Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Oct 20;13(11):737.
doi: 10.3390/toxins13110737.

Modelling the Effects of Weather Conditions on Cereal Grain Contamination with Deoxynivalenol in the Baltic Sea Region

Katarzyna Marzec-Schmidt  1 , Thomas Börjesson  2 , Skaidre Suproniene  3 , Małgorzata Jędryczka  4 , Sigita Janavičienė  3 , Tomasz Góral  5 , Ida Karlsson  6 , Yuliia Kochiieru  3 , Piotr Ochodzki  5 , Audronė Mankevičienė  3 , Kristin Piikki  1
Affiliations

Modelling the Effects of Weather Conditions on Cereal Grain Contamination with Deoxynivalenol in the Baltic Sea Region

Katarzyna Marzec-Schmidt et al. Toxins (Basel). .

Abstract

Fusarium head blight (FHB) is one of the most serious diseases of small-grain cereals worldwide, resulting in yield reduction and an accumulation of the mycotoxin deoxynivalenol (DON) in grain. Weather conditions are known to have a significant effect on the ability of fusaria to infect cereals and produce toxins. In the past 10 years, severe outbreaks of FHB, and grain DON contamination exceeding the EU health safety limits, have occurred in countries in the Baltic Sea region. In this study, extensive data from field trials in Sweden, Poland and Lithuania were analysed to identify the most crucial weather variables for the ability of Fusarium to produce DON. Models were developed for the prediction of DON contamination levels in harvested grain exceeding 200 µg kg-1 for oats, spring barley and spring wheat in Sweden and winter wheat in Poland, and 1250 µg kg-1 for spring wheat in Lithuania. These models were able to predict high DON levels with an accuracy of 70-81%. Relative humidity (RH) and precipitation (PREC) were identified as the weather factors with the greatest influence on DON accumulation in grain, with high RH and PREC around flowering and later in grain development and ripening correlated with high DON levels. High temperatures during grain development and senescence reduced the risk of DON accumulation. The performance of the models, based only on weather variables, was relatively accurate. In future studies, it might be of interest to determine whether inclusion of variables such as pre-crop, agronomic factors and crop resistance to FHB could further improve the performance of the models.

Keywords: Fusarium head blight—FHB; Spearman’s rank correlation coefficient; deoxynivalenol (DON) prediction; machine learning; mycotoxins; phenological development; small grain cereals.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

