About HVAT

The Health Vulnerability and Adaptation Toolkit (HVAT) helps people across Aotearoa New Zealand understand how climate change may affect health in their communities — and where the biggest risks, and opportunities for action, lie.

It links downscaled climate projections to modelled health outcomes, down to suburb-sized areas, so you can explore what a changing climate could mean for the places you care about.

It’s for everyone planning for a changing climate — policymakers, health services, local and regional government, iwi and hapū, community organisations, and researchers. You don’t need to be a data scientist or a climate modeller to use it.

Built by Principal Economics for the Ministry of Health.

What you can find out

• How your area’s climate is projected to change — temperature, rainfall and more, under lower-to-higher emissions futures, down to suburb-sized areas.
• Which health conditions are most climate-sensitive where you live, and how the risk shifts over the century.
• Who is most affected — by age and location — to help prioritise adaptation.
• The evidence behind it — climate indicators, air-quality proxies (heating fuel, vehicle fleet), drinking-water disease, and the exposure–response curves the health projections are built on.

How HVAT is organised

• Find your area — locate your community on the map and see a quick local snapshot of projected warming, population and climate-health signals.
• Policy Explorer — compare scenarios side by side: climate projections, additional hospitalisations, health-risk distributions and population change.
• Data Explorer — the underlying datasets, climate indicators and exposure– response curves.
• Glossary — every term explained in plain language. New to the numbers? Start with Relative Risk, Climate scenarios, and How to read the numbers.

A note on geography. The tool works at three levels: Region, Territorial Authority (TA) — your district or city council area — and SA2, Stats NZ’s small areas of roughly 1,000–4,000 people (think a suburb or rural community).

Climate scenarios

Climate scientists use a standard set of Shared Socio-economic Pathways (SSPs) — storylines of how the world’s emissions and development might unfold. HVAT shows four, so you can see how health outcomes change from a lower-emissions future to a higher-emissions one.

SSP5-8.5 is not yet available in Data Explorer → Climate Model Projections (the source map data is still being prepared); it is included in the health-outcome views and the Policy Explorer climate time-series.

Scope, limitations & data

A few things to keep in mind when reading the numbers:

• It is a planning and prioritisation aid — not a forecast, and not clinical advice.
• The health projections shown here are temperature-related. They do not yet include flooding, mortality, mental health, food-system effects, or slower-building cumulative impacts — so they describe one channel, not the full picture.
• Ranges (the “low–high envelope”) show modelling bounds, not statistical confidence intervals.
• Equity: the model varies risk by age and location; ethnicity and deprivation breakdowns are a priority for future versions.
• Māori data sovereignty: the tool uses publicly available population and health data. We acknowledge Māori data sovereignty principles and are working with partners on appropriate governance and interpretation.

Full method, assumptions and caveats are in the technical documentation below.

Source & contact

Principal Economics (August 2025 draft), Vulnerability and Adaptation Assessment — Technical documentation. Prepared for the Ministry of Health.

Questions or feedback: [email protected] — feedback shapes future updates.

HVAT Glossary

Definitions of the concepts, codes, and data sources used throughout the dashboard. Where definitions come from the technical report they are quoted verbatim from the August 2025 Principal Economics report for the Ministry of Health.

New here? The most useful sections for most people are How to read the numbers (just below), Climate scenarios, Relative Risk, and the page-specific sections further down. You don’t need the technical metric codes unless you want them.

How to read the numbers

A quick guide to interpreting what you see across the dashboard:

• Relative Risk (RR) is a multiplier versus a baseline. RR 1.10 means 10% higher risk; RR below 1 means lower risk — the climate is protective for that group and condition. (See Relative Risk below.)
• “Additional hospitalisations” are counts, and can be negative. A negative value means fewer admissions — in New Zealand the temperature channel often removes more cold-related illness than it adds heat-related illness.
• The “low–high envelope” is a modelling range, not a statistical confidence interval — it shows the spread between the model’s low and high risk bounds.
• Everything here is temperature-related unless stated otherwise; it excludes flooding, mortality, mental health, and longer-term cumulative effects.
• Scenarios and warming periods set which future you’re looking at — see Climate scenarios and Warming periods below.

Geography levels

• Region (Regional Council) — 16 Regional Councils covering New Zealand (e.g. Auckland Region, Wellington Region, Otago Region).
• Territorial Authority (TA) — 67 city and district councils within Regional Councils (e.g. Auckland, Wellington City, Dunedin City).
• SA2 (Statistical Area 2) — Stats NZ’s small-area geography, roughly 2,300 polygons across NZ with typical populations of 1,000–4,000.
• Health Board (DHB) — Drinking Water tab only. Twenty former District Health Board boundaries (e.g. Capital & Coast, MidCentral, Counties Manukau). Health Boards do not align 1:1 with Regional Councils — for example the “Auckland” Health Board is different from the “Auckland Region” Regional Council.

Hierarchy: each SA2 sits inside exactly one TA, and each TA inside exactly one Region. The Find your area page lets you click a map to see this hierarchy for any address.

