1,320 research outputs found
Applied algebra and number theory: essays in honour of Harald Niederreiter on the occasion of his 70th birthday
No full text"Harald Niederreiter's pioneering research in the field of applied algebra and number theory has led to important and substantial breakthroughs in many areas, including finite fields and their application areas as coding theory and cryptography as well as uniform distribution and quasi-Monte Carlo methods. He is author of more than 350 research papers and 10 books"-
Research Partnerships for Mitigating Syndromes of Global Change in Mountain Regions: An Overview of Current Knowledge
No full textMonitoring and modelling for the sustainable management of water resources in tropical mountain basins: The Mount Kenya example
No full textPalaeolimnological Investigations in the Alps: The Long-Term Development of Mountain Lakes
No full textLong-term Responses of Mountain Ecosystems to Environmental Changes: Resilience, Adjustment, and Vulnerability
No full textThe steep environmental gradients of mountain ecosystems over short distances reflect large gradients of several climatic parameters and hence provide excellent possibilities for ecological research on the effects of environmental change. To gain a better understanding of the dynamics of abiotic and biotic parameters of mountain ecosystems, long-term records are required since permanent plots in mountain regions cover in the best case about 50 - 70 years. In order to extend investigations of ecological dynamics beyond these temporal limitations of permanent plots, paleoecological approaches can be used if the sampling resolution can be adapted to ecological research questions, e.g. a sample every 10 years. Paleoecological studies in mountain ecosystems can provide new ecological insights through the combination of different spatial and temporal scales. [f we thus improve our understanding of processes across both steep environmental gradients and different time scales, we may be able to better estimate ecosystem responses to current and future environmental change (Ammann et al. 1993; Lotter et al. 1997).
The complexity of ecological interactions in mountain regions forces us to concentrate on a number of sub-systems - without losing sight of the wider context. Here, we summarize a few case studies on the effects of Holocene climate change and disturbance on the vegetation of the Western Alps. To categorize the main response modes of vegetation to climatic change and disturbance in the Alps we use three classes of ecological behaviour: "resilience", "adjustment", and "vulnerability", We assume a resilient (or elastic) behaviour if vegetation is able to recover to its former state, regaining important ecosystem characteristics, such as floristic composition, biodiversity, species abundances, and biomass (e.g. Küttel 1990; Aber and Melillo 199 1). Conversely, vegetation displacements may occur in response to climatic change and/or disturbance. In some cases, this may culminate in irreversible large-scale processes such as species and/or community extinctions. Such drastic developments indicate high ecosystem vulnerability (or inelasticity or instability, for detailed definitions see Küttel 1990; Aber and Melillo 199 1) to climatic change and/or disturbance. In this sense, the "vulnerability" (or instability) of an ecosystem is expressed by the degree of failure to recover to the original state before disturbance and/or climatic change. Between these two extremes (resilience vs. vulnerability), ecosystem adjustments to climatic change and/or disturbance may occur, including the appearance of new and/or the disappearance of old species. The term "adjustment" is hence used to indicate the response of vegetational communities, which adapted to new environmental conditions without losing their main character. For forest ecosystems, we assume vegetational adjustments (rather than vulnerability) if the dominant (or co-dominant) tree species are not outnumbered or replaced by formerly unimportant plant species or new invaders. Adaptation as a genetic process is not discussed here and will require additional pbylogeographical studies (that incorporate the analysis of ancient DNA) in order to fully understand the distributions of ecotypes
uelischmid/ProForM: ProForM v1.1
No full text<p>Ueli Schmid, Monika Frehner, Harald Bugmann
Forest Ecology, ETH Zürich</p>
<p><strong>Full Changelog</strong>: https://github.com/uelischmid/ProForM/compare/v1.0...v1.1</p>
Supplementary material for Käber et al. "Inferring the regeneration niche from forest inventory data using a dynamic forest model"
No full textThis repository contains scirpts and files used to generate the results of the study titled "Inferring the regeneration niche from forest inventory data using a dynamic forest model" from Yannek Käber, Florian Hartig, and Harald Bugmann.
