Dillant Hopkins Airport lat and long are switched and lat * -1

[hlendway]

I was looking at the airports data and noticed that Dillant Hopkins Airport was at 72.27, i.e. the furthest north. I knew there were airports in Alaska to this seemed incorrect. The lat and lon in this file show lat of 42.89, long of -72.27, which is correct:

It is shown incorrect in the csv:

and also incorrect when you load in R:

I didn't have time to look into where/why this is happening but wanted to report, Vermont is not the furthest north airport :-)

[jwhendy]

The plot thickens:

I can confirm the Dillant Hopkins oddity. It should be at (lat, lon) = ~(43, -72).

> airports %>% filter(name=="Dillant Hopkins Airport")
  faa   name                      lat   lon   alt    tz dst   tzone
1 EEN   Dillant Hopkins Airport  72.3  42.9   149    -5 A     NA   

But picking the first row shows that Landsdowne is actually correct.

> airports %>% filter(name=="Lansdowne Airport")
  faa   name                lat   lon   alt    tz dst   tzone           
1 04G   Lansdowne Airport  41.1 -80.6  1044    -5 A     America/New_York

It's odd that these are both pretty close. I would have suspected some E/W reversal.

[jwhendy]

Which are off from @hlendway 's raw data?

library(tidyverse)
library(nycflights13)
data(airports)

raw <- read.csv("https://raw.githubusercontent.com/jpatokal/openflights/master/data/airports.dat", header=F)
raw_sub <- raw %>% select(c(2, 7, 8))
names(raw_sub) <- c("name", "lat", "lon")

aports_sub <- airports %>% select(name, lat, lon)
data_merged <- merge(raw_sub, aports_sub, by="name")
data_merged <- data_merged %>% transform(lat_delta = lat.x-lat.y, lon_delta = lon.x-lon.y)
data_merged %>% filter(abs(lat_delta) > 1 | abs(lon_delta) > 1)

Results:

                           name    lat.x     lon.x    lat.y      lon.y
1          Capital City Airport 38.18250  -84.9047 40.21714  -76.85136
2          Capital City Airport 40.21710  -76.8515 38.18197  -84.90614
3          Capital City Airport 42.77870  -84.5874 38.18197  -84.90614
4          Capital City Airport 42.77870  -84.5874 40.21714  -76.85136
5       Dillant Hopkins Airport 42.89840  -72.2708 72.27083   42.89833
6     Douglas Municipal Airport 31.34260 -109.5060 31.47674  -82.86057
7     Douglas Municipal Airport 31.47670  -82.8605 31.34260 -109.50645
8    Lawrence Municipal Airport 39.01120  -95.2166 42.71719  -71.12342
9                Linden Airport  5.96592  -58.2703 40.61745  -74.24459
10 Marshfield Municipal Airport 44.63690  -90.1893 42.09825  -70.67214
11   Plymouth Municipal Airport 41.90900  -70.7288 41.36513  -86.30026
12           San Carlos Airport  9.64772  -68.5747 37.51194 -122.24944
13        Shelby County Airport 39.41040  -88.8454 33.17778  -86.78322
14        Shelby County Airport 33.17700  -86.7828 39.41056  -88.84556
15           Tri Cities Airport 42.07850  -76.0963 46.26470 -119.11900
     lat_delta    lon_delta
1   -2.0346400   -8.0533401
2    2.0351260    8.0546374
3    4.5967296    0.3187366
4    2.5615629   -7.7360412
5  -29.3724336 -115.1691307
6   -0.1341386  -26.6454303
7    0.1340970   26.6459579
8   -3.7059944  -24.0931818
9  -34.6515272   15.9742922
10   2.5386519  -19.5171616
11   0.5438693   15.5714594
12 -27.8642237   53.6747446
13   6.2326226   -2.0621757
14  -6.2335579    2.0627572
15  -4.1862010   43.0227040

One could set the filter however makes sense to decide what is actually an error. I also don't have time at the moment to understand where these errors come from, but hopefully this is helpful in identifying suspect cases.