Data structure Mapping data What data you need to entry Exact - PDF document

Attributes and features Data structure

Mapping data

What data you need to entry • Exact location (case specific) • Aggregated data by administrative levels (aggregated)

Aggregated data are usually represented by polygons A field for the definition of the attribute. In other types of files this field is filled with the XY coord of each corner of the polygon

A row: a specific geographical area A column: a variable or field

Exact point representation Each point is represented by Specific XY coordinates

Databases have to be relational • You can produce your data from any database • To be respresented geographically this table needs a field in common to the tables of the map database (name, ID, etc) • In this way the tables can be related (joined) in one single table

Geographic representation Data collected Table for geographic representation

Relational database aggregated data datanorm.rec Disease.rec Levcode (unique) Levcode What What When .dbf Rate Age group Count Geographic. dbf Levcode Levcode (unique) Municipality Municipality (lowest level) Population Province Region Level.rec

Data related and represented

Spatially join point and polygon layers

Spatially join point and polygon layers

Spatially join point and polygon layers

Spatially join point and polygon layers

Respresenting specific location • You need to produce a new layer to be overposed on the map • To do that the layers need to be in the same geographical projection

Layer with same projections • WGS84…….this is the key-ward!!!!!! World Geodetic System

Two layers

Two layers

One for each layer Two different tables

Distribution of IDPs camps in Darfur, Sudan, 2004

Managing administrative boundaries

Managing administrative boundaries • Structure of database • Starting from the lowest administrative representation • Ensure that the database structure remains the same

Province District

Geoprocessing operation to easily manage administrative boudaries • Dissolving features (loose boundaries) • Merging features (preserve your boundaries) • Clipping one feature on another (cutting) • Intersecting two features (overlay) • Creating new administrative areas

Dissolving to create upper administrative levels

Possible representation of the real world Campus.esri.com 2000

GIS, characteristics • Format Vectorial • Format Raster

Vector format . Points Lines Polygons x,y Data are localized by coordinates That represent a specific position on the Earth

Vector format Connectivity Contiguity Adapted From ESRI, Virtual Campus

Vector Format

Limits of vector format • Area in polygons is assumed to be uniform e.g.(population distribution) • Objects represented depends very much from scale representation (the choice of the scale is determined by the initial resolution of epi data) • Adyacency between objects (common boundary

Reccomandation for vector format • Properly mantained database, the minimal mapping unit is indicated

The raster format are represented by grids Raster format y Satellite images x

Campus.esri.com 2000 Cells and their localization Raster format

Raster format Comparison between vector and raster format Campus.esri.com 2000

Campus.esri.com 2000 Position and value of cells Raster format

Data quality • Additional sources of errors in spatial analysis: – Spatial – Resolution – Interpretative – Temporal

Data quality • Spatial errors: – Displacement (data entry, methods to locate • Resolution errors: – Definition of minimal mapping unit • Intepretative error: – Where the « case » is placed • (residence, exposure, onset, notification) • Temporal error: – When (time between exposure and onset)