Data and code for publication: "Spatially and temporally dense measurements reveal meteorological driver of atmospheric mercury variability"
收藏资源简介：
This repository includes all data required to reproduce key figures from Roy et al.: Spatially and temporally dense measurements reveal meteorological driver of atmospheric mercury variability. Subdirectories are organized as follows: data Organized into four subdirectories containing all data required for reproducing results from this manuscript NADP Location of Raw NADP data. Files for four main sites (NJ30,NY06,NY20,VT99) and one supplemental site (NJ54) should be downloaded at AMNet website (https://nadp.slh.wisc.edu/networks/atmospheric-mercury-network/) The combined file required for all analysis (AMNET-NE-h.csv) is created by running site_contact.py. GCdata Contains subdirectories with output from each GEOS-Chem simulation (GC_output) and processed subsets of data (GC_processed). Additionally, Global Mercury Assessment (2018) emissions are included in this directory pblh Location of observationally derived and MERRA2 pblh data. Raw, observationally derived met data described by Zhang et al. (2020) (https://doi.org/10.1029/2020JD032803) can be downloaded at https://doi.org/10.5281/zenodo.3934378. Profiles for EWR, BDL, and MHT required for reproducing analysis. File names will be {site prefix}_20m_interp_profiles.nc In order for profiles to be readable by xarray, the following command should be executed for each site immediately after download: ncrename -v height,h_magl -v time,datetime {site prefix}_20m_interp_profiles.nc {site prefix}_20m_interp_profiles_xr.nc MERRA2 contains meteorological reanalysis data for base and nested runs. Processing scripts used to generate these files are also included in this directory (.sh) boxmod_opt Location of solution space arrays for figures 3 and 4 of main text and S5-S8 of supplemental. Scripts for generating these arrays are located at scripts/preprocess/BulkFlux.py and scripts/preprocess/BulkFlux_June_supplemental.py. scripts Organized into two subdirectories containing scripts for preprocessing data inputs and plotting/analyzing data. preprocess BulkFlux.py: Generates optimal solution space for the selected year and month. Note - will only work if data is available! BulkFlux_June_Supplemental: Generates optimal solution space for supplemental figures S5-S8 for selected year and month. Note - will only work if data is available in pointer directories. function.py: identical to function.py in figures (soft link in original directory). gc_preprocess.sh: Used to select, concatenate files from raw GC output that are then directed to data/GC_output/ gc_preprocess_STND.sh: Same as gc_preprocess.sh, except for STND simulation (needed to read parameters for hybrid sigma vertical coordinates used by GEOS-Chem) GC_UW_FT.py: script for subsampling raw GEOS-Chem output to compile concentrations within the pbl (C_PBL), upwind (C_UW), and the free troposphere (C_FT) figures function.py: Important functions required for processing data prior to plotting EST_figures.ipynb: Plots all figures in main text, along with any percentages discussed therein. EST_supplement.ipynb: Plots all supplemental figures. EST_graphabstract.ipynb: Plots map required for background of graphical abstract figure. results Destination for all figures produced for the main text and supplemental. Note: For simplicity, we have uploaded preprocessed GEOS-Chem output required to reproduce the manuscript and supplemental. Raw GEOS-Chem output is available upon request. Please reach out if you have any questions about the scripts or workflow! Eric Roy - Email: emroy (at) mit (dot) edu
本仓库包含复现Roy等人论文《时空密集测量揭示大气汞变化的气象驱动因子》（*Spatially and temporally dense measurements reveal meteorological driver of atmospheric mercury variability*）关键图表所需的全部数据。 子目录组织如下： ### data 本目录分为四个子目录，包含复现本手稿结果所需的全部数据。 #### NADP 原始NADP（国家大气沉降计划，National Atmospheric Deposition Program）数据的存储位置。四个主要站点（NJ30、NY06、NY20、VT99）与一个补充站点（NJ54）的数据文件需从AMNet网站（https://nadp.slh.wisc.edu/networks/atmospheric-mercury-network/）下载。所有分析所需的合并文件`AMNET-NE-h.csv`需通过运行`site_contact.py`生成。 #### GCdata 包含两个子目录，分别存储各GEOS-Chem（GEOS-Chem大气化学传输模式）模拟的输出结果（`GC_output`）与处理后的数据集子集（`GC_processed`）。此外，本目录还收录了《全球汞评估（2018）》的排放数据。 #### pblh 观测衍生数据与MERRA2（现代气象研究与应用再分析资料第2版）行星边界层高度数据的存储目录。 由Zhang等人（2020）发表的原始观测衍生气象数据（https://doi.org/10.1029/2020JD032803）可在https://doi.org/10.5281/zenodo.3934378下载。复现分析所需的EWR、BDL与MHT站点的廓线文件命名格式为`{site prefix}_20m_interp_profiles.nc`。 为使廓线文件可被xarray（Python科学数据处理库）读取，需在下载后立即对每个站点执行如下命令： ncrename -v height,h_magl -v time,datetime {site prefix}_20m_interp_profiles.nc {site prefix}_20m_interp_profiles_xr.nc MERRA2包含基准模拟与嵌套模拟的气象再分析数据，本目录还附带了用于生成这些文件的处理脚本（`.sh`格式）。 #### boxmod_opt 存储正文中图3、图4以及补充材料图S5-S8的解空间数组。生成这些数组的脚本位于`scripts/preprocess/BulkFlux.py`与`scripts/preprocess/BulkFlux_June_supplemental.py`。 ### scripts 本目录分为两个子目录，分别包含数据预处理脚本与数据绘图、分析脚本。 #### preprocess - `BulkFlux.py`：为选定的年份与月份生成最优解空间。注意：仅当指针目录中存在对应数据时方可运行。 - `BulkFlux_June_supplemental.py`：为补充材料图S5-S8生成选定年份与月份的最优解空间。注意：仅当指针目录中存在对应数据时方可运行。 - `function.py`：与`figures`目录下的`function.py`完全一致（原目录中为软链接）。 - `gc_preprocess.sh`：用于选取并拼接原始GEOS-Chem输出文件，将结果输出至`data/GC_output/`目录。 - `gc_preprocess_STND.sh`：功能与`gc_preprocess.sh`一致，但专为STND模拟编写（用于读取GEOS-Chem使用的混合σ垂直坐标参数）。 - `GC_UW_FT.py`：用于对原始GEOS-Chem输出进行子采样，以整合行星边界层内（C_PBL）、上风区（C_UW）以及自由对流层（C_FT）的浓度数据。 #### figures - `function.py`：绘图前处理数据所需的核心函数。 - `EST_figures.ipynb`：用于绘制正文中所有图表及文中提及的相关百分比统计。 - `EST_supplement.ipynb`：用于绘制所有补充材料图表。 - `EST_graphabstract.ipynb`：用于绘制图形摘要背景所需的地图。 ### results 存储正文中与补充材料所有图表的输出目录。 --- 备注：为简化操作流程，我们已上传复现本论文及补充材料所需的预处理后GEOS-Chem输出数据。原始GEOS-Chem输出数据可通过申请获取。 若对脚本或工作流程有任何疑问，请随时联系我们！ Eric Roy - 电子邮箱：emroy@mit.edu
