An evaluation of void-filling interpolation methods for SRTM data

1 The Digital Elevation Model that has been derived from the February 2000 2 Shuttle Radar Topography Mission (SRTM) has been one of the most 3 important publicly available new spatial datasets in recent years. However, the 4 'finished' grade version of the data (also referred to as Version 2) still contains 5 data voids (some 836,000 km 2) – and other anomalies – that prevent immediate 6 use in many applications. These voids can be filled using a range of 7 interpolation algorithms in conjunction with other sources of elevation data, 8 but there is little guidance on the most appropriate void filling method. This 9 paper describes: (i) a method to fill voids using a variety of interpolators, (ii) a 10 method to determine the most appropriate void filling algorithms using a 11 classification of the voids based on their size and a typology of their 12 surrounding terrain; and (iii) the classification of the most appropriate 13 algorithm for each of the 3,339,913 voids in the SRTM data. Based on a 14 sample of 1,304 artificial but realistic voids across six terrain types and eight 15 void size classes, we found that the choice of void filling algorithm is 16 dependent on both the size and terrain type of the void. Contrary to some 17 previous findings, the best methods can be generalised as: Kriging or Inverse 18 Distance Weighting interpolation for small and medium size voids in relatively 19 flat low-lying areas; Spline interpolation for small and medium sized voids in 20 high altitude and dissected terrain; Triangular Irregular Network or Inverse 21 Distance Weighting interpolation for large voids in very flat areas, and an 22 advanced Spline Method (ANUDEM) for large voids in other terrains.