[R-sig-Geo] multiple lines

[Michael Sumner]

There's a number of questions I have, and a few notes.

You want line segments (two points, start,end) in x,y,time, identified by
boat?

Spatial is not really suited to this kind of multidimensional line data,
but there are specialized tools like spacetime and spatstat, no real
coherence between them but you might find a sub-domain that fits for you.
Explore the Spatiotemporal CRAN Task View for options.

In Spatial, or in sf, this would be a data set of line segment features,
with time start-end and other grouping observations, like boat ID.

There's no need to project your raster, which is costly and lossy, you can
transform the boat data to the CRS of the raster.

I may be able to help more if you clarify a bit.

Cheers, Mike

On Sat, Feb 18, 2017, 03:21 marta azores <martazores at gmail.com> wrote:

> Hello
> I am trying to create multiple lines based on boat positions and calculate
> the distance for each boat track. I have already managed to create a single
> line with all the positions and calculate the distance but I can't managed
> to divided the line by transects which are previously defined based on the
> boat positions. These transects are not consecutive as they can be from
> different days and boats. This is an example of my data:
>
>  dput(effort)
>  structure(list(Ref = c(1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L, 15L,
>  16L), Effort = c(1L, 1L, 1L, 1L, 0L, 1L, 1L, 1L, 0L, 1L, 0L),
>   Trip = structure(c(1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L,
>   2L), .Label = c("P001.01", "P0010.01"), class = "factor"),
>  Code = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
>  1L), .Label = "P1968", class = "factor"), Date = structure(c(1L,
>  1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L), .Label = c("04/05/2010",
>  "19/05/2010"), class = "factor"), Time = structure(c(5L,
>  6L, 7L, 9L, 10L, 1L, 2L, 3L, 4L, 8L, 11L), .Label = c("08:12",
>  "08:25", "08:43", "08:59", "17:30", "17:51", "17:57", "18:02",
>  "18:09", "18:31", "18:39"), class = "factor"), LAT = c(38.67173,
>  38.67035, 38.66965, 38.66908, 38.64717, 38.22252, 38.26693,
>  38.18948, 38.17187, 38.25057, 38.18335), LONG = c(-28.7492,
> -28.64475, -28.70907, -28.72917, -28.63157, -28.90358, -28.87397,
> -29.07098, -28.9542, -28.8876, -28.92818), transect = c(1L, 1L, 1L,
> 1L, 1L, 2L, 2L, 2L, 2L, 4L, 4L)), .Names = c("Ref", "Effort", "Trip",
> "Code", "Date", "Time", "LAT", "LONG", "transect"), class =
> "data.frame", row.names = c(NA, -11L))
>
> The code that I have applied so far is this:
>
> ### boat points the locations of the boats are sort by date and time!
> effort <- read.table(paste0(path_data,"effort_test.csv"), header=TRUE,
>    sep=",", na.strings="NA", dec=".", strip.white=TRUE)
> pos<-as.data.frame(cbind(effort$LAT,effort$LONG, effort$transect))
> str(pos)
> sp::coordinates(pos) <- ~V2+V1
> sp::proj4string(pos) <-CRS("+proj=longlat +datum=WGS84 +no_defs")
> ### project it to somewhat equal area
> effortSP <- sp::spTransform(pos, CRS("+proj=laea +lat_0=30 +lon_0=-30
> +x_0=0   +y_0=0 +a=6370997 +b=6370997 +units=m +no_defs"))# plot boat
> positions
> # cost raster sea and island
> azoTS1<- raster::raster(paste0(path_data,"azoTS1.tif"))
> ### project it to somewhat equal area
> azoTS1<- raster::projectRaster(azoTS1,  crs="+proj=laea +lat_0=30
> +lon_0=-30 +x_0=0 +y_0=0 +a=6370997 +b=6370997 +units=m +no_defs")
> ###scale  real cost file
> azoTS1[azoTS1 ==2]<-100
> plot(azoTS1)
> ### increase the spatial resolution
> costraw<-raster::disaggregate(azoTS1,fact=c(5, 5))
> # prepare it for analysis
> transition<- gdistance::transition(1/costraw, mean, directions=8)
> trCostS16 <- gdistance::geoCorrection(transition, type="c")
> # calculating the first segment of the whole sailing path
> line<- gdistance::shortestPath(trCostS16,
> effortSP at coords[1,],effortSP at coords[2,], output="SpatialLines")
> lines(line)
> ### here we start with the for-loopfor (i in (seq(2,length(effortSP) -
> 1))) {# calculation of the rest of the segements
> nextSegment<- gdistance::shortestPath(trCostS16,
> effortSP at coords[i,],effortSP at coords[i+1,], output="SpatialLines")#
> simple addition combines the single spatialline segements
> line <- nextSegment + line# we plot each new segement
> lines(nextSegment)}
>
>         [[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
>
-- 
Dr. Michael Sumner
Software and Database Engineer
Australian Antarctic Division
203 Channel Highway
Kingston Tasmania 7050 Australia

	[[alternative HTML version deleted]]