Geography Reference
In-Depth Information
COPY ne_cities (gid, scalerank, name, namealt, geom)
FROM stdin;
table has the following fields:
geonameid; name; asciiname; alternatenames (alternatenames, comma separated
varchar(5000)); latitude: latitude in decimal degrees (wgs84); longitude: longitude
in decimal degrees (wgs84); feature class (
OpenStreetMap data
;
Natural Earth data
;
GEONAMES
); feature code (
OpenStreetMap data
;
Natural Earth data
;
GEONAMES
); country code; admin1; admin2; admin3; admin4; population; ele-
vation; dem: digital elevation model; srtm3; timezone: the timezone id; modifi-
cation date: date of last modification in yyyy-MM-dd format. The collated dataset
for maintaining ranking related experiments has been developed using GeoNode
and GeoServer which depicts the names attributes of various locations across the
globe. At a certain zoom level there are predetermined attributes that are called
from the database table according to the rank specified.
Columns of Geonames data:
The main
'
geoname
'
Test Results for the Above Approach for Czech Republic
The columns updated in planet_osm_point table after the dataset joining is shown
below (Fig.
2
). Before joining, the osm_scalerank column for the below places were
0 (zero). We see that scale ranking of all the places for CZ present in NE dataset
have been propagated correctly to OSM dataset. Hence the 12 cities that exist for
CZ as common between NE, OSM, GeoNames are found to be correctly matched
yielding an accuracy of 95-100 %. The geometry, name and geo-id complete and
fuzzy searching and matching around a buffer of 50 km took a minimum of 30 s to
maximum 1 min in a commodity computer with 2 GHz, 2 GB memory. The query
run to provide the part screenshot in Fig.
2
.
select st_astext(way) as geom, name, place, population,
osm_scalerank from planet_osm_point where osm_scalerank !
¼
0
order by name;
The results accomplished are:
Linked together open-source geospatial datasets Natural Earth, Geonames and
Open Street Map.
Joined city rankings from Natural Earth into OSM+Geonames data with fuzzy
matching.
Linked datasets, starting from populated places, integrated attributes such as
administrative class, population, and also how the name is spelt, local form and
English form and so on. Sample datasets used are from Czech country dataset.