Figures

Figure 1
Figure 1
Spearman’s rank correlation coefficient for deoxynivalenol (DON) contamination in Swedish spring oats at harvest and different weather factors estimated for 14-day moving windows during the growing season. Red indicates a positive correlation and blue a negative correlation (both p ≤ 0.01) between DON contamination and a particular weather variable, with a darker colour indicating a higher value of the correlation coefficient. Tmin-daily minimum temperature, Tmean-daily mean temperature, Tmax-daily maximum temperature, RH-mean relative humidity, PREC-precipitation, VPD-vapour pressure deficit.
Figure 2
Figure 2
Spearman´s rank correlation coefficient for deoxynivalenol (DON) contamination in Swedish spring barley at harvest and different weather factors estimated for 14-day moving windows during the growing season. Red indicates a positive correlation and blue a negative correlation (both p ≤ 0.01) between DON contamination and a particular weather variable, with a darker colour indicating a higher value of the correlation coefficient. Tmin-daily minimum temperature, Tmean-daily mean temperature, Tmax-daily maximum temperature, RH-mean relative humidity, PREC-precipitation, VPD-vapour pressure deficit.
Figure 3
Figure 3
Spearman´s rank correlation coefficient for deoxynivalenol (DON) contamination in Swedish spring wheat at harvest and different weather factors estimated for 14-day moving windows during the growing season. Red indicates a positive correlation and blue a negative correlation (both p ≤ 0.01) between DON contamination and a particular weather variable, with a darker colour indicating a higher value of the correlation coefficient. Tmin-daily minimum temperature, Tmean-daily mean temperature, Tmax-daily maximum temperature, RH-mean relative humidity, PREC-precipitation, VPD-vapour pressure deficit.
Figure 4
Figure 4
Spearman´s rank correlation coefficient for deoxynivalenol (DON) contamination in Lithuania grown spring wheat at harvest and different weather factors estimated for 14-day moving windows during the growing season. Red indicates a positive correlation and blue a negative correlation (both p ≤ 0.01) between DON contamination and a particular weather variable, with a darker colour indicating a higher value of the correlation coefficient. Tmean-daily mean temperature, RH-mean relative humidity, PREC-precipitation, VPD-vapour pressure deficit.
Figure 5
Figure 5
Spearman´s rank correlation coefficient for deoxynivalenol (DON) contamination in Polish winter wheat at harvest and different weather factors estimated for 14-day moving windows during the growing season. Red indicates a positive correlation and blue a negative correlation (both p ≤ 0.01) between DON contamination and a particular weather variable, with a darker colour indicating a higher value of the correlation coefficient. Tmean-daily mean temperature, PREC-precipitation.
Figure 6
Figure 6
Variable importance in the Random Forest-based model for Sweden grown spring barley. PREC-precipitation, RH-mean relative humidity, Tmax-daily maximum temperature, Tmean-daily mean temperature, WS-mean wind speed, WD-wind direction. PREC_106-PREC 15.07–28.07, RH _092-RH 01.07–14.07, Tmax_099-Tmax 08.07–21.07, Tmax_106-Tmax 15.07–28.07, Tmean_099-Tmean 08.07–21.07, Tmean_106-Tmean 15.07–28.07, WD_001-WD 01.04–14.04, WD_057-WD 27.05–09.06, WS_008-WS 08.04–21.04, WS_106-WS 15.07–28.07.
Figure 7
Figure 7
Distribution of the minimal depth of the variable and its mean in the Random Forest-based model for Sweden grown spring barley. PREC-precipitation, RH-mean relative humidity, Tmax-daily maximum temperature, Tmean-daily mean temperature, WS-mean wind speed, WD-wind direction. WD_057-WD 27.05–09.06, Tmax_099-Tmax 08.07–21.07, Tmean_099-Tmean 08.07–21.07, Tmean_106-Tmean 15.07–28.07, RH _092-RH 01.07–14.07, WD_036-WD 06.05–19.05, PREC_106-PREC 15.07–28.07, Tmax_106-Tmax 15.07–28.07, WS_106-WS 15.07–28.07, WS_008-WS 08.04–21.04.
Figure 8
Figure 8
