Year 2 (01/07/2002-30/06/2003)

Strategic surveys of peanut products conducted in Central Java have shown that alarming levels of Aspergillus flavus infection and aflatoxin contamination are occurring in raw peanut kernels collected from retailers in traditional markets. These raw kernels had up to 100% infection with Aspergillus flavus, and aflatoxin contamination in the range of 2 to 340 parts per billion (ppb). In contrast, peanut samples collected from farmer's fields, penebas, processors and collectors were generally low in A. flavus infection, and had low aflatoxin contamination (< 15 ppb). Results from the survey clearly indicate that postharvest handling methods employed prior to peanuts being delivered to retailers, and especially at the retailer level in traditional markets, will severely impact on the level of aflatoxin contamination in peanuts in the Indonesian food chain. Any future efforts to reduce or eliminate contamination must necessarily target postharvest handling in the initial stages, as this would have the greatest impact on overall reduction of contamination.

Field and on-farm management and variety trials were conducted during the 2002 dry season (July-November 2002) in eastern Java (RILET) and central Java (GMU) sites. In general these trials have shown there is very little pre-harvest aflatoxin occurring under field conditions in Indonesia, despite the presence of quite severe end-of-season drought conditions and high soil temperatures. In Australia, the aflatoxin minimisation program being implemented via a harvesting management decision-support package has had a significant impact on reducing on-farm aflatoxin for dryland peanut growers. During the 2001-02 season drought and high soil temperatures during the pod-filling period meant aflatoxin risk was extremely high (i.e. 100% risk according to the QDPI a
flatoxin risk model), yet growers who implemented the minimisation strategy (i.e. early harvesting, short cutting--threshing intervals, pre-cleaning, inverted windrows, rapid drying to safe moisture (12%) were able to minimise positive loads deliveries to around 40%, compared to the regional average of nearly 65% positive. The aflatoxin module has been successfully incorporated into the APSIM peanut model, and assessed under Indonesian conditions. A range of medium- to long-term climate data for a number of peanut growing sites throughout Java has been accessed via the web and also from the Indonesian meteorological bureau. Project scientists have begun to determine regional aflatoxin risk for peanuts throughout Java using a newly developed software tool called 'Peanut Whopper Cropper', which allows users (i.e. researchers, industry personnel, policy makers etc) to select peanut cropping scenarios (e.g. location, time of planting, variety, irrigation etc) and assess the associated aflatoxin risk (in terms of probability of exceedance).

Project surveys have confirmed that peanuts are highly contaminated with aflatoxin posing a severe health threat to Indonesians. Many supply-chain members have a poor understanding of the significance of the incidence, health effects and minimisation of aflatoxin. Perhaps fortunately, the main source of contamination identified in Indonesian surveys is in post-harvest, post-farm gate handling. As a result any attempt to rectify the problem efficiently would be best directed at this sector, making the logistics of developing a management package and subsequent extension efforts considerably easier and more practical than at first recognised.

In early 2002, each of the three collaborating institutes were provided with all essential equipment, immunoreagents and an aflatoxin standard needed to conduct the SUNQuik Aflatoxin Test. Significant training activities were also conducted during Year 2, including a workshop at the SEAMEO BIOTROP Institute, Bogor, titled Analysis of Aflatoxin B1 in Peanuts. This workshop sought to provide quality control for ELISA analyses for aflatoxin B1. Close linkages were also made with DFID (UK) project on transfer of technology related to aflatoxin kit production with BALITVET in Bogor.

Year 2 (01/07/2002-30/06/2003)

Strategic surveys of peanut products conducted in Central Java have shown that alarming levels of Aspergillus flavus infection and aflatoxin contamination are occurring in raw peanut kernels collected from retailers in traditional markets. These raw kernels had up to 100% infection with Aspergillus flavus, and aflatoxin contamination in the range of 2 to 340 parts per billion (ppb). In contrast, peanut samples collected from farmer's fields, penebas, processors and collectors were generally low in A. flavus infection, and had low aflatoxin contamination (< 15 ppb). Results from the survey clearly indicate that postharvest handling methods employed prior to peanuts being delivered to retailers, and especially at the retailer level in traditional markets, will severely impact on the level of aflatoxin contamination in peanuts in the Indonesian food chain. Any future efforts to reduce or eliminate contamination must necessarily target postharvest handling in the initial stages, as this would have the greatest impact on overall reduction of contamination.

