As for the trend identification, the literature is full of quantitative studies, but with respect to two-probability distribution function (PDF) description remains at qualitative level. The main purpose of this paper is to convert the two-PDF description into quantitative form for objective and quantitative assessments. Such a goal is achieved on the basis of innovative trend templates, which indicate the arithmetic averages and the standard deviations of the two equal length halves of the given temperature record. Two-PDF explanations are represented by three combinations as changes in the arithmetic averages, standard deviations and in both parameters through their logical connections with the innovative template description. The application of the methodology is given for four mean annual temperature records from Istanbul City, Turkey. Global warming variations are described quantitatively as increases or decreases in the statistical parameters. The proposed methodology can be applied for any temperature record in the world.