There is not yet any clear consensus regarding the magnitude and sign of interactions between elevated CO2 and nutrient availability for crop growth. Reviews of available data indicate that, on average, plants grown at high nutrient supply respond more strongly to elevated CO2 than nutrient-stressed plants (Poorter, 1993, 1998). Nevertheless, the current rise in atmospheric CO2 concentration may help plants cope with soil nutritional deficiencies (Idso and Idso, 1994) and especially with low nitrogen availability (Lloyd and Farquhar, 1996; Drake et al., 1997). Several authors emphasize that a strong increase in biomass production under elevated CO2 cannot be sustained in low fertilizer input systems without an appropriate increase in nutrients assimilation (Comins and McMurtrie, 1993; Gifford, 1994; Schimel, 1998). When other nutrients are not strongly limiting, a decline in nitrogen availability could be prevented by an increase in biological N2 fixation under elevated CO2 (Gifford, 1992, 1994). In fertile grasslands, legumes benefit more from elevated CO2 than nonfixing species, resulting in significant increases in symbiotic N2 fixation (Soussana and Hartwig, 1996; Zanetti et al., 1996).
Plants grown under elevated CO2 generally increase the allocation of photosynthates to roots (Rogers et al., 1996b; Murray, 1997), which increases the capacity and/or activity of belowground carbon sinks (Rogers et al., 1994; Canadell et al., 1996; Körner, 1996), enhancing root turnover (Pregitzer et al., 1995; Loiseau and Soussana, 1999b), rhizodeposition (Cardon, 1996), and mycorrhizal development (Dhillion et al., 1996) in some but not all systems. Some measurements also have shown an increase in soil N cycling (Hungate et al., 1997a), in response to short-term enrichment in CO2, although other studies have shown either no detectable change (Prior et al., 1997) or even a reduction in soil N mineralization (Loiseau and Soussana, 2001). The relationships between C and N turnover in soils after exposure to elevated CO2 therefore are not fully understood, and it is still a matter of debate whether the availability of soil nitrogen for crop plants is reduced after a step increase in atmospheric CO2 concentration.
Soil organic carbon (SOC) stocks result from the balance between inputs and decomposition of soil organic matter (SOM). Residues of cotton (Torbert, et al., 1995), soybean, and sorghum (Henning et al., 1996) display increased C:N ratios from growth under elevated CO2, which may reduce their rate of decomposition in the soil and lead to an increment in ecosystem carbon stocks, similar to that observed in fertile grasslands (Casella and Soussana, 1997; Loiseau and Soussana, 1999a). However, some studies (Newton et al., 1996; Ross et al., 1996, Hungate et al., 1997b) suggest higher carbon turnover rather than a substantial net increase in soil carbon under elevated CO2. Predicted increased air and soil temperatures can be expected to increase the mineralization rate of SOM fractions that are not physically or chemically protected. The degree of protection of SOM varies with several soil-specific factors, including structure, texture, clay mineralogy, and base cation status. This may lead in the long term to negative effects on structural stability, water-holding capacity, and the availability of certain nutrients in the soil (see Reilly et al., 1996). Organic matter decomposition tends to be more responsive than NPP to temperature, especially at low temperatures (Kirschbaum, 2000). Within this range, any warming would stimulate organic matter decomposition (carbon loss) more than NPP (carbon gain); the net response would be a loss of soil carbon. Mineralization rates also are influenced by soil water content. For example, lower soil moisture in Mediterranean regions (see Chapter 3) would compensate temperature increase effects on carbon and nitrogen mineralization (Leiros et al.,1999).
As a result of these interactions with soil processes, experiments that impose
sudden changes in temperature or CO2 and last only a few years are
unlikely to predict the magnitude of long-term responses in crop productivity,
soil nutrients (Thornley and Cannell, 1997), and carbon sequestration (Luo and
Reynolds, 1999). This may implyin agreement with Walker et al. (1999)that
the actual impact of elevated CO2 on crop yields in farmers'
fields could be less than in earlier estimates that did not take into account
limitations of nutrient availability and plant-soil interactions.
Land management will continue to be the principal determinant of SOM content
and susceptibility to erosion during the next few decades, but changes in vegetation
cover resulting from short-term changes in weather and near-term changes in
climate are likely to affect SOM dynamics and erosion, especially in semi-arid
regions (Valentin, 1996; Gregory et al., 1999).
The severity, frequency, and extent of erosion are likely to be altered by changes (see Table 3-10) in rainfall amount and intensity and changes in wind (Gregory et al., 1999). Models demonstrate that rill erosion is directly related to the amount of precipitation but that wind erosion increases sharply above a threshold windspeed. In the U.S. corn belt, a 20% increase in mean windspeed greatly increases the frequency with which the threshold is exceeded and thus the frequency of erosion events (Gregory et al., 1999). Thus, the frequency and intensity of storms would have substantial effects on the amount of erosion expected from water and wind (Gregory et al., 1999). Different conclusions might be reached for different regions. Thus, before predictions can be made, it is important to evaluate models for erosion and SOM dynamics (Smith et al., 1997). By reducing the water-holding capacity and organic matter contents of soils, erosion tends to increase the magnitude of nutrient and water stress. Hence, in drought-prone and low-nutrient environments such as marginal croplands, soil erosion is likely (high confidence) to aggravate the detrimental effects of a rise in air temperature on crop yields.
Modest progress has been made in understanding of pest (weeds, insects, pathogens) response to climate change since the SAR. Oerke et al. (1995) estimate preharvest losses to pests in major food and cash crops to be 42% of global potential production. Rosenzweig et al. (2000) suggest that ranges of several important crop pests in the United States have expanded since the 1970s, including soybean cyst nematode and corn gray leaf blight; these expansions are consistent with enabling climate trends, although there are competing explanations. Promising work linking generic pest damage mechanisms with crop models is reported by Teng et al. (1996). For example, Luo et al. (1995) linked the BLASTSIM and CERES-RICE models to simulate the effects of climate change on rice leaf blast epidemics. They found that elevated temperature increases maximum blast severity and epidemics in cool subtropical zones; it inhibits blast development in warm humid subtropics. Such model linkages have been used to examine climate change impacts on weed-crop competition (e.g., for rice-weed interactions see Graf et al., 1990) and insect pests (Venette and Hutchison, 1999; Sutherst et al., 2000). Any direct yield gain caused by increased CO2 could be partly offset by losses caused by phytophagous insects, pathogens, and weeds. Fifteen studies of crop plants showed consistent decreases in tissue nitrogen in high CO2 treatments; the decreases were as much as 30%. This reduction in tissue quality resulted in increased feeding damage by pest species by as much as 80% (Lincoln et al., 1984, 1986; Osbrink et al., 1987; Coviella and Trumble, 1999). Conversly, seeds and their herbivores appear unaffected (Akey et al., 1988). In general, leaf chewers (e.g., lepidoptera) tend to perform poorly (Osbrink et al., 1987; Akey and Kimball, 1989; Tripp et al., 1992; Boutaleb Joutei et al., 2000), whereas suckers (e.g., aphids) tend to show large population increases (Heagle et al., 1994; Awmack et al., 1997a; Bezemer and Jones, 1998)indicating that pest outbreaks may be less severe for some species but worse for others under high CO2. It is important to consider these biotic constraints in studies on crop yield under climate change. Nearly all previous climate change studies excluded pests (Coakley et al., 1999).
Other reports in this collection