with an average wind speed in excess of 15 kilometres an hour.

Were there periods of time that spraying occurred when the wind speed did exceed 20 kilometres per hour?-- No.

No? I see. And if that did occur would you agree that that would be unacceptable?-- If it did actually occur-----

Yes?-- -----I would think that's probably above the acceptable limit, yes.

Because I'm putting to you again even with a normal mix of chemicals in the normal rate and concentrations, not that we had here, the acceptable range is four to 15 kilometres per hour?-- It depends on what you're applying.[30]

So, you'd agree that if there were times when you were exceeding 20 kilometres per hour that would have been unacceptable?-- Oh, yes. I agree.

And that would have posed an unacceptable risk of spray drift, wouldn't it?-- Not necessarily.[31]"

[91] Having considered the evidence of the pilots, it is clear that an understanding of the actual weather conditions operating on the day is vital to the question as to whether the sixth and eighth defendants were negligent.

[92] There are a number of weather stations in the area surrounding the plaintiff's properties. Those stations are located at Hopelands, Alderton and Miles. They are all calibrated weather stations that log weather recordings every 15 minutes. The maximum wind speed at Hopelands during the spraying period was between 23 and 35 kms per hour (kph) and at Alderton it was between 17 and 29 kph.

[93] The evidence indicates that the Department of Primary Industries (DPI) preferred wind speed is between four to 15 kph for aerial spraying. On the basis of the weather station recordings, the wind speeds were above the preferred upper limit of 15 kph by 8.00 am.

[94] In relation to the wind direction, the Hopelands' weather station recorded wind direction as generally north/north-easterly and Alderton recorded generally winds from a northerly to north-easterly direction.

[95] In relation to humidity, Hopelands recorded the humidity dropped to 50 per cent by 11.30 am and Alderton dropped to 50 per cent at 10.45 am.

[96] The temperature was above 30 degrees at Alderton at 10.30 am and at Hopelands at 11.00 am.

[97] In my view the records of the three weather stations are relatively consistent with each other. Those records however are not consistent with the pilots' log books of the prevailing weather conditions at the time they sprayed. The flight logs indicate that the herbicide applications to the two properties occurred between 6.40 am and 1.30 pm on 15 December 2005 and that two fixed wing aircraft were used. The pilots' logs indicate that temperatures during that period ranged from 24 to 32 degrees Celsius and the pilots logged the wind as easterly and south-easterly to north-easterly. The wind speed recorded by the pilots was between 03 to 05 knots,[32] with relative humidity readings between 55 per cent at 7.30 am and 75 per cent by 9.00 am.

[98] The recordings from the weather stations which surround the relevant properties do not support the recordings of the pilots. I therefore need to determine which evidence I prefer.

[99] Each pilot kept his own log and those log books[33] indicate that one plane did a total of six loads and the other did eight loads. Those logs also record such things as temperature and humidity, as well as wind speed and direction. The difficulty with the pilots' records in the logs is that they are not consistent with each other. I adopt Mr Howe's summary which indicates the extent to the inconsistency as set out below:

[100] It is clear that the readings for temperature, humidity and wind speed and direction vary greatly even though the logs show the recordings were essentially made by each pilot within minutes of each other. Hill has conceded that those logs are not an accurate record of the conditions on the day. His evidence was as follows:[34]

"And you've been, even before the commencement of this trial, advocating and presenting a case that what you did was correct, because of what's in your job cards; isn't that the case?-- Oh, I guess so.

Mmm, but the job card doesn't correctly record, even on your own case, what occurred, does it?-- Not exactly, no."

[101] Accordingly, it is clear that the weather station records and the pilots' records are not consistent. In the circumstances, therefore, given the obvious inconsistency between the log books of the two pilots, as well as the concession that the log books are not strictly accurate, the evidence I must rely on is the objective evidence from the three weather stations. That is the only clear evidence of the weather conditions on the day.

[102] On the basis of that information and the DPI guidelines as to when it is appropriate to spray, it is clear that the spraying on 15 December 2005 did not occur in optimal weather conditions. On the basis of the temperature, wind speed and humidity during that day as recorded by the weather stations, I consider that for a significant period of time the weather conditions during spraying were, in fact, inappropriate for aerial spraying.

[103] The pilots both gave evidence that they sprayed at a height about 1.5 to two metres above wattle regrowth. The evidence indicates that the wattle regrowth was some three to four metres above ground. The pilots' evidence was that they were releasing above the wattle regrowth and then either shutting off over the trees or flying around them.[35] This arguably gives a release height of some four to six metres.

[104] The evidence of Hill in this regard was as follows:[36]

"The company policy is I would like to see them flying at one half to two metres above the canopy.

