[R-sig-Geo] Climate data in R

Tue Aug 2 23:34:50 CEST 2016

Without knowing what part of the globe you need climate data for, nor what
data you need (30 year "normals", bioclim parameters, monthly values,
daily, extremes?), I can't give a useful specific answer.

What follows is an initial draft of a web page on climate data in R.  It's
somewhat US and North America centric.

The number of APIs providing climate data is growing rapidly.  The downside
is that most data providers are region/continent specific.  NOAA NDCD,
ACIS, Mesowest, PRISM, SNOTEL, SNODAS, and others are North America or
CONUS, there are different APIs and packages for EU, UK, NZ, and Canadian
climate.  The few worldwide services tend to have lower station densities
than region-specific services, especially for older dates.

Some APIs and overviews:
ACIS (NOAA):  http://www.rcc-acis.org/docs_webservices.html
NOAA Co-ops: http://tidesandcurrents.noaa.gov/api/
PRISM http://www.prism.oregonstate.edu/index.phtml
MesoWest: http://www.rcc-acis.org/docs_webservices.html
http://synopticlabs.org/api/
http://www.esrl.noaa.gov/psd/data/gridded/
https://climatedataguide.ucar.edu/climate-data/global-temperature-data-sets-overview-comparison-table
http://www.cgiar-csi.org/data/uea-cru-ts-v3-10-01-historic-climate-database

All of those APIs can be hit from R; some have R packages available to make
it easier.  [And, more importantly for my use cases, those packages isolate
my code that hits the package functions from changes in the underlying
API.  Instead of each of us having to keep up with changing APIs and fixing
our broken code every 6-12 months, package maintainers put out new versions
that work with the new APIs but use the same parameters and return the same
objects.  If my code from last year fails to fetch data this year, I
usually can simply update the package.  Thanks, maintainers!]

For R packages, start with https://github.com/ropensci/opendata#climate and
http://ropensci.org/packages/

Most non-raster APIs have 2 steps: a station finder that you query by name
or location, then a data fetcher pulling data from the found stations.

Raster datasets tend to be huge, and tend to be organized as 1 file per
date.  There's a PRISM API that lets you select locations and retrieve data
for a range of dates for requested rectangular AOAs.

My 2 lessons learned as an ecologist using climate data (not a real
climatologist) are:

1: Spatial interpolation is non-trivial.  PRISM uses not just distances
from stations of record, but also topography: elevation for temperatures
and precipitation and side of the mountain (very important for
precipitation).  Don't just use the nearest station or inverse-distance
krige unless you are in a broad flat plain or steppe.  [Sorry, Ed!]  There
are a couple of raster products for daily data (not just the monthly
PRISM), but they tend to be available a few weeks after the fact.

2: Station records are instrumentation values, not actual temperatures &
precipitation amounts.  Almost all climate data providers QA and filter the
values, because sensors drift and fail.  However, the appropriate amount of
filtering is dependent on your use.  Some blips in the raw data are
instrumentation errors, some are short extreme events; some are of a
magnitude & duration where it can't be determined which of those they are.
The appropriate or optimal amount of filtering for uses such as crop
production models is much greater than the appropriate amount for some
ecological applications where short intense events can be important.  Even
within larger agencies like NOAA, different regional climatologists filter
the data to different extents.  To do valid science, if extreme events
matter for your use, you need to know the preprocessing on your datastream,
lest you get fooled into thinking that region-specific filtering is
climatological signal.

If anyone has comments, additions, or corrections to the above, please send
them to me!

If the OP has a specific data type in a specific region, I might have a
specific suggestion.

Tom 2

Tom Philippi
Quantitative Ecologist & Data Therapist
Inventory and Monitoring Program
US National Park Service

On Mon, Aug 1, 2016 at 12:49 PM, Isaque Daniel <isaquedanielre at hotmail.com>
wrote:

> Hi,
>
> You can follow the example in:
> http://stackoverflow.com/questions/34619218/extract-raster-values-from-stack-to-points-in-for-loop
>
>
> Is really simple, using coordinates and extract functions of raster
> package.
>
>
> Best
>
> Isaque
>
>
>
> ------------------------------------------------------------------------------------------------------------------
> Eng. Agr. Isaque Daniel Rocha Eberhardt
> Mestre em Sensoriamento Remoto - Instituto Nacional de Pesquisas Espaciais
> (INPE)
> Doutorando em Transportes - Universidade de Bras??lia (UNB)
> Mobile: +55 (061) 99015658
>
> ------------------------------------------------------------------------------------------------------------------
> Agronomist engineer
> Master in Remote Sensing - National  Institute for Space Research (INPE) -
> Brazil
> PHD Student in Transport - Bras??lia University (UNB)
>
>
> ________________________________
> De: R-sig-Geo <r-sig-geo-bounces at r-project.org> em nome de Kenny Bell <
> kmb56 at berkeley.edu>
> Enviado: segunda-feira, 1 de agosto de 2016 19:47
> Para: M. Edward (Ed) Borasky
> Cc: R-sig-geo mailing list
> Assunto: Re: [R-sig-Geo] Climate data in R
>
> See https://cran.r-project.org/web/packages/prism/index.html for
> CRAN - Package prism<
> https://cran.r-project.org/web/packages/prism/index.html>
> cran.r-project.org
> prism: Access Data from the Oregon State Prism Climate Project. Allows
> users to access the Oregon State Prism climate data. Using the web service
> API data can easily ...
>
>
> interpolated US weather data
>
> On Mon, Aug 1, 2016 at 12:43 PM, M. Edward (Ed) Borasky <znmeb at znmeb.net>
> wrote:
>
> > I have a similar question. Assume I've downloaded the climate data for
> all
> > the stations in the US state of Oregon, including their latitude,
> longitude
> > and elevation, using rnoaa. How do I *interpolate* climate values for an
> > arbitrary latitude / longitude / elevation inside a triangle defined by
> the
> > nearest stations?
> >
> > On Mon, Aug 1, 2016 at 3:24 AM Miluji Sb <milujisb at gmail.com> wrote:
> >
> > > Dear all,
> > >
> > > I have a set of coordinates. Is it possible to extract climate data
> > > (temperature and precipitation) by coordinates using the R packages
> such
> > as
> > > rnoaa?
> > >
> > > For example;
> > >
> > > out <- ncdc(datasetid='ANNUAL', stationid='GHCND:USW00014895',
> > > datatypeid='TEMP')
> > >
> > > But instead of stationid can I pass a list of coordinates through it?
> > > Thanks a lot!
> > >
> > > Sincerely,
> > >
> > > Milu
> > >
> > >         [[alternative HTML version deleted]]
> > >
> > > _______________________________________________
> > > R-sig-Geo mailing list
> > > R-sig-Geo at r-project.org
> > > https://stat.ethz.ch/mailman/listinfo/r-sig-geo
> > >
> > --
> > How many people can stand on the shoulders of a giant before the giant
> > collapses?
> >
> >         [[alternative HTML version deleted]]
> >
> > _______________________________________________
> > R-sig-Geo mailing list
> > R-sig-Geo at r-project.org
> > https://stat.ethz.ch/mailman/listinfo/r-sig-geo
> >
>
>
>
> --
> Kendon Bell
> Email: kmb56 at berkeley.edu
> Phone: (510) 612-3375
>
> Ph.D. Candidate
> Department of Agricultural & Resource Economics
> University of California, Berkeley
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>

	[[alternative HTML version deleted]]