Variable importance in the Random Forest-based model for Sweden grown spring wheat. PREC-precipitation, RH-relative humidity, Tmax-daily maximum temperature, WS-wind speed, WD-wind direction, VPD-vapour pressure deficit. PREC_022-PREC 22.04–05.05, PREC_085-PREC 24.06–07.07, RH _029-RH 29.04–12.05, RH_036-RH 06.05–19.05, Tmax_099-Tmax 08.07–21.07, VPD_036-VPD 06.05–19.05, WS_008-WS 08.04–21.04, WS_050-WS 20.05–02.06, WS_057-WS 27.05–09.06, WS_092-WS 01.07–14.07.
Figure 9
Figure 9
Distribution of the minimal depth of the variable and its mean in the Random Forest-based model for Sweden grown spring wheat. PREC-precipitation, RH-relative humidity, Tmax-daily maximum temperature, WS-wind speed, WD-wind direction, VPD-vapour pressure deficit. RH_036-RH 06.05–19.05, PREC_106-PREC 15.07–28.07, WS_050-WS 20.05–02.06, WD_099-WD 08.07–21.07, WS_057-WS 27.05–09.06, WS_092-WS 01.07–14.07, VPD_036-VPD 06.05–19.05, PREC_085-PREC 24.06–07.07, Tmax_099-Tmax 08.07–21.07, RH _001-RH 01.04–14.04.
Figure 10
Figure 10
Variable importance in the Random Forest-based model for Lithuania grown spring wheat. PREC-precipitation, Tmean-daily mean temperature. PREC_022-PREC 22.04–05.05, Tmean_008-Tmean 08.04–21.04, Tmean_015-Tmean 15.04–28.04, Tmean_022-Tmean 22.04–05.05, Tmean_029-Tmean 29.04–12.05, Tmean_36-Tmean 06.05–19.05, Tmean_085-Tmean 24.06–07.07, Tmean_092-Tmean 01.07–14.07, Tmean_099-Tmean 08.07–21.07, Tmean_106-Tmean 15.07–28.07.
Figure 11
Figure 11
Distribution of the minimal depth of the variable and its mean in the Random Forest-based model for Lithuania grown spring wheat. Tmean-daily mean temperature, PREC-precipitation. Tmean_008-Tmean 08.04–21.04, Tmean_099-Tmean 08.07–21.07, Tmean_106-Tmean 15.07–28.07, Tmean_015-Tmean 15.04–28.04, Tmean_001-Tmean 01.04–14.04, PREC_022-PREC 22.04–05.05, Tmean_036-Tmean 06.05–19.05, Tmean_085-Tmean 24.06–07.07, PREC_071-PREC 10.06–23.06, Tmean_022-Tmean 22.04–05.05.
Figure 12
Figure 12
Variable importance in Random Forest-based model for Poland grown winter wheat. PREC-precipitation, Tmean-daily mean temperature. PREC_029-PREC 29.05–11.06, PREC_036-PREC 05.06–18.06, PREC_050-PREC 19.06–02.07, PREC_057-PREC 26.06–09.07, PREC_064-PREC 03.07–16.07, PREC_092-PREC 31.07–13.08, Tmean_015-Tmean 15.05–28.05, Tmean_057-Tmean 26.06–09.07, Tmean092-Tmean 31.07–13.08, Tmean_099-Tmean 08.08–21.08.
Figure 13
Figure 13
Distribution of the minimal depth of the variable and its mean in the Random Forest-based model for Poland grown winter wheat. PREC-precipitation, Tmean-daily mean temperature. PREC_057-PREC 26.06–09.07, Tmean_099-Tmean 08.08–21.08, PREC_092-PREC 31.07–13.08, PREC_064-PREC 03.07–16.07, Tmean_057-Tmean 26.06–09.07, PREC_050-PREC 19.06–02.07, PREC_036-PREC 05.06–18.06, Tmean_015-Tmean 15.05–28.05, PREC_029-PREC 29.05–11.06, Tmean092-Tmean 31.07–13.08.
Figure 14
Figure 14
Location of field trials conducted in Sweden (SE), Poland (PL) and Lithuania (LT).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Moretti A., Pascale M., Logrieco A.F. Mycotoxin risks under a climate change scenario in Europe. Trends Food Sci. Technol. 2019;84:38–40. doi: 10.1016/j.tifs.2018.03.008. - DOI
    1. Perrone G., Ferrara M., Medina A., Pascale M., Magan N. Toxigenic fungi and mycotoxins in a climate change scenario: Ecology, genomics, distribution, prediction and prevention of the risk. Microorganisms. 2020;8:1496. doi: 10.3390/microorganisms8101496. - DOI - PMC - PubMed
    1. Bottalico A., Perrone G. Toxigeneic Fusarium species and mycotoxins associated with head blight in small-grain cereals in Europe. Eur. J. Plant Pathol. 2002;108:611–624. doi: 10.1023/A:1020635214971. - DOI
    1. Moss M.O., Thrane H. Fusarium taxonomy with relation to trichothecene formation. Toxicol. Lett. 2004;153:23–28. doi: 10.1016/j.toxlet.2004.04.021. - DOI - PubMed
    1. Kosiak B., Torp M., Skjerve E., Thrane U. The prevalence and distribution of Fusarium species in Norwegian cereals: A survey. Acta Agric. Scand. Sect. B Soil Plant Sci. 2007;53:168–176. doi: 10.1080/09064710310018118. - DOI

Publication types