Supplementary data for "Less extreme and earlier outbursts of ice-dammed lakes since 1900"

This respository contains supplementary and source data for the study for<em><strong> Less extreme and earlier outbursts of ice-dammed lakes since 1900</strong></em> by Georg Veh and co-authors. We investigate trends in the peak discharge <em>Q</em>_p, flood volume <em>V</em>₀, timing <em>doy</em> (day of year) and elevation <em>Z</em> of glacier lake outburst floods (GLOFs), and focus on ice-dammed lakes. We use Bayesian hierarchical models, implemented in the package <em>brms</em> in in the statistical programming software <em>R</em>. More information on these data, including detailed scripts to process them, are available at available at https://github.com/geveh/IceDamFailures, and archived at 10.5281/zenodo.7326865. We provide the following data: <strong>1 The GLOF database</strong> <em>glofdatabase_2022_05_30.ods:</em> OpenOffice table with all reported GLOFs. Compiliation as of May 30, 2022 <em>Parameter_Readme.ods:</em> Readme file describing all parameters (i.e. columns) in glofdatabase_2022_05_30.ods <strong>2 Data on glacier elevation changes</strong> <em>dh_pergla_cut.7z:</em> zipped csv tables of cumulative elevation change (in m) for glaciers with repeat GLOFs between 2000 and 2019 <strong>3 Preprocessed GLOF data as R objects</strong> all_glofs_tibble.RDS: R-object with a preprocessed table of all reported GLOFs all_glofs_qp_tibble.RDS: R-object in table format of lakes with repeat GLOFs and reported peak discharge <em>Q</em>_p all_glofs_V0_tibble.RDS: R-object in table format of lakes with repeat GLOFs and reported flood volume <em>V</em>₀ glofs_ice_with_z.RDS: R-object of first reported GLOF from a given lake and its elevation <em>Z</em> <br> <strong>4 Output from Bayesian hierarchial models</strong> <em><strong>a) Regional models</strong></em> qp_models.RDS: R-object with regional quantile regression models of <em>Q</em>_p versus time for the 50th and 90th for 4 time periods V0_models.RDS: R-object with regional quantile regression models of <em>V</em>₀ versus time for the 50th and 90th for 4 time periods doy_trends_per_region.RDS: R-object with regression models of <em>doy</em> versus time for all dated GLOFs in the six regions Z_trends_per_region.RDS: R-object with a hierarchical regression models of elevation <em>Z</em> versus time for dated GLOFs in the six regions between 1900 and 2021 Regional_glacier_and_melt_volumes.rds: R-object containing the total volume of glacier volume and volume change between 2000 and 2019 in 100-m elevation bins <em><strong>b) Local models</strong></em> <em>V0_model_median_local.RDS:</em> R-object with regional quantile regression models of median <em>V</em>₀ versus time <em>qp_model_median_local.RDS:</em> R-object with local quantile