Limits...
Impacts of IOD, ENSO and ENSO Modoki on the Australian Winter Wheat Yields in Recent Decades.

Yuan C, Yamagata T - Sci Rep (2015)

Bottom Line: It is found that IOD plays a dominant role in the recent three decades; the wheat yield is reduced (increased) by -28.4% (12.8%) in the positive (negative) IOD years.In contrast, the ENSO Modoki may have its distinct impacts on the wheat yield variations, but they are much smaller compared to those of IOD.The present study may lead to a possible scheme for predicting wheat yield variations in Australia in advance by use of simple climate mode indices.

View Article: PubMed Central - PubMed

Affiliation: Key Laboratory of Meteorological Disaster of Ministry of Education, Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science &Technology, Nanjing 210044, China.

ABSTRACT
Impacts of the Indian Ocean Dipole (IOD), two different types of El Niño/Southern Oscillation (ENSO): canonical ENSO and ENSO Modoki, on the year-to-year winter wheat yield variations in Australia have been investigated. It is found that IOD plays a dominant role in the recent three decades; the wheat yield is reduced (increased) by -28.4% (12.8%) in the positive (negative) IOD years. Although the canonical ENSO appears to be responsible for the wheat yield variations, its influences are largely counted by IOD owing to their frequent co-occurrence. In contrast, the ENSO Modoki may have its distinct impacts on the wheat yield variations, but they are much smaller compared to those of IOD. Both the observed April-May and the predicted September-November IOD indices by the SINTEX-F ocean-atmosphere coupled model initialized on April 1st just before the sowing season explain ~15% of the observed year-to-year wheat yield variances. The present study may lead to a possible scheme for predicting wheat yield variations in Australia in advance by use of simple climate mode indices.

No MeSH data available.


Related in: MedlinePlus

Time series of (a) the Australian winter wheat yields, (b) their year-to-year anomalous percentages (%) and (c) correlation coefficients with the three-month-running mean indices of IOD (DMI, red line), canonical ENSO (Niño3, blue line) and ENSO Modoki (EMI, gray line). The correlation coefficients significant at the 95% confidence level are marked by the open squares in c. The September-November DMI (red, ˚C), November-January Niño3 (blue, ˚C) and June-August EMI (gray, ˚C) are superimposed in b after multiplied by 15. Years in the X axis in a-b denote the years when the wheat is sowed.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4663488&req=5

f1: Time series of (a) the Australian winter wheat yields, (b) their year-to-year anomalous percentages (%) and (c) correlation coefficients with the three-month-running mean indices of IOD (DMI, red line), canonical ENSO (Niño3, blue line) and ENSO Modoki (EMI, gray line). The correlation coefficients significant at the 95% confidence level are marked by the open squares in c. The September-November DMI (red, ˚C), November-January Niño3 (blue, ˚C) and June-August EMI (gray, ˚C) are superimposed in b after multiplied by 15. Years in the X axis in a-b denote the years when the wheat is sowed.

Mentions: As shown in Fig. 1, the Australian wheat yield shows an apparent linear trend in the recent three decades probably owing to improvement in the agricultural technology. Besides, it undergoes considerable year-to-year variations. Here, the year-to-year variations are referred to the anomalous percentage deviated from the five-year running mean7. We have calculated the linear correlation coefficients with the seasonal IOD indices (DMI), canonical El Niño indices (Niño3) and ENSO Modoki indices (EMI) before and during the growing seasons. Results show that the year-to-year wheat yield variations have the highest coefficient of −0.64 with DMI in September-November (SON) at the 99% confidence level by the two-tailed t test. With Niño3, the highest coefficient is −0.49 in November-January (NDJ) at the same confidence level. With EMI, it is −0.39 in June-August (JJA), significant at the 95% confidence level. We note that the tropical climate modes may impact the wheat continuously throughout the whole growing seasons. The seasons of SON, NDJ and JJA are selected here not because IOD and ENSO have the strongest influences in these specific seasons, but because the indices in these specific seasons may represent the integrated impacts of the tropical climate modes on the wheat yields. It is apparent that IOD has the highest correlation coefficient with the year-to-year wheat yield variations, which implies the most important role of IOD and seems to be consistent with the recent remarkable impacts of IOD on the Australian precipitations910. Although the present study has focused on the wheat yield variations at the national level, we note that IOD also shows higher correlation coefficients than ENSO at the provincial levels (Supplementary Figs 1–2).


