To reconstruct sub-regional European climate over the past centuries, several efforts have been made using historical datasets. However, only scattered information at low spatial and temporal resolution have been produced to date for the Mediterranean area. This paper has exploited, for Southern and Central Italy (Mediterranean Sub-Regional Area), an unprecedented historical dataset as an attempt to model seasonal (winter and summer) air temperatures in pre-instrumental time (back to 1500). Combining information derived from proxy documentary data and large-scale simulation, a statistical methodology in the form of multiscale-temperature regression (MTR)-model was developed to adapt larger-scale estimations to the sub-regional temperature pattern. The modelled response lacks essentially of autocorrelations among the residuals (marginal or any significance in the Durbin-Watson statistic), and agrees well with the independent data from the validation sample (Nash-Sutcliffe efficiency coefficient >0.60). The advantage of the approach is not merely increased accuracy in estimation. Rather, it relies on the ability to extract (and exploit) the right information to replicate coherent temperature series in historical times.