Search

Your search keyword '"Dormann, Carsten F."' showing total 764 results
Start Over Author "Dormann, Carsten F."
764 results on '"Dormann, Carsten F."'

1. Advances in Land Surface Model-based Forecasting: A comparative study of LSTM, Gradient Boosting, and Feedforward Neural Network Models as prognostic state emulators

Author

Wesselkamp, Marieke, Chantry, Matthew, Pinnington, Ewan, Choulga, Margarita, Boussetta, Souhail, Kalweit, Maria, Boedecker, Joschka, Dormann, Carsten F., Pappenberger, Florian, and Balsamo, Gianpaolo

Subjects
Physics - Atmospheric and Oceanic Physics, Computer Science - Machine Learning
Abstract

Most useful weather prediction for the public is near the surface. The processes that are most relevant for near-surface weather prediction are also those that are most interactive and exhibit positive feedback or have key role in energy partitioning. Land surface models (LSMs) consider these processes together with surface heterogeneity and forecast water, carbon and energy fluxes, and coupled with an atmospheric model provide boundary and initial conditions. This numerical parametrization of atmospheric boundaries being computationally expensive, statistical surrogate models are increasingly used to accelerated progress in experimental research. We evaluated the efficiency of three surrogate models in speeding up experimental research by simulating land surface processes, which are integral to forecasting water, carbon, and energy fluxes in coupled atmospheric models. Specifically, we compared the performance of a Long-Short Term Memory (LSTM) encoder-decoder network, extreme gradient boosting, and a feed-forward neural network within a physics-informed multi-objective framework. This framework emulates key states of the ECMWF's Integrated Forecasting System (IFS) land surface scheme, ECLand, across continental and global scales. Our findings indicate that while all models on average demonstrate high accuracy over the forecast period, the LSTM network excels in continental long-range predictions when carefully tuned, the XGB scores consistently high across tasks and the MLP provides an excellent implementation-time-accuracy trade-off. The runtime reduction achieved by the emulators in comparison to the full numerical models are significant, offering a faster, yet reliable alternative for conducting numerical experiments on land surfaces.

Published
2024

5. Process-guidance improves predictive performance of neural networks for carbon turnover in ecosystems

Author

Wesselkamp, Marieke, Moser, Niklas, Kalweit, Maria, Boedecker, Joschka, and Dormann, Carsten F.

Subjects
Statistics - Applications
Abstract

Despite deep-learning being state-of-the-art for data-driven model predictions, it has not yet found frequent application in ecology. Given the low sample size typical in many environmental research fields, the default choice for the modelling of ecosystems and its functions remain process-based models. The process understanding coded in these models complements the sparse data and neural networks can detect hidden dynamics even in noisy data. Embedding the process model in the neural network adds information to learn from, improving interpretability and predictive performance of the combined model towards the data-only neural networks and the mechanism-only process model. At the example of carbon fluxes in forest ecosystems, we compare different approaches of guiding a neural network towards process model theory. Evaluation of the results under four classical prediction scenarios supports decision-making on the appropriate choice of a process-guided neural network.

Published
2022

9. A standard protocol for reporting species distribution models

Author

Zurell, Damaris, Franklin, Janet, Koenig, Christian, Bouchet, Phil J, Dormann, Carsten F, Elith, Jane, Fandos, Guillermo, Feng, Xiao, Guillera-Arroita, Gurutzeta, Guisan, Antoine, Lahoz-Monfort, Jose J, Leitao, Pedro J, Park, Daniel S, Peterson, A Townsend, Rapacciuolo, Giovanni, Schmatz, Dirk R, Schroeder, Boris, Serra-Diaz, Josep M, Thuiller, Wilfried, Yates, Katherine L, Zimmermann, Niklaus E, and Merow, Cory

Subjects
biodiversity assessment, ecological niche model, habitat suitability model, reproducibility, Shiny, transparency, Ecology, Ecological Applications, Environmental Science and Management
Published
2020