Field and on-farm management and variety trials were conducted during the 2002 dry season (July-November 2002) in eastern Java (RILET) and central Java (GMU) sites. In general these trials have shown there is very little pre-harvest aflatoxin occurring under field conditions in Indonesia, despite the presence of quite severe end-of-season drought conditions and high soil temperatures. In Australia, the aflatoxin minimisation program being implemented via a harvesting management decision-support package has had a significant impact on reducing on-farm aflatoxin for dryland peanut growers. During the 2001-02 season drought and high soil temperatures during the pod-filling period meant aflatoxin risk was extremely high (i.e. 100% risk according to the QDPI a
flatoxin risk model), yet growers who implemented the minimisation strategy (i.e. early harvesting, short cutting--threshing intervals, pre-cleaning, inverted windrows, rapid drying to safe moisture (12%) were able to minimise positive loads deliveries to around 40%, compared to the regional average of nearly 65% positive. The aflatoxin module has been successfully incorporated into the APSIM peanut model, and assessed under Indonesian conditions. A range of medium- to long-term climate data for a number of peanut growing sites throughout Java has been accessed via the web and also from the Indonesian meteorological bureau. Project scientists have begun to determine regional aflatoxin risk for peanuts throughout Java using a newly developed software tool called 'Peanut Whopper Cropper', which allows users (i.e. researchers, industry personnel, policy makers etc) to select peanut cropping scenarios (e.g. location, time of planting, variety, irrigation etc) and assess the associated aflatoxin risk (in terms of probability of exceedance).

Project surveys have confirmed that peanuts are highly contaminated with aflatoxin posing a severe health threat to Indonesians. Many supply-chain members have a poor understanding of the significance of the incidence, health effects and minimisation of aflatoxin. Perhaps fortunately, the main source of contamination identified in Indonesian surveys is in post-harvest, post-farm gate handling. As a result any attempt to rectify the problem efficiently would be best directed at this sector, making the logistics of developing a management package and subsequent extension efforts considerably easier and more practical than at first recognised.

In early 2002, each of the three collaborating institutes were provided with all essential equipment, immunoreagents and an aflatoxin standard needed to conduct the SUNQuik Aflatoxin Test. Significant training activities were also conducted during Year 2, including a workshop at the SEAMEO BIOTROP Institute, Bogor, titled Analysis of Aflatoxin B1 in Peanuts. This workshop sought to provide quality control for ELISA analyses for aflatoxin B1. Close linkages were also made with DFID (UK) project on transfer of technology related to aflatoxin kit production with BALITVET in Bogor.

Aflatoxin is a human carcinogen that contaminates peanuts and hence is a major food quality problem throughout the world. The broad aim of the project is to minimise, and eventually eliminate, aflatoxin contamination in Indonesian and Australian peanuts through research, development and extension of appropriate on-farm and post-harvest management practices. The strategies to be employed include a survey of critical hazard points for aflatoxin incidence in peanut products from 'paddock to plate' in the Indonesian food chain. As well, on-farm agronomic management strategies that have been demonstrated to minimise aflatoxin under Australian conditions will be evaluated and extended in Indonesian farming systems. The utility of the bio-control approach to aflatoxin minimisation using non-toxigenic strains of the Aspergillus flavus fungus, will also be tested under large scale field trials in Australia, with preliminary investigations carried out in Indonesia. Implementation of aflatoxin control strategies in Indonesia will be fostered by the development of a simple, rapid and low cost aflatoxin ELISA based analytical system. As well, involvement of socio-economic experts in the project will ensure control strategies are successfully implemented on-farm.

