While many Australian cotton growing valleys had the fortune of relatively low Helicoverpa spp. activity during the 1997/98 season, this was not the universal experience. Extremely high pest activity on the Darling Downs has reinforced the concerns that our current over-reliance on insecticides for the management of Helicoverpa on cotton and grain crops is unsustainable. Just as important is the realisation that within the agro ecosystem, action must be taken to attempt to maintain Helicoverpa spp. populations at more manageable levels. While densities of up to 10 eggs per metre can be managed satisfactorily, densities of 50 or more eggs per metre pose serious difficulties, especially if activity persists at this level for several days or even weeks. Helicoverpa spp. populations fluctuate in response to various factors. Where a succession of wild or cultivated hosts are available, successive generations can develop through the spring, summer and autumn months. If progressive population increases take place for whatever reasons, the end result can be serious management difficulties. Such was the case on the Darling Downs during the 1997/98 season, where the earlier than normal appearance of substantial numbers of H armigera and their persistence at high levels throughout the season resulted in high insecticide use and some control difficulties across cotton and grain crops

Madame Chairperson, ladies and gentleman, good morning and welcome to my presentation on Profitability and Sustainability.

Thielaviopsis basicola, a phytopathogenic filamentous fungus, is the causative agent of black root rot in a variety of host plants, including the economically important crop, cotton. The method of Agrobacterium tumefaciens mediated transformation (ATMT) was chosen to investigate the molecular interactions that exist between T. basicola and cotton. ATMT has long been used to generate transgenic plants and has more recently become a popular method for random insertional mutagenesis in the transformation of filamentous fungi. Generation of a large number of reduced pathogenicity mutants using this technique will aid to elucidate the identification of key pathogenic genes providing a better understanding of the molecular interactions between T. basicola and cotton, governing the pathogenesis of black root rot.

Development of an efficient ATMT protocol, designed specifically for transforming T. basicola, required optimisation of the experimental conditions prior to, during and after transformation. Transformation efficiency was found to be dependent upon the duration and temperature of pre­ cultivation, co-cultivation and selection. The number of A. tumefaciens cells and the status of the T. basico/a cells were also found to have significant influence on the efficiency of transformation. A consistently high rate of transformation efficiency was achieved by employing the hypervirulent strain AGLI, carrying the binary vector p8Ht2, which contains the modified bacterial Hygromycin B phosphotransferase hph gene under the control of the Aspergillus nidulans trpC promoter. The

media used during co-cultivation and the method of selection also played an important role in

optimising the ATMT protocol for T. basicola. Optimal conditions of transformation led to the

production of 300-770 Hygromycin B resistant (HygR) putative transformants per I x 106 conidia of

T. basicola.

All I 0 HygR putative transformants tested remained mitotically stable, maintaining their Hygromycin B resistance after five generations on non-selective medium. Primary pathogenicity screenings indicated that three of the I 0 mitotically stable HygR putative transformants had reduced pathogenicity, showing decreased virulence towards infected cotton seedlings when compared to the WT. Vegetative growth tests of these same 10 HygR putative transformants, displayed varying growth by comparison to the WT; with six showing reduced growth and four growing at a similar rate to the WT. Colony morphology also indicated that at least seven of the HygR putative transformants differed in colour, texture, and number of chlamydospores compared to the WT.

Further genetic testing will be required to confirm that single and random insertion of the T-DNA

occurs in the T. basicola genome.

Southern blot analysis on three of the five T. basicola reduced pathogenicity mutants generated by PEG/CaC(z, revealed that in p737 and p888, more than one insertion of pGpdGFP took place at multiple loci in the fungal genome; a common occurrence when using this method of transformation. The reduced pathogenicity mutant p 16 instead had a single insert of the plasmid pGpdGFP integrated at a locus in the fungal genome, which suggests that further attempts could be made to recover the tagged pathogenicity gene from this mutant. Phenotypic analyses of all five PEG mutants, as well as 20 HygR putative ATMT transformants, indicated that T. basicola most likely has some pathogenicity genes that are similar to those found in other filamentous fungi; including genes involved in the formation of infection structures and hydrophobins, spore development and germination, regulation and biosynthesis of melanin, cuticle and cell wall degrading hydrolytic enzymes, and regulatory proteins, including transcription factors, receptors, G proteins, and enzymes.

