The Future

Seasonal to interannual prediction is no longer the marginal activity that it was only a few years ago. With the ability to observe the tropical Pacific Ocean both at the surface and at depth in real time, to observe the other oceans in increasing detail through satellites and moored buoys and through the ARGO system coming on stream, and with the rapidly increasing computing power available, global seasonal to interannual prediction has become an operational activity at a number of major meteorological centres. Workstations and powerful Personal Computers now allow the smallest research groups the opportunities to run RCMs, and even the global models. Empirical models can be developed and run by all. In the future further developments will continue along similar lines. Assimilation of ocean data into the coupled models, lead by the GODAE project, will improve with the implementation of ARGO and further moored buoys systems. Increasingly complex coupled models will be run in larger ensembles. The benefits of RCMs will be examined in detail. And improved empirical models will be developed within a framework that permits contrast of forecast skill with the numerical models and optimal consensus forecast building.

Nevertheless, substantial leaps in skill are unlikely to be gained. Rather a steady improvement in forecast quality will be achieved. Gains will be obtained through the use of improved quality models, through improved oceanic observations and related improved assimilation of ocean data, and perhaps through the use of land surface initialisation of models following results from the GEWEX project. RCMs may prove to provide additional valuable information for some parts of the world.

Predictability gains will be limited by the chaotic nature of the atmosphere-ocean system. The gains listed above will contribute to improvements in predicted probability distributions, with the larger ensembles providing more detail on the outlying parts of these distributions. These outlying parts contain information on the more extreme events and so are key to agricultural decision making. RCMs may be incorporated to help provide probability distributions on reduced spatial and temporal scales.

Alongside the developments in prediction science will be a marked intensification of activities related to decision making processes in agriculture, and it is in this area that the major advances are likely to come in the short term. The use of crop models will be developed and should become integrated into the ensemble outputs of the dynamical models. If necessary, weather generators will be used to obtain crop model output from empirical predictions. A focus in research will be on the integration of all prediction and historical information and its optimal use in agriculture. In support of these activities will be increased and imaginative use of existing information, both climatological and agricultural. This in turn will lead to development of the improved, coordinated climatologi-cal/agricultural data bases necessary to provide the information required. Further experiments of the type run in Algeria whereby crop patches were treated with or without use of forecast information will be used widely to assess the value of the forecasts.17

All evidence to date suggests that the potential benefits of the science are substantial, to both agriculture and to many other economic sectors, but that the effort involved in extracting optimum benefit is equally substantial and requires coordinated multidisciplinary research. Further activities towards development of such co-ordination will continue under the umbrella of the various multinational organisations in the United Nations system. These will cover not only climate science, but also all involved areas through to the social sciences. Activities such as the current Regional Climate Outlook Forums will provide a model for part of this coordination, but this will be extended through initiatives such as the Regional Climate Centres (Nicholls, 2002) and Res Agricola (Meinke and Stone, 2004). Over and above all of this will be an integrated capacity building activity designed to introduce all concerned to the opportunities and potential pitfalls of the developing science.

0 0

Post a comment