Skip to content

Latest commit

 

History

History
102 lines (66 loc) · 4.48 KB

README.md

File metadata and controls

102 lines (66 loc) · 4.48 KB
Raw

ReportingBias

This repository contains two scripts, one that preprocesses a raw data table of reported GLOFs (preprocessing.R), and a second one (assessing_bias.R) that

  • estimates temporal and regional trends in reported GLOFs;
  • estimates GLOF trends from the predictors 'annual air temperatures' and 'research activity'; and
  • hind- and forecasts annual GLOF counts, based on these two predictors.

The codes are written in the statistical programming language R (https://www.r-project.org/) and run from the Graphical User Interface RStudio (https://rstudio.com) under a Microsoft Windows 10 operating system. To successfully run the codes, please install both R and RStudio on your machine.

The R codes also depend on a number of packages, listed at the beginning of both scripts. Please install those before running the codes. The comments within the scripts provide further details on model dependencies and usage of functions.

Scripts

preprocessing.R

Preparation script to obtain annual statistics of

  • GLOF counts by dam type;
  • glaciological research activity;
  • temperature; and
  • precipitation

from the original Open-Office spreadsheet 'Global_GLOF_database_2021_06_09.ods', available at this page (see a detailed description in the section on Input data). The script produces the R-Data object regional_glof_stats.rds, which is already available on this page.

assessing_bias.R

Main script to

  • find change points in time series of reported GLOFs, air temperatures, and glacier surveys;
  • estimate trends in GLOF reporting for each study region and dam type;
  • predict annual GLOF counts from air temperatures and the number of glacier surveys;
  • to hind- and forecast the number of GLOFs before and after the global break in GLOF reporting.
  • reproduce all figures in the manuscript.

Input data

Global_GLOF_database_2021_06_09.ods

Open-Office spreadsheet as of 09 June 2021 with seven sheets named after the regions, for which we obtained historical GLOF occurrences. Each sheet has 32 columns containing the attributes that we were able to collect for each GLOF. Empty cells mean 'No Data'. The first row is the column name, followed by two rows with further description of the content and the data structure. The content of the columns 'Major_RGI_Region', 'Mountain_range_Region', 'Glacier', 'RGI_Glacier_Id', and 'RGI_Glacier_Area' is from the Randolph Glacier Inventory, V6.0 (https://www.glims.org/RGI/rgi60_dl.html).

Region_extents.zip

Extents of study regions in a WGS 84 / World Mercator projection

regional_glof_stats.rds

R-Data object (a list with 8 entries) containing regional annual statistics of GLOF occurrences, temperatures, and research activity. Description of the column names:

  • 'year': Year;
  • 'freq': Total number of reported GLOFs per year, including GLOFs from volcanic eruptions;
  • 'moraine': Number of moraine-dam failures per year;
  • 'ice': Number of ice-dam failures per year;
  • 'other': Number of GLOFs from other (bedrock, water pockets, supraglacial) or unknown sources;
  • 'volc': Number of GLOFs from subglacial lakes beneath ice-covered volcanoes;
  • 'mb_meas': Annual number of glacier surveys measuring in-situ mass balances from the WGMS database;
  • 'front_meas': Annual number of glacier surveys measuring in-situ front variations;
  • 'dch_meas': Annual number of glacier surveys measuring geodetic mass balances (includes also remote sensing studies);
  • 'all_meas': Annual sum of mb_meas, front_meas, and dch_meas;
  • 'mb_and_front': Annual sum of mb_meas and front_meas;
  • 'region': Name of the study region;
  • 'year_scale': Standardised years (zero mean and unit standard deviation);
  • 'temp_mean': Mean annual air temperature extracted from the CRU TS 4.05 dataset from all lakes that produced at least one GLOF in a given region;
  • 'temp_q25': 25th percentile of annual air temperatures in a given region;
  • 'temp_q75': 75th percentile of annual air temperatures in a given region;
  • 'pre_sum': total amount of precipitation in a given region.

References

Veh, G., Lützow, N., Kharlamova; V., Petrakov, D., Hugonnet, R. & Korup, O.: Trends, breaks, and biases in the frequency of reported glacier lake outburst floods. Earth's Future (accepted)

See also

http://glofs.geoecology.uni-potsdam.de

Contact

Georg Veh

Working group on natural hazards

University of Potsdam

georg.veh@uni-potsdam.de

https://www.uni-potsdam.de/de/umwelt/forschung/ag-naturgefahren.html