regression models of median <em>Q</em>_p versus time <em>doy_trends_per_glacier.RDS: </em>R-object with regression models of <em>doy</em> versus time for lakes with repeat GLOFs <em>local_Qp_vs_dhdt_model.RDS:</em> R-Object containing a hierarchical model of local changes in <em>Q</em>_p versus glacier elevation change <em>local_V0_vs_dhdt_model.RDS:</em> R-Object containing a hierarchical model of local changes in <em>V</em>₀ versus glacier elevation change <strong>4 GIS Data</strong> <em>Ice_dammed_lakes_Zenodo.7z: </em>zipped folder containing manually mapped outlines of ice-dammed lakes. <em>Region_extents.7z:</em> zipped folder containing the outlines of the study regions. <strong>5 Figures</strong> <em>Qp_local.pdf:</em> PDF figure showing temporal trends of median Qp for individual glacier lakes <em>V0_local.pdf: </em>PDF figure showing temporal trends of median V0 for individual glacier lakes <em>doy_local.pdf:</em> PDF figure showing temporal trends in GLOF timing for individual glacier lakes<br>本仓库包含Georg Veh及其合著者题为《1900年以来冰碛湖爆发程度更温和且时间提前》的研究的补充与源数据。 本研究聚焦冰碛湖，探究冰川湖溃决洪水（glacier lake outburst floods, GLOFs）的峰值流量$Q_p$、洪水体积$V_0$、发生时间doy（一年中日序数）以及高程$Z$的变化趋势。研究采用贝叶斯分层模型（Bayesian hierarchical models），该模型通过统计编程软件R中的brms包实现。更多相关数据信息（包括处理数据的详细脚本）可在https://github.com/geveh/IceDamFailures获取，并已存档于10.5281/zenodo.7326865。 本仓库提供如下数据： 1. **冰川湖溃决洪水数据库** - `glofdatabase_2022_05_30.ods`：包含所有已报道冰川湖溃决洪水的OpenOffice表格，汇编截至2022年5月30日 - `Parameter_Readme.ods`：说明`glofdatabase_2022_05_30.ods`中所有参数（即列项）的自述文件 2. **冰川高程变化数据** `dh_pergla_cut.7z`：包含2000年至2019年间存在重复溃决事件的冰川的累积高程变化（单位：米）的压缩CSV表格 3. **预处理为R对象的冰川湖溃决洪水数据** - `all_glofs_tibble.RDS`：包含所有已报道冰川湖溃决洪水的预处理表格的R对象 - `all_glofs_qp_tibble.RDS`：包含存在重复溃决事件且报告了峰值流量$Q_p$的湖泊的表格格式R对象 - `all_glofs_V0_tibble.RDS`：包含存在重复溃决事件且报告了洪水体积$V_0$的湖泊的表格格式R对象 - `glofs_ice_with_z.RDS`：包含某一湖泊首次报道的溃决洪水及其对应高程$Z$的R对象 4. **贝叶斯分层模型输出** a) **区域模型** - `qp_models.RDS`：包含针对4个时间段的50%与90%分位数，探究$Q_p$随时间变化的区域分位数回归模型的R对象 - `V0_models.RDS`：包含针对4个时间段的50%与90%分位数，探究$V_0$随时间变化的区域分位数回归模型的R对象 - `doy_trends_per_region.RDS`：包含针对六个区域内所有有年代记录的冰川湖溃决洪水，探究doy随时间变化的回归模型的R对象 - `Z_trends_per_region.RDS`：包含针对1900年至2021年间六个区域内有年代记录的冰川湖溃决洪水，探究高程$Z$随时间变化的分层回归模型的R对象 - `Regional_glacier_and_melt_volumes.rds`：包含以100米高程分箱统计的2000年至2019年间冰川总体积与体积变化量的R对象 b) **局地模型** - `V0_model_median_local.RDS`：包含探究中位$V_0$随时间变化的区域分位数回归模型的R对象 - `qp_model_median_local.RDS`：包含探究中位$Q_p$随时间变化的局部分位数回归模型的R对象 - `doy_trends_per_glacier.RDS`：包含针对存在重复溃决事件的湖泊，探究doy随时间变化的回归模型的R对象 - `local_Qp_vs_dhdt_model.RDS`：包含探究$Q_p$局地变化与冰川高程变化关系的分层模型的R对象 - `local_V0_vs_dhdt_model.RDS`：包含探究$V_0$局地变化与冰川高程变化关系的分层模型的R对象 5. **GIS数据** - `Ice_dammed_lakes_Zenodo.7z`：包含人工绘制的冰碛湖轮廓的压缩文件夹 - `Region_extents.7z`：包含研究区域轮廓的压缩文件夹 6. **附图** - `Qp_local.pdf`：展示单个冰川湖中位$Q_p$时间变化趋势的PDF图 - `V0_local.pdf`：展示单个冰川湖中位$V_0$时间变化趋势的PDF图 - `doy_local.pdf`：展示单个冰川湖冰川湖溃决洪水发生时间变化趋势的PDF图