Bachelor

Fakultät 08
Automated Extraction of Time-Variable Water Surfaces with Cloud-Based Google Earth Engine
Kategorie:
Abschlussarbeit
Status:
abgeschlossen
Themengruppe:
Fernerkundung, GIS
BetreuerIn:
Extern/e BetreuerIn:
M.Sc. Christian Schwatke (DGFI-TUM)
Extern/e AutorIn:
Daniel Scherer
Jahr:
2017
Lake Tawakoni water surface area and elevation time series with error plots
Lake Tawakoni water surface area and elevation time series with error plots

Automatische Extraktion zeitvariabler Wasserflächen mittels cloud-basierter Google Earth Engine


Cloud-based data storage and geospatial processing as provided by the Google Earth Engine brings new possibilities to remote sensing. Long lasting downloads of large satellite imagery raster files, which forced the user to work on a small area of interest or caused delays before processing are no longer required, as the data is stored and processed in the cloud. Additionally, this makes the usage very flexible, as the user can work with this service on any device connected to the Internet. In this thesis, the Automated Water Area Extraction Tool (AWAX) was developed to provide water surface area time-series to interested users independently of their remote sensing knowledge and experience. AWAX utilizes the cloud-based Google Earth Engine to extract the area and extent of time-variable water surfaces on a monthly base automatically. Preprocessed images taken by the Landsat satellite missions from 1982 to 2017 are used to generate the best possible cloud-free composites of the target for every month since 1982. Four different indices - Modified Normalized Difference Water Index (MNDWI), New Water Index (NWI), Automated Water Extraction Index for Non-Shadow Areas (AWEInsh) and Automated Water Extraction Index for Shadow Areas (AWEIsh) - are applied to the composite and adaptive thresholds are used to classify water. The result is a probability map for water, from which the targets extent and area is derived. Inland water bodies with significant water level variability (Lake Tawakoni, Salton Sea, Lake Okeechobee, Lake Beysehir and Lake Tawakoni) are used as study areas for the automated water area extraction. The extracted area is compared with water level elevation products. AWAX is able to detect seasonal changes of the water surface extent for study areas with high correlation coefficients. At targets with low correlation coefficients still enough data can be extracted to interpolate the area for different water levels.


The bachelor thesis was written in cooperation with the Deutsches Geodätisches Forschungsinstitut der Technischen Universität München (DGFI-TUM).

Suche

Suchen Sie
... Abschlussarbeiten, Projekte oder Publikationen zu einem bestimmten Thema? Dann verwenden Sie das Suchformular.

... oder von einem bestimmten Autor oder Betreuer? Dann können Sie auch auf den persönlichen Seiten nachsehen.