What would be the release height above the ground?-- Well, that obviously becomes around that five - four and a half to six metres above the ground.

Did you ever make a release as high as 15 metres?-- Never.

Did you go around the individual trees?-- Where it's appropriate individual trees you would go around. If the trees got to the point where there's clump where you couldn't fit through the gap or something you'd be going over them. On the lower tree clumps there that may be only four to five metres above your spray height you might just turn and fly over if they're wide enough and then descend back down and start spraying on the other side. If they're a big, tall skinny tree line you may have to - in which case happens quite regularly, you'll spray up to it, pull up. You've got no chance of pushing so you do an orbit and come back on the other side. It's pretty well like effectively you're nearly spraying two paddocks.

You shut off at the top?-- Shut off at the deck, pull off. To finish the paddock off you would do a rundown parallel to the tree line either side to get the area that you've obviously missed on your climb out."

[105] The reports of both Bullen and Somervaille assumed a much higher release height in the order of 10 to 15 metres (Bullen) or 15 to 20 metres (Somervaille).

[106] It is clear from the photographs which were tendered that the terrain varied, as did the wattle regrowth and the tree height. There were some tall trees on the properties sprayed and the estimates are that they could have been up to 18 metres tall.

[107] I therefore find it difficult to accept that, given the terrain and the height of some of the trees, the spraying occurred uniformly at such a low level. Furthermore, given the damage to some of the tall trees, I consider that the release height varied and at times may have been considerably higher than six metres. I do not consider that there was a uniform release height of 1.5 to two metres above the wattle regrowth and that the release height had to have been considerably higher than six metres to have caused damage to the trees, which appear in the photographs which are in evidence.[37]

[108] I consider that the evidence supports a finding that the release height was above six metres. Accordingly, I consider that the height of three metres used by Tremain and Preston in the AVPMA model to be inaccurate.

[109] The quantity of spray is obviously a factor and the evidence is that two planes were used which involved 50 spray lines. Bullen's 10 July 2007 report also states that the herbicides were applied to 300 hectares of Wallumba and 120 hectares on Sherwood. The pilots sprayed for a period of time of about five hours each.

[110] I consider that the sixth and eighth defendants were aware that aerial spraying of the chemicals was likely to cause harm to susceptible crops and trees within the proximity of the spraying. The label set that out clearly. There is no doubt that the chemicals were dangerous and in fact the eighth defendant had earlier refused to spray as the chemical application proposed was "off label".

"... drift is on the forefront every agricultural pilot's mind. It's knowing what susceptible crops are in the district, where the wind's going, what we're spraying, everything - drift is one of the major issues that we've got consider whenever we go and spray a crop ...".

[112] The question of the likely amounts of product that could move away from an aerial application occupied a great deal of time during the trial. In view of the importance of this issue and the disparity of views, an order was made on 8 November 2009 for a Joint Report to address this issue and for evidence on this issue to be given concurrently at the resumed hearing on 11 February 2010.

[113] The defendants' experts Tremain and Preston had quoted model outputs generated from a model known as APVMA but, as I have already indicated, I do not consider the release height used in the model is accurate, given the conditions operating in this case. Furthermore, the model is only valid to 800 metres.

[114] Agdisp uses two models to predict movement of spray materials. Initially it uses the Lagrangian Model and it has been validated out to 800 metres and is considered to be highly accurate. Beyond 800 metres it transfers data on the airborne portion of the spray material to a Gaussian Diffusion Model, which predicts movement of that material out to 20 kms from the release point and, whilst not as absolutely accurate, it was argued that it is a useful guide to potential deposition.

[115] Initially there was disagreement as to the reliability and value of this different model to demonstrate the likelihood of Metsulfuron being deposited 20 kms from the point of release.

[116] Gordon believes that the Agdisp model is reasonably accurate and can serve as a useful guide as to whether chemicals could have deposited at distances downwind up to 20 km from the point of release. He stated:[39]

"The whole point of the modelling exercise as far as I was concerned was to determine under a general set of parameters is it possible for the chemical to move that distance? I would not attempt to use a model to predict a quantity. My original model when I prepared that report was purely - the only reason I used the Agdisp model was the other two experts reports, one of Mr Tremain and one of Mr Preston, used or tried to put into evidence that the model predictor couldn't go 20 kilometres from the original report. So I used the model to show well, yes, it can, but I do not believe the model should be used for the purpose of trying to quantify an amount that would land at that distance. So, there will be a range of parameters and I'd probably agree with all of them. I don't think you can encompass the range of parameters that are likely to have occurred on that day and I cannot tell you what they were, therefore, I could not use the model to predict a quantity with any certainty. That quantity, we predict, is probably just useful as a guide."

