18.104.22.168 Research tasks not yet undertaken – ongoing uncertainties
Several uncertainties remain unresolved since the TAR. Better knowledge in several research areas is critical to improve our ability to predict the magnitude, and often even the direction, of future climate change impacts on crops, as well as to better define risk thresholds and the potential for surprises, at local, regional and global scales.
In terms of experimentation, there is still a lack of knowledge of CO2 and climate responses for many crops other than cereals, including many of importance to the rural poor, such as root crops, millet, brassica, etc., with few exceptions, e.g., peanut (Varaprasad et al., 2003) and coconut (Dash et al., 2002). Importantly, research on the combined effects of elevated CO2 and climate change on pests, weeds and disease is still insufficient, though research networks have long been put into place and a few studies have been published (Chakraborty and Datta, 2003; Runion, 2003; Salinari et al., 2006). Impacts of climate change alone on pest ranges and activity are also being increasingly analysed (e.g., Bale et al., 2002; Todd et al., 2002; Rafoss and Saethre, 2003; Cocu et al., 2005; Salinari et al., 2006). Finally, the true strength of the effect of elevated CO2 on crop yields at field to regional scales, its interactions with higher temperatures and modified precipitation regimes, as well as the CO2 levels beyond which saturation may occur, remain largely unknown.
In terms of modelling, calls by the TAR to enhance crop model inter-comparison studies have remained unheeded; in fact, such activity has been performed with much less frequency after the TAR than before. It is important that uncertainties related to crop-model simulations of key processes, including their spatial-temporal resolution, be better evaluated, as findings of integrated studies will remain dependent upon the particular crop model used. It is still unclear how the implementation of plot-level experimental data on CO2 responses compares across models; especially when simulations of several key limiting factors, such as soil and water quality, pests, weeds, diseases and the like, remain either unresolved experimentally or untested in models (Tubiello and Ewert, 2002). Finally, the TAR concluded that the economic, trade and technological assumptions used in many of the integrated assessment models to project food security under climate change were poorly tested against observed data. This remains the situation today (see also Section 5.6.5).