2. TerraPop Goals
Lower barriers to conducting interdisciplinary human-environment interactions research by making data with different formats from different scientific domains easily interoperable Provide an organizational and technical framework to preserve, integrate, disseminate, and analyze global-scale spatiotemporal data describing population and the environment.

3. Source Data
Domains & Formats
Population Microdata
Area-Level Data

4. Terra Populus Data Domains
Microdata
Land cover
Individuals and households
Population
Environment
Land use
Disparate scientific domains – interrelated processes
Multiple data formats
Areal Data
Climate

5. Trent Alexander, CIC Conference Presentation on IPUMS, 10/11/07
Age
Birthplace
Sex
Mother’s birthplace
Relationship
Race
Occupation
PopulationMicrodata
Structure
Geographic and housing
characteristics
Rows
Household records
Person records within households
Columns
Variables
Key points –
Preserve richness of microdata
Location based on administrative units

6. Microdata Availability
Thanks to attending partners:
Czech Republic*
Mexico*
Brazil
University of Barcelona
Slovenia*
Netherlands*
Spain*
Italy*
Poland*
*anticipating 2010 or 2011 data
Croatia – legislation approved?
Norway & Sweden NAPP only
Bulgaria no data provided
Finland, Slovakia, Kosovo
Denmark?

7. Area-level Data Sources
Census tables, especially where microdata is unavailable
Other types of surveys, data
Agricultural surveys
Economic surveys, data
Election data
Legal information

8. Environmental Data (Rasters) TerraPop Prototype
Land cover data from satellite images
(Global Land Cover 2000)
Agricultural land use data from satellites and government records (Global Landscapes Initiative)
Climate data from weather stations (WorldClim)

9. Location-Based Integration
Microdata  Area-level  Raster

10. Location-Based Integration
Microdata
Integration across domains, formats hinges on geography
Users get any type of data in format useful to them
Requires boundary files, boundaries harmonized over time
Rasters
Area-level data

11. Location-Based Integration
Microdata
Individuals and households
with their environmental
and social context
Integration across domains, formats hinges on geography
Users get any type of data in format useful to them
Requires boundary files, boundaries harmonized over time
Rasters
Area-level data

12. Location-Based Integration
Microdata
Summarized environmental and population
County ID G G G G G G G
County ID
Mean Ann. Temp.
Max. Ann. Precip.
Rent, Rural
Rent, Urban
Own, Rural
Own, Urban G
21.2
768
3129
1063
637
365 G
23.4
589
2949
1075
1469
717 G
24.3
867
3418
1589
1108
617 G
21.5
943
1882
425
202
142 G
24.1
2416
572
426
197 G
24.4
697
2560
934
950
563 G
25.6
701
2126
653
321
215
County ID
Mean Ann. Temp.
Max. Ann. Precip. G
21.2
768 G
23.4
589 G
24.3
867 G
21.5
943 G
24.1 G
24.4
697 G
25.6
701
characteristics for administrative districts
Integration across domains, formats hinges on geography
Users get any type of data in format useful to them
Requires boundary files, boundaries harmonized over time
Rasters
Area-level data

13. Location-Based Integration
Microdata
Rasters of population and environment data
Integration across domains, formats hinges on geography
Users get any type of data in format useful to them
Requires boundary files, boundaries harmonized over time
Rasters
Area-level data

14. Boundaries are Key
Linkages across data formats rely on administrative unit boundaries
Particular needs
Lower level boundaries
Historical boundaries

15. Administrative Unit Boundary Processing
Obtaining
Linking to microdata
Temporal harmonization
regionalization

16. Obtaining Boundary Data
Potential sources of digital data
National Statistical Offices
Global Administrative Areas data (e.g. SALB, GAUL)
Digitizing from images or paper maps
Challenges
Lower level and historical data
Date mismatches with census data
Code matching to microdata

17. Digitizing Boundaries Leveraging available digital data
Script input
Existing digital data
Rough digitized boundaries
Script output
Relevant boundaries from digital data
Relationship between digital and digitized units
Advantages
Preserve accuracy and detail
Flag areas needing more work

18. Code Matching
Codes link boundaries to microdata records, connect people to places
Boundary data may or may not include codes
Approach
Name matching, when possible
Map observations – digitizing script captures codes
Research on boundary changes
Boundary
shape
attributes
IPUMS
microdata

19. Temporal Harmonization
Purpose
Create consistent units for time-series analysis
Top-down strategy
Start with first administrative level units
Harmonize 2nd level units within 1st level “containers”
Script to create “least common denominator” units
Applicable when maps from multiple years are available
Creates aggregate units encompassing areas with boundary changes
Constructs source-harmonized crosswalk

20. “Erase” interior boundaries applicable to only one census
Apply harmonized codes
Also aids in code matching
Crosswalk

21. Regionalization
Confidentiality concerns require minimum 20,000 population in each unit disseminated
REDCAP tool
Constructs regions by combining units
Regions meet minimum population threshold
Contiguity constrained
Combines units that are similar in terms of a selected variable
Currently in testing phase
REDCAP Algorithms and parameters
Optimization variables (e.g., pop. density, education, occupation)
Testing on Malawi TAs, Brazil 2000 municipios
Guo’s NSF grant number is , in case they ask

22. Regionalization - Lilongwe, Malawi
Units < 20K combined with neighbors to meet threshold
Specific aggregation depends on
Optimization variable
Algorithm
Need to check this map on the projector….may be tough to see

23. Beyond Administrative Boundaries
Arbitrary boundaries
rasterization

24. Arbitrary Boundaries Watersheds, buffers around features, etc.
Near-term
Summarize rasters to user-supplied boundaries
Identify administrative units intersecting user-supplied boundaries
Future
Reallocation based on uniform distribution assumption
Reallocation based on other assumptions

25. Rasterization Prototype - All cells in unit get the same value
Use lowest level units available
Rates only, not counts
Future – Distribute based on ancillary data
Requires research on available methods
May provide as service – users select:
Ancillary data
Weights
Spatial distribution parameters

26. Acknowledgements IPUMS-International Participating Countries
Brazil
Bulgaria
Czech Republic
Germany
Italy
Ireland
IPUMS-International Supporters & Partners
Eurostat
Universitat Autònoma de Barcelona
Mexico
Netherlands
Poland
Slovenia
Spain