No description available.

No description available.

The objective of the thesis is the model-based analysis of spatial patterns of decomposition properties on the forested slopes of the montane level (ca. 1200-2200 m a.s.l.) in a study area in the Italian Alps (Val di Sole / Val di Rabbi, Autonomous Province of Trento). The analysis includes humus forms and enchytraeid assemblages as well as pH values, activities of extracellular enzymes and C/N ratios of the topsoil. The first aim is to develop, test and apply data-based techniques for spatial modelling of soil ecological parameters. This methodological approach is based on the concept of digital soil mapping. The second aim is to reveal the relationships between humus forms, soil organisms and soil microbiological parameters in the study area. The third aim is to analyze if the spatial patterns of indicators of decomposition differ between the landscape scale and the slope scale. At the landscape scale, sample data from six sites are used, covering three elevation levels at both north- and south-facing slopes. A knowledge-based approach that combines a decision tree analysis with the construction of fuzzy membership functions is introduced for spatial modelling. According to the sampling design, elevation and slope exposure are the explanatory variables. The investigations at the slope scale refer to one north-facing and one south-facing slope, with 30 sites occurring on each slope. These sites have been derived using conditioned Latin Hypercube Sampling, and thus reasonably represent the environmental conditions within the study area. Predictive maps have been produced in a purely data-based approach with random forests. At both scales, the models indicate a high variability of spatial decomposition patterns depending on the elevation and the slope exposure. In general, sites at high elevation on north-facing slopes almost exclusively exhibit the humus forms Moder and Mor. Sites on south-facing slopes and at low elevation exhibit also Mull and Amphimull. The predictions of those enchytraeid species characterized as Mull and Moder indicators match the occurrence of the corresponding humus forms well. Furthermore, referencing the mineral topsoil, the predictive models show increasing pH values, an increasing leucine-aminopeptidase activity, an increasing ratio alkaline/acid phosphomonoesterase activity and a decreasing C/N ratio from north-facing to south-facing slopes and from high to low elevation. The predicted spatial patterns of indicators of decomposition are basically similar at both scales. However, the patterns are predicted in more detail at the slope scale because of a larger data basis and a higher spatial precision of the environmental covariates. These factors enable the observation of additional correlations between the spatial patterns of indicators of decomposition and environmental influences, for example slope angle and curvature. Both the corresponding results and broad model evaluations have shown that the applied methods are generally suitable for modelling spatial patterns of indicators of decomposition in a heterogeneous high mountain environment. The overall results suggest that the humus form can be used as indicator of organic matter decomposition processes in the investigated high mountain area.

The Wadden Sea off the coast of the southern North Sea is the largest coherent area of tidal flats worldwide. As a highly productive ecosystem it is of global importance, e.g. as nursery for fish and as a feeding and resting area for 10 – 12 million migratory birds following the East Atlantic Flyway. The outstanding ecological significance of this region corresponds to a high level of protection by EU directives and national law and by inscription as UNESCO World Heritage Site, all of which requires regular monitoring and assessment. Apart from the ecological aspects, the Wadden Sea is also of great importance for coastal protection. To survey the extensive, often inaccessible tidal area, remote sensing is essential and while mainly airborne techniques have been carried out for decades, now high-resolution satellite-borne sensors open up new possibilities relevant for monitoring and long-term ecological research. Especially satellite synthetic aperture radar (SAR) sensors offer a high availability of acquisitions as they operate largely independently of daylight and weather. The aim of the studies presented here was to explore the use of data from the TerraSAR-X satellite to record geomorphological structures and habitats for Wadden Sea Monitoring. More than 100 TerraSAR-X acquisitions from 2009 to 2016 were analyzed to distinguish various and variable surface types continuously influenced by tidal dynamics in the main study area, the tidal flats near the island of Norderney. Visual image interpretation supported by extensive in-situ verification proved to be a suitable and unsophisticated approach which is unspecific enough to identify mussel beds, fields of shell-detritus, gully structures, mud fields, and intertidal bedforms in the upper flats of the East Frisian Islands. The method proved to be robust against changes in geometry of acquisition and environmental influences. Several time series of TerraSAR-X data enabled to follow inter-annual and seasonal dynamics as well as event effects (Adolph et al. 2018). The high-frequency TerraSAR-X data revealed novel evidence of an intertidal bedform shift in an easterly direction during the study period. To this aim, an unsupervised ISODATA classification of textural parameters was developed to vectorize and compare the bedforms positions in a GIS (Adolph et al. 2017a). The same intertidal bedform area was chosen as test-site for comparison of different remote sensing methods, namely airborne lidar, satellite-based radar (TerraSAR-X) and electro-optical sensors (RapidEye) (Adolph et al. 2017b). High-resolution SAR data offer a relevant component for Wadden Sea Monitoring and Research, as they provide reliable, regular data with a high repetition rate. In particular, habitats with noticeable surface roughness, specific structures and textures are well reflected. Geomorphic Structures and their dynamics can be observed indirectly via detection of residual water trapped within. A comprehensive concept for Wadden Sea Monitoring however, requires automatized classifications and an integrative, multi-sensor approach (SAR, LIDAR, multi-spectral data, drones) in which different and complementary information, coverage and resolutions (spatial and temporal) contribute to an overall picture. The studies were carried out as part of the joint research project “Scientific monitoring concepts for the German Bight” (WIMO), jointly funded by the Ministry of Environment, Energy and Climate Protection (NMU) and the Ministry of Science and Culture (NMWK) of the Federal State of Lower Saxony. The findings have been published in Geo-Marine Letters 37/2 (2017) and in Remote Sensing 10/7 (2018).

