@article{RosenbaumRaatzWeithoffetal.2019, author = {Rosenbaum, Benjamin and Raatz, Michael and Weithoff, Guntram and Fussmann, Gregor F. and Gaedke, Ursula}, title = {Estimating parameters from multiple time series of population dynamics using bayesian inference}, series = {Frontiers in ecology and evolution}, volume = {6}, journal = {Frontiers in ecology and evolution}, publisher = {Frontiers Research Foundation}, address = {Lausanne}, issn = {2296-701X}, doi = {10.3389/fevo.2018.00234}, pages = {14}, year = {2019}, abstract = {Empirical time series of interacting entities, e.g., species abundances, are highly useful to study ecological mechanisms. Mathematical models are valuable tools to further elucidate those mechanisms and underlying processes. However, obtaining an agreement between model predictions and experimental observations remains a demanding task. As models always abstract from reality one parameter often summarizes several properties. Parameter measurements are performed in additional experiments independent of the ones delivering the time series. Transferring these parameter values to different settings may result in incorrect parametrizations. On top of that, the properties of organisms and thus the respective parameter values may vary considerably. These issues limit the use of a priori model parametrizations. In this study, we present a method suited for a direct estimation of model parameters and their variability from experimental time series data. We combine numerical simulations of a continuous-time dynamical population model with Bayesian inference, using a hierarchical framework that allows for variability of individual parameters. The method is applied to a comprehensive set of time series from a laboratory predator-prey system that features both steady states and cyclic population dynamics. Our model predictions are able to reproduce both steady states and cyclic dynamics of the data. Additionally to the direct estimates of the parameter values, the Bayesian approach also provides their uncertainties. We found that fitting cyclic population dynamics, which contain more information on the process rates than steady states, yields more precise parameter estimates. We detected significant variability among parameters of different time series and identified the variation in the maximum growth rate of the prey as a source for the transition from steady states to cyclic dynamics. By lending more flexibility to the model, our approach facilitates parametrizations and shows more easily which patterns in time series can be explained also by simple models. Applying Bayesian inference and dynamical population models in conjunction may help to quantify the profound variability in organismal properties in nature.}, language = {en} } @article{SelleKnorrLischeid2019, author = {Selle, Benny and Knorr, Klaus-Holger and Lischeid, Gunnar}, title = {Mobilisation and transport of dissolved organic carbon and iron in peat catchments-Insights from the Lehstenbach stream in Germany using generalised additive models}, series = {Hydrological processes}, volume = {33}, journal = {Hydrological processes}, number = {25}, publisher = {Wiley}, address = {Hoboken}, issn = {0885-6087}, doi = {10.1002/hyp.13552}, pages = {3213 -- 3225}, year = {2019}, abstract = {During the last decades, increasing exports of both dissolved organic carbon (DOC) and iron were observed from peat catchments in North America and Europe with potential consequences for water quality of streamwater and carbon storages of soils. As mobilisation and transport processes of DOC and iron in peat catchments are only partly understood, the purpose of this study was to elucidate these processes in an intensively monitored and studied system. Specifically, it was hypothesised that dissimilatory iron reduction in riparian peatland soils mobilises DOC initially adsorbed to iron minerals. During stormflow conditions, both DOC and iron will be transported into the stream network. Ferrous iron may be reoxidised at redox interfaces on its way to the stream, and subsequently, ferric iron could be transported together with DOC as complexes. To test these hypotheses, generalised additive models (GAMs) were applied to 14 years of weekly time series of discharge and concentrations of selected solutes measured in a German headwater stream called Lehstenbach. This stream drains a 4.19-km(2) forested mountain catchment; one third of which is covered by riparian peatland soils. We interpreted results of different types of GAM in the way that (a) iron reduction drove the mobilisation of DOC from peatland soils and that (b) both iron and DOC were transported as complexes after their joint mobilisation to and within the steam. It was speculated that low nitrate availability in the uppermost wetland soil layer, particularly during the growing season, promoted iron reduction and thus the mobilisation of DOC. However, the influence of nitrate on the DOC mobilisation remains relatively uncertain. This influence could be further investigated using methods similar to the GAM analysis conducted here for other catchments with long-term data as well as detailed measurements of the relevant species in riparian wetland soils and the adjacent stream network.}, language = {en} } @article{PennekampIlesGarlandetal.2019, author = {Pennekamp, Frank and Iles, Alison C. and Garland, Joshua and Brennan, Georgina and Brose, Ulrich and Gaedke, Ursula and Jacob, Ute and Kratina, Pavel and Matthews, Blake and Munch, Stephan and Novak, Mark and Palamara, Gian Marco and Rall, Bjorn C. and Rosenbaum, Benjamin and Tabi, Andrea and Ward, Colette and Williams, Richard and Ye, Hao and Petchey, Owen L.}, title = {The intrinsic predictability of ecological time series and its potential to guide forecasting}, series = {Ecological monographs : a publication of the Ecological Society of America.}, volume = {89}, journal = {Ecological monographs : a publication of the Ecological Society of America.