11. Outstanding Challenges in the Transferability of Ecological Models

Author

Yates, Katherine L, Bouchet, Phil J, Caley, M Julian, Mengersen, Kerrie, Randin, Christophe F, Parnell, Stephen, Fielding, Alan H, Bamford, Andrew J, Ban, Stephen, Barbosa, A Márcia, Dormann, Carsten F, Elith, Jane, Embling, Clare B, Ervin, Gary N, Fisher, Rebecca, Gould, Susan, Graf, Roland F, Gregr, Edward J, Halpin, Patrick N, Heikkinen, Risto K, Heinänen, Stefan, Jones, Alice R, Krishnakumar, Periyadan K, Lauria, Valentina, Lozano-Montes, Hector, Mannocci, Laura, Mellin, Camille, Mesgaran, Mohsen B, Moreno-Amat, Elena, Mormede, Sophie, Novaczek, Emilie, Oppel, Steffen, Crespo, Guillermo Ortuño, Peterson, A Townsend, Rapacciuolo, Giovanni, Roberts, Jason J, Ross, Rebecca E, Scales, Kylie L, Schoeman, David, Snelgrove, Paul, Sundblad, Göran, Thuiller, Wilfried, Torres, Leigh G, Verbruggen, Heroen, Wang, Lifei, Wenger, Seth, Whittingham, Mark J, Zharikov, Yuri, Zurell, Damaris, and Sequeira, Ana MM

Subjects
Ecology, Models, Biological, Predictive modeling, extrapolation, generality, habitat models, model transfers, species distribution models, uncertainty, Environmental Sciences, Biological Sciences, Evolutionary Biology
Abstract

Predictive models are central to many scientific disciplines and vital for informing management in a rapidly changing world. However, limited understanding of the accuracy and precision of models transferred to novel conditions (their 'transferability') undermines confidence in their predictions. Here, 50 experts identified priority knowledge gaps which, if filled, will most improve model transfers. These are summarized into six technical and six fundamental challenges, which underlie the combined need to intensify research on the determinants of ecological predictability, including species traits and data quality, and develop best practices for transferring models. Of high importance is the identification of a widely applicable set of transferability metrics, with appropriate tools to quantify the sources and impacts of prediction uncertainty under novel conditions.

Published
2018

14. Risk response towards roads is consistent across multiple species in a temperate forest ecosystem.

Author

Luca Bastianelli, Matteo, von Hoermann, Christian, Kirchner, Katrin, Signer, Johannes, Dupke, Claudia, Henrich, Maik, Wielgus, Elodie, Fiderer, Christian, Belotti, Elisa, Bufka, Luděk, Ciuti, Simone, Dormann, Carsten F., Kuemmerle, Tobias, Storch, Ilse, Grilo, Clara, and Heurich, Marco

Subjects
TEMPERATE forest ecology, OMNIVORES, WILD boar, BIOTIC communities, PREDATION, LYNX, RED deer, ECOLOGICAL disturbances, HABITAT selection
Abstract

Roads can have diverse impacts on wildlife species, and while some species may adapt effectively, others may not. Studying multiple species' responses to the same infrastructure in a given area can help understand this variation and reveal the effects of disturbance on the ecology of wildlife communities. This study investigates the behavioural responses of four species with distinctive ecological and behavioural traits to roads in the protected Bohemian Forest Ecosystem in central Europe: European roe deer Capreolus capreolus, a solitary herbivore; red deer Cervus elaphus a gregarious herbivore; wild boar Sus scrofa, a gregarious omnivore and Eurasian lynx Lynx lynx, a solitary large carnivore. We used GPS data gathered from each species to study movement behaviour and habitat selection in relation to roads using an integrated step selection analysis. For all species and sexes, we predicted increased movement rates in response to roads, selection of vegetation cover near roads and open areas after road crossings, and increased road avoidance during the day. We found remarkably similar behavioural responses towards roads across species. The behavioural adaptations to road exposure, such as increased movement rates and selection for vegetation cover, were analogous to responses to natural predation risk. Roads were more strongly avoided during daytime, when traffic volume was high. Road crossings were more frequent at twilight and at night within open areas offering food resources. Gregarious animals exposed to roads favoured stronger road avoidance over faster movements. Ungulates crossed roads more at twilight, coinciding with commuter traffic during winter. Despite differences in the ecology and behaviour of the four species, our results showed similar adaptations towards a common threat. The continuous expansion of the global transportation network should be accompanied by efforts to understand and minimise the impact of roads on wildlife to assist wildlife management and ensure conservation. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

