08. September 2025, 08:30 until 09:30

Master's thesis defense Konstantin Falkner

Other

Assessing Data Volume Redundancy in the Sentinel-2 Tiling Grid System

Grid-based spatial reference systems serve as fundamental frameworks for the efficient han¬dling of Earth Observation (EO) raster imagery. By projecting and partitioning the Earth’s surface onto regular planar grids and tiles, such Tiling Grid Systems (TGS) enable consis¬tent georeferencing, simplify the management of large datasets, and support the creation of spatio-temporal datacubes. This facilitates efficient access, processing, and integration of EO data across different timeframes and sensors. The Sentinel-2 (S-2) mission employs a TGS based on the Universal Transverse Mercator projections and Military Grid Reference System tiles. However, its design results in significant overlapping tile coverage, causing regions of the Earth’s surface to be mapped multiple times. This redundancy leads to data duplication, increasing storage, processing, and dissemination demands, which ultimately reduces the efficiency of S-2 imagery usage. The objective of this thesis is to quantify the extent of this redundancy across global landmasses using S-2 datasets. By assessing the volumes of the datasets stored in both the S-2 TGS, including data duplicates, and in a ref¬erence framework nearly free of redundancy, the system’s efficiency is evaluated through direct comparison. Datasets with varying properties and different experimental setups are employed to investigate how redundancy depends on the type of data as well as geographic regions. The results align with previous studies based on land surface area calculations and reveal that the S-2 TGS causes substantial overheads of up to 32% additional data volume. These findings underscore the necessity of optimised spatial reference frameworks to support the increasing requirements of modern EO data.

Calendar entry

Event details

Event location
Sem. DA grün 02 A (access via 2nd floor yellow)
1040 Wien
Wiedner Hauptstraße 8
Organiser
TU Wien
Public
Yes
Entrance fee
No
Registration required
No