| In the current processes of rural cadastral revision, the geometric and radiometric quality of the digital orthophotographs play an increasingly relevant role. The on-screen digitalisation of the geographic entities is conditioned by many factors (resolution of the orthophoto, precision in the georreferenciation, radiometric density, viewing zoom,...), which ultimately result in certain imprecision. The combination of these factors produces errors not only in the correct location of the rural plots, but also, and very importantly, in the value of their cadastral surface.The cadastral surface of the rural estate is derived from the on-screen digitalisation of the perimeter, and its comparison with the rural surface obtained in topographical works shows some surface errors of even 10%. On the other hand, there are many cities that only have paper cartographies largely inconsistent in terms of the surface registered in the Cadastral Database and the surface registered in the Cadastral Map.Thus, the four main variables studied are: the cadastral surface obtained in the on-screen digitalisation of Digital Ortofotographs, the surface obtained in the digitalisation on the paper cadastral map, the surface registered in the Cadastral Database and the rural surface obtained in topographical works.The comparison of all these variables shows some surface errors and, this doctoral thesis studies precisely the best parameters to define the cadastral surface errors in rural real estate. Apart from the traditional relative surface error [ESR], another parameter called Index of Boundary Uncertainty [IBU] is proposed. This parameter distributes the absolute surface error in the area where this error occurs, that is to say, in the perimeter, being a parameter more independent of the size of the plots. The correlation between both parameters is very high, particularly for error values lower than 5%. Nevertheless, it is always necessary to take into account the conditions under which the experimental part of the research is performed. |