[117] Gordon considered, however, that the model can provide a reasonable prediction of potential deposition and that the modelled predictions are likely to be within an order of magnitude of what the actual deposition is likely to have been (that is, they may be approximately 10 times higher or lower than the Agdisp prediction).

[118] Tremain, however, believes the Agdisp model far field predictions overestimate the actual amount of deposition on the subject cotton fields for the following reasons:

● The distance to the subject cotton fields was generally more than 20 km.

● The average wind speed was less than five metres per second according to the Miles, Trenmore and Alderton weather records.

● The Agdisp predictions are greater than those calculated by using the chemical aloft at 800 metres spread out over the entire spray plume. As these latter calculations also overestimate actual deposition and are more reliable than the far field Agdisp predictions, the Agdisp predictions must also be overestimated.

[119] In particular, Tremain argues that the Agdisp estimates, based on a canopy height of 0.5 or 1.0 metres, substantially overestimate deposition because they do not accurately reflect the high proportion of chemical that is deposited on the spray block with a greater canopy height.

[120] He also argues that the actual deposition cannot be approximately 10 times higher than the Agdisp predictions, as suggested by Gordon. Whilst the amount of chemical airborne within 800 metres of the spray block is relatively accurately modelled by Agdisp and has been validated for larger far field depositions than predicted to have occurred, the amount of airborne chemical would have been much higher than was actually present. The Agdisp calculations show that most of the airborne chemical is deposited before the 20 km mark. Tremain considers that there is simply insufficient airborne chemical remaining to result in actual depositions which are significantly higher than predicted values. He also argues that the calculations using the chemical aloft at 800 metres spread out over the entire spray plume overestimate the actual amount of deposition on the subject cotton fields.

[121] Accordingly, a joint report was commissioned and a concurrent evidence session was proposed on the resumption of the trial in February 2010 for the experts to answer the following question:

"Do the experts agree that the Agdisp Model predicts that it is likely a quantity of herbicides, predominantly Metsulfuron could have been deposited a distance 20 kilometres from the point of release?"

[122] The experts agreed on the following facts for the purposes of the Joint Expert Report of 18 December and the concurrent evidence to be given on the 11 February 2010 in relation to the Agdisp model.

1. A windspeed of five metres per second;
2. Spray quality of fine to fine medium;
3. An average release height of six metres;
4. A canopy height of 0.5 to one metre at the lower end of the range to four metres at the higher end of the range;
5. Spray lines being 50 lines.

[123] Whilst acknowledging that the Agdisp model has been validated to 800 metres, the experts ultimately conceded in the Joint Report that the Agdisp model does confirm that some chemicals could have been deposited 20 kms from the site of aerial spraying. Gordon and Tremain specifically agreed with this conclusion at the concurrent evidence session, but disagreed as to the quantities which could have been deposited.

[124] I consider, therefore, that the Agdisp model was a useful guide to potential deposition. I also consider that the agreed facts were a useful basis from which to estimate whether conclusions could be drawn. Accordingly I agree with Gordon's statement in his report dated 31 July 2009, that the Agdisp model does indeed indicate the possibility of the deposit of some chemical at a distance of 20 kms.

[125] I am therefore satisfied that it was more probable than not that a quantity of the chemicals sprayed on 15 December 2005 at Sherwood and Wallumba reached the plaintiff's cotton fields at Noonameena and Elgin.

[126] I also accept that the evidence is that Brushwet would increase the uptake of the chemicals sprayed and that Brushwet was used at 10 times the recommended label rate. It was also used in combination with Metsulfuron which was sprayed at 14 times the recommended label rate.

[127] I also consider that there can be subclinical damage which occurs at less than 1/10th of a gram.[40] This is clear from the evidence of Dr Somervaille and in Exhibit 27 he states that the worst damage would be caused by .5 of a gram. He said that there was insufficient information to accurately work back from residue data to project spray drift rate that deposited onto the cotton crop.

[128] Bullen gave a substantially lower figure at which one can have damage and that is 0.0017 grams per hectare.[41] Preston accepted that with 1/10th of a gram you can still get injury with a susceptible species[42] and he stated he would defer to Dr Somervaille concerning subclinical damage.[43] Furthermore Dr Preston has not done any studies on the effect of such rates on dry land cotton.[44]

[129] Given the variables which were operating at the time and the disparity of views as to the level of chemical which could be deposited, I do not consider that a reliable estimate of the amount of chemical which could have been deposited can be made using the Agdisp model. Whilst I cannot make a finding as to the level of chemical which could have been deposited, I accept that some of the herbicide was deposited.