The project has made significant progress in several areas, and highlights of outputs to date are summarised below:
Objective 1 - Following on from the results of the Year 1 and 2 surveys, which indicated there was significant potential for aflatoxin contamination from imported peanuts in the Indonesian food chain, a survey was conducted in Year 3 to monitor aflatoxin contamination in imported peanuts. The imported peanut delivery chain can be classified into three levels, i.e. importer, wholesaler and retailer. Aflatoxin B1 contents of two peanut kernel samples collected from an importer was low (6.4 and 4.6 ppb, respectively). Aflatoxin B1 contents of kernels collected from wholesalers as well as retailers however varied, ranging < 3.6 - 330.2 ppb for retailers and < 3.6 - 38.8 ppb for wholesalers, again indicating that aflatoxin seems to build up in the peanut food chain in Indonesia. Further surveys of aflatoxin levels in imported peanuts are planned between July and October, 2004.
Objective 2 - Experiments investigating the effect of harvesting time and differing seed storage treatments on kernel aflatoxin contamination in general showed only low levels of pre- and post harvest aflatoxin, and little evidence of dramatic aflatoxin build up in post harvest storage at both ILETRI and GMU sites. Overall these results show that the level of pre-harvest aflatoxin in peanuts grown in farmers' fields, and their subsequent on-farm storage, was generally low. The significant point however is that some aflatoxin is always present, thus indicating that the A. flavus fungus is present in some kernels (especially shrivelled and damaged kernels) and these must therefore pose a significant risk for aflatoxin production further on the in peanut food chain (ie at the wholesaler and retail level). In Australia, the aflatoxin minimisation program was implemented via the web based AFLOMAN decision support package which approximately 12 growers used during the 2003-04 season. Seasonal conditions during 2003-04 were very good, and little on-farm aflatoxin occurred.
Objective 3 - The aflatoxin module was successfully incorporated into the APSIM peanut model, and assessed under Indonesian conditions. A range of medium to long-term climate data for a number of peanut growing sites throughout Java has been accessed via the web and also from the Indonesian met bureau. The newly developed software tool called "Peanut Whopper Cropper", which allows users (ie researchers, industry personnel, policy makers etc..) to assess aflatoxin risk (in terms of probability of exceedance) throughout Indonesian locations, has been tested and training on its use provided to our Indonesian collaborators.
Objective 4 - Aflatoxin bio-control studies continued in field experiments during 2003/04 and again demonstrated good build up of non-toxigenic strains of Aspergillus flavus on barley inoculum applied post-planting. The season was not conducive to aflatoxin production, so there was no effect of inoculum treatment on aflatoxin production. Controlled environment experiments have shown there are a number of highly competitive non-toxigenic strains of Aspergillus flavus/parasiticus available for further evaluation under field conditions in Australia. Training of Indonesian collaborators in a range of techniques appropriate to biocontrol research was provided during November 2003.

Objective 5 - Results from the socio-economic surveys clearly indicates there are currently few practical incentives for the Indonesian peanut industry to address the issue of aflatoxin contamination. Aflatoxin and many other aspects of food safety are not mandated as important parts of the Indonesian food production and processing chain, either due to limited legislative or regulatory controls or a practical inability to implement, monitor and police food safety standards at an operational level. It appears that major processors may have the greatest capacity to handle and process product in a safe manner, by possessing the greatest potential to influence product pricing and intake specifications via the 'tebasan' system or via purchases they make of intermediaries in other regionally important production areas. At a local level, logistical considerations will constrain any farmer oriented education process due to shear numbers, lack of literacy and a poor understanding about the relatively complex processes involved in minimising aflatoxin in such a simple production system as currently exists.
Aflatoxin ELISA - By the third year of the project it is clear that our Indonesian collaborators have available a convenient means of analysing aflatoxin in peanut samples - for both the field surveys in the market supply chain and for agronomic experiments investigating management options. The adaptation and application of the SUNQuik aflatoxin B1 test has provided a simple and effective screening analysis with minimal cost, skills and maintenance. Compared to other produce such as maize, the assay of raw and roasted peanuts has relatively low matrix interferences and quantitative assay of aflatoxin B1 is possible.

