30 Day Map Challenge Day 3 and 4

Note: for those interested in the code used to generate each visualization, you can see it by clicking the “Show/Hide All Code” button at the top right of the page or the “Code” button above and to the right of each visualization.

Day 3 - Polygons

Mobile network cells can be modeled using Voronoi Diagrams. Typically, this is done for mobility analysis with cellular signaling data, as shown in this paper.

library(tidyverse)
library(sf)
library(httr)
library(ggvoronoi)
library(ggmap)

ign_geoserver_base_url <- "https://wms.ign.gob.ar/geoserver/ows"

ign_geoserver_query <- list(service="wfs",
                            version="1.1.0",
                            request="GetFeature",
                            typeNames="ign:departamento",
                            cql_filter="in1='02007'")

ign_geoserver_url <- modify_url(url = ign_geoserver_base_url,
                                query = ign_geoserver_query)

comuna1_sf <- read_sf(ign_geoserver_url)

comuna1_centroid <- st_centroid(comuna1_sf)
comuna1_bbox <- st_bbox(comuna1_sf)

idecom_geoserver_base_url <- "https://www.idecom.gob.ar/geoserver/ows"

idecom_geoserver_query <- list(service="wfs",
                               version="1.3.0",
                               request="GetFeature",
                               typeNames="publico:Antenas103-NO4G-5-3",
                               cql_filter=paste0("BBOX(the_geom,", paste(comuna1_bbox[c(2,1,4,3)], collapse = ","), ")", " AND Banda='7'"))

idecom_geoserver_url <- modify_url(url=idecom_geoserver_base_url, 
                                   query = idecom_geoserver_query)

radiobases_4g_comuna1 <- read_sf(idecom_geoserver_url)

radiobases_4g_comuna1 <- radiobases_4g_comuna1 %>% 
  as_Spatial() %>% 
  sp::remove.duplicates() %>% 
  st_as_sf() %>% 
  na.omit()

register_google(Sys.getenv("GOOGLE_MAPS_API_KEY"))
comuna1_map <- get_googlemap(center = c(lon=-58.37869433449463, 
                                        lat=-34.606838945727), 
                             zoom = 15)

ggmap(comuna1_map, 
      base_layer = ggplot(data = na.omit(st_drop_geometry(radiobases_4g_comuna1)), 
                          aes(Longitud, Latitud))) +
  geom_path(stat = "voronoi", alpha = 0.6, size = 0.4) +
  geom_point(color = "blue", size = 0.7) +
  theme_minimal() +
  theme(axis.text = element_blank(),
        axis.title = element_blank())

Day 4 - Hexagons

Hexagons can also be used to plot 4G radio bases, using color intensity to show their density.

library(hexbin)

ign_geoserver_base_url <- "https://wms.ign.gob.ar/geoserver/ows"

ign_geoserver_query <- list(service="wfs",
                            version="1.1.0",
                            request="GetFeature",
                            typeNames="ign:provincia",
                            CQL_FILTER="in1='02'")

ign_geoserver_url <- modify_url(url = ign_geoserver_base_url,
                                query = ign_geoserver_query)
caba <- read_sf(ign_geoserver_url)

caba_bb <- st_bbox(caba)

idecom_geoserver_base_url <- "https://www.idecom.gob.ar/geoserver/ows"

idecom_geoserver_query <- list(service="wfs",
                               version="1.3.0",
                               request="GetFeature",
                               typeNames="publico:Antenas103-NO4G-5-3",
                               cql_filter=paste0("BBOX(the_geom,", paste(caba_bb[c(2,1,4,3)], collapse = ","), ")"))

idecom_geoserver_url <- modify_url(url=idecom_geoserver_base_url, 
                                   query = idecom_geoserver_query)

radiobases_4g_caba <- read_sf(idecom_geoserver_url)

if(st_crs(caba)!=st_crs(radiobases_4g_caba)) {
  radiobases_4g_caba <- st_transform(radiobases_4g_caba, crs = st_crs(caba))
}
radiobases_4g_caba <- st_join(radiobases_4g_caba, caba, left=FALSE)

caba %>% 
  ggplot() +
  geom_sf() +
  geom_hex(data = radiobases_4g_caba, aes(Longitud, Latitud)) +
  theme_bw() +
  coord_sf(datum = NA) +
  scale_fill_viridis_c(direction = -1, name = "Radiobases 4G") +
  theme(axis.title.x = element_blank(),
        axis.title.y = element_blank())