An important requirement for successful management of Helicoverpa spp. In Australian cotton is a reliable and effective monitoring system. Timely pest management decisions must be based upon knowledge of the pest population densities with a reasonable degree of accuracy. Effective sampling for Helicoverpa spp life stages can be time consuming and onerous, and any technique which reduces the amount of time spent scouting is welcomed, provided accuracy is not compromised. Whatever sampling scheme is used must be based on an understanding of the distribution and feeding behaviour of Helicoverpa life stages and then be associated with a well validated threshold to allow management decisions. Here we briefly review sampling techniques for cotton and present our first field results aimed at assessing the validity of these techniques with INGARD cotton.

Although cotton is grown on fertile soils, commonly nutrient deficiencies become apparent due to many factors. Cotton has a high demand for many nutrients which are taken up over a period of weeks (Table I). Nutrient deficiency (or excess) can reduce crop yields. Nitrogen deficiency, for example, is easily detected, but usually by the time the deficiency is recognised and remedied, the crop has suffered some nutritional stress.

1) The fit of new insecticides into integrated pest management. Selection of insecticides can have a

big influence on both control of the target pest as well as on beneficials, and the risk of secondary

pest outbreaks. We found that low rates of fipronil provided strong efficacy against mirids, with

or without salt, and were significantly more selective that the full rates of fipronil against

beneficials, though still with a risk of flaring mites. Low rates of indoxacarb alone provided poor

efficacy against mirids but the addition of salt or canopy oil boosted this to efficacy equivalent of

the full rate with low risk to beneficials or risk of flaring mites. Altacor (rynaxypyr) a new

insecticide for Helicoverpa control was highly efficacious and selective against many beneficials

indicating a good IPM fit. These results have been made available to industry via the Cotton Pest

Management Guide, to assist pest manager in spray choices, and to industry to help in

registration, thereby ensuring availability of new insecticides or uses of insecticides to industry.

2) Defining the pest status of emerging pests

a) Management of thrips on seedling cotton. Tobacco thrips, Thrips tabaci, is still the dominant

species. It was controlled moderately well at some sites but poorly at others, which could

indicate resistance. At some locations the western flower thrips, Frankliniella occidentalis, is

also abundant and poorly controlled by available options indicating insecticide resistance.

Control of thrips is problematic because damage is often cosmetic, plants will recover without

loss, and because thrips are also predators of spider mites. Nevertheless in cooler regions, where

control is justified, management of WFT may be difficult. Monitoring of thrips population

composition early in the season and determination of resistance profiles for both WFT and T.

tabaci has been initiated in conjunction with Dr Grant Herron (NSWDPI) and CSD.

b) Late season pest damage from thrips and jassids. These pests often build to levels causing

significant damage to leaves on maturing cotton. We found that late season damage to leaves is

only likely to reduce yield in crops with high yield potential and if the damage is very severe and

prolonged before cut-out. High yielding crops are likely more affected but even they show strong

compensation at yield levels up to 14 b/ha.

c) The effect of mirid sprays on secondary pests. We found that controlling mirids with the most

popular insecticide (fipronil) increases the risk of causing mite outbreaks, which would then

require additional control. Our results also suggest that in some situations Bollgard II crops are

more at risk – this deserves further investigation. Nevertheless, the results highlight the need to

have good mite sampling protocols in place in Bollgard II crops especially if OP’s, SP’s or

fipronil are used to control mirids.

3) Develop a new aphid sampling strategy and thresholds. These were developed and extended to

industry. They will provide a more rational basis for deciding when the occurrence of this pest

justifies control and when beneficials are providing adequate control. This information has been

linked with new information on the aphid borne disease cotton bunchy top, to provide pest

managers with a holistic approach to managing both the pest and the disease.