[130] The duty which the defendants owed is set out in Bonic v Fieldair (Deniliquin) Pty Ltd & Ors:[45]

"[23] All the defendants would have been aware that aerial spraying of the weedicide 2, 4-D was likely to cause harm to any occupier of a property in the proximity which had susceptible vines, trees and plants on it, if adequate care to avoid harm was not taken. The labels alone made that plain. It cannot be in doubt that it was known to the fourth defendants that the weedicide was a dangerous chemical and that care in its use must be taken to avoid harm.

[24] In Burnie Port Authority v General Jones Pty Ltd, the majority made the point that, where activities were carried out which involved the handling or storing of dangerous goods, the duty to take care would not necessarily be discharged by the employment of a competent independent contractor and that each person had a duty to ensure that reasonable care was taken. At 550, the majority said:

'It has long been recognised that there are certain categories of case in which a duty to take reasonable care to avoid a foreseeable risk of injury to another will not be discharged merely by the employment of a qualified and ostensibly competent independent contractor. In those categories of case, the nature of the relationship of proximity gives rise to a duty of care of a special and "more stringent" kind, namely a "duty to ensure that reasonable care is taken": see Kondis v State Transport Authority [1984] HCA 61; (1984) 154 CLR 672 at 686. Put differently, the requirement of reasonable care in those categories of case extends to seeing that care is taken.

Viewed from the perspective of the person to whom the duty is owed, the relationship of proximity giving rise to the non-delegable duty of care in such cases is marked by special dependence or vulnerability on the part of that person: The Commonwealth v Introvigne [1982] HCA 40; (1982) 150 CLR 258 at 271, per Mason J.'

[25] In the present case, the nature of the chemicals being sprayed was such that all persons responsible for the operation, that is the four defendants, had a non-delegable duty to ensure that properties in the vicinity of the Rendell land were not damaged by the spraying."

[131] I consider that the reasonable man in the position of the defendants would have foreseen that their conduct in spraying off label concentrations of herbicides in inappropriate weather conditions involved a risk of injury to the plaintiff. I am satisfied that, given the nature and combination of the chemicals being sprayed, the sixth and eighth defendants had a non-delegable duty to ensure that the susceptible cotton crops growing on the plaintiff's properties, which they admitted they were fully aware of, were not damaged by the spraying.

[132] I consider that the defendants breached the duty of care they owed to the plaintiff by spraying the quantities and combinations of chemicals in the weather conditions which prevailed on 15 December 2005.

Was there a sufficient quantity of chemical drift to cause damage?

[133] Having established that the defendants breached their duty of care to the plaintiff the question which arises next is whether the breach of that duty resulted in the chemicals damaging the cotton crops. The question is really whether it is more probable than not that the chemicals which did drift actually caused the herbicide damage observed at Noonameena and Elgin. Or put another way, is it more probable than not that all of the damage was due to the drift of chemicals sprayed on 15 December? The defendants point to the existence of other chemicals which they submit may also have been operating to cause the damage.

[134] There was evidence that some of the fields at Noonameena and Elgin had been sprayed with 2,4-D and the defendants argue that this may have caused some of the damage to the plaintiff's cotton. I am satisfied, however, that the evidence indicates that previous use of this chemical had never before resulted in yield loss. Preston's evidence in this regard was that he did not consider it was a "chief factor" or that it was responsible for "the strong symptoms that were in the photographs". There was also clear evidence that it was the Metsulfuron which was responsible for the significant damage observed and not an auxin, such as 2,4-D, Picloram or Triclopyr. On the basis of that evidence, I am not satisfied that 2,4-D was responsible for the crop loss.

[135] I also consider that the evidence of Ed Offner, the agronomist, also supported the plaintiff's case. His evidence was that on 5 December he noticed some damage to the cotton and it appeared to be minor 2,4-D damage. He considered it improved by 12 December 2005. His evidence was that on 19 December there was some worsening of the crop.[46] The symptoms were a yellowing and that by 22 December what he detected was a yellowing which was getting worse.[47] His evidence was that he had not seen this type of symptom before. This also corroborates Ashley Geldard's evidence, "it was yellowing, and it was getting worse, it was not consistent with 2,4-D damage".