Wir befinden uns inmitten einer globalen Biodiversitätskrise mit steigender Aussterberate, die die natürliche Hintergrundrate bereits jetzt um das Tausendfache überschreitet. In terrestri-schen Lebensräumen geht vom Landnutzungswandel die größte Gefahr für die Biodiversität aus. Besonders die ursprünglich hohe Vielfalt im mitteleuropäischen Grünland ist seit der In-dustrialisierung durch Nutzungsintensivierung und -aufgabe stark zurück gegangen. Traditionell genutztes, beweidetes Grünland ist in Mitteleuropa selten geworden. Dennoch deuten einige aktuelle Studien auf seine große Bedeutung für den Erhalt der Biodiversität in der Agrarland-schaft hin. Allerdings fehlen umfassende, systematische Untersuchungen für viele traditionelle Grünlandsysteme. Um diese Wissenslücke zu verringern, untersuchte ich in meiner Dissertation die Relevanz voralpiner, traditionell bewirtschafteter Allmendweiden als Refugium für bedrohte Zielarten (Baumpieper [Anthus trivialis], Blaukernauge [Minois dryas]) sowie für arten- und individuenreiche Zoozönosen (Heuschrecken [Orthoptera]) der mitteleuropäischen Agrarlandschaft. Gepaarte Vergleiche von Allmendweiden und angrenzenden Flächen mit identischen Standortbedingun-gen (Referenzflächen) ermöglichten den Vergleich der Arten bzw. -gruppe anhand von Umwelt-parametern in den Vegetationstypen Grünland auf Mineralboden und offenen Mooren. Der Un-tersuchungsraum in Südbayern stellt den Verbreitungsschwerpunkt für Allmendweiden und intakte bis naturnahe Moorökosysteme in Deutschland dar. Meine Analysen zeigen, dass sich der Landnutzungswandel drastisch auf die Referenzflä-chen, die früher überwiegend Teil der Allmendweiden waren, auswirkte: Im Vergleich zu den Weiden weisen sie (i) homogene Strukturen auf Landschafts- und Habitatebene, (ii) geringe Anteile nährstoffarmer Habitate und (iii) eine geringe Besiedlung von Vögeln und Insekten auf. Somit sind sie weitgehend ungeeignet für den Erhalt der Biodiversität. Die einzige Ausnahme innerhalb der Referenzflächen bilden traditionell genutzte Heu- und Streuwiesen. In den Allmendweiden sorgt die traditionelle Beweidung (0.5–2.0 Großvieheinheiten/ha von Mai–Oktober) für eine hohe strukturelle Vielfalt mit nährstoffarmen Habitaten und fließende Übergänge zwischen Wald und Offenland. So können saisonal bzw. innerhalb des Lebenszyklus wechselnde Ansprüche vieler typischer Agrarlandschaftsarten im eng verzahnten Mosaik wert-voller Habitate erfüllt werden. Der Erhalt der Allmendweiden ist für das Überleben von Agrar-landschaftsarten im Untersuchungsgebiet essenziell. Eine Ausdehnung der Allmendweiden-Nutzung auf angrenzende Flächen ist unbedingt zu empfehlen.

No description available.