4) Sporadic pests. Information and publication to help manage the pale cotton stainer was completed

and will help industry to manage this pest in the future. In particular thresholds for lint damage

are now available. These indicate that for stained locks the threshold is >50% of bolls with all

locks damage, and for tight-locked bolls it’s > 20% of bolls with all locks damaged. We also

studied the feeding behaviour and have a better understanding of the damage symptoms.

This project has provided new information to help pest managers to make better decisions about

management of emerging pests. Outcomes have been largely delivered to industry through a range of

presentations, published resources and the WWW. Benefits to the industry are more rational decisions

on the need to control pests, and awareness of risks for different control options. This will contribute

to reduced pesticide use with flow-on economic, social and environmental benefits.

With the advent of Bt cotton, mirids have become more of a pest in cotton and are attracting a number of insecticide applications. This has the potential to disrupt IPM in cotton and increase the risk of mite, aphid or whitefly outbreaks.

The aim of this project was to identify factors which could influence mirid damage to cotton. In particular, the project looked at current mirid management methods, tested the effectiveness of potential predators and identified other interspecific interactions which could reduce mirid damage to cotton.

We found that pest managers that only sprayed for mirids once the numbers had exceeded the recommended threshold suffered no yield loss, and if anything it was beneficial to the profitability of the field. Nevertheless, we did find that pest managers were less likely to use the beatsheet threshold than the visual survey threshold, indicating that more extension work is needed in this area.

The project was able to identify a number of predators that could reduce mirid numbers and affect mirid feeding behaviour. In particular the plain brown lynx spider, which is very common in cotton, was a very efficient predator of mirids. More work is needed to confirm their effectiveness under field conditions.

The project also showed that mirids may not attack cotton if alternative foods, such as Helicoverpa eggs, are available. Thus a Helicoverpa egg lay in a Bt crop could be advantageous if the field has a heavy mirid infestation because it could reduce the likelihood that the mirids present will attack the cotton. This finding needs to be confirmed under field conditions.

The reproductive status of the mirid had little effect on mirid damage, but damage caused by mites and aphids overrode any damage caused by mirids. This indicates that if pest managers have to choose between mirid and mite control, they should be more concerned about controlling the mites.

The results of this project indicate that insecticide applications to control mirids can be kept to a minimum; 54% of the sprays in the mirid survey were applied to mirid numbers below threshold. If these sprays were irradiated, there would be a huge saving in insecticides, no cost in yield, and a large advantage in terms of the development of IPM in cotton. The results of this project indicate that this should be the goal of mirid management in cotton.

The aim of this project was to identify factors which could influence mirid damage to cotton. In particular, the project looked at current mirid management methods, tested the effectiveness of potential predators and identified other interspecific interactions which could reduce mirid damage to cotton.

Black root rot, caused by Thielaviopsis basicola, is a significant disease threat to cotton, especially in cooler areas and seasons. In just over a decade it has come to affect more than half of the cotton farms in southern Queensland and New South Wales and it is currently found to be present in every surveyed farm. While management strategies based on cultural practices can reduce the severity of the disease and of crop losses, yield can still drop by up to 40% annually and further loss can occur due to increased susceptibility of black root rotinfected plants to other diseases. Thus, there is considerable scope for new disease controlmethods based on an improved knowledge of the biology of the pathogen and its interactions with cotton. The black root rot fungus occurs as strains that are specific to particular host plants, and must establish a special relationship with living cells of the host root before root rotting can occur. This suggests that there are highly specific biochemical and genetic interactions between the fungus and cotton that are involved in the infection progress. The longer term aim of our multidisciplinary group of researchers is to identify key factors in the molecular interactions between T. basicola and cotton roots, to determine whether such interactions could be exploited in disease management. Identification of host-specific interactions during infection could be used to find components of resistance that will increase the efficiency of breeding varieties with enhanced resistance.In the seed project completed in June 2007, our group developed methods for investigating the interactions between the fungal pathogen and its cotton host, as well as tools for genetic manipulation of the pathogen and for proteome analyses of both the pathogen and cotton roots. The main outcomes in technique developments were (1) the establishment of a genetic transformation protocol for the production of a large number of fungal mutants and the establishment of a procedure to select those mutants affected only in their pathogenicity towards cotton, (2) the establishment of extraction protocols for the purification of both T. basicola and cotton root proteins, (3) the development of a method to produce two dimensional protein electrophoresis maps for both T. basicola and cotton roots, which resulted in successful production of reference protein maps for both the fungus and cotton roots and (4) the development and optimisation of reliable systems for the study of T. basicola interactions with different host plants, which allowed the comparison of the interactions of cotton with pathogenic versus non-pathogenic strains of T. basicola.The techniques developed in the seed project will be adopted in current and future research with the aim of identifying factors responsible for changes in pathogenicity, especially those involved in host-specific interactions. In addition, the protein maps (total cell proteins, also called proteome) will be used in order to identify proteins (and thus, genes) involved in the infection process and to find if the disease could be blocked by reducing plant stimulants that enhance the pathogen, or by inducing plant resistance to the disease. The long term objectives of this project were to find out whether there are host-specific triggers to certain stages of the pathogen infection and whether it is feasible to exploit key steps in the infection process for the development of resistance or other control measures, such as soil amendments.The development of tools in this seed project allow research into the T. basicola-cotton interactions, which could lead to discoveries towards ways of controlling the disease and thus increasing cotton yields, by either breeding cotton towards disease resistance or by producing more effective soil amendments against T. basicola.