[136] Another point raised by the defendants, but not actually pleaded, is that the damage was caused by previous Metsulfuron used by the plaintiff. However, it is clear that there is damage in Elgin 3 where Metsulfuron has never been used. The agreed findings of the Joint Expert Report of August 2009 included a finding that "Assuming that Elgin 3 was not sprayed with a SU (residual soil sulfonylurea) in previous seasons, residual levels of SU would not have been the cause of the symptoms observed in Elgin3". I also accept that if there was damage from the plaintiff's use that it should have been seen in strips given there was a history of 17 metre strips being used. Furthermore, if Metsulfuron in the soil residue was an issue, then one would have expected to see these symptoms early in the crop and that was clearly not the case. The symptoms were not observed until after nodes six to eight. I consider that the damage to the plaintiff's cotton was not caused by previous use of Metsulfuron.

[137] The defendants also referred to the absence of Picloram and Triclopyr in the samples. I do not consider, however, that this is inconsistent with the plaintiff's case and can be explained by the rates and quantities that were used, as explained by Somervaille and Bullen in their reports.[48] Furthermore, it is clear that symptoms consistent with damage from these herbicides were actually observed in the crops. Preston himself indicated that herbicides can be detoxified through metabolism and that this occurs at different rates with different plants. Furthermore, the means of detection and quantification limits for Metsulfuron were five times lower than the other herbicides.

[138] The defendants also argued that the insufficiency of samples, the pattern of damage in the crops and the gradient of damage does not support a finding that the chemicals in the drift caused the damage. The defendants also referred to the evidence of the owners of some of the neighbouring properties that their properties had not been damaged and that their trees and vegetables were unaffected.

[139] First, I have to indicate that I consider that the photographic evidence clearly and overwhelmingly established extensive damage to the cotton crops. I consider that the photographic evidence was not only detailed and contemporaneous but that is was convincing evidence of the damage caused by herbicides.

[140] Not only were there contemporary photographs but there was samples taken shortly after the spraying event. The samples substantiated the fact that herbicide damage, as indicated by the photographs, had in fact occurred. Whilst there were not a lot of samples taken, this was due to the cost of the samples. I note that Bullen stated that the cost per sample was in excess of $2,000. I consider that the samples taken in fact support the plaintiff's case. I note Bullen's view that, "There were however, adequate samples in my opinion to show a gradient source of spraydrift and cause of damage by presence in cotton sample".

[141] The defendants also rely on the fact that the symptoms occurred more quickly than would normally be expected from Metsulfuron and that this militates against a finding that Metsulfuron was to blame. However, I accept Somervaille's evidence that in summer months you can get rapid expression of symptoms. Furthermore, Preston conceded that it was potentially possible to pick up symptoms in four days if it was detected by an educated observer.

[142] On balance, I am satisfied that the samples taken not only support the presence of herbicides but support the plaintiff's claim that the damage was caused by the herbicides sprayed on 15 December 2005 at Sherwood and Wallumba. On my analysis of the evidence, I consider that it supports a finding that there was damage to the wilga trees from a north-east direction. I am therefore satisfied that the pattern of damage supports a finding that there was a drift of chemicals from the north.

[143] Accordingly, whilst I am satisfied that there was a drift of an unknown quantity of chemicals from the spray event on 15 December 2005, the critical question is whether I am satisfied that there is sufficient evidence to support a finding that it is more likely than not that the actual damage to the crops was caused by the drift of these specific but unquantifiable chemicals from the spray event on 15 December 2005.

[144] The relevant questions which should be posed in answering such a question were recently posed in the New South Wales Court of Appeal decision in Flounders v Millar:[49]

"16 In Bendix, Mason P emphasised (at 318) that a party who negligently exposes a plaintiff to a risk of injury will not be liable unless the plaintiff can persuade the trier of fact that it was probable that the risk came home. His Honour said that the basal principle remains: '[t]he law never gives judgment in favour of a plaintiff when the only finding is equally consistent with liability and non-liability'. Beazley JA held (at 339) that the onus remains on the plaintiff to prove causation on the balance of probabilities and stated: '[t]he onus is not discharged by establishing that a particular matter cannot be excluded as a cause of the injury'. Stein JA dissented on the issue of causation.

17 Mason P's statement in Bendix echoed what was said in Luxton v Vines. His Honour concluded that to discharge the onus of proof in regard to causation, it was not sufficient for a plaintiff to prove that the defendant negligently exposed the plaintiff to a risk of injury. Liability depended upon the plaintiff persuading the trier of fact that it was probable that the risk came home.

18 In Wallaby Grip v Macleay Area Health Service (1998) 17 NSWCCR 355, Beazley JA (with whom Sheller JA and Sheppard AJA agreed) followed Bendix, saying (at 362 to 363, [16]) that it 'applied well-known principles of causation' and that '[a] plaintiff must show more than that it is possible that injury was caused by the defendant'.