15. Crop pests and predators exhibit inconsistent responses to surrounding landscape composition.

Author

Karp, Daniel S, Chaplin-Kramer, Rebecca, Meehan, Timothy D, Martin, Emily A, DeClerck, Fabrice, Grab, Heather, Gratton, Claudio, Hunt, Lauren, Larsen, Ashley E, Martínez-Salinas, Alejandra, O'Rourke, Megan E, Rusch, Adrien, Poveda, Katja, Jonsson, Mattias, Rosenheim, Jay A, Schellhorn, Nancy A, Tscharntke, Teja, Wratten, Stephen D, Zhang, Wei, Iverson, Aaron L, Adler, Lynn S, Albrecht, Matthias, Alignier, Audrey, Angelella, Gina M, Zubair Anjum, Muhammad, Avelino, Jacques, Batáry, Péter, Baveco, Johannes M, Bianchi, Felix JJA, Birkhofer, Klaus, Bohnenblust, Eric W, Bommarco, Riccardo, Brewer, Michael J, Caballero-López, Berta, Carrière, Yves, Carvalheiro, Luísa G, Cayuela, Luis, Centrella, Mary, Ćetković, Aleksandar, Henri, Dominic Charles, Chabert, Ariane, Costamagna, Alejandro C, De la Mora, Aldo, de Kraker, Joop, Desneux, Nicolas, Diehl, Eva, Diekötter, Tim, Dormann, Carsten F, Eckberg, James O, Entling, Martin H, Fiedler, Daniela, Franck, Pierre, Frank van Veen, FJ, Frank, Thomas, Gagic, Vesna, Garratt, Michael PD, Getachew, Awraris, Gonthier, David J, Goodell, Peter B, Graziosi, Ignazio, Groves, Russell L, Gurr, Geoff M, Hajian-Forooshani, Zachary, Heimpel, George E, Herrmann, John D, Huseth, Anders S, Inclán, Diego J, Ingrao, Adam J, Iv, Phirun, Jacot, Katja, Johnson, Gregg A, Jones, Laura, Kaiser, Marina, Kaser, Joe M, Keasar, Tamar, Kim, Tania N, Kishinevsky, Miriam, Landis, Douglas A, Lavandero, Blas, Lavigne, Claire, Le Ralec, Anne, Lemessa, Debissa, Letourneau, Deborah K, Liere, Heidi, Lu, Yanhui, Lubin, Yael, Luttermoser, Tim, Maas, Bea, Mace, Kevi, Madeira, Filipe, Mader, Viktoria, Cortesero, Anne Marie, Marini, Lorenzo, Martinez, Eliana, Martinson, Holly M, Menozzi, Philippe, Mitchell, Matthew GE, Miyashita, Tadashi, Molina, Gonzalo AR, and Molina-Montenegro, Marco A

Subjects
Animals, Crops, Agricultural, Ecosystem, Pest Control, Biological, Models, Biological, agroecology, biodiversity, biological control, ecosystem services, natural enemies, Crops, Agricultural, Pest Control, Biological, Models
Abstract

The idea that noncrop habitat enhances pest control and represents a win-win opportunity to conserve biodiversity and bolster yields has emerged as an agroecological paradigm. However, while noncrop habitat in landscapes surrounding farms sometimes benefits pest predators, natural enemy responses remain heterogeneous across studies and effects on pests are inconclusive. The observed heterogeneity in species responses to noncrop habitat may be biological in origin or could result from variation in how habitat and biocontrol are measured. Here, we use a pest-control database encompassing 132 studies and 6,759 sites worldwide to model natural enemy and pest abundances, predation rates, and crop damage as a function of landscape composition. Our results showed that although landscape composition explained significant variation within studies, pest and enemy abundances, predation rates, crop damage, and yields each exhibited different responses across studies, sometimes increasing and sometimes decreasing in landscapes with more noncrop habitat but overall showing no consistent trend. Thus, models that used landscape-composition variables to predict pest-control dynamics demonstrated little potential to explain variation across studies, though prediction did improve when comparing studies with similar crop and landscape features. Overall, our work shows that surrounding noncrop habitat does not consistently improve pest management, meaning habitat conservation may bolster production in some systems and depress yields in others. Future efforts to develop tools that inform farmers when habitat conservation truly represents a win-win would benefit from increased understanding of how landscape effects are modulated by local farm management and the biology of pests and their enemies.

