An official website of the United States government.

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Modelling, Predicting and Mapping the Emergence of Aflatoxins in Cereals in the EU Due to Climate Change


The impact of climate change has been identified as an emerging issue for food and feed safety. With its mandate to identify emerging risks in food and feed production, EFSA’s Emerging Risks Unit identified changing patterns in mycotoxin contamination in cereals such as wheat, maize and rice, due to climate change as a potential emerging hazard. In particular, aflatoxins (AFs) which are frequent in tropical and sub-tropical areas may become a concern in the EU. Aspergillus flavus and A. parasiticus, the main AF producers, are xerophilic fungi. With climate change and expected increasing temperature and decreasing rain, these fungi may find conditions that are more suitable for their development.
An inventory and modelling of the factors influencing the emergence of AFs in maize, wheat and rice crops in EU due to climate change, as well as the production of maps to highlight predicted AF contamination in these crops was requested. Therefore, the aim of the current study was to evaluate the scientific literature related to AF contamination in wheat, maize and rice, and to develop predictive models and draw maps of potential AF contamination in these crops in EU.

More information

A literature review was performed to provide an accurate “state of the art” summary regarding the role of ecological factors on the growth and metabolic activity of A. flavus and A. parasiticus. Other factors dealing with aspergillus-crop interactions, crop phenology, climate change effects/scenarios and predictive models were also investigated. This review followed the principles of the systematic literature review as described by EFSA (2011). The main sources of information were CAB abstracts on OVIDSP; ISI Web of Sciences and Scopus. In these databases it was possible to review papers from 1978 onwards. All relevant papers were assessed in full (title, abstract and text) to extract all the required information. These papers were stored in an EndNote database and all the data describing modelling, meteorology and climate change scenarios, crop phenology predictions and AF indexes were stored in MS Excel tables.

Based on these results, together with data on A. flavus produced by the consortium during the course of this project, a predictive model was developed for A. flavus growth and AFB1 production on maize. Aspergillus parasiticus and aflatoxins other than AFB1 were not included in the modelling because data were limited and not sufficient for further use in predictive modelling.

The A. flavus-AFB1 model was then linked to crop phenology data in maize, wheat and rice. The crop phenology database developed by the Joint Research Centre of the European Commission (JRC) at the EU scale was used and further adapted by the MODMAP consortium (i.e. the authors of the MODMAP project) to the specific needs of the project. The A. flavus-AFB1 model, linked with crop phenology, was based on daily meteorological data to provide AF indexes, also designated as meteorological risk indexes for AFB1 contamination in maize, wheat and rice during harvest at the European level. Daily meteorological data were obtained by the consortium using the LARS weather generator. A series of 100 years, intended as 100 runs of the LARS model, was produced for the three selected climate change scenarios (i.e. actual, +2°C and +5°C). For the mapping of these data, crucial years were selected based on the output data, considering temperature and rainfall variability between years.

The risk of AF contamination was predicted in each of the three climate change scenarios using A. flavus-AFB1 model, predicted crop flowering and harvest dates, and meterological data. Results on climate, crop phenology and AF risks were used for statistical analysis and mapping.

The literature search pointed out a lack of data on the ecology of A. parasiticus and its role in AF contamination in maize, and a lack of data as well on AF contamination in wheat at harvest. No literature was found on AF contamination in rice at harvest and on wheat and rice interaction with A. flavus.

Surveys on AF contamination in maize were found to be focused on AFB1 whereas few data were available regarding AFB2 and AFGs. Based on the data collected in Italy, AFB1 represented 90% of the total of AFs detected in maize in the last ten years, suggesting that predictions of AFB1 risk provided a reasonably good picture of the risk of total AFs.

Predictions showed a reduction in season length and an advance in flowering and harvest dates for all the crops considered; this could allow an enlargement of the crop growing areas towards the north of EU, mainly for maize and rice, because earlier ripening would then be possible in these areas.

According to the results of this project, the risk for AFs contamination is expected to increase in maize, mainly in the +2°C scenario. In this scenario, a clear increase in AF risk was shown in typical European agricultural areas such as in the centre and south of Spain, the south of Italy and in the Balkans, including Turkey (European Turkey only). Finally, a 5 and 10 days advance was predicted in flowering and harvesting dates respectively, implying no change in the agricultural practices management.

The +5°C scenario depicted a completely different situation with a considerable enlargement of AF risk areas and an overall decrease in AF risks, therefore, the extended areas of south-east Europe would present medium to low risks. In this scenario, a consistent advance in flowering and harvesting dates was estimated at 10 and 15-20 days, respectively, implying possible changes in the agricultural practices management.

In summary, in the +2°C scenario, higher levels of contamination are expected in the areas where maize is currently grown whereas in the +5°C scenario, levels of contamination are predicted to be lower but risks are expected to be wider and enlarge towards northern EU countries.

Considering the different climate change scenarios and matching them with the calculated AF risks and actual data on crop production, the following scenarios were depicted: (i) high AF risk in the southern EU countries that are not in the current main maize crop production area (i.e. in central and south of Spain, central and south of Italy, Greece, north and south-east of Portugal, Bulgaria, Albania and Cyprus); (ii) low and medium AF risks in the four main maize productive countries (i.e. in Romania, France, Hungary and north east of Italy, all accounting for 73% of the present total of the EU-27 production); (iii) low AF risks in northern European countries, that currently are in the safeguard zone regarding AF risk due to their climatic conditions; (iv) very low AF risks in Europe for wheat and rice.

The risk maps produced by the A. flavus/AF model were proposed to be used as a communication tool to reinforce prevention of AF risks by identifying priority locations for intervention. The predictions confirmed that maize is the cereal crop of concern and that both human and animal populations could be exposed to a high AF risk, at least in some EU regions. Wheat would present a negligible AF risk and rice no risk at all. However, the estimated AF risk cannot be quantitatively correlated to EU legal maximum levels for AF contamination. Nevertheless, it is suggested to gather data to understand the impact of the application of good agricultural practices in the field and good post-harvest management practices to control and prevent potential AF risks.

See the <a href="; target="_blank"> Scientific Report </a> submitted to EFSA.

Battilani, P
Università Cattolica del Sacro Cuore
Start date
End date
Project number