19 In E M Baldwin & Son Pty Limited v Plane & Anor (1998) 17 NSWCCR 434, Fitzgerald AJA (with whom Meagher JA and, in substance, Beazley JA agreed) referred to the relevant authorities, including those I have cited, and said (at 473, [77]):

'All relevant circumstances, including an increase in the risk to the plaintiff from the defendant's breach of duty and the character and sequence of events, must be considered in deciding whether a defendant's breach of duty which is a possible cause of the plaintiff's damage probably materially contributed to that damage. Circumstances are relevant for this purpose if they assist in establishing or strengthening a causal connection between the defendant's breach of duty and the plaintiff's damage according to expert opinion or by the application of logic, common sense or experience.'"

[145] Counsel for the sixth and eighth defendants provided further written submissions on 19 March 2010 regarding the High Court decision of Amaca v Ellis.[50] That decision related to a claim in negligence brought by the estate of Mr Cotton, who had died of lung cancer after being a smoker for over 26 years and also being exposed to respirable asbestos fibres during the course of his employment for some 15 years. No scientific or medical evidence could particularise what the cause of the lung cancer was. The central question in the case related to causation and whether it was more probable than not that the defendant's negligence (exposing Mr Cotton to asbestos) was the cause of the lung cancer. The claim ultimately failed and it was held per curiam that although exposure to asbestos may have been a cause of the cancer, it was not a probable cause.[51]

[146] Counsel for the sixth and eighth defendants relied on the decision in Amaca to reinforce the importance of causation and the onus on the plaintiff in this regard. Counsel for the defendants argued that the plaintiff's submission that the defendants were negligent was a conclusion based solely on the order of events, without ruling out other factors which could explain the connection. They submit that in the present case, as was the case in Amaca, there is insufficient evidence to support a positive finding of causal connection.

[147] I agree with Counsel for the plaintiff, that Amaca merely applies settled legal principles to the particular facts of that case. I do not consider that Amaca is of any particular assistance, given the particular factual circumstances operating here. Unlike the situation in Amaca, in this case there is no evidence of competing causes that could explain the crop damage.

[148] In order to come to a conclusion on the issue of causation, I will summarise the factual matters of which I am satisfied.

[149] I consider the evidence supports the following findings:

(i) The sixth and eighth defendants conducted an aerial spraying operation on 15 December 2005 on the properties Sherwood and Wallumba to kill wattle regrowth on 420 hectares of land.

(iii) Brushwet would not have a direct impact on the cotton but may enhance the uptake of chemicals such as Metsulfuron and Grazon into cotton plants.

(iv) Metsulfuron (Group B) and Grazon (Group I) would, if applied in sufficient quantities, adversely impact cotton.

(v) The chemicals were not mixed to the correct ratios.

(vi) The correct quantities of chemicals were not used. Metsulfuron was applied in excess of 14 times the correct rate. Brushwet was applied in excess of 10 times the correct rate.

(vii) For a significant period of time the weather conditions were inappropriate for aerial spraying.

(viii) The chemicals were not appropriate for spraying from fixed wing aircraft.

(ix) The weather station information indicates that for a significant period of time the wind was from a north-easterly direction.

(x) There is damage to tree lines between Sherwood and Wallumba and the plaintiff's properties on the northern side of the trees consistent with damage from the chemicals in question.

(xi) There was spray drift from the chemical sprayed on 15 December to property Toston which should not have occurred if the conditions had been suitable.

(xii) The terrain and tree height varied, with some trees as high as 18 metres.

(xiii) The release height of the spray also varied and was higher than six metres at times.

(xiv) All experts agree that it is theoretically possible for a quantity of herbicide, predominately Metsulfuron, to have drifted 20 kms from the spray site.

(xv) Three of the five experts agreed that other than the aerial application at Sherwood and Wullumba, the claimed extensive area of damage to the Elgin and Noonameena cotton fields could have only have come from a ground rig application spraying if a very large area was spraying under still or inversion conditions.

(xvi) The conditions on the day were not still or inverse conditions.

(xvii) It is not possible to estimate with any accuracy the exact quantity of Metsulfuron which drifted.

(xviii) The cotton crops at Elgin and Noonameena were damaged by herbicides in the 2005/06 season.

(xix) The yellowing and stunting symptoms observed on the cotton are typical of an exposure to a Group B herbicide which includes Metsulfuron.

(xx) Auxin (Group I) like symptoms were also prevalent in the cotton crop and the herbicide Grazon (Picker) is in Group I.

(xxi) The damage to the plaintiff's cotton was not caused by previous use of Metsulfuron or the application of 2,4-D earlier in the season

[150] As the case law clearly establishes, it is not sufficient for a plaintiff to prove that the defendants negligently exposed the plaintiff to a risk of injury. It is clear that I need to be persuaded that it was probable that the risk came home.