Scripts and data files
File
Description
./binaries
This folder contains compiled versions of the c# applications used.
./binaries/forclim
ForClim model used for calibration as binaries. Its source files are in ./forclim.
./binaries/xmltodat
xmltodat is a c# program which compiles binaries from stand data input files. This program is used to reduce read and write overhead when during model calibration. Its source files are in ./xmltodat.
./cluster
This folder contains scripts and files used for model calibration and simulations on a High Performance Computing cluster.
./cluster/calculate_marginalLL_cluster.R
This script calculates the marginal likelihood based on simulations.
./cluster/calibr_setup
This folder contains all scripts which are required for model calibration on the cluster. These files are created with ./cluster/calibr_setup/create_cluster_file.R
./cluster/calibr_setup/calibrate.R
Main script for running the model calibration with BayesianTools. This script may work on a single machine or utilizes various computing nodes and cores on a HPC cluster.
./cluster/calibr_setup/create_cluster_file.R
This files copies and extracts files from other directories to create a set of files and scripts that are needed for model calibraiton. There is no need to run this script when all files are already within this directory. If you still want to run this file make sure that 7z (https://www.7-zip.org/download.html) is installed and added to the system path.
./cluster/calibr_setup/climate_data.csv
Climate input data for each calibration and validation site.
./cluster/calibr_setup/dat_100runs_1cmdbhclass
Site input files created with "xmltodat".
./cluster/calibr_setup/forclim_linux.zip; ./cluster/calibr_setup/forclim_win.zip
ForClim binaries used on the cluster for model calibration.
./cluster/calibr_setup/forclimr.zip
ForClimR R-Package. Basic R-Interface for ForClim.
./cluster/calibr_setup/functions/default_functions.R
R functions for model calibration and parallelisation on the HPC cluster.
./cluster/calibr_setup/posterior_likelihoods.R
Run ForClim simulations on the HPC cluster for calculating the likelihood base on the posterior distribution.
./cluster/calibr_steup/posterior_simulations.R
Run ForClim simulations on the HPC cluster for obtaining posterior samples.
./cluster/calibr_setup/sim_mat_dt.csv
Parameter values of the posterior distribution.
./cluster/calibr_setup/worker_script.R
This script is called by ./cluster/calibr_setup/calibrate.R as a worker on different computing nodes.
./cluster/calibr_setup/zips
Zip files copied here with ./cluster/calibr_setup/create_cluster_file.R. These will be extracted when ./cluster/calibr_setup/calibrate.R is running.
./data/calibration_output
This folder contains the calibraiton output as .RDS files for the different model variants 1 (simple) and 11 (complex).
./data/calibration_output/calibration_V9b_v1/bt_out_round1916.rds
BayesianTools output object for the calibration of variant 1 (simple).
./data/calibration_output/calibration_V9b_v1/bt_out_round1916.pdf
Traceplot from the model calibration of variant 11 (complex).
./data/calibration_output/calibration_V9b_v11/bt_out_round2009.rds
BayesianTools output object for the calibration of variant 11 (complex).
./data/calibration_output/calibration_V9b_v11/bt_out_round2009.pdf
Traceplot from the model calibration of variant 11 (complex).
./data/processed_data/chelsa/chelsa_means_unique_plot_id2.csv
Chelsa climate data for each site.
./data/processed_data/forclim-stand-files
Observational data
./data/processed_data/forclim-stand-files/climate_data.csv
Chelsa climate data for each site.
./data/processed_data/forclim-stand-files/dat_100runs_1cmdbhclass
Initial stand conditions as dat and csv files. The dat files are copied to ./cluster/calibr_setup/dat_100runs_1cmdbhclass for the calibration. Dat files contain the exact same information storred in the csv files. But the dat files are compiled to binaries with the xmltodat application for performance issues
./data/processed_data/forclim-stand-files/obs_data.csv ./data/processed_data/forclim-stand-files/obs_data7.csv ./data/processed_data/forclim-stand-files/obs_data10.csv
These files store the observed recruitment rates at the inventory specific dbh (obs_data.csv), all sites with dbh <= 7cm, and dbh <= 10 cm.
