Using Dasymetric Mapping to Develop a Population

Report
Using Dasymetric Mapping to
Develop a Population Grid for
Hazard Risk Assessments
BEN ANDERSON
PROJECT MANAGER
UNIVERSITY OF LOUISVILLE
CENTER FOR HAZARDS RESEARCH AND
POLICY DEVELOPMENT
[email protected]
Presentation Outline
 Census Data Aggregation Levels
 Problem Statement
 Dasymetric Mapping
 Population Grid
Objectives
 Introduction to Dasymetric Mapping
 Application of Dasymetric Mapping
 Underlying need for Standardizing Area
 Use in Risk Assessment
Census Data Aggregation Levels
 Aggregation Levels
 Country
 State
 County
 Tract
 Block Group
 Block
 Tracts, Block Groups, and Blocks are aggregation
levels that are designed to be similar in population
but not area.
Census Tracts
 Census Tracts are a statistical subdivision of a county
 Tracts are designed to have between 1,500 and
8,000 persons
 Tracts can change from Census to Census as the
population changes.
 In Kentucky Counties have between 1 tract and 191
tracts in the 2010 Census
Census Block Groups
 Census Block Groups are a statistical subdivision of
Census Tracts
 Block Groups are designed to have between 600 and
3,000 people with an optimum size of 1,500
 Block Groups can change from Census to Census
 Block Groups are the smallest level which the Census
bureau publishes sample data

As of the 2010 Census all data excluding population count, sex,
age, race, or ownership status is sample data.
Census Blocks
 Census Blocks are a statistical subdivision of Block
Groups
 Blocks in urban areas are often literally a city block,
in rural areas blocks can be much larger

In KY Blocks range from 94.22 to less than .001 square
kilometers.
 Blocks are the smallest subdivision that the Census
releases full count data on.
Comparing Risk
 Center for Hazards Research has done Hazard Risk
Assessments Down to the block level for Kentucky.
 CHR’s latest state plan relied on count data within
blocks to develop a risk score.


CHR’s Risk score is a combination of Exposure and Hazard
Risk
Using pure count data results in a rural bias for increased risk

Increased exposure
• Larger blocks may have a higher population but lower density
• Larger blocks may also contain more assets: Roads, Rail, Bridges.

Increased hazard risk
• Larger Blocks also have more area exposed to a hazard and may
potentially be affected by more incidents due to the increased area
Dasymetric Mapping
 Method of mapping population within an
aggregation area using population data and land
cover data
http://pubs.usgs.gov/fs/2008/3010/fs2008-3010.pdf
USGS Dasymetric Mapping Tool
 A free tool which simplifies the Dasymetric process
 Requires Land Cover data to be broken down into 4
Classes


User defined breaks
Suggests –
High Density
 Low Density
 Non-Urban Inhabited
 Uninhabited

 Link: http://geography.wr.usgs.gov/science/dasymetric/
 Or Google: Dasymetric Mapping USGS
USGS Dasymetric Tool Key functions
 Empirical Sampling

Empirical sampling is used to determine the fraction of the census
unit's population that should be allocated to each inhabited class for
the study area
 Areal Weighting

The ‘population density fraction’ must be adjusted by the percentage
of the block-group’s total area that each ‘inhabited class’ occupies. A
ratio is calculated for each ‘inhabited class’ representing the
percentage of area that an ‘inhabited class’ actually occupies within a
block group to the expected percentage. The area ratio is used to
adjust the ‘population density fraction’ accounting for the variation
of both the population density and area for the different ‘inhabited
classes’ for each block group.
Source: http://geography.wr.usgs.gov/science/dasymetric/data/ToolDescription.docx
Land Cover Data Issues
 Low Resolution

In non urban areas, there may not be a differentiation in land class
between a residence and the surroundings

Resolution is typically able to differentiate roads from surroundings in
rural areas
 Assumption is population typically lives near roads
 Developed Areas that are uninhabited can show as high
density development

Use Feature Classes to reclassify raster areas to eliminate developed
areas that are uninhabited
Highways
 Airport


Block level data often finite enough that industrial areas are
separated from residential areas and show no population
Military Grid Reference System
 MGRS could provide an alternative aggregation level
to the Census Block level

Each unit is identical in size


Allows better comparisons between units in different parts of the
state
Population and demographic data is not calculated at MGRS
level
Assign proportionally based on area
 Assign using Dasymetric mapping based on area

Grid Advantages For Population Mapping
 Grid will enable a like to like comparison of areas
across the state rural or urban


Population is compared in an equal area
Better and easier to view visuals every polygon is equal area

A group of small highly populated blocks (Downtown areas) will
now be as visible as suburban areas.
Kentucky State Hazard Mitigation Plan Risk
Assessment
Hazard Vulnerability Score = Exposure Score X Risk
Score
 Risk Score = Probability of an event X actual
consequences (loss) and the % area of each unit that
is probable to be affected by an event.

The % area is calculated for hazards that have a defined and
predictable spatial extent. For example; Flooding (DFIRM),
Karst (KGS), Land Subsidence (KGS), and Landslide (KGS)
Kentucky State Hazard Mitigation Plan Exposure
Score
 Exposure Score= Population Rank + Property Rank
+ Essential Facilities Rank + Utility Rank +
Transportation Rank + Government-Owned
Facilities Rank + Hazardous Materials Rank
 Included raw counts and provided a rank (0-3) for
each one and then each was added together and
ranked again
Next Steps and Issues
 Build state wide risk assessment using dasymetric
modeling and an equal area grid
 Need Better Land Use Data

LIDAR


With Buildings
Building Footprint data
 Better understanding of where population is
 Better comparison of different areas
 Takes a Census count of population and creates an
estimate

similar documents