[R-sig-Geo] ordering points into a line

[Ben Tupper]

You make it look so easy!  That is just what I was trying to noodle
out.  I need to spend more time digging into the
[s2](https://r-spatial.github.io/s2/index.html) package.  Thank you!


On Wed, Jun 5, 2024 at 1:09 PM Dewey Dunnington <dewey using dunnington.ca> wrote:
>
> Hi Ben,
>
> I had a bit of fun with this [1]...the act of following a shortest path
> and finding the endpoints are sort of related (in that if you follow the
> shortest path setting a maximum distance threshold, eventually you will
> fail to find any more points, and you've found (maybe) and endpoint!).
>
> I hope this is helpful!
>
> Cheers,
>
> -dewey
>
> [1] https://gist.github.com/paleolimbot/0be47836de5008f308959923dac02c5b
>
> On 2024-06-05 13:20, Ben Tupper wrote:
> > Hello,
> >
> > NOAA's Ocean Prediction Center (https://ocean.weather.gov/) serves the
> > US Navy's almost-daily estimate of the locations of the north and
> > south walls of the Gulf Stream in the North Atlantic
> > (https://ocean.weather.gov/gulf_stream_latest.txt). The also serve an
> > archive via a FTP server
> > (https://ftp.opc.ncep.noaa.gov/grids/experimental/GStream) for
> > downloads by year.  The dataset provides manually estimated of
> > locations along the so-called north and south walls of the Gulf
> > Stream, although there is ongoing research and debate about the best
> > way to define the Gulf Stream. My colleagues and I have been slowly
> > pulling together a small package to make working with this data easier
> > (https://github.com/BigelowLab/gstream).
> >
> > This dataset provides an interesting spatial puzzle because the points
> > for a specific wall on a given date are not ordered.  Assuming that
> > the points could be ordered into LINESTRING, the problem is how to do
> > it.
> >
> > We probably can use sfnetworks package
> > (https://luukvdmeer.github.io/sfnetworks/) extensive tools to follow a
> > shortest path... but first one of the two endpoints must be found.
> > Eeek! How does one do that programmatically?  The plot attached shows
> > the points for the north wall on 2020-12-19.  Also shown are the first
> > and last points in the record which are clearly not the endpoints of a
> > path we might assemble from the points.
> >
> > #### START
> > suppressPackageStartupMessages({
> >   library(sf)
> >   library(readr)
> >   library(dplyr)
> > })
> >
> >
> > file =
> > "https://raw.githubusercontent.com/BigelowLab/gstream/main/inst/examples/2020-12-19-north.csv"
> > x = readr::read_csv(file, col_types = "nn") |>
> >   sf::st_as_sf(coords = c("X", "Y"), crs = 4326) |>
> >   sf::st_as_sf()
> >
> >
> > p = sf::st_cast(dplyr::select(x, attr(x, "sf_column")), "POINT")
> > plot(sf::st_geometry(x), type = "b", axes = TRUE, reset = FALSE)
> > plot(dplyr::slice(p,1), col = "orange", add = TRUE, pch = 2, cex = 1.5)
> > plot(dplyr::slice(p,nrow(x)), col = "orange", add = TRUE, pch = 6, cex
> > = 1.5)
> > ### END
> >
> > --
> >
> > Ben Tupper he/him
> >
> > Research Associate
> >
> > Bigelow Laboratory for Ocean Sciences
> >
> > –––––––––––––––––––––––––––––––––––––––––––––––––––
> >
> > E  btupper using bigelow.org
> >
> > O  207-315-2567
> >
> > –––––––––––––––––––––––––––––––––––––––––––––––––––
> >
> > Bold Science for Our Blue Planet  |  BIGELOW.ORG
> >
> > 60 BIGELOW DRIVE  |  EAST BOOTHBAY, MAINE 04544 USA
> >
> > _______________________________________________
> > R-sig-Geo mailing list
> > R-sig-Geo using r-project.org
> > https://stat.ethz.ch/mailman/listinfo/r-sig-geo