./data/processed_data/forclim-stand-files/sites.csv
Site information relevant for calibration: dbh thresholds, plot area, train/test site
./data/processed_data/posterior/aggregated_posterior_parameters.csv
Aggregated values of posterior estimates. Credible intervalls, median estimate, etc.
./data/processed_data/posterior/default_simulations_dbh0.csv ./data/processed_data/posterior/default_simulations_dbh7.csv ./data/processed_data/posterior/default_simulations_dbh10.csv
Simulation output for default parameter configuration for varying dbh thresholds (*dbh0.csv); dbh = 7cm; and dbh = 10cm.
./data/processed_data/posterior/full_posterior_parameters.csv
Posterior distribution of non-species specific parameters for both model variants.
./data/processed_data/posterior/ll_postsim.csv
Marginal likelihood derived from test data. (This data is created from very large cluster output files with the script ./scripts/analysis/compare_test_train.R)
./data/processed_data/posterior/posterior_simulations_dbh0.csv ./data/processed_data/posterior/posterior_simulations_dbh7.csv ./data/processed_data/posterior/posterior_simulations_dbh10.csv
Simulation output for posterio parameter distribution for varying dbh thresholds (*dbh0.csv); dbh = 7cm; and dbh = 10cm.
./data/processed_data/posterior/sim_mat_dt_v1.csv ./data/processed_data/posterior/sim_mat_dt_v11.csv
Posterior parameter distribution used as input on the cluster.
./forclim
ForClim source files.
./forclimr
R-Package for interfacing ForClim with R.
./scripts
Scripts used for data preparation and analysis.
./scripts/analysis
This folder contains all scripts which are used to create figures and graphs.
./scripts/analysis/posterior_graphs2.R
Script for Figures 1, 2, 3, A1, A2 and Tables A1, A2, A3
./scripts/analysis/error_drivers2.R
Scripts for Figures 4, 5, A3, A4
./scripts/analysis/compare_test_train.R
Script for Table 1
./scripts/build
R scripts used to compile C# applications.
./scripts/data_preparation
This folder stores scripts used to derive the variables used in the models. Raw data used by these scripts consists of extremely large files from CHELSA (https://chelsa-climate.org), ISRIC Data Hub (https://data.isric.org), and EU-DEM (https://land.copernicus.eu/imagery-in-situ/eu-dem/eu-dem-v1.1). The origin of these files are described in detail in the subfolder scripts/data_preparation/site_level_data
ForClim Model
The implementation and model code of the forest gap model ForClim used in this study was slightly modified to increase performance and reduce read and write actions on the HPC. In addition the functionality for a model initialisation phase as described in the main text was added. The model itself was not changed and corresponds to open access version ForClim v.4.0.1.656 which is publicly available under https://ites-fe.ethz.ch/openaccess/products/forclim. Thus model description of Bugmann (1994) and Huber et al. (2020) is valid for the model used in this study.
A detailled description of the regeneration module used in this study is given in Appendix B.
Climate and Forest Inventory Data
Further information on the forest inventory data used for deriving the regeneraiton rates is presented in Käber et al. (2023)
Literature
Bugmann, H. (1994). On the ecology of mountainous forests in a changing climate: A simulation study [PhD Thesis].
Käber, Y., Bigler, C., HilleRisLambers, J., Hobi, M., Nagel, T. A., Aakala, T., Blaschke, M., Brang, P., Brzeziecki, B., Carrer, M., Cateau, E., Frank, G., Fraver, S., Idoate-Lacasia, J., Holik, J., Kucbel, S., Leyman, A., Meyer, P., Motta, R., … Bugmann, H. (2023). Sheltered or suppressed? Tree regeneration in unmanaged European forests. Journal of Ecology, n/a(n/a). https://doi.org/10.1111/1365-2745.14181
Huber, N., Bugmann, H., & Lafond, V. (2020). Capturing ecological processes in dynamic forest models: Why there is no silver bullet to cope with complexity. Ecosphere, 11(5). https://doi.org/10.1002/ecs2.3109This repository contains observational forest data that was collected by the members of the EuFoRIa Network (www.euforia-project.org).
