The Regional Impacts of Climate Change

	Other reports in this collection

5.3.2.1. Agricultural Production

Although agriculture on a global basis could remain unchanged relative to base production in a changed climate-as a result of technological improvements and more efficient agricultural practices (IPCC 1996, WG II, Chapter 13)-there will be large regional discrepancies; these differences may be less acute in Europe, where agriculture is highly advanced.

Agricultural production can be described in terms of amount and quality. The reactions of an individual crop to global change will depend on the balance of shorter cycles resulting from increased air temperatures, shorter periods to accumulate yield products (at least in the case of determinate crops), higher potential yields resulting from increased assimilation of CO₂, and increased water-use efficiency resulting from enhanced regulation of transpiration with elevated CO₂. This equilibrium could be quite subtle in the highly mechanized context of European agriculture.

Projections of future European production derive from controlled-conditions experiments and from feeding crop simulation models, with climatic observations adjusted according to scenarios of future climate. Quantitative results of simulations therefore are highly dependent on the type of climate scenario used (especially in terms of available precipitation), though qualitative indications (trends) generally are constant among published results.

Most simulation and experimental studies so far have used expected fluctuations of mean values for climate variables, but increasing emphasis is being put on possible consequences of a more variable interannual and intra-annual climate (i.e., within and between years).

5.3.2.1.1. Annual crops

In the following paragraphs, crops are classified according to their main growing period: winter (i.e., late autumn, winter, spring, and sometimes early summer) or summer (i.e., spring, summer, and early autumn). For Europe, winter crops include winter cereals, peas, and rapeseed; summer crops include maize (forage and grain), sunflowers, potatoes, sugar beet, and spring cereals. The results below need to be tempered by the physiological characteristics of crops-for instance, the shortening of the crop season resulting from increasing temperature is far less marked on nondeterminate crops (peas, potatoes) than on determinate crops (wheat).

Suitable areas

As a result of increasing air temperatures in winter, the risks associated with damaging frosts will be reduced as a whole. This factor will allow expansion of winter cereals and probably other winter crops (CLAIRE, 1996) in large areas such as southern Fennoscandia, western Russia, and the Alpine regions-topography, physiography, soils, and socioeconomic circumstances permitting. Increasing spring temperatures will accelerate soil temperature increases and extend suitable zones for most summer crops, allowing a reasonable length for their growth season. In the case of sunflowers (and probably for grain maize), the area of suitability would extend eastward to Belarus and northward to northern Germany and southern Fennoscandia for a specified climate change. As an illustrative example, the northern limit of reliable spring cereal cultivation in Finland has been estimated to shift northward by 100-150 km for each 1�C of warming (Carter et al., 1996). For a representative range of temperature scenarios, the rate of northward shift is about 10-80 km per decade (Saarikko and Carter, 1996). Some northward shift also is simulated for vegetable crops like onion (CLAIRE, 1996). As far as temperature is concerned, there is no region that would become completely unsuitable for agriculture-even in southern Europe under high-temperature climate change scenarios. Studies currently are being carried out to determine what damage can occur in crops at extremely high temperatures.

Crop duration

Expected increases in temperature will cause faster rates of development as a whole (see "Adaptive Responses" below) and shorten the length of growing periods for determinate crops, consequently shortening the length of the grain-filling period. The total growing season for these crops may be reduced by 15-30 days, depending on the climate scenarios used; in this respect, crop duration could be more reduced in central and eastern Europe (4 weeks) than in western Europe (3 weeks). Cereal harvest dates therefore would occur sooner. Nevertheless, some indication is given by Miglietta et al. (1995) that a lack of cold days could reduce vernalization effects and consequently lengthen the first part of the growing season for winter cereals. Because of faster rates of development, nondeterminate crops would develop more potential harvestable organs (theoretically during a longer period) because no development event would stop the process of production; increasing temperatures, however, probably would favor an increased senescence rate of the crop and tend to mitigate the beneficial effect of increases in potential organs by reducing the length of crop photosynthetic life.

Temperature increases in spring and summer will accelerate the course of crop development more crucially on short-cycle crops that are sown in spring than on winter crops. This general rule must be adapted, however, to local conditions (see, e.g., Saarikko and Carter, 1996)-who show by simulation that the sowing-to-leading phase in spring cereals in Finland declines by about 1 day per 1�C warming, and the heading-to-yellow ripening shortens by 2-4 days per 1�C warming). A simulation exercise on sunflowers shows a reduction of the crop cycle by 10-50 days with UKTR3140 scenario and 10-70 days with UKTR6675 (see Harrison and Butterfield, 1996 for extensive results on sunflower development under different GCM-based scenarios). This quite important change is likely to affect most of Europe, with a gradient from the southWest (low reduction in Spain and Italy, where cycles are already short) to the northeast (Poland and Russia). This pattern can be extrapolated to all determinate summer crops. Nondeterminate crops would experience a faster rate of development as well, which would induce earlier senescence.

Crop yields

Accounting for the enhancement of growth resulting from increasing CO₂ concentrations, the potential yield of winter crops (assuming that neither precipitation nor irrigation is limiting) would increase almost everywhere (with central or southern Europe experiencing the highest winter wheat yield boosts, depending on the climatic scenario). If water limitations are considered, crop response apparently would depend on the scenario chosen for the time evolution of CO₂ concentrations. In the case of winter wheat, there is some indication that the rate of increase in yields across Europe could be 0.2-0.36 T/ha/decade under the IS92a emission scenario and 0.13 T/ha/decade with the IS92d emission scenario, under both the UKTR3140 and the UKTR6675 climate scenarios. The largest increases would occur in central and eastern Europe (regardless of changes in management practices that may occur in some countries as a result of changes in economic structure) and in southern Europe (see Table 5-2). All winter crops probably would follow the pattern of winter wheat yield changes. The largest increases per country might occur in northern Europe because of increased possibilities for taking winter cereals into cultivation.

For summer crops, determinate crop yields would be affected by the shortened crop cycle and reduced time to assimilate supply and grain-filling periods. On the other hand, improvements in the rate of dry-matter production can be expected from enhanced CO₂ concentrations. This effect has been illustrated for sunflowers (CLAIRE, 1996), where simulation models suggest a compensation between the negative effects of temperature and the positive effects of carbon fertilization. Under the IS92a emission scenario, yield benefits occur in central and eastern Europe and losses occur in western Europe-but all of them are small. With IS92d emissions, the gains are even smaller, but the losses are amplified (see Table 5-3). The same pattern can be expected for all determinate summer crops (even if the fertilizing effect of atmospheric CO₂ is less beneficial to C₄ crops such as maize). Similar results were obtained with onion, whose simulated yield increased 4-8% with UKTR3140 and GFDL2534 (IS92a emissions) and 15% (resp. 7%) with UKTR6675 and GFDL5564, IS92a (resp. IS92c)-CO₂ effects again counteracting a negative effect of reduced growth period on yield (see claire, 1996). For nondeterminate crops, Peiris et al. (1996) expect potato yields to increase by as much as 35% under northern conditions as a result of the lengthening of the growing season, regardless of CO₂ fertilization effects.

(continues on next page...)