Several ACIAR bilateral projects over the past 6 years have addressed aspects of aflatoxin assessment with significant progress made in developing integrated control strategies. Extensive surveys of aflatoxin contamination at various levels of the Indonesian peanut distribution chain in the initial phase of this project found that the main source of contamination lies in the wholesaler and retail sectors, and strongly suggests that aflatoxin builds up during poor storage conditions. In the extension of this project, more detailed and controlled studies of environmental conditions experienced in wholesalers, wet markets and supermarkets have been established to better undersand the dynamics of the observed aflatoxin buildup. Early results from these studies confirm that A. flavus is highly prevalent in kernel samples stored in wholesaler and wet markets, with temperature and relative humidity conditions being highly conducive for aflatoxin production. Studies next year will repeat the monitoring experiments with peanut pods derived from crops grown under high and low aflatoxin risk conditions, to assess the impact of pre-storage conditions on subsequent aflatoxin production. Results to date have enhanced our understanding of factors leading to the post-harvest build up of aflatoxin in peanuts during storage in wet markets in Indonesia.
In Australia, an artificial drying system has been developed to investigate the effect of airflow rate and temperature on drying efficiency and a range of quality of peanuts. Critical drying parameters such as airflow rates, temperature and humidity have been monitored using dataloggers throughout the drying cycle. Collaboration with colleagues at the University of NSW has enabled us to adapt their grain drying model for peanuts. Preliminary results from our model analyses have shown very good agreement between simulated and observed rates of drying. It is encouraging that the drying model will be able to be customised to simulate peanut drying and further developed to define the optimum combination of parameters for an efficient on-farm drying system, as well as including a range of peanut quality parameters.
Simulation modelling approaches are continuing to integrate the interaction between crop, soil, environment and A. flavus to enable assessment of the probability of aflatoxin formation at various stages during harvest and storage of peanuts. This information will assist in the development and improved predictive ability of the APSIM Peanut-Aflatoxin model for post harvest management of aflatoxin. Once fully validated, the model can be used for scenario analysis and development of decision support tools to minimise aflatoxin incidence in both Indonesian and Australian peanut industries. Basic studies on the ecology of the etiella soil insect, which is closely associated with aflatoxin contamination, have shown the life cycle of this insect is strongly dependent on temperature conditions. Response functions relating etiella growth to potential pod infection are being developed for incorporation into the APSIM peanut model.
In the project extension the team have been active in developing a program of aflatoxin information and awareness in Indonesia to publicise project results and outputs to government, industry and consumers. 2 brochures containing general information about benefits of peanut and aflatoxin risk implications to human health, and how to minimise aflatoxin risk in peanut preparations by screening out bad kernels and using only good kernels for processing, are being prepared. A National Aflatoxin Awareness Workshop is being organised for Jan 2006 at Malang and will involve project collaborators presenting information to peanut industry stake holders and officials from Govt health Department. A QA system for aflatoxin analysis for our Indonesian collaborators has been established by our UNSW team member as part of the capacity building activity. A workshop was conducted in January 2005 to validate a number of analytical performances in Indonesia and in turn, to generate data that are necessary for establishing the Control Charts.

Activities under the extension project are focussed as sub-objectives under two of the specific objectives of the original project vis:

Objective (3). To develop a detailed understanding of the soil, plant and environmental factors influencing Aspergillus flavus invasion and aflatoxin production, and incorporate this knowledge into the APSIM Peanut model for scenario analysis and development of decision support tools.

Focus under the extension will be

(a) To develop an enhanced understanding of factors leading to the post-harvest build up of aflatoxin in peanuts during storage in wet markets in Indonesia and in artificial dryers in Australia.

and

(b). To further develop, validate and improve the predictive ability of the APSIM Peanut-Aflatoxin model for scenario analysis and development of decision support tools by:-

(i) using agronomic data generated in Indonesian studies to validate the model performance in Indonesia
(ii) incorporating an Etiella soil insect module to allow prediction of potential aflatoxin risk associated with this pest.
(iii) incorporating environmental factors affecting post-harvest aflatoxin build up, as developed in Objective 1

Objective (5) To foster the implementation of aflatoxin monitoring and control strategies.

Focus under the extension will be:

(a). To foster an aflatoxin information and awareness program in Indonesia and Australia

(b) To establish on-going and reliable aflatoxin analytical systems using ELISA methodology in Indonesia.