[151] On the basis of the findings I have made which are set out above, I am satisfied that it is more probable than not that the damage to the plaintiff's cotton was caused by the spray event on 15 December 2005 at Sherwood and Wallumba.

[152] The historical cotton yield averages for 2003/2004 and 2004/2005 for Cullingral were 4.06 and 1.46 bales per hectare respectively. The Noonameena cotton averages for the same years were 3.82 and 1.75 bales per hectare respectively. The Elgin farm averages for those seasons were 3.82 and 1.75 bales per hectare.

[153] The following graph[52] sets out the relative yields in 2004, 2005 and 2006.

[154] In my view the graph indicates that while the yields were similar on all three properties in 2004 and 2005 there was a major difference in the yield as between the properties in 2006.

[155] Cotton was harvested by Columboola Cotton staff in the harvest period on all the fields at Noonameena, Elgin and Cullingral between 30 March 2006 and 8 June 2006. This was an extended harvest period due to the different growth rates between the affected fields and Cullingral and led to some extra production costs.

[156] The following graph[53] indicates the "in crop" rainfall during the relevant period on the cotton fields at Noonameena, Elgin and Cullingral.

[157] Somervaille states that the exact extent of the effect of the herbicide on the cotton at Noonameena and Elgin is not known as there is no unaffected cotton in or adjacent to those fields. He concludes, however, that:[54]

"a good indication of the levels of production may be provided by boll counts and lint production from a Collumboola property upwind from the site of application of metsulfuron and Grazon DS that was subjected to similar management to the crops on Noonameena and Elgin and experienced comparable weather conditions for much of the life of the crop."

[158] Bullen has produced a Quantum Report dated 20 March 2007 together with an addendum contained in his report of 30 July 2009. Those reports address the plaintiff's loss.

[159] Tremain however, in his report of 27 March 2009 and his addendum notes of 3 July and 22 July 2009, argues that many of Bullen's assumptions are incorrect. He disputes the starting moisture levels and argues that there is no substantiation for some of the costs. He also argues that the loss of lint quality was not as great as argued. He concedes, however, that the fields claimed have been checked against aerial photographs and "appear to be relatively accurate". He also argues that Cullingral is not a suitable "control property".

[160] Tremain also states that Bullen quantified spray damage in two ways, namely the number of affected nodes and the phytotoxicity scores. He notes that the number of nodes affected in each field varied from 3.2 to 14.1, but that despite that range, the relative yields did not vary. He argues that if the claimed reduced yield was caused by chemical damage, there should be a strong relationship between the number of nodes affected and the relative yield. Tremain also notes that the phytotoxicity score varied from 0.39 to 2.03 but despite that range, the average yield of the worst affected field was almost identical to the least affected field. He therefore concludes that the yield reduction was due to factors other than chemical damage.

[161] However, as Bullen points out[55]:

"Mr Tremain conducts comparison of relative cotton yields to average nodes affected and average phytotoxicity scores. It was the conclusion of Mr Tremain from chart 1 and chart 2 that crops with no nodes affected or no phytotoxicity score are expected to yield between 54 and 52%.

This is based on comparing individual field yields whereas quantification of loss conducted by Freemans Claims and Insurance Services, Agriculture Division, in the Quantum Report of the 20th March 2007 is clearly based on farm map average analysis methodology comparing Noonameena and Elgin farm average to Cullingral farm average.

It is therefore mathematically correct that Mr Tremain has identified that approximately 50% yield potential of non damaged fields can be expected as the yields portrayed were in fact a farm average over damage levels.

Subsequently in re-calculation of the claim in Section 6.8, Mr Tremain has applied a 54% of potential yield to farm average yield figures. This has the obvious mathematical effect of applying a 54% discount to already averaged yields."

[163] Tremain's report also assumes Elgin 3 was undamaged. It is clear that some fields visually appeared to be less damaged than others however I accept that all fields in Noonameena and Elgin were damaged given the extensive evidence contained in Bullen's reports. Tremain also conceded in his oral evidence that there was damage, in fact, in all the paddocks at both Noonameena and Elgin but that some paddocks were "less affected".[56] He also agreed that there were different subsoil moisture levels operating and that it can affect yield.

[164] It seems to me that the quantification of loss by Bullen was clearly based on whole farm averages and that a farm average comparison methodology was used to establish both potential yields and for the purpose of loss quantification. I accept that the APSIM model[57] demonstrates that yield potential for Noonameena and Elgin was equal to that for Cullingral. I accept the evidence that there is a huge disparity in the actual yields predicted by the APSIM model for Noonameena and Elgin compared to potential yields. It is clear that the APSIM model actually predicted a potential yield at Cullingral that was close to the actual yield obtained. It seems, in fact, that the actual yield at Cullingral was higher than the yield predicted.

