GeoNames Forum

There are too many of these cities in Brazil named Almas. Is there a way to check if any of these are real and not just errors? This particular set if from Dec 31, 2008.

DSG CC1 ADM1 FULL_NAME_ND DMS_LAT DMS_LONG REGION JOG
PPL BR 05 Almas -123800 -450600 Bahia SD23-02
PPL BR 05 Almas -101800 -412400 Bahia SC24-09
PPL BR 05 Almas -90600 -401100 Bahia SC24-06
PPL BR 05 Almas -120900 -391300 Bahia SD24-02
PPL BR 05 Almas -105100 -421500 Bahia SC23-12
PPL BR 20 Almas (2) -60700 -411300 Piaui SB24-09
PPL BR 20 Almas -33445 -420139 Piaui SA23-16
PPL BR 20 Almas (1) -60100 -414000 Piaui SB24-09
PPL BR 20 Almas -61700 -410600 Piaui SB24-09
PPL BR 30 Almas -75100 -375300 Pernambuco SB24-15
PPL BR 06 Almas -25342 -411633 Ceara SA24-09
PPL BR 06 Almas -50900 -405400 Ceara SB24-05
PPL BR 13 Almas -34649 -424945 Maranhão SA23-16
PPL BR 06 Almas -52800 -402100 Ceara SB24-06
PPL BR 06 Almas -55800 -403400 Ceara SB24-05
PPL BR 15 Almas -190900 -443800 Minas Gerais SE23-15
PPL BR 17 Almas -63900 -374100 Paraiba SB24-11
PPL BR 31 Almas -113734 -470606 Tocantins SC23-13

Just because a name occurs many time in a country does not necessarily mean the places don't exist or are duplicates, in the first place it means it is a very popular place name in a country.

Marc

I would be the first in line to NOT dispute the fact that there can be multiple singularities in an area with the same name. In general, what I have noted most about duplicate names for a given country is that they generally carry different ADM1 signatures or they are within a few miles of each other which to me indicates that it may be a large city with perhaps more than one administrative district within a city. Admittedly this is speculative on my part and a result of working with so many cities over a long period of time.

On the other hand, when there is more than one point within a given region, the wide diversity in longitude and latitude generally does not co-exist. Take Bahia as an example:
dms_lat dms_long
-123800 -450600
-120900 -391300
-101800 -412400
-90600 -401100
-105100 -421500

The city looks to cover a latitude form 9 S to almost 13 S - seems suspect is all.

And the one in Ceara also seems to have a wide diversity also:
dms_lat dms_long
-55800 -403400
-52800 -402100
-25342 -411633
-50900 -405400

If one of these truly stands out as erroneous it would be the one at 2S53'42. Seems like most of these fall in the 5 S latitude area, but for the one.

Is that a separate Almas in Ceara? I added the second administrative district to the query to see if it actually had one and could not find a solution there. While I can identify it by longitude and latitude as a spatially geometric point on a map, it is difficult to indicate whether the city belongs to Almas, Ceara or to Almas, Ceara, if you get my point. How do they keep track of this city in Brazil?

The combination of cityname and admin region is not unique. You have to find another way to display it to your users. For instance you can reverse geocode the ambiguous citynames and add the nearest larger city to the name so that your users know which one is meant. This is also the way a human would disambiguate place names.

Best

Marc