This project has addressed several aspects of aflatoxin contamination in both Indonesian and Australian food chains with the ultimate aim of developing integrated strategies that can minimise its effect for peanut consumers. Extensive surveys of aflatoxin contamination at various levels of the Indonesian peanut distribution chain in the initial phase of this project found that the main source of contamination lies in the wholesaler and retail sectors, and strongly suggests that aflatoxin builds up during poor storage conditions. In this extension project more detailed and controlled studies of environmental conditions experienced in wholesalers, wet markets and supermarkets were established to better undersand the dynamics of the observed aflatoxin buildup. Results from these studies confirm that A. flavus is present in the majority of kernel samples stored in wholesaler and wet markets. Monitoring of temperature and relative humidity conditions under these storage conditions has confirmed that these enviromental conditions are highly conducive for aflatoxin production. Studies this year have monitored peanut pods dervied from crops grown under high and low aflatoxin risk conditions, to assess the impact of pre-storage conditions on subsequent aflatoxin production. While afaltoxin levels in these samples were relatively low, these studies in W and E Java clearly showed that there was an increase in A. flavus levels, proportion of damaged kernels and aflatoxin concentration with increasing length of storage (up to 6 months). These results, in combination with the earlier monitoring studies, clearly show that aflatoxin can increase under the poor storage conditions typical of the wet markets where consumers buy peanuts. The monitoring studies also continued to show that alarming aflatoxin levels (up to several thousand ppb) are present in peanut sauces (e.g. Satee sauces, Sambel Pacel, Bumbu Pacel, Bumbu Gado Gado) which are consumed in large quantities by Indonesian consumers. It appears that very low quality product (with high aflatoxin levels) is being used by sauce processors, as they are able to easily hide suspect looking kernels by grinding. Also, the studies have clearly shown that higher aflatoxin levels are present in damaged kernels, yet these kernels are often still sold in the market at a lower price to consumers who are either unaware that they are a health risk, or to consumers of a lower socio-economic class.

In Australia, an artificial drying system has been developed to investigate the effect of airflow rate and temperature on drying efficiency and a range of quality of peanuts. Collaboration with Drs Rob Discoll and George Srzednicki at the University of NSW has enabled the project team to adapt their grain drying model for peanuts. Drying studies have generated valuable data to allow us to validate the UNSW drying model under Queensland conditions, with very close agreement between measured and predicted values under our controlled drying facility, as well as in commercial drying bins. There is enormous scope for applying the model for the improvement of current drying practices in Australia. For instance, the model has shown that the strategy of product mixing followed by low air speed drying could result in significant efficiencies in energy and labour use and hence potential cost savings (of up to 40%) for peanut growers. Both growers and the Peanut Company of Australia are very interested in the outcomes of this research which shows considerable promise to improve the efficiency of post-harvest drying and hence the minimisation of aflatoxin build up in storage.

Simulation modelling approaches are continuing to integrate the interaction between crop, soil, environment and A. flavus to enable assessment of the probability of aflatoxin formation at various stages during harvest and storage of peanuts. This information will assist in the development and improved predictive ability of the APSIM Peanut-Aflatoxin model for post harvest management of aflatoxin. The DPI&F team has now developed a partially validated model for use in Indonesia using local growth and yield data developed in the previous agronomic studies. The model has been used for scenario analysis using the 'Whopper Cropper' graphical interface, with a CD developed and distributed to indoensian collaborators at the recent review meeting in Surabaya in July 2006. Basic studies on the ecology of the etiella soil insect, which is closely associated with aflatoxin contamination, have shown the life cycle of this insect is strongly dependent on temperature conditions. Response functions relating etiella growth to potential pod infection are being developed for incorporation into the APSIM peanut model.

The extension project has been active in developing a program of aflatoxin information and awareness in Indonesia to publicise project results and outputs to government, industry and consumers. 2 brochures and 4 posters containing general information about benefits of peanut and aflatoxin risk implications to human health, and how to minimise aflatoxin risk in peanut preparations by screening out bad kernels and using only good kernels for processing, have been produced and widely distributed to a range of peanut processors, regulatory bodies and consumers via a multitude of field days and other communication forums (see links within the report). A significant highlight of this extension effort has been the holding of two "Aflatoxin Forum Indonesia (AFI-1 and AF-2) which were co-organised by Gadjah mada University (in Yogyakarta) and ILETRI staff. These for a have involved many key Government and industry representatives who have heard about the many outputs from the ACIAR aflatoxin project. In particular, a number of senior officials from the Indonesian Food and Drug Agency, who are responsible for the regulation of afaltoxin in the Indonesian food chain, were very interested in following up on many of the project outputs as they develop policy for the regulation of mycotoxins in Indonesian food systems.

The validity of BALITVET ELISA for several commodities including corn, animal feeds and peanuts is now confirmed. BALITVET has conducted several training workshops on the use of ELISA techniques for monitoring aflatoxin contaminations in Indonesia. The BALITVET ELISA are now supplied to several local feed factories, research institutes and Universities for aflatoxin B1 analysis of agricultural commodities, animal feeds and food products. It is believed that this ELISA will be an important resource for aflatoxin monitoring program in Indonesia.