Selection of adaptation measures

The adaptation strategies evaluated here originate from discussions with stakeholders from the seven basins about the most likely and interesting options to explore (see earlier). Farmers' main options to adapt are: conversion from irrigation to rain-fed or vice versa, installation of drainage, a change in cropping patterns, salinity control and intensification in general. The latter is of paramount importance for developing countries, since the gap between obtainable and existing yields is still enormous (Fig. 3.6). Water managers at the basin level can take decisions that can minimize the adverse impact of climate change or, in cases where the impact is positive, make better use of these positive impacts. While farmers' adaptation is limited to only the agricultural sector, water managers' decisions are related to water allocation between and within

Germany Ukraine United Turkey Ethiopia


Fig. 3.6. Exploitable yield gaps for wheat: actual versus obtainable yields for some selected countries. (Source: Fisher et al., 2002.)

different sectors. Adaptation strategies as studied at the field scale should also be considered in a basin-wide context. Reduced irrigation, as explored for a couple of crop-basin combinations, is not an adaptation strategy at the field scale to overcome the negative impact of climate change, but can be upscaled to a basin-wide strategy.

The number of crops to evaluate was limited to two and the adaptation strategies were limited to three. This was essential to minimize the number of years to evaluate, which is with the current defined options already over 15,000 years.


Overall, the general picture is that crop yields will be higher in the future (Fig. 3.7), but that variation in yields between years (Fig. 3.8) will increase as well. This indicates that concerns are predominantly related to variations in food security expressed by farmers' income, and not to total food production. It should be noted that most of this increase in total production is an effect of the increase in CO2, as presented in Table 3.1. There is still considerable debate on the validity of these data, as discussed earlier.

In terms of water resources, the total amount of water consumed (Fig. 3.9) is essential. A distinction has been made between total consumed water and so-called productively consumed water. The latter relates only to crop transpiration, while the former includes soil evaporation as well. This distinction is important since transpiration is associated with a beneficial consumption of water (crop per drop), while the latter can be seen as a real loss of water. However, a certain amount of soil evaporation it is not completely unavoidable. Important to the discussion on this productivity is that this is only relevant in cases where water is scarce. In this study percolation is not considered as water consumption, since most of this water is groundwater recharge and can be reused.















MIimiimimii in



1111 M 111111 M M111 M 111111 M


Mllllt Illllll




11111 n M 11 ii 11111 ii11ii11111ii



11111n1111111111111111111111 ii





^2010-2039 CD 2070-2099

Changes in yield (%)

Fig. 3.7. Changes in yield for the periods 2010-2039 and 2070-2099 as compared to the baseline 1961-1990. Values for HadCM3 A2 climate change projections and business as usual strategy.

Fig. 3.8. Changes in variation in yield for the periods 2010-2039 and 2070-2099 as compared to the baseline 1961-1990. Values for HadCM3 A2 climate change projections and business as usual strategy.
Fig. 3.9. Changes in water consumption for the periods 2010-2039 and 2070-2099 as compared to the baseline 1961-1990. Values for HadCM3 A2 climate change projections and business as usual strategy.

Water Productivity (WP), expressed as dollars gross return per m3 of water consumed (total consumed, so actual evapotranspiration), for each crop-basin combination was also calculated (Fig. 3.10). This is affected by local crop prices, but since these data were hard to obtain we have elected to use average world market prices over the last 5 years. Deviations are possible, since for some crops special varieties are grown that have higher local market values than world market ones. Overall, WP values range from a low $0.01/m3 to almost $1/m3.

A detailed description of results for the seven basins can be found elsewhere (Droogers and van Dam, 2004) and we will focus here on the inter-comparison of the seven basins, with some specific conclusions per basin.

Regarding the field scale inter-comparisons, one of the most striking conclusions that the overall picture of the impact of climate change on crop yields is positive (Fig. 3.7). In the business as usual option, expected yields are higher for all basins except one basin-crop combination for the period 2010-2039 and two basin-crop combinations for the period 2070-2099, as can be seen from Fig. 3.7. However, there is a price to pay for these positive impacts, which is that more water is consumed (Fig. 3.9) and, especially for the end of this century, this increase is expected to be substantial.