[165] I also consider that the evidence indicates that the soil test moisture levels before the event in question show that Noonameena and Elgin had slightly better soil moisture content than Cullingral. The nutrition factor in Elgin, Noonameena and Cullingral was also clearly similar.[58]

[166] I accept that Cullingral is an appropriate comparable property, it is being managed in the same way, has similar variability and similar soils. Offner's impression was that Cullingral was superior but that he did not know what the yield differences were. The soil moisture records also indicate that Noonameena and Elgin were slightly superior. The APSIM model showed that yield results should have been at least equal if not better. As Tremain conceded, the issue of the quantification of the loss is a difficult issue. Ultimately I accept Bullen's methodology. Cullingral actually yielded 1.91 bales per hectare but Noonameena and Elgin yielded 0.967 bales per hectare. I accept that this is a loss of 0.943 bales per hectare.

[167] Clearly underpinning this analysis is the fact that I am satisfied that an event which explains the difference between the yield predicted for Cullingral and the yield achieved and the yields predicted for Noonameena and Elgin and the yields actually achieved is that Cullingral never suffered spray drift damage.

[168] Bullen's report of 20 March 2007 sets out the loss. He is an experienced loss assessor who has conducted many such assessments in the Condamine region of Queensland. He has examined all the records and compared the fields before and after December 2005. He has inspected the properties and the crops. Significantly, Bullen carried out plant mapping assessments in February and March 2006. He also examined the Harvest Protocol for the 2006 cotton harvest for Columboola Cotton. It is clear, therefore, that I prefer Bullen's analysis to that of Tremain. In my view, he has carried out a detailed analysis and fully explained his methodology which, as I have indicated, ultimately I accept.

[169] On that basis, I essentially accept Bullen's figures for assessed potential yields as set out in Exhibits 2 and 4.[59] The spray drift damaged area was 1291.52 hectares and the assessed production loss was 0.943 bales per hectare. The assessed lint production loss was 1,216.96 bales at a figure of $395 per bale, which is a total loss of $480,692.

[170] The quality penalties are also set out. They were due to poorer quality cotton produced at both Noonameena and Elgin. The average discount for fields at Cullingral was $4.90 per bale. Based on that discount, an assessment was made of the difference between the actual and potential discounts for each field at Noonameena and Elgin. The total additional discounting for lint quality attributable to spray drift damage compared to Cullingral was $23,644. Given the smaller crop, there were some harvest and production costs saved which were quantified as $37,148.74.

[171] Accordingly, I consider that the total quantifiable loss is $467,187.45

[172] The plaintiff is also entitled to interest at the rate of 10 per cent to the date of judgment.

[173] As I indicated during the trial, I will also now hear from Counsel in relation to the issue of the apportionment of liability under the Civil Liability Act 2003 (Qld) and the effect of the settlement by the plaintiff with the other defendants. I will hear submissions from Counsel as to the degree of responsibility which is to be shared between the sixth and eighth defendants.

[174] I will also hear from Counsel as to costs and the form of the orders.

[4] Bonic v Fieldair (Deniliquin) Pty Ltd & Ors [1999] NSWSC 636.

[5] [1980] HCA 12; 146 CLR 40 at 47-48.

[6] Batiste v State of Queensland [2001] QCA 275 at [10]; Civil Liability Act 2003 (Qld), s 12.

[7] Flounders v Millar [2007] NSWCA 238 at [35].

[8] [1999] NSWSC 636 at [25].

[9] Exhibits, 2-7, 23, 24, 25, 27, 40, 47, 48, 49.

[13] Transcript day 4, p 108, ll 15-35.

[17] Report dated 10 July 2007 at 6.2.3.

[24] Transcript day 6, p 52 l 56 to p 53 l 5.

[30] Transcript day 8, p 38, ll 41 to p 39 l 2.

[32] One knot is equal to 1.852 kph. Therefore three to five knots is approximately 5.556 to 9.26 kph.

[45] [1999] NSWSC 636 [23-25].

[49] [2007] NSWCA 238.

[50] [2010] HCA 5; (2010) 263 ALR 576.

[51] [2010] HCA 5; (2010) 263 ALR 576 at [14].

[58] Exhibit 3, para 6.11.3 of Bullen's Report.

[59] Exhibit 2 Table 4 and Exhibit 4, paras 21.4 and 21.5.