טכניון מכון טכנולוגי לישראל
הטכניון מכון טכנולוגי לישראל - בית הספר ללימודי מוסמכים  
M.Sc Thesis
M.Sc StudentRonen Grinstein
SubjectResearch and Implementation of Transformations between
Horizontal Geodetic Datums for Geographic
Databases
DepartmentDepartment of Civil and Environmental Engineering
Supervisor Dr. Gabay Yair


Abstract

     

          Definition of a geodetic datum translates to attribution of spatial coordinates to a point. Coordinate transformation have been executed for many decades but only in recent years, when numerous geodetic datums have been defined and calculated, they became an integral part of the mapping process in engineering, industry, photogrammetry, remote sensing and geodesy. The motivation for the research originated from experience with geodetic datum transformations and common GIS software, and familiarity with the relevant features that are usually inflexible, limited and inadequate for certain situations. Transfer of geographic data, between different organizations and countries, from different periods, utilizing various datums in the foundation of their databases, requires an enhancement of current standards and tools.


          Definition of a datum transformation is based on points with known coordinates in two datums and a geodetic or mathematical model, which describes a group of parameters and their relationship or a method of interpolation. The research distinguishes between global models that are defined by a single set of parameters and local models that are defined by a group of common points and a method of interpolating differences between the datum coordinates. Global models analyzed are Bursa-Wolf 7-parameter, Molodensky 3-parameter and Molodensky-Badekas 4-parameter transformations. Local models analyzed are transformation by radial proximity and transformation by Delaunay triangulation.


          Method: Two geodetic datums were selected: the old Israel datum and the new Israel datum. A data set of 31 common control points was selected for calculation of parameters and for local transformations. Another set of 321 points, of the same quality, was selected for the analysis of accuracy and comparison between models. An attempt was made to predict the appropriate model for the data, by preliminary analysis of the coordinates. Then parameters were calculated for the global models, search and interpolation methods were selected for the local transformations and finally, the 321 points were transformed using all the models. The transformed coordinates were compared to known coordinates in the destination datum and the differences were presented graphically and analyzed.


          Conclusions: The research describes a detailed comparison between transformations by global and local models and between the models and precise reference data. Utilization of local models for a low-accuracy datum situation, which yielded good results comparing to conservative models, is presented. Finally, recommendations for selection of an appropriate model for geographic database transformations and fairly simple implementation procedures for GIS software are presented.