lhmet / inmetr

DEPRECATED A R-package to Import Historical Data from Brazilian Meteorological Stations

DEPRECATED - no longer actively maintained

Due to changes in access to the BDMEP-INMET data acquisition system, {inmetr} is no longer supported. See https://portal.inmet.gov.br/noticias/inmet-lan%C3%A7a-novo-portal and consider using https://bdmep.inmet.gov.br/ instead.

inmetr: Historical Data from Brazilian Meteorological Stations in R

Overview

inmetr provide access to historical data measured by meteorological stations available in the Meteorological Database for Education and Research (BDMEP) from National Institute of Meteorology (Instituto Nacional de Meteorologia - INMET), Brazil.

Installation

inmetr is easy to install from Git Hub using the devtools package.

library(devtools)

install_github('lhmet/inmetr')

Load package

library(inmetr)

Stations ID

To search a meteorological station from INMET we can use metadata of INMET stations included in inmetr package as bdmep_meta.

head(bdmep_meta)
tail(bdmep_meta)

bdmep_meta is a data frame providing the id of stations, a numeric code defined by OMM. This id is a necessary argument to bdmep_import() function which allows to download data from meteorological stations into R.

Here, we show how to find the OMM code for meteorological stations at two cities (randomly sampled).

#stations <- c("Santa Maria", "Macapá")
stations <- c("Rio de Janeiro", "Goiás")
# random sample of two stations names 
#stations <- sample(bdmep_meta$name, 2)
stations_rows <- pmatch(stations, bdmep_meta$name)
bdmep_meta[stations_rows, ]
stns_codes <- bdmep_meta[stations_rows, "id"] 
stns_codes

Import data

Now we can import data for the two cities from 1961 to the current date.

start_date <- "01/01/1961"
end_date <- format(Sys.Date(), "%d/%m/%Y")
met_data <- bdmep_import(id = stns_codes,
                         sdate = start_date, 
                         edate = end_date, 
                         email = "your-email",
                         passwd = "your-password",
                         verbose = TRUE)

# check de start date
head(met_data)

# check de end date
tail(met_data)

You can save data in a CSV file setting argument destdir = "path/to/write/files" in bdmep_import function. Data will be save one file per station.

A description of meteorological variables can be obtained by:

bdmep_description()

Eventually, if the request failed a message will be prompted with the HTTP status code, for example:

------------------------
station: `r stns_codes[1]`
Bad Gateway (HTTP 502).
------------------------
station: `r stns_codes[2]`
Bad Gateway (HTTP 502).

In this case the outcome data frame will be filled with NA, except for request_status which will return information on the request status.

To cite this software

citation("inmetr")

To cite package 'inmetr' in publications use:

  Tatsch, J.D. 2020. inmetr R package (v 0.2.5): Historical Data from
  Brazilian Meteorological Stations in R. Zenodo.
  https://doi.org/10.5281/zenodo.580813.

A BibTeX entry for LaTeX users is

  @Manual{,
    title = {inmetr: Historical Data from Brazilian Meteorological Stations in R},
    author = {Jonatan Tatsch},
    year = {2020},
    note = {R package version 0.2.5},
    doi = {https://doi.org/10.5281/zenodo.580813},
    institution = {Universidade Federal de Santa Maria-UFSM},
    url = {https://github.com/lhmet/inmetr},
  }

inmetr's Issues

Failed to connect to www.inmet.gov.br port 80: Timed out

Hello!

This code it's awesome! It's work to get some information about two stations.

However, when I try to get informations about 10 stations, I received this text:

Error in curl::curl_fetch_memory(url, handle = handle) :
Failed to connect to www.inmet.gov.br port 80: Timed out

Now I can't get mor informations.

Do you know how can I solve this?

Thanks!

Downloading daily or monthly data

Hello,
Is there a way to tweek a function or a configuration available to download daily or monthly data instead of hourly?

Thanks!

Error in (rowheader + 1):(length(x) - 1) : argument of length 0

Hi,

I'm trying to run the README.md example but I got this error wen trying to import data:

------------------------------
station: 83936
OK (HTTP 200).
Error in (rowheader + 1):(length(x) - 1) : argument of length 0

My code is:

stations <- c("Santa Maria", "Macapá")
stations_rows <- pmatch(stations, bdmep_meta$name)
stns_codes <- bdmep_meta[stations_rows, "id"] 
# the line that returns this error
met_data <- bdmep_import(id      = stns_codes,
                         sdate   = "01/01/2012", 
                         edate   = "01/02/2012",
                         destdir = 'C:/',
                         email   = "[email protected]",
                         passwd  = "666",
                         verbose = T)

Is this problem related to the code (in the example)? or with any change in the location or format of the data (I saw that the site has changed its address)?

Thank you,
Daniel

Stations missing from inmetr::bdmep_meta

Some stations are not listed in inmetr::bdmep_meta dataset. The stations are missing from original file (http://www.inmet.gov.br/webcdp/climatologia/normais/imagens/normais/planilhas/Relac_Est_Meteo_NC.xls) as well, but they are listed at http://www.inmet.gov.br/sim/sonabra/index.php. The meteorological information is in BDMEP and can be downloaded through the package bypassing "all(id %in% inmetr::bdmep_meta$id)" in bdmep_import.

Can't seem to import BDMEP data

I'm getting this message when I try to import:

---

station: 83579
Bad Request (HTTP 400).

my script:
library(inmetr)
head(bdmep_meta)
sort(unique(bdmep_meta$name))
stations<-c("Araxá")
stations_rows<-pmatch(stations, bdmep_meta$name)
bdmep_meta[stations_rows,]

stns_cods<- bdmep_meta[stations_rows,'id']
start_date <- "01/01/1961"
end_date <- format(Sys.Date(),' %d/%m/%Y')
met_data <- bdmep_import(id=stns_cods, sdate = start_date, edate = end_date, email= "myemail", passwd='mypass', verbose=T)
(just to be clear, my e-mail and password work fine on the BDMEP website)

I get bad request for all stations I've tried.
I followed the instructions, which were very clear... what I am doing wrong?
Thanks

NA values are replaced with station id only when downloaded as .csv

When I use destdir = "path/to/write/files" and open the files downloaded in the specified location. All of the NA values are replaced by the station id.

Erro: `replacement` must be a character vector

Erro ao tentar puxar os dados da estação da cidade do Rio de Janeiro.

info <- bdmep_stations()

start_date <- "01/01/1961"
end_date <- format(Sys.Date(), "%d/%m/%Y")
sm <- import_bdmep(id = 83743,
                   sdate = start_date, 
                   edate = end_date, 
                   email = "[email protected]",
                   passwd = "minhasenha")
Login sucessfull.
Erro: `replacement` must be a character vector