Impacts of IOD, ENSO and ENSO Modoki on the Australian Winter Wheat Yields in Recent Decades.

Yuan C, Yamagata T - Sci Rep (2015)

Time series of (a) the Australian winter wheat yields, (b) their year-to-year anomalous percentages (%) and (c) correlation coefficients with the three-month-running mean indices of IOD (DMI, red line), canonical ENSO (Niño3, blue line) and ENSO Modoki (EMI, gray line). The correlation coefficients significant at the 95% confidence level are marked by the open squares in c. The September-November DMI (red, ˚C), November-January Niño3 (blue, ˚C) and June-August EMI (gray, ˚C) are superimposed in b after multiplied by 15. Years in the X axis in a-b denote the years when the wheat is sowed.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4663488&req=5

f1: Time series of (a) the Australian winter wheat yields, (b) their year-to-year anomalous percentages (%) and (c) correlation coefficients with the three-month-running mean indices of IOD (DMI, red line), canonical ENSO (Niño3, blue line) and ENSO Modoki (EMI, gray line). The correlation coefficients significant at the 95% confidence level are marked by the open squares in c. The September-November DMI (red, ˚C), November-January Niño3 (blue, ˚C) and June-August EMI (gray, ˚C) are superimposed in b after multiplied by 15. Years in the X axis in a-b denote the years when the wheat is sowed.
Mentions: As shown in Fig. 1, the Australian wheat yield shows an apparent linear trend in the recent three decades probably owing to improvement in the agricultural technology. Besides, it undergoes considerable year-to-year variations. Here, the year-to-year variations are referred to the anomalous percentage deviated from the five-year running mean7. We have calculated the linear correlation coefficients with the seasonal IOD indices (DMI), canonical El Niño indices (Niño3) and ENSO Modoki indices (EMI) before and during the growing seasons. Results show that the year-to-year wheat yield variations have the highest coefficient of −0.64 with DMI in September-November (SON) at the 99% confidence level by the two-tailed t test. With Niño3, the highest coefficient is −0.49 in November-January (NDJ) at the same confidence level. With EMI, it is −0.39 in June-August (JJA), significant at the 95% confidence level. We note that the tropical climate modes may impact the wheat continuously throughout the whole growing seasons. The seasons of SON, NDJ and JJA are selected here not because IOD and ENSO have the strongest influences in these specific seasons, but because the indices in these specific seasons may represent the integrated impacts of the tropical climate modes on the wheat yields. It is apparent that IOD has the highest correlation coefficient with the year-to-year wheat yield variations, which implies the most important role of IOD and seems to be consistent with the recent remarkable impacts of IOD on the Australian precipitations910. Although the present study has focused on the wheat yield variations at the national level, we note that IOD also shows higher correlation coefficients than ENSO at the provincial levels (Supplementary Figs 1–2).

Bottom Line: It is found that IOD plays a dominant role in the recent three decades; the wheat yield is reduced (increased) by -28.4% (12.8%) in the positive (negative) IOD years.In contrast, the ENSO Modoki may have its distinct impacts on the wheat yield variations, but they are much smaller compared to those of IOD.The present study may lead to a possible scheme for predicting wheat yield variations in Australia in advance by use of simple climate mode indices.

View Article: PubMed Central - PubMed

Affiliation: Key Laboratory of Meteorological Disaster of Ministry of Education, Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science &Technology, Nanjing 210044, China.

ABSTRACT
Impacts of the Indian Ocean Dipole (IOD), two different types of El Niño/Southern Oscillation (ENSO): canonical ENSO and ENSO Modoki, on the year-to-year winter wheat yield variations in Australia have been investigated. It is found that IOD plays a dominant role in the recent three decades; the wheat yield is reduced (increased) by -28.4% (12.8%) in the positive (negative) IOD years. Although the canonical ENSO appears to be responsible for the wheat yield variations, its influences are largely counted by IOD owing to their frequent co-occurrence. In contrast, the ENSO Modoki may have its distinct impacts on the wheat yield variations, but they are much smaller compared to those of IOD. Both the observed April-May and the predicted September-November IOD indices by the SINTEX-F ocean-atmosphere coupled model initialized on April 1st just before the sowing season explain ~15% of the observed year-to-year wheat yield variances. The present study may lead to a possible scheme for predicting wheat yield variations in Australia in advance by use of simple climate mode indices.

No MeSH data available.


Related in: MedlinePlus