This article is dedicated to the Tout&TS-scatter plot studies using temperature domain (TD) method, which allows the complete (Tout'I; TS'i)-point set corresponding to the whole heating season (HS) represent as HS-TD which one can divided into several smaller sub-TDj with respect that sub-TD1 sub-TD2 sub-TD3 ... =HS-TD. Thus, the regression model of the whole HS surrounding Tout&TS-scatter plot can be set up additively, combining separate regression models of the appropriate sub-TDj. Deeper and more detailed picture of those "j-th" (Tout'I; TS'i)-point subsets associated with mentioned above sub-TDj has been derived by calculating their frequency distribution functions (FDF). On the FDF-base was developed temperature spectroscopy of such DH parameters as outside temperature, Tout, supply and return water temperature, TS and TR, etc. Differential and cumulative (integral) FDF, namely, dFDF(Tout)&dFDF(TS) and cFDF(Tout)&cFDF(TS), corresponding to different sub-TDj have been calculated in at least 3%-high resolution because temperature class intervals of 1- and 2-degrees in length were used. In Tout&TS-scatter plot studies the design of regression model and assessment of its fitness have always been a topical issue. Therefore the model fitness was measured by the coefficient of determination, R^2: the higher the R^2 value, the higher the fitness of the regression model.