Summary

The main finding of this report is the following:

  1. The top five cities that have the most mosques are:
  1. Abidjan, Cote d’lvoire (220)
  2. Omdurman, Sudan (135)
  3. Khartoum, Sudan (134)
  4. Khartoum Bahri, Sudan (127)
  5. Vina, Cameroon (126)

Among these cities, (ii) - (iv) are very close together around the Khartoum metropolitan area. I suggest we do the pilot study in Khartoum metropolitan area.

  1. The new selection criteria filtered out many more places, leaving 5389 mosques for id purposes (including Western Africa and Libya). The selection process looked at the name column and retained all rows whose name is a variant of the word “mosque” in many languages.

Analysis

Use a more restrictive criteria to select mosques in OSM data. The only other column that is instrumental to our selection process is the name column. I first summarized all phrases relevant to “mosque” in the dataset and keep only data whose name contains an variant of the phrase “mosque” (accents and capitalization are accounted for already)

Of the 63808 place of worship, 14859 are coded as “Muslim”. I then subset these data to keep the locations whose name contain any of the following phrases:

  1. “mosquee”, “mosque”, “mesquita”, “centre islamique”,
  2. “مسجد”, “مصلى”,“جامع”,
  3. “masjid”, “masdjid”, “moschee”, “mosqee”, “moquee”, “masjad”

The first group includes deaccented phrases in French, English, and Portuguese. The second group includes mosques of different sort in Arabic. The third group catches misspelled variants in the dataset.

Libya and Western Africa are filtered out due to a lack of administrative district data

After going through these filters, 4287 mosques remain.

I then count how many mosques are in each city (which is defined here as the 2nd administrative level, under country(0th) and province(1st)

The five districts that have the most mosques are:

  1. Abidjan, Cote d’lvoire (220)
  2. Omdurman, Sudan (135)
  3. Khartoum, Sudan (134)
  4. Khartoum Bahri, Sudan (127)
  5. Vina, Cameroon (126)

The data is displayed below. I also included an visualization that plot all the mosques and the number of mosques aggregated to city level. Names of the top 10 cities are plotted in the graph.

(df_top_10 <- df_num_mosque %>% arrange(desc(num_msq)) %>% head(10))
## Simple feature collection with 10 features and 9 fields
## Geometry type: MULTIPOLYGON
## Dimension:     XY
## Bounding box:  xmin: -16.61846 ymin: 5.227361 xmax: 32.67232 ymax: 31.35986
## Geodetic CRS:  WGS 84
##       UID num_msq cntry_c       cntry_n    GID_1           prvnc_n      GID_2
## 1  135048     220     CIV Côte d'Ivoire  CIV.1_1           Abidjan  CIV.1.1_1
## 2  129860     135     SDN         Sudan  SDN.7_1          Khartoum  SDN.7.4_1
## 3  129936     134     SDN         Sudan  SDN.7_1          Khartoum  SDN.7.3_1
## 4  129897     127     SDN         Sudan  SDN.7_1          Khartoum  SDN.7.2_1
## 5  134164     126     CMR      Cameroon  CMR.1_1          Adamaoua  CMR.1.5_1
## 6  133878      73     CIV Côte d'Ivoire CIV.11_1 Vallée du Bandama CIV.11.1_1
## 7  131024      63     NER         Niger  NER.5_1            Niamey  NER.5.1_1
## 8  129940      44     SDN         Sudan  SDN.7_1          Khartoum  SDN.7.7_1
## 9  128028      42     EGY         Egypt  EGY.6_1   Al Iskandariyah  EGY.6.7_1
## 10 130072      39     SEN       Senegal SEN.13_1             Thiès SEN.13.3_1
##           city_nm     osm_id                       geometry
## 1         Abidjan  768587526 MULTIPOLYGON (((-3.810719 5...
## 2        Omdurman 1868388768 MULTIPOLYGON (((32.48051 15...
## 3        Khartoum   49326464 MULTIPOLYGON (((32.64031 15...
## 4  Khartoum Bahri 1784074017 MULTIPOLYGON (((32.66424 15...
## 5            Vina 5934688932 MULTIPOLYGON (((13.77939 7....
## 6           Gbeke 4434492409 MULTIPOLYGON (((-4.888226 7...
## 7          Niamey 3156009709 MULTIPOLYGON (((2.233916 13...
## 8        Um Badda   49219852 MULTIPOLYGON (((32.37406 15...
## 9     Al-Muntazah 1449499523 MULTIPOLYGON (((30.10653 31...
## 10      Tivaouane 4732867160 MULTIPOLYGON (((-16.49635 1...
tmap_mode("view")
# plot each mosques and total number of mosques aggregated up to each city level
tm_shape(df_num_mosque) + 
tm_sf(col="num_msq", id="city_nm", alpha=0.4) + 
tm_shape(df_top_10) + 
tm_text("city_nm", col="green", size=1.5)  + 
tm_shape(sf_mosque) + 
tm_dots(size=0.005, col="blue", id="name")