[R-sig-Geo] Reprojection doubt

[Michael Sumner]

It probably doesn't matter at all. Your data is longitude,latitude on datum
WGS84? Those strings are equivalent (some softwares will flip the order of
lon,lat as that's what the standard dictates but easy to spot that
usually).

raster::projection(), raster::crs(), sp::proj4string() all are fairly
equivalent functions, they just get the projection metadata - they are just
overlapping and inconsistent because of historical vagarie, they differ in
ideological stance and how much checking and churn they do to the metadata
and in efforts to interpret your intentions. But on the face of it I don't
think you need to worry about your code here.

Just keep an eye on whether your data looks right, before and after
reprojection. It takes care, but the details are yours alone for checking
unless you can pin down a reproducible example that raises questions.

Good luck.

fwiw you seem to not be doing anything with ncdf4 (or ncdf4::nc_open) here.
the raster package *uses* ncdf4 on your behalf via raster()/stack()/brick()
functions.

Cheers, Mike.



On Fri, Nov 6, 2020 at 12:11 PM Pietro Andre Telatin Paschoalino <
Pietro_telato using hotmail.com> wrote:

> Dear,
>
> I have a question that is possibly extremely easy for you.
>
> I'm extracting data from layers of grid to a shape file.
>
> To do this, first, I know that I need to reproject the coordinates
> reference system of my shape to the same CRS as my grid. The code for this
> is:
>
> library(ncdf4)
>
> fn<-file.path(path of my data)
>
> ncfile <- nc_open(fn)
>
> rasbrick <- stack(fn)
>
> rasbrick <- brick[[70:81]] (the layers that I want)
>
> shape2<- spTransform(shape, crs(rasbrick)) in lowercase, right?  It is my
> first question.
>
> The rest of the extraction code works without any problems. If I open this
> file in qgis the CRS is:
>
> OGC:CRS84 - WGS 84 (CRS84) - Geographic
>
> The problem is that if I write a raster of a layer using:
>
> rasbrick <- brick[[70]]
>
> writeRaster(rasbrick, "test.tif", format="GTiff")
>
> and open the raster in qgis the CRS is:
>
> EPSG:4326 - WGS 84 - Geographic
>
> So, my second question is why on the raster it looks like EPSG and on the
> shape it looks like OGC? And could this cause any problems in extracting
> data from layers for the shape? If yes, what I need to change to reproject
> the shape for the correct CRS?
>
> Thanks.
>
> Pietro Andre Telatin Paschoalino
> Doutorando em Ciências Econômicas da Universidade Estadual de Maringá -
> PCE.
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo using r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>


-- 
Michael Sumner
Software and Database Engineer
Australian Antarctic Division
Hobart, Australia
e-mail: mdsumner using gmail.com

	[[alternative HTML version deleted]]