}, number = {2}, publisher = {Wiley}, address = {Hoboken}, issn = {0012-9615}, doi = {10.1002/ecm.1359}, pages = {17}, year = {2019}, language = {en} } @misc{ŚlęzakBurneckiMetzler2019, author = {Ślęzak, Jakub and Burnecki, Krzysztof and Metzler, Ralf}, title = {Random coefficient autoregressive processes describe Brownian yet non-Gaussian diffusion in heterogeneous systems}, series = {Postprints der Universit{\"a}t Potsdam Mathematisch-Naturwissenschaftliche Reihe}, journal = {Postprints der Universit{\"a}t Potsdam Mathematisch-Naturwissenschaftliche Reihe}, number = {765}, issn = {1866-8372}, doi = {10.25932/publishup-43792}, url = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus4-437923}, pages = {18}, year = {2019}, abstract = {Many studies on biological and soft matter systems report the joint presence of a linear mean-squared displacement and a non-Gaussian probability density exhibiting, for instance, exponential or stretched-Gaussian tails. This phenomenon is ascribed to the heterogeneity of the medium and is captured by random parameter models such as 'superstatistics' or 'diffusing diffusivity'. Independently, scientists working in the area of time series analysis and statistics have studied a class of discrete-time processes with similar properties, namely, random coefficient autoregressive models. In this work we try to reconcile these two approaches and thus provide a bridge between physical stochastic processes and autoregressive models.Westart from the basic Langevin equation of motion with time-varying damping or diffusion coefficients and establish the link to random coefficient autoregressive processes. By exploring that link we gain access to efficient statistical methods which can help to identify data exhibiting Brownian yet non-Gaussian diffusion.}, language = {en} } @article{ŚlęzakBurneckiMetzler2019, author = {Ślęzak, Jakub and Burnecki, Krzysztof and Metzler, Ralf}, title = {Random coefficient autoregressive processes describe Brownian yet non-Gaussian diffusion in heterogeneous systems}, series = {New Journal of Physics}, volume = {21}, journal = {New Journal of Physics}, publisher = {Deutsche Physikalische Gesellschaft ; IOP, Institute of Physics}, address = {Bad Honnef und London}, issn = {1367-2630}, doi = {10.1088/1367-2630/ab3366}, pages = {18}, year = {2019}, abstract = {Many studies on biological and soft matter systems report the joint presence of a linear mean-squared displacement and a non-Gaussian probability density exhibiting, for instance, exponential or stretched-Gaussian tails. This phenomenon is ascribed to the heterogeneity of the medium and is captured by random parameter models such as 'superstatistics' or 'diffusing diffusivity'. Independently, scientists working in the area of time series analysis and statistics have studied a class of discrete-time processes with similar properties, namely, random coefficient autoregressive models. In this work we try to reconcile these two approaches and thus provide a bridge between physical stochastic processes and autoregressive models.Westart from the basic Langevin equation of motion with time-varying damping or diffusion coefficients and establish the link to random coefficient autoregressive processes. By exploring that link we gain access to efficient statistical methods which can help to identify data exhibiting Brownian yet non-Gaussian diffusion.}, language = {en} } @article{LoiblBookhagenValadeetal.2019, author = {Loibl, David and Bookhagen, Bodo and Valade, Sebastien and Schneider, Christoph}, title = {OSARIS, the "Open Source SAR Investigation System" for Automatized Parallel InSAR Processing of Sentinel-1 Time Series Data With Special Emphasis on Cryosphere Applications}, series = {Frontiers in Earth Science}, volume = {7}, journal = {Frontiers in Earth Science}, publisher = {Frontiers Media}, address = {Lausanne}, issn = {2296-6463}, doi = {10.3389/feart.2019.00172}, pages = {20}, year = {2019}, abstract = {With the advent of the two Sentinel-1 (S1) satellites, Synthetic Aperture Radar (SAR) data with high temporal and spatial resolution are freely available. This provides a promising framework to facilitate detailed investigations of surface instabilities and movements on large scales with high temporal resolution, but also poses substantial processing challenges because of storage and computation requirements. Methods are needed to efficiently detect short term changes in dynamic environments. Approaches considering pair-wise processing of a series of consecutive scenes to retain maximum temporal resolution in conjunction with time series analyses are required. Here we present OSARIS, the "Open Source SAR Investigation System," as a framework to process large stacks of S1 data on high-performance computing clusters. Based on Generic Mapping Tools SAR, shell scripts, and the workload manager Slurm, OSARIS provides an open and modular framework combining parallelization of high-performance C programs, flexible processing schemes, convenient configuration, and generation of geocoded stacks of analysis-ready base data, including amplitude, phase, coherence, and unwrapped interferograms. Time series analyses can be conducted by applying automated modules to the data stacks. The capabilities of OSARIS are demonstrated in a case study from the northwestern Tien Shan, Central Asia. After merging of slices, a total of 80 scene pairs were processed from 174 total input scenes. The coherence time series exhibits pronounced seasonal variability, with relatively high coherence values prevailing during the summer months in the nival zone. As an example of a time series analysis module, we present OSARIS' "Unstable Coherence Metric" which identifies pixels affected by significant drops from high to low coherence values. Measurements of motion provided by LOSD measurements require careful evaluation because interferometric phase unwrapping is prone to errors. Here, OSARIS provides a series of modules to detect and mask unwrapping errors, correct for atmospheric disturbances, and remove large-scale trends. Wall clock processing time for the case study (area ~9,000 km2) was ~12 h 4 min on a machine with 400 cores and 2 TB RAM. In total, ~12 d 10 h 44 min (~96\%) were saved through parallelization. A comparison of selected OSARIS datasets to results from two state-of-the-art SAR processing suites, ISCE and SNAP, shows that OSARIS provides products of competitive quality despite its high level of automatization. OSARIS thus facilitates efficient S1-based region-wide investigations of surface movement events over multiple years.}, language = {en} }