The overall effect of climate change on food production for the seven basins appears to be positive and one can ask whether adaptation strategies, as defined earlier, are required. The first reason is that although the impact of climate change is positive, it might be that with some adaptation strategies the positive impact can be even higher. An example is Volta Basin where more precipitation can be expected in the future, so irrigation can be intensified. A more important consideration to imple-

Fig. 3.10. Changes in water productivity for the periods 2010-2039 and 2070-2099 as compared with the baseline 1961-1990. Values for HadCM3 A2 climate change projections and business as usual strategy.

ment adaptation strategies at field scale is that at the basin scale changes will occur that will require responses at the field scale. An example is the Walawe Basin, where less water will become available for agriculture as a result of economic changes, which will result in lower water availability for agriculture. In other words, the adaptation strategies as defined will not lead automatically to higher yields.

Figure 3.11 depicts in summary what the impact of the defined adaptation strategies will be, compared to the business as usual strategy, for the period 2010-2039. In other words, the impact of climate change without any adaptation is reflected in Figs 3.7-3.11, and shows a positive trend, but what will happen if adaptation strategies are implemented? As mentioned before, some of the adaptation strategies will result in lower yields, compared to the business as usual situation, but these adaptations might be required due to basin-scale changes. However, most of the adaptation strategies analysed here will still generate higher yields in 2010-2039 compared to 1961-1990, but the business as usual approach performs better in half of the locations.

The following sections will provide for each basin a summary of the main results of the adaptation assessment at field scale. Figures 3.7-3.11 form the basis for these discussions, where the following points are important to consider.

• The baseline period (1961-1990) is used as reference for the future periods of 2010-2039 and 2070-2099.

• For each basin two crops are considered.

• For each basin two adaptation strategies are compared to the reference (business as usual).

Fig. 3.11. Impact of adaptation strategies as defined in Table 3.2 compared with the business as usual option. The changes in yield show what the impact is if the adaptation strategy had been implemented for the period 2010-2039 (HadCM3/ A2).


For the baseline scenario, rice production will increase substantially in the future, especially according to the A2 scenario. The B2 scenario shows only a small increase for the distant future, but at the cost of a high variability. Water consumption is increasing by 100—150 mm. Since the Mekong receives a substantial amount of precipitation during the main rice season, the option to cultivate rice without irrigation was tested. Somewhat surprisingly, the predicted yield was almost similar and only a small increase in variation could be seen for the 1961-1990 period. Average annual precipitation during these years was about 1800 mm. Also for the future this non-irrigated rice seems to be an option, but variation in yield will increase substantially. Although the SWAP model runs on a daily basis, rainfall was only available on a monthly basis, so drier spells within a month were not included. Finally, what the impact of a shorter growing season would be was explored. Obviously, this assumes that crop varieties exist, or will be developed, with a short growing season of about 120 days. It appears that this option will result in the highest water productivity, but not in the highest crop yield in kg/ha. This might lead to a potential conflict of interest, where farmers are less interested in water productivity than water managers are.

Maize was introduced recently as crop to be planted after the flood season in December and January. Maize is not irrigated and, because of the residual water in the soil and supplementary irrigation, it is still possible to obtain a reasonable yield. The option to grow maize in the summer season (June-September) results in very low yields according to the model and many years will have no yield at all. This is somewhat surprising, since during the summer season average rainfall is around 1000 mm.

Table 3.2. Basic crop information and adaptation strategies considered for each basin and crop combination.



Adaptation strategy 1

Adaptation strategy 2



Stop irrigation

Short season


More irrigation

Summer season



Lower groundwater

Short season


Lower groundwater

Spring season



Irrigation max. 900 mm

Deficit irrigation


Stop irrigation

Increased irrigation

Syr Darya


Reduce soil evaporation

Increased irrigation


Reduce soil evaporation

Increased irrigation



Stop irrigation







Short season

Reduced irrigation


Short season

Reduced irrigation



Reduced salinity

Reduced irrigation


Reduced salinity

Increased salinity

A closer look at the model results indicates that crop yield is hampered by water surplus, and fields are completely saturated during prolonged periods. Since we have used the same soil type as for rice, except for the puddling layer, drainage is inadequate and therefore less favourable for maize. Obviously an enhanced drainage capacity would diminish these low yield potentials.

0 0

Post a comment