Refine
Document Type
- Doctoral Thesis (4) (remove)
Language
- English (4)
Has Fulltext
- yes (4)
Is part of the Bibliography
- no (4)
Keywords
- Fernerkundung (4) (remove)
Institute
- Institut für Geographie und Geologie (4) (remove)
Seas and oceans are essential for the global ecosystem. Entire societies, economies and countless livelihoods rely on their good environmental status. Yet, pressures on marine environments are increasing. An extensive assessment and monitoring of marine habitats is a vital precondition for understanding these systems and their sustainable conservation. Remote sensing methods can temporally accelerate the mapping, improve the spatial resolution and support the interpretation of large areas. Hydroacoustic becomes the method of choice for areas deeper than the coastal zone as optical signals are limited by strong attenuation in the water column. Apart from depth measurements for the creation of bathymetric charts, the recording of backscatter strength is useful for the characterization of the seafloor surface. The direct influence of the inhabiting benthic community on the backscattered signal is rarely considered, although it can be utilized for the detection of benthic life. Information about habitat-specific backscatter responses or a hydroacoustic remote sensing catalog for benthic habitats is missing so far.
The multibeam echosounder (MBES) has the advantage of recording both, bathymetry and backscatter strength simultaneously with related incidence angle. Further, recent technological developments allow to change between frequencies. Angular range curves supported the quantification of backscatter strength of different frequencies. Acoustic data sets were complemented by ground truthing in form of sedimentological and biological samples as well as video profiles. Study areas were located offshore the island of Sylt in the North Sea as well as in vicinity to Oder Bank and close to the coast offshore Hohe Düne/Rostock, both in the Baltic Sea. Investigated habitats included sand areas inhabited by tubeworms, loose mussel clusters on top of sand areas, seagrass meadows, coarse sand and gravel areas, and a reef covered by mussels.
Multifrequency backscatter maps, combining frequencies between 200 kHz and 700 kHz, illustrate small-scale features at the seafloor not visible in monofrequent maps. Key habitats showed a specific backscatter response, which can partly be related to macrobenthic flora and fauna. Data sets recorded with a (partly calibrated) MBES in three different month (May, August, October) revealed that backscatter strength can further detect spatial as well as temporal habitat dynamics. Alterations in the sediment composition at the seafloor surface of the ecologically valuable coarse sand and gravel areas were caused by seasonal changes in local hydrodynamics.
A newly developed 3D seismic lander has the ability to support hydroacoustic remote sensing as an additional, non-destructive ground truthing method utilizing a high frequency of 130 kHz to image the shallow subsurface. Buried objects, e.g., stones, shells, fruit gummy worms, as well as sediment disturbances could be detected and visualized in a laboratory experiment. The 3D seismic lander is likely to improve the investigation of volume scatter contribution to backscatter strength and is potentially applicable for the imaging of bioturbation.
The achievement and monitoring of a good environmental status on continental shelf seas requires
the use of acoustic remote sensing techniques due to their range. The interpretation of acoustic signals
for the identification of benthic communities, however, is still in its infancy. In this thesis, the results
of two field campaigns conducted in a sandy environment off the shore of Sylt Island (North Sea)
utilizing ship- and lander-based acoustic and optical remote sensing techniques are discussed. The
objective of the thesis is a better knowledge of the impact of the polychaete Lanice conchilega on
physical seafloor properties, especially roughness at a cm to mm scale, which is relevant for
understanding acoustic scatter. The results show a clear impact of L. conchilega on roughness even in
sparse populations of less than 2% coverage. However, these sparsely populated areas could not be
reliably identified with acoustic data; a denser population of L. conchilega provided a clearer signal for
the acoustic remote sensing methods. The results are promising regarding the broader use of acoustic
remote sensing techniques for environmental monitoring in selected habitats, although the
determination of minimum population thresholds that can be identified will require further studies.
A large portion of the earth's surface is covered with various vegetation classes (i.e. grassland, wetland and agricultural area, forest) of many diverse species and canopy configurations. The ability to assess and to monitor canopy parameters, such as biomass, leaf area index, and vegetation water content, is of vital importance to the study of different agronomic processes. Remote sensing techniques provide a unique capability towards probing different vegetation types and canopy by operating at different bands, observation angle etc. Over the past decades, significant progress has been made in remote sensing techniques of land processes specially vegetation characteristics through development of advanced ground-based, airborne and space-borne microwave sensors, methods and approaches such as theoretical, semi-empirical and empirical models, needed for analyzing the data. These activities have sharply increased in recent years since the launch of different active and passive satellites and sensors. Remote Sensing (RS) science and techniques combined with ground truth data can provide new tools for advanced agricultural crop applications. It has been demonstrated that RS has the ability to estimate biophysical parameters of agricultural crops over time at local, regional, and global scales. In this study, RS images in visible/near infrared (VIS/NIR) domain as well as microwave domain combined with ground truth data were used to assess biophysical parameters of agricultural crop during their whole growing season at Durable Environmental Multidisciplinary Monitoring Information Network (DEMMIN) test site in North East Germany. Ground truth studies were carried out for 31 weeks during 17th April – 13th November 2013 over three crop lands including winter wheat, barley and canola. Landsat 8 OLI, Landsat 7 ETM+ were used for the VIS/NIR studies and TerraSAR-X synthetic aperture radar (SAR) images were used to study biophysical parameters of agricultural crops in microwave part of electromagnetic spectrum. The analysis was conducted by calculating different vegetation indices (VIs) to estimate the biomass (fresh and dry), LAI, and vegetation water content (VWC) of three crops using Landsat 8 OLI and Landsat 7 ETM+ combined with ground truth data. A new concept of Soil Line retrieval from Landsat 8 image was also developed to estimate plant biophysical parameters using soil line related vegetation indices in optical domain of electromagnetic spectrum. Different approaches including univariate, multivariate stepwise regression and semi-empirical water cloud model was also used to estimate the biophysical parameters of agricultural crop using TerraSAR-X data in microwave domain of electromagnetic spectrum. Perhaps the most important conclusion of this study is that the RS approach can provide useful information about estimating agricultural crop parameters over time and local scale, which can therefore provide valuable information to aid the agronomy community.
Kurze Inhaltszusammenfassung in einer weiteren Sprache (Englisch): Muong Lay is one of the important social and economic areas in Northwestern Vietnam. Landslides occur frequently in the area and seriously affect local livelihoods and living conditions. Therefore, the problem of landslide hazard and mitigation for a sustainable development of this area is significant. The spatial analysis of landslide hazard assessment in the mountainous regions in Muong Lay is important to address this development challenge. This study focuses on the application of GIS and RS to landslide hazard assessment, especially for support of GIS modeling to landslide hazard susceptibility for Muong Lay area. By using Remote sensing with LandSat TM image and aerial photos of scale 1:50,000 and using statistical models with GIS-software’s ENVI3.4, ILWIS3.0, PCI9.0 and ARC/GIS9.1, the study tries to evaluate and estimate the landslide in relation with naturally different elements of natural conditions such as geology, geomorphology, geology-engineering, tectonics, hydrology, rainfall, etc… Especial, the study firstly aims to produce the causal factor maps by verifying digital data. These factors then will be applied in a methodology based on statistical methods such as: “bivariate statistical analysis” and “multivariate statistical analysis” approach to calculate the susceptibility level of each class of each factor to landslide. The integration of Geographic Information System (GIS) and Remote Rensing (RS)for landslide hazard zonation and assessment is a valid approach. In these researches various methods for image integration and information extraction have been analysed and evaluated in detail.