An estimate of global temperature increase by means of a fuzzy logic model

The climate change scenarios developed by the Intergovernmental Panel on Climate Change (IPCC) indicate a wide range of future concentration of greenhouse gases and the corresponding range of temperature increases. From these data, it can be inferred that higher temperature increases are directly related to higher emission levels of greenhouse gases and to the increase in their atmospheric concentrations. It is also evident that lower temperature increases are related to smaller amounts of emissions and, therefore, to lower greenhouse gases concentrations. In this work, simple linguistic rules are extracted by means of visual inspection of the IPCCs Fourth Assessment Report. These rules describe the relations between the greenhouse gases emissions, their concentrations, the radiative forcing associated with concentrations, and the corresponding temperature changes as would be obtained by expert opinion. These rules are used to build a fuzzy model, which uses emission and concentration values of greenhouse gases as input variables and gives, as output, the temperature increase projected at year 2100. A second fuzzy model based on Zadeh´s extension principle is also build using temperature values obtained from a simple, deterministic climate system model. Both fuzzy models are very attractive because their simplicity and capability to integrate the uncertainties associated to the input and output variables. These simple models contain all the information of much more complex determinist models, characteristics that make easier to understand the behavior of the system and help to produce climate change scenarios that could be more meaningful for policy-makers.