全球变化知识资源中心

By the middle of the twenty-first century, it is probable that climate change will result in more frequent wheat crop failures accross Europe¹. There are many reasons why the frequency and severity of crop failures might increase in the future, albeit with large regional differences. Some adaptive measures to minimize yield losses show more promise than others, yet none of them seem to be sufficient to fully avoid the problem.

Assessments of the potential impact of climate change on agriculture have flooded the scientific literature over the past decade. They range from detailed laboratory or field experiments^{2, 3} to global impact studies⁴. In the majority of cases, these analyses account for probable crop physiological responses to either temperature, rainfall or CO₂. These factors are generally evaluated progressively, rarely considering changes in the extremes of climatic variables (such as intensive rainfall events or heatwaves) or the combined effect of extreme events on crop physiology and crop management practice. Thus, most studies examine climate change impacts from a monocausal, crop physiological perspective. Yet, as every farmer knows, producing an economically viable yield requires the effective management of a multitude of potential perils, combined with a fair amount of skill and luck.

As they report in Nature Climate Change, Trnka and colleagues¹ take a refreshingly different approach to this problem. Not only do they avoid the common monocausal trap, they also resist the temptation to 'over-quantify' climate change impacts on wheat yields by, for instance, using highly parameterized production models. Instead, the authors only simulate the bare essentials — the probable changes in crop development (phenology) over the next 50 years from the present period (1981–2010) to the middle of the twenty-first century (2051–2070) — for 14 locations across Europe. These simulations provide the necessary input dates (sowing, anthesis and maturity) for a carefully designed, multi-peril risk assessment.

Trnka et al.¹ selected 14 case study locations across 13 countries covering the major wheat-producing regions of the European Union (including some possible future production sites), ranging from Seville in Spain to Jyvaskyla in Finland and from Rothamsted in the UK to Athens in Greece. For each site they evaluated the changing frequencies in the occurrence of 11 separate risk factors, considering each factor individually as well as combinations of factors. These risk factors include direct climate impacts on crop growth such as frost, heat stress, water logging and drought, but also climatic impacts on crop management such as adverse conditions during either sowing or harvesting.

While several previous studies suggested a possible northern shift of European wheat production due to changes in thermal suitability, the study by Trnka et al. strikes a more nuanced note by considering the probability of multiple risk factors, either in isolation or by considering their combined probabilities. The study found that overall, and in spite of the uncertainties associated with climate change projections, the 'adverse event frequency' is more likely to increase than decrease across all 14 European sites. This raises a cautionary flag for policy makers as it indicates that the problem a changing climate poses is complex and cannot be solved by simply moving production north, in accordance with the changing temperature gradients.

© LIV FRIIS-LARSEN / ALAMY

1. Trnka, M. et al. Nature Clim. Change 4, 637–643 (2014).
2. Ahuja, I. et al. Trends Plant Sci. 15, 664–674 (2010).
   • CAS
   • PubMed
   • Article
3. de Oliveira, E. D. et al. Funct. Plant Biol. 40, 160–171 (2013).
   • CAS
   • Article
4. Nelson, G. C. et al. Food Security, Farming, and Climate Change to 2050: Scenarios, Results, Policy Options (IFPRI, 2010).
5. Zhang, T., Zhu, J. & Yang, X. Agricul. For. Meteorol. 148, 1412–1418 (2008).
   • Article

Download references