Assessment of spatial interpolation techniques for generating an accurate digital elevation surface using combined radar and LiDAR Elevation data. Proceedings of the 8th International Conference on Microwaves

the process of generating a digital elevation surface (DSM) is still long-lasting task with the development in remote sensing (RS) technology and geographic information system (GIS). The procedure is very significant as it provides a true representation of topography in digital format which can be used for analysis or visualisation or both. DSM can be generated through spatial interpolation technique which is a process of estimating the values of a specific attribute at unsampled locations based on the values of the attributes at the sampled locations. This study was conducted to test and analyze the interpolation techniques for deriving a DSM from combined use of radar and LiDAR data in order to demonstrate the level of confidence with which the interpolation techniques can generate a better interpolated continuous surface, and improve the elevation accuracy of DSM extracted by individual data. We used point maps generated from Geoscience Laser Altimetry System (GLAS) onboard Ice-Cloud-Elevation satellite (ICESat) and RADARSAT Antarctic Mapping Project (RAMP) data. Different interpolation techniques were applied to these datasets. Deterministic interpolation techniques such as inverse distance weighted (IDW), global polynomial interpolation (GP), local polynomial interpolation(LP), radial basis function (RBF) and stochastic interpolation techniques such as simple kriging(SK), ordinary kriging (OK), universal kriging (UK), disjunctive kriging (DK) and Co-Kriging were used. A set of 20 ground survey points were used for accuracy assessment to calculate the elevation differences between DSM and accurate ground survey (GPS) points. Accuracy assessment suggests that the DK interpolation provides the most accurate elevation for RAMPbased point elevation data, while RBF and SK works superior for GLAS point elevation data interpolation. It is also evident that OK and UK provide superior results for RAMP+GLAS based point data. In conclusion, the work suggests that DK interpolation techniques provide the most accurate elevation surface as compared to other interpolation techniques used for RAMP-based point elevation data.