A replacement for rnoaa

[jfunction]

According to the Notice in the README:

NOAA has changed not only the base URL of the API but reorganized the structure and endpoints of the data and API. The current rnoaa package is still functional with the NOAA API v2 but the data available in the API end at approximately 2022-09-15. A new NOAA weather package is planned to replace rnoaa functionality where possible. At that point the rnoaa package will be archived.

Any more information on this? I'm embarking on a project where we are growing a community of practice in Climate Sensitive Infectious Disease modelling and am eager to know what to recommend to R users. Clearly the plan is for rnoaa to go, but does the new package have a name? Maybe an alpha or beta version? Thanks.

[djhocking]

I'm actually just starting the formal package after testing some of the new API data pulls. I have to check cran availability before finalizing the name and pushing a copy to github. This can be the push I need to get it up and running.

Are there specific datasets you're looking for? I can prioritize trying to get those into an alpha version on GitHub for you.

[jfunction]

This is awesome news! Regarding datasets I would find it difficult to speak for everyone especially since my background is more in the disease modelling field. That said, I think something with global extent would be ideal since the community is global and perhaps temperature and rainfall are good candidates for first datasets since they are the most well studied.

[ericpgreen]

Hey @djhocking, stopped by to ask the same question as @jfunction. Last year I created the package feb2 about groundhog day predictions for my r course. Here's what I use:

ghcnd_stations()
meteo_nearby_stations()
ghcnd_search()
isd_stations_search()

Steps I take with noaa data to construct the classification of long winter or early spring:

• Identify the 10 weather stations with TMAX data closest to each prognosticator’s city
• Use rnoaa::ghcnd_search() to obtain daily historical high temperatures for each station
• Calculate the mean daily high temperature for each prognosticator’s city by averaging over the 10 closest weather stations
• Calculate the mean monthly high temperature for each prognosticator’s city by averaging over the daily data
• Calculate the 15-year rolling mean high monthly temperature for each prognosticator’s city

Thanks for the great package and efforts to adapt it!

[kross0926]

I was wondering if the new package was pushed to GitHub?

[djhocking]

Sorry the development stalled on the replacement. Even the base functions had to be started completely from scratch and I ran out of free time. I'll try to at least get some basic functionality posted soon even if it's not as complete or robust as I'd like.

[ellenmonster]

Hi! I maintain an RShiny dashboard for some National Park Service users that pulls daily precipitation data using rnoaa. I'm currently upgrading code to replace packages that have recently retired or soon will retire. Do you have an idea if/when a replacement package for rnoaa might be available? We are just trying to figure out best options moving forward at this point. Thank you!

[adamhsparks]

For folks looking for daily weather data, I maintain another rOpenSci package, {GSODR} that has daily weather data from weather stations that are largely duplicated and then some with the GHCN that {rnoaa} has access to.

https://github.com/ropensci/GSODR

@sckott and I did talk about merging the packages but in the end decided against it for various reasons. They don't really duplicate each other as they access different data sets, but the different data sets do have duplicated data...