README.md

ecmwfr2

Goals

Submit ERA5 downloading tasks in batch and download them with modern parallel downloading tool aria2c.

解决cdsapi的问题：

• 不能并行提交任务

• 单线程下载速度过慢

实现思路

• 先提交所有任务

• 写一个死循环，不断刷新和下载数据，直到所有任务完成

  # 见examples
  Rscript.exe download.R

• aria2c并行下载，跑满30M（CUG网络）

Installation

remotes::install_github("rpkgs/ecmwfr2")

Usage

授权

1. 登陆下面的网址，即可显示api key，形式如下：

   https://cds.climate.copernicus.eu/api-how-to

   url: https://cds.climate.copernicus.eu/api/v2
   key: 12106:faa165eb-2d80-4843-9c80-2d5e90adf***

2. 将api保存至~/.cdsapirc

Examples

• 1. 下载1959-2021 ERA5 hourly地表气象数据

  library(ecmwfr2)
  login()
  # ecmwfr::wf_set_key(user, key, service = "cds") # second option
  
  param <- list(
    product_type = "reanalysis",
    format = "netcdf",
    variable = c("2m_dewpoint_temperature", "2m_temperature", "skin_temperature"),
    time = c("00:00", "06:00", "12:00", "18:00"),
    year = c("2010"),
    month = sprintf("%02d", 1:12),
    day = sprintf("%02d", 1:31)
  )
  
  dsname = "reanalysis-era5-single-levels"
  
  vars <- c(
    "2m_temperature",
    "2m_dewpoint_temperature",
    "skin_temperature"
  )
  
  # c_request(param, dsname)
  down_vars(vars, param, dsname)

Known issues

1. keyring 授权不适用于wsl