Published
2018

20. Computing AIC for black-box models using Generalised Degrees of Freedom: a comparison with cross-validation

Author

Hauenstein, Severin, Dormann, Carsten F., and Wood, Simon N

Subjects
Statistics - Machine Learning
Abstract

Generalised Degrees of Freedom (GDF), as defined by Ye (1998 JASA 93:120-131), represent the sensitivity of model fits to perturbations of the data. As such they can be computed for any statistical model, making it possible, in principle, to derive the number of parameters in machine-learning approaches. Defined originally for normally distributed data only, we here investigate the potential of this approach for Bernoulli-data. GDF-values for models of simulated and real data are compared to model complexity-estimates from cross-validation. Similarly, we computed GDF-based AICc for randomForest, neural networks and boosted regression trees and demonstrated its similarity to cross-validation. GDF-estimates for binary data were unstable and inconsistently sensitive to the number of data points perturbed simultaneously, while at the same time being extremely computer-intensive in their calculation. Repeated 10-fold cross-validation was more robust, based on fewer assumptions and faster to compute. Our findings suggest that the GDF-approach does not readily transfer to Bernoulli data and a wider range of regression approaches., Comment: accompanying R-code on github

Published
2016

25. Model averaging in ecology : a review of Bayesian, information-theoretic, and tactical approaches for predictive inference

Author

Dormann, Carsten F., Calabrese, Justin M., Guillera-Arroita, Gurutzeta, Matechou, Eleni, Bahn, Volker, Bartoń, Kamil, Beale, Colin M., Ciuti, Simone, Elith, Jane, Gerstner, Katharina, Guelat, Jérôme, Keil, Petr, Lahoz-Monfort, José J., Pollock, Laura J., Reineking, Björn, Roberts, David R., Schröder, Boris, Thuiller, Wilfried, Warton, David I., Wintle, Brendan A., Wood, Simon N., Wüest, Rafael O., and Hartig, Florian

Published
2018

36. Hypothesen und Tests

Author

Dormann, Carsten F., Dette, Holger, Series editor, Härdle, Wolfgang, Series editor, and Dormann, Carsten F.

Published
2017
Full Text
View/download PDF

40. Does 'model-free' forecasting really outperform the 'true' model? A reply to Perretti et al

Author

Hartig, Florian and Dormann, Carsten F.

Subjects
Quantitative Biology - Populations and Evolution, Nonlinear Sciences - Chaotic Dynamics, Statistics - Applications
Abstract

Estimating population models from uncertain observations is an important problem in ecology. Perretti et al. observed that standard Bayesian state-space solutions to this problem may provide biased parameter estimates when the underlying dynamics are chaotic. Consequently, forecasts based on these estimates showed poor predictive accuracy compared to simple "model-free" methods, which lead Perretti et al. to conclude that "Model-free forecasting outperforms the correct mechanistic model for simulated and experimental data". However, a simple modification of the statistical methods also suffices to remove the bias and reverse their results., Comment: Letter submitted to PNAS, with additional supplementary information. R code included in the latex source

Published
2013
Full Text
View/download PDF

41. Detecting modules in quantitative bipartite networks: the QuaBiMo algorithm

Author

Dormann, Carsten F. and Strauss, Rouven

Subjects
Quantitative Biology - Quantitative Methods, Quantitative Biology - Populations and Evolution
Abstract

Ecological networks are often composed of different sub-communities (often referred to as modules). Identifying such modules has the potential to develop a better understanding of the assembly of ecological communities and to investigate functional overlap or specialisation. The most informative form of networks are quantitative or weighted networks. Here we introduce an algorithm to identify modules in quantitative bipartite (or two-mode) networks. It is based on the hierarchical random graphs concept of Clauset et al. (2008 Nature 453: 98-101) and is extended to include quantitative information and adapted to work with bipartite graphs. We define the algorithm, which we call QuaBiMo, sketch its performance on simulated data and illustrate its potential usefulness with a case study., Comment: 19 pages, 10 figures, 4 tables

Published
2013
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results