R/Gis: How to Subset a Shapefile by a Lat-Long Bounding Box

R/GIS: How to subset a shapefile by a lat-long bounding box?

You could try the following:

library(rgeos)
rivers <- readWKT("MULTILINESTRING((15 5, 1 20, 200 25), (-5 -8,-10 -8,-15 -4), (0 10,100 5,20 230))")
bbx <- readWKT("POLYGON((0 40, 20 40, 20 0, 0 0, 0 40))") 

rivers.cut <- gIntersection(rivers, bbx)

plot(rivers, col="grey")
plot(bbx, add=T, lty=2)
plot(rivers.cut, add=T, col="blue")

Subset Spatial Features by Bounding Box or Polygon

Looks like you're pretty close. I added the missing step of turning the two diagonal points from xlim & ylim into an sf object, and using their bounding box to subset the rivers.

library(sf)

# Download river shapefile here: https://www.weather.gov/gis/Rivers

# Import river data as SF
my_rivers <- st_read(dsn = '/home/x/Downloads/rivers/', layer = 'rv16my07') 
#> Reading layer `rv16my07' from data source `/home/x/Downloads/rivers' using driver `ESRI Shapefile'
#> Simple feature collection with 61122 features and 17 fields
#> Geometry type: MULTILINESTRING
#> Dimension:     XY
#> Bounding box:  xmin: -124.7068 ymin: 25.83636 xmax: -67.11324 ymax: 52.80121
#> Geodetic CRS:  NAD83

# The rivers data comes with a crs, so this step wasn't needed.
# Add a common CRS to the river dataset
#st_crs(my_rivers) <-  CRS('+proj=longlat')

# Set x and y limits for the plot, then make the points an sf object,
# set the crs as the same for my_rivers
ylims <- c(30.2, 31.4)
xlims <- c(-88.3, -87)
box_coords <- tibble(x = xlims, y = ylims) %>% 
  st_as_sf(coords = c("x", "y")) %>% 
  st_set_crs(st_crs(my_rivers))

#get the bounding box of the two x & y coordintates, make sfc
bounding_box <- st_bbox(box_coords) %>% st_as_sfc()

river_subset <- st_intersection(my_rivers, bounding_box)

head(river_subset)
#> Simple feature collection with 6 features and 17 fields
#> Geometry type: LINESTRING
#> Dimension:     XY
#> Bounding box:  xmin: -87.07318 ymin: 30.60249 xmax: -87 ymax: 30.83638
#> Geodetic CRS:  NAD83
#>      IHABBSRF_I          RR     HUC TYPE PMILE                  PNAME OWNAME
#> 8616       8616 03140104001 3140104    T   0.0           BLACKWATER R      0
#> 8617       8617 03140104002 3140104    R   0.5           BLACKWATER R      0
#> 8618       8618 03140104003 3140104    R   4.2           BLACKWATER R      0
#> 8630       8630 03140104015 3140104    R   7.8       BIG COLDWATER CR      0
#> 8631       8631 03140104016 3140104    R  17.3 BIG COLDWATER CR  E FK      0
#> 8634       8634 03140104019 3140104    R  17.3 BIG COLDWATER CR  W FK      0
#>           PNMCD OWNMCD       DSRR   DSHUC USDIR LEV J TERMID TRMBLV K
#> 8616 3140104001   <NA> 3140105007 3140105     R   1 0    205      1 0
#> 8617 3140104001   <NA> 3140104001 3140104     R   1 1    205      1 0
#> 8618 3140104001   <NA> 3140104002 3140104     R   1 1    205      1 0
#> 8630 3140104007   <NA> 3140104003 3140104     R   2 1    205      1 0
#> 8631 3140104008   <NA> 3140104015 3140104     R   2 2    205      1 0
#> 8634 3140104010   <NA> 3140104015 3140104     L   3 2    205      1 0
#>                            geometry
#> 8616 LINESTRING (-87.02298 30.60...
#> 8617 LINESTRING (-87.02928 30.60...
#> 8618 LINESTRING (-87.00626 30.64...
#> 8630 LINESTRING (-87 30.76254, -...
#> 8631 LINESTRING (-87.02458 30.78...
#> 8634 LINESTRING (-87.02458 30.78...

#Plot
ggplot() + 
  geom_sf(data = river_subset) + 
  geom_sf(data = bounding_box, fill = NA, color = 'red')

^{Created on 2022-03-30 by the reprex package (v2.0.1)}

Subset class sfc_LINESTRING & sfc objects within a bbox

Here's an example using an sf object from tigris, just for reproducibility. I'm using towns in New Haven County, Connecticut, plotting it the way it comes in. Then I crop it to a bounding box I made up, using st_crop, which I believe was added fairly recently to sf. If I had the bbox as a shape, instead of a vector of coordinates, I could have used st_intersection.

I don't have a linestring object handy, but I'd assume it works the same way.

library(tidyverse)
library(sf)

# selecting just to limit the amount of data in my sf
ct_sf <- tigris::county_subdivisions(state = "09", county = "09", cb = T, class = "sf") %>%
  select(NAME, geometry)

plot(ct_sf)

crop_bbox <- c(xmin = -73, ymin = 41.2, xmax = -72.7, ymax = 41.5)
ct_cropped <- st_crop(ct_sf, crop_bbox)

plot(ct_cropped)

---

Related Topics