TY - JOUR A1 - Vorpahl, Peter A1 - Elsenbeer, Helmut A1 - Märker, Michael A1 - Schröder-Esselbach, Boris T1 - How can statistical models help to determine driving factors of landslides? JF - Ecological modelling : international journal on ecological modelling and engineering and systems ecolog N2 - 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. KW - Landslides KW - Tropical montane forests KW - Statistical modeling KW - Model comparison KW - Artificial neuronal network KW - Classification trees KW - Random forests KW - Boosted regression trees KW - Generalized linear models KW - Multivariate adaptive regression splines KW - Maximum entropy method KW - Weighted model ensembles Y1 - 2012 U6 - https://doi.org/10.1016/j.ecolmodel.2011.12.007 SN - 0304-3800 SN - 1872-7026 VL - 239 IS - 7 SP - 27 EP - 39 PB - Elsevier CY - Amsterdam ER - TY - JOUR A1 - Radchuk, Viktoriia A1 - Johst, Karin A1 - Groeneveld, Jürgen A1 - Turlure, Camille A1 - Grimm, Volker A1 - Schtickzelle, Nicolas T1 - Appropriate resolution in time and model structure for population viability analysis: Insights from a butterfly metapopulation JF - : an international journal N2 - The importance of a careful choice of the appropriate scale for studying ecological phenomena has been stressed repeatedly. However, issues of spatial scale in metapopulation dynamics received much more attention compared to temporal scale. Moreover, multiple calls were made to carefully choose the appropriate model structure for Population Viability Analysis (PVA). We assessed the effect of using coarser resolution in time and model structure on population dynamics. For this purpose, we compared outcomes of two PVA models differing in their time step: daily individual-based model (dIBM) and yearly stage-based model (ySBM), loaded with empirical data on a well-known metapopulation of the butterfly Boloria eunomia. Both models included the same environmental drivers of population dynamics that were previously identified as being the most important for this species. Under temperature change scenarios, both models yielded the same qualitative scenario ranking, but they quite substantially differed quantitatively with dIBM being more pessimistic in absolute viability measures. We showed that these differences stemmed from inter-individual heterogeneity in dIBM allowing for phenological shifts of individual appearance. We conclude that a finer temporal resolution and an individual-based model structure allow capturing the essential mechanisms necessary to go beyond mere PVA scenario ranking. We encourage researchers to carefully chose the temporal resolution and structure of their model aiming at (1) depicting the processes important for (meta)population dynamics of the species and (2) implementing the environmental change scenarios expected for their study system in the future, using the temporal resolution at which such changes are predicted to operate. KW - Temporal grain KW - Model complexity KW - Model comparison KW - Population dynamics KW - Individual-based model KW - Stage-based model Y1 - 2014 U6 - https://doi.org/10.1016/j.biocon.2013.12.004 SN - 0006-3207 SN - 1873-2917 VL - 169 SP - 345 EP - 354 PB - Elsevier CY - Oxford ER - TY - JOUR A1 - Mühlenbruch, Kristin A1 - Kuxhaus, Olga A1 - Pencina, Michael J. A1 - Boeing, Heiner A1 - Liero, Hannelore A1 - Schulze, Matthias Bernd T1 - A confidence ellipse for the Net Reclassification Improvement JF - European journal of epidemiology N2 - The Net Reclassification Improvement (NRI) has become a popular metric for evaluating improvement in disease prediction models through the past years. The concept is relatively straightforward but usage and interpretation has been different across studies. While no thresholds exist for evaluating the degree of improvement, many studies have relied solely on the significance of the NRI estimate. However, recent studies recommend that statistical testing with the NRI should be avoided. We propose using confidence ellipses around the estimated values of event and non-event NRIs which might provide the best measure of variability around the point estimates. Our developments are illustrated using practical examples from EPIC-Potsdam study. KW - Risk assessment KW - Risk model KW - Model comparison KW - Reclassification KW - Confidence intervals Y1 - 2015 U6 - https://doi.org/10.1007/s10654-015-0001-1 SN - 0393-2990 SN - 1573-7284 VL - 30 IS - 4 SP - 299 EP - 304 PB - Springer CY - Dordrecht ER -