IB Geography D.5.4 Representing Patterns of Risk and Vulnerability Notes

Risk and Vulnerability Require Different Data Layers

1. Hazard is not risk because a physical-probability surface does not show who is exposed or how susceptible they are to loss.
2. Indicator scale should match the phenomenon, with building-level data for structural vulnerability and neighbourhood data for service access.
3. Temporal alignment matters because hazard, census and infrastructure layers from different years may describe conditions that never existed together.
4. A hazard layer represents the physical process; it may map expected shaking, lava paths, ashfall, tsunami inundation or slope instability.

1. An exposure layer locates people and assets; population, buildings, roads, hospitals and utilities show what lies within the hazard footprint.
2. A vulnerability layer represents susceptibility; building quality, age, income, disability and access to services help explain unequal potential loss.
3. A capacity layer records resources for action; shelters, evacuation routes, emergency services, insurance and communication coverage can reduce consequences.
4. A risk representation combines these layers; it is an analytical model rather than a direct measurement of future disaster.

Different Graphics Answer Different Questions

1. Normalization converts totals to rates or densities where area or population differs, allowing fairer comparison between administrative units.
2. Bivariate display can show two variables together, such as hazard intensity and social vulnerability, but the legend must remain readable.
3. Uncertainty display uses ranges, transparency or confidence classes so modelled values are not mistaken for precise observations.
4. Point symbols show locations; earthquake epicentres, volcanoes, shelters and damaged facilities can be plotted precisely.
5. Proportional symbols compare magnitude; larger circles may represent deaths, displaced population or economic loss at named locations.
6. Choropleth maps compare standardised areas; rates or percentages should normally be used because administrative units differ in population and size.
7. Isolines and hazard zones represent continuous variation; contours can show shaking intensity, ash thickness or expected inundation across space.
8. Time-series graphs show change; event frequency, displacement or recovery indicators can be tracked before and after a disaster.
9. Scatter graphs test association; relationships such as income and mortality can be explored without assuming that correlation proves causation.

GIS Overlays Build a Transparent Risk Model

1. Weighting expresses a judgement about indicator importance, so alternative weights should be tested and documented.
2. Standardisation places variables measured in different units onto a comparable scale before they are combined.
3. Sensitivity analysis reveals whether a small change in weights, thresholds or missing data moves a place into a different risk class.
4. Ground-truthing compares the model with field observation and community knowledge to detect inaccessible routes, informal settlements or outdated records.
5. Step 1 maps hazard probability or intensity; the physical layer should state its time period, threshold and uncertainty.
6. Step 2 adds exposure; population and asset data identify who and what occupy each hazard zone.
7. Step 3 adds vulnerability and capacity; weights may represent construction, deprivation, mobility or emergency access.
8. Step 4 combines the layers; the output ranks places for planning but depends on the chosen indicators and weights.
9. Step 5 validates the pattern; past impacts, field observations and local knowledge reveal omissions or misleading classifications.

Design Choices Can Distort Spatial Patterns

1. Modifiable areal unit problem means a pattern can change when the same data are grouped into different boundaries or spatial units.
2. Ecological fallacy occurs when an average for an area is incorrectly applied to every household or person within it.
3. Visual hierarchy can exaggerate importance when colour, symbol size or map extent draws attention to one pattern while obscuring another.
4. Absolute totals favour populous places; rates reveal relative vulnerability but can exaggerate small populations, so both may be needed.
5. Large administrative units conceal local variation; an average can hide high-risk neighbourhoods beside safer areas.
6. Class intervals shape visual contrast; equal intervals, quantiles and natural breaks can produce different impressions from the same data.
7. Colour order should match meaning; a light-to-dark sequential scheme communicates increasing risk more clearly than unrelated colours.
8. Data date and completeness matter; outdated censuses, informal settlements and unreported losses can bias the displayed pattern.
9. Scale affects interpretation; a national map supports comparison while a street-level evacuation map supports action.

Choose the Graphic to Match the Claim

1. Process map is appropriate for pathways such as tsunami inundation or lahar channels, while a choropleth is appropriate for rates attached to areas.
2. Comparison design should keep classification and symbol scales consistent across paired maps so visual differences reflect data rather than styling.
3. Source note should state date, unit, denominator and method so the reader can judge whether the evidence supports the claim.
4. Use a hazard map for physical extent; do not describe it as a vulnerability map unless social conditions are included.
5. Use a choropleth for comparable area rates; state the denominator and avoid mapping raw totals when areas have unequal populations.
6. Use symbols for events or assets; legends must explain size, colour and units without requiring guesswork.
7. Use paired maps to show interaction; placing hazard intensity beside vulnerability can reveal why similar events produce unequal impacts.
8. Use graphs to support temporal or statistical claims; connect each pattern to a plausible process and acknowledge alternative explanations.

Representation Should Lead to a Defensible Decision

1. Decision threshold translates mapped evidence into action such as inspection, evacuation, zoning or resource allocation.
2. False precision should be avoided where coarse or uncertain data cannot justify a sharp boundary between safe and unsafe.
3. Review cycle updates the representation after urban growth, infrastructure change or a new event alters the assumptions.
4. Identify the decision first; evacuation, zoning, infrastructure protection and aid targeting require different spatial detail.
5. Select indicators that match the decision; a national mortality rate cannot locate a blocked evacuation route.
6. Make assumptions visible; titles, legends, dates, sources and units allow readers to judge the evidence.
7. Compare model output with lived experience; community knowledge can identify mobility barriers and informal assets absent from official datasets.
8. Revise the representation when conditions change; new monitoring, urban growth and recovery alter both risk and vulnerability.