使用 geom_sf() 繪制地理數據

Question

我正在嘗試學習（或更具體地說，記住並學習實施）R 中的空間數據分析。我在處理地理數據方面遇到了一些基本困難，這妨礙了任何方法的學習。

切入正題，當我試圖產生 map 的財富時，我得到了一些明顯奇怪的東西：財富和填充之間沒有美學聯系，許多國家缺失。 我希望由於部分完整的數據集而丟失一些，但不會像這個那么多。 其次，所有具有匹配值的填充的國家似乎都具有相同的值——見下文。

我怎么到這里了？ 我有一個df數據集，其中包含國家財富數據（來自瑞士信貸）和一些關於 gdp 增長的數據，我在下面有dput() ：

我下載並解壓縮了世界行政數據的 shapefile：

download.file(
  url = "https://public.opendatasoft.com/api/explore/v2.1/catalog/datasets/world-administrative-boundaries/exports/shp?lang=en&timezone=Europe%2FLondon",
  destfile = "world_administrative_boundaries.zip")
 
unzip(
  zipfile = "world_administrative_boundaries.zip",
  exdir = "world_administrative_boundaries"
  )

這很好用。 順便說一句，我將感謝國家和（對於英國）區域和更低級別的空間數據來源的建議，以及用於連接數據的 R 包的建議。 但無論如何...

首先我加載了一些包：

library(tidyverse)
library(sf)

然后我將形狀文件讀入 R，並將數據集與我的財富數據合並。

world_adm0_sf <-st_read("world_administrative_boundaries/world-administrative-boundaries.shp")

world_adm0_sf <- inner_join(world_adm0_sf, df, by = "iso3")

那應該給我一個部分數據集——但不像看起來繪制的那樣部分。 然后我使用 ggplot2：

ggplot(data = world_adm0_sf, aes(fill=wealth)) +
    geom_sf()+
    scale_fill_steps() +
    theme_light()
    
ggsave(filename = "Wealth.png", 
     device = "png",
     width = 10,
     height = 4)

這會產生上面的 plot。

這都是初步的。 我的真正目標是創建一個鄰接矩陣，這樣我就可以在數據上運行一些空間模型。 我想我可以這樣做：

library(spdep)
w_world_adm0  <- poly2nb(world_adm0_sf) %>% nb2mat(style = "W")

但這會出現以下錯誤： Error in poly2nb(world_adm0_sf): Empty geometries found 。 我只能想象這與繪圖問題有關。