Deep drainage below the root zone is still the least understood component of the water balance, especially in cracking clay soils. It represents a waste of a valuable resource and can leach nitrogen out of the root zone. It has the potential to cause watertables to rise, with the accompanying risk of salinity. Drainage can move contaminants, such as salt and agrochemicals, into the groundwater.

The lysimeter facility at the Australian Cotton Research Institute, near Narrabri NSW, was used to study drainage, its contaminants and its interaction with groundwater in a heavy clay soil under a furrow-irrigated cotton – wheat rotation from 2006 to 2011.

Drainage during the cotton seasons varied from 0 – 74 mm, under wheat it was negligible and under fallow it was 23 mm. Drainage occurred in two forms: matrix drainage and by-pass drainage. The former occurs when the water storage capacity of the soil is filled due to prolonged rainy periods with any extra water becoming drainage. Drainage rates are not high (<0.5 mm/day) but can continue for periods of a month.

By-pass drainage occurs after furrow irrigation when water flows rapidly down macropores and by-passes the matrix of the subsoil. Peak drainage rates are reached 25 hours after irrigation and can reach more than 3 mm/day. The rate then declines exponentially over a week to about 0.5 mm/day. The amount of by-pass drainage appears to be controlled by the soil water deficit in the upper metre of soil. Drainage increases as the 0 – 0.5 m layer becomes drier, possibly due to greater cracking. However, larger deficits in the 0.5 – 1.0 m layer decrease drainage and appear important in mitigating by-pass drainage.

The risk of by-pass drainage is greatest when irrigation is necessary early in the cotton season, when the crop is too small to create a subsoil deficit between irrigations, especially if the subsoil was already wet before sowing.

The risk of matrix drainage can be minimized by managing both the rotation and irrigation scheduling to ensure there is sufficient deficit to accommodate likely inputs of water and irrigation at any time of year. Nevertheless there will always be times of above average rainfall when the profile is filled to capacity and drainage occurs.

However, some drainage is necessary to leach salts from the irrigation water that accumulate in the root zone. The electrical conductivity (EC) of matrix drainage is greater than by-pass drainage, suggesting matrix drainage is more efficient at leaching salt.

In addition to salt, drainage leaches nitrogen from the topsoil. During the 2008/09 cotton season approximately 9.5 kg N/ha – equivalent to 6% of that applied as fertilizer – was lost in drainage.

Seasonal drainage from the root zone appears to recharge the watertable at 16 m depth within weeks, although this result is still tentative. There is continuous downwards leakage of salty water from the upper, watertable aquifer into the lower confined aquifer, from which water is extracted for a variety of uses. This leakage is exacerbated by pumping from the lower aquifer.

The lysimeter was also used to test less expensive methods of estimating drainage. A barrel lysimeter installed near the lysimeter facility overestimated drainage, whereas the chloride mass balance method underestimated drainage.