Import CAR UNICEF WASH

From OpenStreetMap Wiki
Jump to navigation Jump to search

Goals

This is the third import for Central African Republic done as a part of the EUROSHA project. Its goal is to include the UNICEF data for WASH for half of Central African Republic into OSM.

Schedule

  1. Preparation, discussion - April 2013
  2. Import - expected start in April 2013

Import Data

Data description

The data has been collected on the field by UNICEF during various surveys. The dataset consists of 1937 nodes for 6 regions (prefectures) of CAR: Mambere-Kadei (173 nodes), Nana-Gribizi (252 nodes), Nana-Mambere (206 nodes), Ouaka (195 nodes), Ouham (565 nodes) and Ouham-Pende (546 nodes).

The dataset contains following field attributes: ID, WPT, CODE, WASH_CODE, LAT, LONG, PAYS, PREF, S_PREF, COMMUNE, VILLE_VILL, QUARTIER, POP, TYPE_POINT, TYPE_POINT_SHORT, FONCTIONNE, COMITE_GES, RESID_ART_, LOCALISATI, ANNEE_INST, ANNEE_INST, DERNIERE_M, COMMENTAIR

Background

ODbL Compliance verified: YES
The authorization given by UNICEF CAR was approved by the OSMF License Working Group as ODbL compatible.

Import Type

This is a import done manually in JOSM, through a specific Tasking Manager job, according to a worflow document.

Data Preparation

Data Reduction & Simplification

The data will be reduced only to the keys listed below.

Tagging Plans

Considering the accuracy of the location is sometimes coarse and the objects have to be moved, a fixme will be created with the UNICEF pair of coordinates as values so that the contributors can have a track of the original location.

In the spreadsheet below, No values are not counted within the number of objects. The other fields are considered as not relevant and so will not be tagged.

Unicef Key Unicef value (number) OSM tag
all objects drinking_water=yes
ID not tagged because not relevant
WPT not tagged because not relevant
CODE not tagged because not relevant
WASH_CODE not tagged because not relevant
LAT_DD not tagged because not relevant
LONG_DD not tagged because not relevant
PAYS not tagged because not relevant
PREF, SOUS_PREF, COMM, VILLE_VILL, QUARTIER addrːfull=*
VILLE_VILL addr:city=*
TYPE_POINT Forage (1681) man_made=water_well, pump=yes
TYPE_POINT Puits moderne (101) man_made=water_well, water_well:structure=modern, pump=no
TYPE_POINT Puits traditionnel (96) man_made=water_well, water_well:structure=traditional, pump=no
TYPE_POINT Reseau moderne (33) amenity=drinking_water
TYPE_POINT Source amenagee (21) natural=spring
TYPE_POINT Source non amenagee (11) natural=spring
FONCTIONNE Oui (1641) operational_status=open
FONCTIONNE Non (298) operational_status=closed
COMITE_GES Oui (1623) operator=yes
COMITE_GES Non (316) operator=no

Changeset Tags

We will use the following changeset tags.

Data Transformation

Data were received as a XLS sheet. Reduction, simplification, manual cleaning and tag transformation will be done directly in a copy of the XLS file. The two fields of coordinates will be renamed X and Y. Then the file will be converted into CSV and divided into 6 files based on the region. The 6 CSV files will be opened with JOSM as CSV files with the opendata plugin. Finally the files will be saved in the OSM format and ready for distribution/import.

Data Merge Workflow

Team Approach

The importation team consists of 5 EUROSHA volunteers: FRosenkranc, LenkaP, fedebasa, Jorieke V and MorganeG. The preparation is done by SeverinGeo.

References

The import will be discussed in the import list.

Workflow

Please refer to the workflow page.

Reverse plan

In case of any trouble, JOSM reverter will be used.

Conflation

Unicef dataset has 1937 nodes out of which all should be imported. In the 6 regions of CAR where import will be happening, the only WASH facilities mapped are those already mapped by Eurosha volunteers in Bangui and cities taken by the rebels in December 2012. These cities are: Mbrès (region Nana-Grěbizi), Kaga-Bandoro(region Nana-Grěbizi), Batangafo (region Ouham) and Kabo (region Ouham). Only data in these cities will be manually merged.