Refine
Year of publication
Document Type
- Article (35)
- Review (3)
- Conference Proceeding (2)
- Doctoral Thesis (1)
- Other (1)
- Postprint (1)
- Preprint (1)
Keywords
- ecological niche modelling (2)
- Artificial neuronal network (1)
- Biodiversity monitoring (1)
- Biotic interactions (1)
- Boosted regression trees (1)
- Boosted regression trees (BRT) (1)
- Boosting (1)
- Borneo (1)
- Borrelia afzelii (1)
- Borrelia lusitaniae (1)
- Cerrado (1)
- Classification trees (1)
- Climate change (1)
- Clonal plants (1)
- Coastal regions (1)
- Database (1)
- Deforestation (1)
- Dispersal (1)
- Distribution patterns (1)
- Driving forces (1)
- Earthworms (1)
- Ecosystem engineer (1)
- Ecosystem function (1)
- Environmental niche (1)
- Feedback biotic-abiotic (1)
- Forest regeneration (1)
- Functional diversity (1)
- Functional effect (1)
- Functional traits (1)
- Generalized linear models (1)
- Hypothesis generation (1)
- Land use (1)
- Land-cover change (1)
- Landscape (1)
- Landslides (1)
- Lyme disease (1)
- Management (1)
- Mato Grosso (1)
- Maximum entropy method (1)
- Mboost (1)
- Model comparison (1)
- Modeling (1)
- Multivariate adaptive regression splines (1)
- Non-stationarity (1)
- Oligochaeta (1)
- Orthoptera (1)
- Plant growth (1)
- Population dynamics (1)
- Predictive vegetation mapping (1)
- Random forests (1)
- SDM (1)
- Shrubland regeneration (1)
- Site ecology (1)
- Soil hydrology (1)
- Southeast Asia (1)
- Species distribution modelling (1)
- Species distribution models (1)
- Statistical modeling (1)
- Tropical montane forests (1)
- Weighted model ensembles (1)
- biodiversity conservation (1)
- biomass (1)
- carnivora (1)
- cash crops (1)
- climate change (1)
- community dynamics (1)
- conservation planning (1)
- drylands (1)
- ecohydrological modelling (1)
- ecohydrology (1)
- ecological forecasts (1)
- ecological theory (1)
- ecosystem change (1)
- extrapolation (1)
- feedback (1)
- geographical range shifts (1)
- global environmental change (1)
- hyporheic zone (1)
- individual based modeling (1)
- inflated response curves (1)
- land degradation (1)
- land use change (1)
- landscape genetics (1)
- lizard (1)
- long distance movement (1)
- maximum entropy (MaxEnt) (1)
- mechanistic model (1)
- mechanistic models (1)
- meso-scale ecosystems (1)
- migration (1)
- mobile links (1)
- novel environment (1)
- organic layer (1)
- parameterization (1)
- pesticides (1)
- plant-animal-soil-system (1)
- process-based model (1)
- process-based statistics (1)
- river networks (1)
- sampling bias (1)
- sampling space (1)
- shallow translational landslides (1)
- soil parameters (1)
- species coexistence (1)
- species distribution model (1)
- species distribution modelling (1)
- species distribution models (1)
- stratification (1)
- tropical montane forest (1)
- uncertainty (1)
- validation (1)
- viverridae (1)
- water quality (1)
- zooprophylaxis (1)
Interdisziplinäres Zentrum für Musterdynamik und Angewandte Fernerkundung Workshop vom 9. - 10. Februar 2006
Gütemaße für Habitatmodelle
(2004)
The globally threatened Aquatic Warbler Acrocephalus paludicola is an umbrella species for fen mires and is at risk of extinction in its westernmost breeding population due to severe habitat loss. We used boosted regression trees to model Aquatic Warbler habitat selection in order to make recommendations for effective management of the last remnant habitats. Habitat data were collected in the years 2004-2006 in all remaining breeding sites in Pomerania (eastern Germany and western Poland) as well as in recently abandoned sites. Models were validated using data from similar Aquatic Warbler habitats in Lithuania. The probability of occurrence of Aquatic Warblers in late May/early June was positively associated with low isolation from other occupied sites, less eutrophic conditions, a high proportion of area mown early in the preceding year, high availability of vegetation 60-70 cm high, high prey abundance and high habitat heterogeneity. Early summer land management is needed in the more productive sites to prevent habitat deterioration by succession to higher and denser vegetation. As this also poses a serious threat to broods, management that creates a mosaic of early and late used patches is recommended to preserve and restore productive Aquatic Warbler sites. In less productive sites, winter mowing can maintain suitable habitat conditions. Aquatic Warbler-friendly land use supports a variety of other threatened plant and animal species typical of fens and sedge meadows and can meet the economic interests of local land users.
Das Teilprojekt Landschafts- und Autökologie (LÖK) hat den Schwerpunkt auf die Erarbeitung einer e- Learning-Einheit zur Habitatmodellierung im allgemeinen und dem Verfahren der logistischen Regression im speziellen gelegt. In den sechs Lernmodulen der Lerneinheit werden alle für eine erfolgreiche Modellierung der Habitateignung erforderlichen Arbeitsschritte sequentiell behandelt. Die wesentlichen Schritte werden mit interaktiven Aufgaben vertieft, in welchen an entscheidenden Stellen WebGIS eingesetzt wird. Der räumliche Bezug wird in der Regel über WebGIS- Anwendungen zu einer virtuellen Landschaft hergestellt, die in das GIMOLUS-System integriert ist. Die erforderlichen Datensätze für die Analyse von Art-Habitat- Beziehungen werden bereitgestellt oder können interaktiv aus der virtuellen Landschaft erzeugt werden.
Landslides are a hazard for humans and artificial structures. From an ecological point of view, they represent an important ecosystem disturbance, especially in tropical montane forests. Here, shallow translational landslides are a frequent natural phenomenon and one local determinant of high levels of biodiversity. In this paper, we apply weighted ensembles of advanced phenomenological models from statistics and machine learning to analyze the driving factors of natural landslides in a tropical montane forest in South Ecuador. We exclusively interpret terrain attributes, derived from a digital elevation model, as proxies to several driving factors of landslides and use them as predictors in our models which are trained on a set of five historical landslide inventories. We check the model generality by transferring them in time and use three common performance criteria (i.e. AUC, explained deviance and slope of model calibration curve) to, on the one hand, compare several state-of-the-art model approaches and on the other hand, to create weighted model ensembles. Our results suggest that it is important to consider more than one single performance criterion.
Approaching our main question, we compare responses of weighted model ensembles that were trained on distinct functional units of landslides (i.e. initiation, transport and deposition zones). This way, we are able to show that it is quite possible to deduce driving factors of landslides, if the consistency between the training data and the processes is maintained. Opening the 'black box' of statistical models by interpreting univariate model response curves and relative importance of single predictors regarding their plausibility, we provide a means to verify this consistency.
With the exception of classification tree analysis, all techniques performed comparably well in our case study while being outperformed by weighted model ensembles. Univariate response curves of models trained on distinct functional units of landslides exposed different shapes following our expectations. Our results indicate the occurrence of landslides to be mainly controlled by factors related to the general position along a slope (i.e. ridge, open slope or valley) while landslide initiation seems to be favored by small scale convexities on otherwise plain open slopes.