Volume XLII-4/W14
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-4/W14, 3–9, 2019
https://doi.org/10.5194/isprs-archives-XLII-4-W14-3-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-4/W14, 3–9, 2019
https://doi.org/10.5194/isprs-archives-XLII-4-W14-3-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

  23 Aug 2019

23 Aug 2019

A TENSOR BASED FRAMEWORK FOR LARGE SCALE SPATIO-TEMPORAL RASTER DATA PROCESSING

S. Bhattacharya, C. Braun, and U. Leopold S. Bhattacharya et al.
  • Environmental Informatics Unit, Department for Environmental Research and Innovation (ERIN), Luxembourg Institute of Science and Technology (LIST), Luxembourg

Keywords: Tensor, Tensorflow, Raster, Spatio-temporal simulation, Scalability

Abstract. In this paper, we address the curse of dimensionality and scalability issues while managing vast volumes of multidimensional raster data in the renewable energy modeling process in an appropriate spatial and temporal context. Tensor representation provides a convenient way to capture inter-dependencies along multiple dimensions. In this direction, we propose a sophisticated way of handling large-scale multi-layered spatio-temporal data, adopted for raster-based geographic information systems (GIS). We chose Tensorflow, an open source software library developed by Google using data flow graphs, and the tensor data structure. We provide a comprehensive performance evaluation of the proposed model against r.sun in GRASS GIS. Benchmarking shows that the tensor-based approach outperforms by up to 60%, concerning overall execution time for high-resolution datasets and fine-grained time intervals for daily sums of solar irradiation [Wh.m-2.day-1].