Importance of impacts scenarios for the adaptation of agriculture to climate change

The great possibility that the climate is already changing, and the most drastic way possible, increases the challenge of agricultural engineering, especially in environmentally vulnerable areas and in regions where agriculture has a high economic and social importance. Knowledge of potential impacts that may be caused by changes in water and thermal regimes in coming decades is increasingly strategic, as they allow the development of techniques to adapt agriculture to climate change and therefore minimizes the risk of undesirable impacts, for example, in food and nutritional security. Thus, the main objective of this paper is to describe a way to generate impacts scenarios caused by anomalies of precipitation and temperature in the definition of climate risk zoning of an agricultural crop very important in the tropics, such as the sugar cane, especially in central-southern Brazil, which is one of its main world producers. A key point here is the choice of the climate model to be used, considering that 23 different models were used in the fourth IPCC report published in 2007. The number and range of available models requires the definition of criteria for choosing the most suitable for the preparation of the impacts scenarios. One way proposed and used in this work is based on the definition of two groups of models according to 27 technical attributes of them. The clustering of 23 models in two groups, with a model representing each group (UKMO_HadCM3 and MIROC3.2_medres), assists the generation and comparison of impacts scenarios, making them more representative and useful. Another important aspect in the generation of impacts scenarios is the estimate of the relative importance of the anomalies of precipitation and temperature, which are the most commonly used. To assess the relative importance of the anomalies are generated scenarios considering an anomaly at a time and both together. The impacts scenarios for a high emission of greenhouse gases (A2), from 2010 to 2039, were more drastic for the sugar cane in central-southern Brazil using the UKMO_HadCM3 model than the MIROC3.2_medres model. These impacts scenarios, however, were less drastic than those generated for the arabica coffee in the same simulation conditions, reinforcing the increased vulnerability of this agricultural crop to climate change than the sugar cane. The inclusion of other restrictions on the climate risk zoning improves the quality of the generated scenarios and expands its usefulness for agricultural engineering.