非常感謝任何幫助。

structure(list(iso3 = c("USA", "CHN", "JPN", "DEU", "GBR", "FRA", 
"IND", "CAN", "ITA", "AUS", "ESP", "NLD", "CHE", "MEX", "BEL", 
"IDN", "BRA", "SWE", "SAU", "DNK", "AUT", "SGP", "NZL", "ISR", 
"POL", "NOR", "THA", "PRT", "BGD", "ARE", "PHL", "VNM", "ZAF", 
"IRL", "GRC", "FIN", "PAK", "CHL", "UKR", "ROU", "COL", "MYS", 
"HUN", "KWT", "KAZ", "PER", "QAT", "LKA", "MAR", "KEN", "LBN", 
"LUX", "ARG", "DZA", "BGR", "ECU", "ETH", "HRV", "BLR", "JOR", 
"SVN", "SRB", "TUN", "CRI", "SLV", "OMN", "URY", "PAN", "LTU", 
"BHR", "LVA", "ISL", "AZE", "CYP", "BOL", "BIH", "EST", "ALB", 
"NPL", "KHM", "MUS", "COD", "MLT", "NIC", "TTO", "ARM", "GEO", 
"CMR", "JAM", "LBY", "SEN", "MNE", "NAM", "MDG", "BWA", "ZMB", 
"TJK", "MLI", "GAB", "MWI", "GNQ", "MOZ", "BRB", "MNG", "LBR", 
"FJI", "TCD", "GUY", "ERI", "COG", "SYC", "SLE", "BLZ", "COM", 
"DJI", "CAF", "LSO", "GNB"), wealth = c(126300.403036252, 73939.8891311991, 
27260.570352125, 17389.799264483, 15675.9889849248, 16282.4392222225, 
12700.3877756376, 10585.7395730159, 12176.0329952741, 9268.27874790906, 
8481.28808374314, 5436.51408079177, 4672.6030405074, 3774.9451374071, 
3543.77178519827, 3086.57900830581, 3034.93997199139, 2741.53111378823, 
1905.26613759989, 1782.18132955766, 1855.14333493139, 1636.83125031141, 
1288.20550646464, 1307.54537310394, 1464.60908917529, 1345.59016547028, 
1392.90190541805, 1337.19890038495, 898.958801638123, 926.368205722374, 
933.191002686083, 874.858045943263, 903.040059653359, 882.20086484466, 
962.807508534614, 836.536239906252, 663.532991342168, 737.397506705501, 
517.376031972012, 584.466282600888, 586.072886162324, 565.720706897012, 
450.495646279881, 413.524330775955, 466.674062703931, 427.957436973251, 
347.37129513471, 368.206981773813, 338.261747852825, 328.180531919524, 
307.493974123627, 305.899138540934, 260.992240660559, 254.245993196906, 
215.860597024656, 207.22935523629, 221.394916106677, 205.419322153595, 
164.189635248546, 184.189116509732, 180.001925261347, 161.244568189131, 
160.306734507052, 165.337454268781, 145.171877463561, 144.726277337691, 
150.407054934631, 129.783182270863, 131.132956518345, 114.623545266656, 
114.486922170596, 98.9661166555199, 98.3003532984958, 101.44441788665, 
87.1461534054738, 74.9746836821397, 77.795657106148, 66.1431475890217, 
71.2009574127775, 55.4559618547982, 59.5703711670758, 48.0465708689348, 
57.4375916937781, 51.4806937778165, 46.9537143621449, 44.8053089422735, 
43.2099298782149, 40.5284167581835, 41.1425896096207, 70.5563794908187, 
39.5376722680099, 31.4416134859147, 25.4319082241779, 28.7033124617245, 
23.5162145038624, 21.8465824967143, 21.5778619134227, 20.8494437096404, 
17.2032811219951, 18.5250206627168, 14.6881425292255, 13.2498106184224, 
15.5574142939735, 12.7783060223608, 11.2938871435042, 9.82612644104218, 
8.61312574821742, 6.08951482556896, 5.04939300180385, 4.01486292583383, 
3.2236344186979, 2.88763002016324, 2.54563458904526, 2.47536481004851, 
1.96914002009599, 1.93613437718668, 1.90500798026385, 1.64313029137334
), gdp_growth = c(5.94548476122172, 8.1097925807738, 1.65707111732294, 
2.62698726672524, 7.52491037367686, 6.81658913649873, 8.68122873083119, 
4.54088730942124, 6.73727712298924, 2.23621243944461, 5.5195949732682, 
4.86321914933848, 4.22171657654073, 4.72056529768081, 6.13386596356445, 
3.69124011191289, 4.61942162067312, 5.07508472544021, 3.24092976848051, 
4.85570758570233, 4.55685091308877, 7.61396261502647, 3.71401011995567, 
8.61199034846989, 6.84771336241919, 3.88093603636767, 1.53383558441951, 
5.482391708458, 6.93867508910159, 3.91629608037678, 5.70315596840621, 
2.56155114232494, 4.9130966818804, 13.5882471079549, 8.43442550167381, 
2.97322935640895, 6.48708677384843, 11.6682247016219, 3.40000000000001, 
5.10012819668945, 10.6770130738261, 3.09215964982188, 7.11731531479809, 
NA, 4.29999999999981, 13.3495090777977, 1.59075695426887, 3.32676381781447, 
7.9296679419552, 7.51735538703886, -7.0012393603983, 5.10245544656804, 
10.3982494646904, 3.50000000049758, 7.63460971569079, 4.23525124172572, 
5.63730300819225, 13.0722049215979, 2.29987941172813, 2.2281384129623, 
8.2110625918128, 7.54991253537386, 4.32277635659793, 7.57956529907922, 
10.2784761168053, 3.09227668737446, 4.36700931644991, 15.3358739290709, 
5.98103211278502, 2.22535124229128, 4.06807386332741, 4.40304478596893, 
5.60003727280338, 6.63892696813528, 6.1058042252093, 7.54559112979621, 
8.01346292400842, 8.51550112324915, 4.24694014145481, 3.02638936277746, 
3.70361112396466, 6.20015401075946, 10.2977635999018, 10.3443407049189, 
-1.02654016223218, 5.69999999999997, 10.4655372923886, 3.64991688648905, 
4.60000000000001, 31.3724222267248, 6.06449596752137, 12.4343589851741, 
2.65605081999696, 4.40234000007614, 11.3700218626674, 4.59873403878439, 
9.2001904308498, 3.05261356942125, 1.462286510102, 2.75141640762955, 
-0.945382916517076, 2.36398841807251, -0.189272866293138, 1.63676957584143, 
4.98671283481617, -5.11160739321285, -1.1999905246095, 20.0600109474921, 
NA, -2.20000260638274, 7.8618824199665, 4.10470008884118, 15.2261306532663, 
2.11219557695527, 4.81247738295502, 0.900000006878756, 1.34762315508203, 
3.80000242656607)), row.names = c(NA, -118L), class = c("tbl_df", 
"tbl", "data.frame"))

Answer 1

部分問題是您使用的行政邊界 shapefile 不完整。 如果你在內部連接之前單獨輸入 plot，你可以看到這個：

library(tidyverse)
library(sf)

world_adm0_sf <- st_read(paste0("world_administrative_boundaries/",
                                "world-administrative-boundaries.shp"))

ggplot(data = world_adm0_sf) +
  geom_sf() +
  theme_light()

相反，我們可以使用方便的 rnaturalearth package 來獲取世界上所有國家的 shapefile，包括它們的 ISO 代碼：

library(rnaturalearth)

ne_sf <- ne_countries(returnclass = "sf")

ggplot(data = ne_sf) +
  geom_sf() +
  theme_light()

我們可以在這個 map 的頂部做我們的加入和 plot 數據，這樣沒有數據可用的國家仍然可見。 我們應該選擇一個差異很大的色標來增加不同國家財富之間的對比：

world_adm0_sf <- inner_join(ne_sf, df, by = c(iso_a3 = "iso3"))

ggplot(data = world_adm0_sf) +
  geom_sf(data = ne_sf) +
  geom_sf(aes(fill = wealth)) +
  scale_fill_distiller(palette = "Spectral") +
  theme_light()