Virtual Special Issue on food security – greater than anticipated impacts of near‐term global atmospheric change on rice and wheat

Working Group II (WGII) of the Intergovernmental Panel on Climate Change (IPCC), concluded in their fourth assessment report (AR4) that despite climate change and increasing world population, crop surpluses at the global level would continue, resulting in a ‘small decline in real world food (cereal) prices’ over this century (www.ipcc.ch/publications_and_data/publications_and_data_reports.shtml). In 2009, Assistant Director-General Hafez Ghanem of the UN Food and Agriculture Organization (FAO) stated ‘FAO is cautiously optimistic about the world's potential to feed itself by 2050’ (www.fao.org/news/story/en/item/35571/icode/). This is despite the fact that with increasing affluence and a third more people it was estimated that 70% more food would be needed by 2009. Underpinning this optimism is the fact that yields have increased steadily at around 1.5–2% p.a. for almost 60 years, and that this might be expected to continue, or even accelerate in response to increased prices (IPCC AR4). While it has to be hoped that this optimism is realized, the year-on-year increases achieved with the two most important food crops to the world ceased for wheat and slowed for rice in the first years of this Century (Fig. 1). Further, research subsequent to the IPCC assessment shows aspects of global atmospheric change anticipated for mid-century that will drive down yields. This research is consolidated in this Virtual Special Issue of Global Change Biology comprising 11 articles that focus on post AR4 assessments of the response of rice and of wheat to rising carbon dioxide, tropospheric ozone and temperature (Ainsworth, 2008; Biswas et al., 2008; Feng et al., 2008, 2011; Shimono et al., 2008; Taub et al., 2008; Liu et al., 2010; Asseng et al., 2011; Wall et al., 2011; Zhu et al., 2011; Pleijel & Uddling, 2012). Atmospheric [CO2] is expected to rise to about 550 ppm by mid-century. In theory this would be sufficient to increase yield by almost 30%. Ainsworth (2008) conducted a meta-analysis of studies of the impacts of rising [CO2] and of tropospheric [O3] on rice. This work showed that while yield increases approaching 30%, and of the scale anticipated by IPCC WGII, were seen in some studies in which [CO2] was elevated in artificial enclosed environments, ‘the more closely the fumigation mimicked field conditions, the smaller was the stimulation of yield by elevated [CO2].’ The average increase in yield under Free Air Concentration Enrichment (FACE) of [CO2], where elevation is under entirely open-air field conditions, was just 12%. Indeed, results from the large-scale FACE experiment in Japan, in which [CO2] was elevated to ca. 580 ppm, showed that rice yield was increased on average by just 11.5% (Shimono et al., 2008). Across the studies surveyed by Ainsworth (2008), combining elevated temperature with the elevation of [CO2] negated any increase in rice yield. In a separate meta-analysis, growth under elevated [CO2] was found to decrease the grain protein content of both rice and wheat by ca. 10% (Taub et al., 2008). Pleijel & Uddling (2012) confirm this effect in an extensive meta-analysis of the impact of elevated [CO2] and [O3] on wheat, incorporating 57 experiments which cover 24 cultivars and 112 independent experimental treatments. Here, too yield increase by elevated [CO2] in wheat was negatively correlated with protein content of the grain. Pre-industrial concentrations of ozone in the troposphere ([O3]) were less than 10 ppb. Today, levels in much of the northern hemisphere are 40–60 ppb, with higher levels in E. and S. Asia, and may increase another 20–80% by mid-century according to the IPCC AR4 Working Group I. Tropospheric O3 is a secondary pollutant resulting from the interaction of sunlight, gaseous hydrocarbons with nitrogen oxides. While the primary pollutant may be formed in cities, O3 is often formed hundreds and thousands of miles from the primary source, elevating levels over important rural areas of cereal production. Across prior studies, rice exposed to elevated [O3] at an average of 62 ppb, showed a 14% decrease in grain yield (Ainsworth, 2008). In a parallel meta-analysis Feng et al. (2008) show an even greater impact of elevated [O3] on wheat across all prior studies with an 18% yield loss for an average [O3] of just 43 ppb. Although grain protein content was elevated by 7%, when combined with the yield mass loss, this still resulted in significantly less protein per hectare. In another study, Feng et al. (2011) show a significant difference between the physiological responses of two Chinese winter wheat cultivars to elevated [O3] of 57 ppb using a large-scale FACE system. In a different study at the same site, Zhu et al. (2011) compared the yield of four winter wheat cultivars to elevated [O3] over three growing seasons. Elevated [O3] was found to lower grain yield on average by 20%, in agreement with prior studies in other regions of wheat production. Given that [O3] has been rising for 50 years, it might be expected that plant breeders would have inadvertently selected for increased tolerance. However, Biswas et al. (2008) analyzed the response of 20 cultivars of winter wheat released over the past 60 years to O3. Rather than be more resistant to O3, the more modern cultivars were the most sensitive. This was attributed to the higher stomatal conductance which supports their higher photosynthetic rates, but also allows more ozone to enter and damage the leaf. While the negative effect of [O3] and the smaller than expected stimulation of yield by [CO2] were not included in the IPCC 2007 assessment, the impacts of rising temperature were. Temperature is expected to rise by 1–2 °C by mid-century. These changes were thought to have a mild impact on grain production. However, the analysis of temperature variability on wheat yields by Asseng et al. (2011) suggests, alarmingly, that these effects may be much larger than previously thought. Using current variation in growing season temperature across the main wheat growing area of Australia (±2 °C), these authors show a 50% loss of yield in the warmer years. This is attributed to accelerated leaf senescence during grain-filling at >34 °C. Since conditions in the Australian wheat growing areas reflect those of the major regions of the world, such yield reductions could be expected globally if the predicted increase in temperatures during grain-filling projected for mid-century are realized (Asseng et al., 2011). Elevated ozone similarly causes premature leaf senescence during grain-filling (Feng et al., 2008, 2011); yield losses might be even greater where these two changes combine. In a parallel study, Liu et al. (2010) show from 25 years of crop data that warmer weather on the North China Plain during the vegetative growth phase has shortened the growing season of given winter wheat cultivars, with negative effects on yield. However, in this case, cultivars more suited to this increase have been adopted, offsetting the yield loss due to warming during the vegetative phase. Wall et al. (2011) used a crop temperature feed-back controlled array of infra-red lamps to provide an open-air elevation of temperature in a wheat crop. They found that a 1.3 °C elevation of the crop temperature lowered soil volumetric water content by 14%. This demonstrates that the damaging direct effects of temperature on pollination and grain-filling are compounded with decreased soil moisture and increased crop water stress. The stagnation in increase in the yield of rice and wheat per hectare over the past decade has been attributed to the fact that means to improvement used in, and following, the green revolution are approaching their biological limits. This VSI suggests that warming and rising tropospheric ozone may already have played a part. While rising CO2 may partly offset these yield losses it is driving down grain quality, in terms of protein content. The very real concern now must be whether increasing temperature and ozone, particularly in E. and S. Asia, will begin to drive down the gains of the late 20th century rather than provide the year-on-year increases that had been anticipated.