Conclusions

One of the most striking conclusions of this study is that the overall impact of climate change on crop yields is positive. Figure 3.7 illustrates that in the business as usual strategy expected yields are higher for all basins except for one basin-crop combination in the period 2010-2039 and for two basin-crop combinations in the period 2070-2099. Also variation in crop yields is going down for more than half of the basin-crop combinations (Fig. 3.8), despite the increase in climate variability in the HadCM3 projections. However, there is a price to pay for these positive impacts and that is that more water is consumed (Fig. 3.9) and, especially for the end of this century, this increase is expected to be substantial.

Water productivity (Fig. 3.10) shows a mixed picture, with some basin-crop com binations having 10-20% higher values and others 50% lower values. It is important that water productivity values are a function of market prices, which we have assumed constant here. Although this is highly unlikely, it provides good insight into the combined yield and water consumption processes.

Future crop yields are potentially higher as a result of enhanced CO2 levels. At a basin scale it has to be evaluated whether the increased amount of water required for the crops is available. As mentioned in the introduction, this chapter only describes the field scale component of the ADAPT project and linkages to the basin scale will be described in the other chapters. The adaptation strategies as explored here can be used in basin-scale studies to evaluate alternative options for dealing with increased water consumption.

One of the dominant factors in the analysis in this chapter is the impact of elevated CO2 levels on crop production. This impact, and especially the long-term impacts and feedback, is still under debate. However, results of numerous experimental studies as described in the methods section indicate that a doubling in CO2 levels can indeed induce higher yields by up to 50%.

We employed the relation of Doorenbos and Kassam (1979) to derive relative crop yield as a function of relative transpiration. In order to apply this methodology, the potential crop growth in a certain weather year had to be estimated from the potential transpiration (Droogers and van Dam, 2004). The adopted methodology has the tendency to overestimate the variation of crop yields in a sequence of years. Further studies with simple but reliable crop growth routines are required.

The increase in variation in crop yields is not as dramatic as expected. One of the most important reasons for this is the way irrigation scheduling was included in the simulations. The irrigation scheduling option in SWAP implies that if a crop experienced stress beyond a defined threshold value, irrigation was considered to take place. The consequence is that during dry years, through either low rainfall or high temperatures, more irrigation would be applied than during wet years. This is only a realistic assumption when sufficient storage capacity exists in a basin. This scheduled irrigation approach guarantees that irrigation timing is always optimal, ensuring that results are not affected by this timing but only by the total water applied.

0 0

Post a comment