The individual contribution of each EuFoRIa partner who is acknowledged here.
EuFoRIa has been invaluable for this study, and we are grateful for their support and trust. In particular we thank Thomas A. Nagel (University of Ljubljana, Ljubljana, Slovenia), Tuomas Aakala (University of Eastern Finland, Joensuu, Finland), Markus Blaschke (Bavarian State Institute for Forestry, Freising, Germany), Bogdan Brzeziecki (Warsaw University of Life Sciences, Warszawa, Poland), Marco Carrer (University of Padova, Legnaro, Italy), Eugenie Cateau (Reserves Naturelles de France, Quetigny, France), Georg Frank (Austrian Federal Research Centre for Forests, Natural Hazards and Landscape (BFW), Wien, Austria), Shawn Fraver (School of Forest Resources, University of Maine, Orono, Maine, USA), Jan Holik (Silva Tarouca Research Institute, Brno, Czech Republic), Stanislav Kucbel (Department of Silviculture, Faculty of Forestry, Technical University Zvolen, Slovakia), Anja Leyman (Research Institute for Nature and Forest, Brussels, Belgium), Peter Meyer (Northwest German Forest Research Institute, Göttingen, Germany), Renzo Motta (University of Torino, Torino, Italy), Pavel Samonil (Silva Tarouca Research Institute, Brno, Czech Republic), Lucia Seebach (Forest Research Institute of Baden-Württemberg, Freiburg, Germany), Jonas Stillhard (Swiss Federal Institute for Forest, Snow and Landscape Research, Birmensdorf, Switzerland), Miroslav Svoboda (Czech University of Life Sciences, Prague, Czech Republic), Jerzy Szwagrzyk (University of Agriculture, Krakow, Poland), Kris Vandekerkhove (Research Institute for Nature and Forest, Brussels, Belgium), Ondrej Vostarek (Czech University of Life Sciences, Prague, Czech Republic), Kamil Kral (Czech University of Life Sciences, Prague, Czech Republic), Tzvetan Zlatanov (Institute of Biodiversity and Ecosystem Research, Bulgarian Academy of Sciences, Sofia, Bulgaria)
Carbon budget of Swiss forests: Evaluation and application of process models for assessing the future impact of management and environmental change
No full text2. Networks, states and empires in the Baltic Region
No full textMain author Kristian Gerner. Harald Runblom author of boxes on The Vikings and The Hansa.</p
Debris-flow activity along a torrent in the Swiss Alps: Minimum frequency of events and implications for forest dynamics
No full textThis study reports on a tree-ring-based reconstruction of geomorphic activity and illustrates impacts of such processes on tree germination along a debris-flow torrent in the Swiss Alps. Analysis included the identification of growth disturbances and the assessment of germination dates for 28 trees along the channel of the Geisstriftbach torrent (Valais, Swiss Alps). Provided that recolonizing trees indicate the minimum time elapsed since the last deposition, germination dates suggest that a devastating debris-flow event in the 1880s had cleared the surface and scoured the currently active channel. This interpretation is supported by two topographic maps showing a dislocation of the channel. Analyzing the age structure of trees along the channel in more detail, we observe higher tree ages with increasing distance from the cone apex. In addition, dendrogeomorphic methods allowed for the reconstruction of 13 debris-flow events between AD 1913 and 2006. In combination with geomorphic mapping, the spatial distribution of trees affected by individual events was assessed and a minimum frequency of previous debris-flow events reconstructed. Although the present study was based on a limited set of tree-ring records, it illustrates that tree-ring analysis in combination with cartographic methods holds much promise for dating minimum ages of surfaces cleared by destructive events as well as for determining the spatio-temporal impacts of past debris-flow activity
- …
