Last active
December 7, 2017 14:31
-
-
Save jsta/e486f337be6d5bcdb3aeb1335959de52 to your computer and use it in GitHub Desktop.
Counting the number of GLEON lake sites that resolve to a Wikipedia page
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # GLEON lakes according to Wikipedia | |
| library(rvest) | |
| library(wikilake) | |
| library(sf) | |
| library(maps) | |
| # get data #### | |
| gleon_lakes <- read_html("http://gleon.org/lakes") | |
| gleon_lakes <- html_nodes(gleon_lakes, ".views-field") | |
| gleon_lakes <- html_text(gleon_lakes) | |
| gleon_lake_names <- gleon_lakes[seq(1, length(gleon_lakes), by = 3)] | |
| gleon_lake_names <- gsub("\n", "", gleon_lake_names) | |
| gleon_lake_names <- trimws(gleon_lake_names)[-1] | |
| res <- lapply(gleon_lake_names, function(x) tryCatch(wikilake::lake_wiki(x), error = function(err){NA})) | |
| # clean missing lakes #### | |
| res_clean <- res[unlist(lapply(res, function(x) !is.null(x)))] | |
| res_clean <- res_clean[unlist(lapply(res_clean, function(x) length(x) > 1))] | |
| res_clean <- res_clean[unlist(lapply(res_clean, function(x) !is.na(x[,"Lat"])))] | |
| length(res_clean) / length(gleon_lake_names) # proportion of gleon lakes that resolve to wikipedia pages | |
| ##### collapse list to data.frame | |
| res_df_names <- unique(unlist(lapply(res_clean, names))) | |
| res_df <- data.frame(matrix(NA, nrow = length(res_clean), | |
| ncol = length(res_df_names))) | |
| names(res_df) <- res_df_names | |
| for(i in seq_len(length(res_clean))){ | |
| # print(i) # debugging | |
| dt_pad <- data.frame(matrix(NA, nrow = 1, | |
| ncol = length(res_df_names) - ncol(res_clean[[i]])), | |
| stringsAsFactors = FALSE) | |
| names(dt_pad) <- res_df_names[!(res_df_names %in% names(res_clean[[i]]))] | |
| dt <- cbind(res_clean[[i]], dt_pad) | |
| dt <- dt[,res_df_names] | |
| res_df[i,] <- dt | |
| } | |
| # Keep only common columns ##### | |
| good_cols <- data.frame(as.numeric(as.character(apply(milakes, | |
| 2, function(x) sum(!is.na(x)))))) | |
| good_cols <- cbind(good_cols, names(milakes)) | |
| good_cols <- good_cols[good_cols[,1] > 20 ,2] | |
| res_final <- res_df[,good_cols] | |
| # saveRDS(res_final, "gleon-lakes_wikipedia.rds") | |
| res_final <- readRDS("gleon-lakes_wikipedia.rds") | |
| world <- sf::st_as_sf(map("world", plot = FALSE, fill = TRUE)) | |
| res_sf <- st_as_sf(res_final, coords = c("Lon", "Lat"), crs = 4326) | |
| plot(world$geometry) | |
| plot(res_sf$geometry, add = TRUE, pch = 21, col = "red") | |
| nrow(res_sf) / length(gleon_lake_names) # proportion found on Wikipedia | |
| knitr::kable(gleon_lake_names[!(gleon_lake_names %in% res_sf$Name)]) # missing lakes | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
According to this analysis only 38.5 % of GLEON lakes have a Wikipedia page.
Ignoring the fact that some missing information is due to simple naming differences, here are the missing lakes: