Ancient maps comparison and georeferencing techniques: a case study

Two 16th century coeval maps of Po river delta area (Northern Adriatic Sea) were studied. They were made by the land surveyor Ottavio Fabri, who worked to both maps alone or together with his colleagues, Gerolamo Pontara and others. It has to be pointed out that Fabri was also author of a specific textbook concerning the field survey methodology by means of a new topographical instrument which he invented (Squadra mobile or zoppa). Stated that quite evident differences exist in the details drawn in the two maps, no a merely qualitative but quantitative comparison between the available samples must be attempt to define which is the most reliable in between. Being pregeodetic documents, a lot of problems were involved in the analysis. Different commercial softwares were used to test the reliability of georeferencing method in order to understand their effectiveness in highlighting the kinds of errors recorded in the maps. Global and local georeferencing algorithm were used in order to compare: i) ancient maps with today IGM.I sheets (direct-georeferencing); ii) each ancient map with the other (“auto-referencing”); today IGM.I map with the ancient one (“reversed-referencing”). The use of these terms is informal. The main focused problems were: i) the inability of the available softwares to record the true, original author errors; ii) the impossibility to restore the planform of the eroded landscape morphological tracts due to the current inexistence of landmarks. To overcome these problems a three-step analysis is suggested, consisting in: i) attempting the recognition of the original survey techniques and of the used ancient topographical instruments, with their accuracy; ii) evaluating the true differences between the ancient map and the preserved environmental context (i.e. mapping errors, usually not coincident with the residual errors resulting from the relocation of the landmarks due to standard geometrical transformation applied); iii) performing an extrapolation method to define the ancient landscape tracts not yet existing. Up to now some positive elements were acquired concerning point i) and ii). In particular, the possible starting base for the ancient field surveys was easily recognized on the maps; furthermore, four field survey operational areas were probably recognizable based on the analysis of the residuals and backsights original length. The usual technique adopted by the ancient land-surveyors was that of reversed–intersection applied to the main topographic base. The highest errors were recorded in the coastal areas, according to a possible adoption of a point-by-point survey technique, developed along selected pathways in the less accessible areas.