Climate scenarios (SSPs)

The dashboard uses four Shared Socioeconomic Pathways from the IPCC’s CMIP6 ensemble.

• SSP1-2.6 — Low emissions. Strong global cooperation; net-zero CO₂ by about 2050; warming stays below 2 °C.
• SSP2-4.5 — Middle emissions. Current social, economic and technological trends continue; warming reaches ~2.7 °C by 2100. (Default on some pages.)
• SSP3-7.0 — High emissions. Limited climate policy; CO₂ roughly doubles by 2100; warming reaches ~3.6 °C.
• SSP5-8.5 — Very high emissions. Continued fossil-fuel growth and rapid development; warming exceeds 4 °C by 2100.

Coverage gap: the Data Explorer → Climate Model Projections page only covers SSP1-2.6, SSP2-4.5, and SSP3-7.0 (SSP5-8.5 is not yet in the source data file). The V&A health-outcome and Time-Series pages include all four.

Climate models and MMM

The Policy Explorer → Time Series page lets you pick from seven climate models:

• MMM (multi-model mean) (default) — Mean across the six individual CMIP6 models below. Reduces individual-model bias and is recommended as the default for policy analysis.
• ACCESS-CM2 — Australian Community Climate and Earth System Simulator.
• AWI-CM-1-1-MR — Alfred Wegener Institute Climate Model.
• CNRM-CM6-1 — Centre National de Recherches Météorologiques (France).
• EC-Earth3 — European consortium climate model.
• GFDL-ESM4 — NOAA Geophysical Fluid Dynamics Laboratory Earth System Model.
• NorESM2-MM — Norwegian Earth System Model.

The Health Impacts, Additional Hospitalisations, and SA2 maps all use MMM internally — there is no model selector on those pages.

Warming periods

The V&A model summarises projections across three policy-relevant horizons:

• Near term — 2021–2040
• Mid century — 2041–2060
• End century — 2081–2100

Each period represents the climate state averaged across its 20-year window for the selected scenario and model. The 2061–2080 window is intentionally not summarised separately.

Relative Risk (RR)

A risk multiplier used throughout the V&A model.

• RR = 1 — risk unchanged from the reference period (baseline).
• RR > 1 — elevated risk. Example: RR = 1.5 means 50% higher risk than baseline.
• RR < 1 — the climate driver is protective (lower risk than baseline) for that population and condition. Many conditions show RR < 1 in some scenario/season/age combinations — these appear as the protective (bluish) band in the Health Impacts chart.

The “elevated-risk population” calculations sum population within SA2s where RR > 1.

V&A model

Vulnerability and Adaptation model. The Principal Economics model that links climate projections (temperature, precipitation, wind, humidity, etc.) to health outcomes (hospital admissions, mortality, exposure) for New Zealand populations, segmented by geography (Region / TA / SA2) and demographics (age, ethnicity, deprivation, disability).

Outputs:

• Exposure–response curves — Hospital Admissions tab.
• Population at elevated risk — Policy Explorer → Health Impacts.
• Additional hospitalisations — Policy Explorer → Additional Hospitalisations.

Source: Principal Economics (August 2025 draft), Vulnerability and Adaptation Assessment — Technical documentation, prepared for the Ministry of Health.

Data limitations & known gaps

The model is a useful planning aid, not a complete picture. Known gaps (and priorities for future versions):

• Temperature channel only. Flooding, mortality, mental health, vector-borne and food-system effects, and slower cumulative impacts are not yet included.
• Equity dimensions. Risk varies by age and location, but ethnicity and deprivation are not yet broken out in the interactive risk outputs.
• Seasonal averaging can smooth out short, extreme heat events.
• Coverage. Some source datasets are still being completed (e.g. SSP5-8.5 climate maps; selected ECI areas).

Māori data sovereignty

HVAT uses publicly available, aggregated population and health data (no individual records). We acknowledge Māori data sovereignty principles and are working with partners on appropriate governance and interpretation of the outputs. See the technical documentation for more.

Additional Hospitalisations page

(Policy Explorer → Additional Hospitalisations)

• Net change in hospitalisations — Projected change in temperature-related hospital admissions per year for the selected area and age, versus today. Negative values mean fewer admissions (in New Zealand, warming removes more cold-related illness than it adds heat-related illness).
• Low–high envelope — The range across the model’s low and high risk bounds. It is not a statistical confidence interval.
• National context — The same measure summed across all regions, for the same age and season as the local view.

ECI Indicators

(ECI Indicators tab)

Composite extreme-climate indices. Each is a summary of multiple thresholds or daily counts. Higher scores indicate more frequent or more intense extreme events.

• Extreme Heat (eci_heat) — Days above 25 °C plus consecutive hot-day spells (3, 5, or 10 days). Tracks heatwave frequency and duration. Strong predictor of heat-related hospital admissions.
• Extreme Cold (eci_cold) — Frost mornings (<0 °C), cold nights (<5 °C), ice days (max <0 °C) and cold days (max <10 °C). Linked to winter illness and energy demand.
• Extreme Dry (eci_dry) — Six metrics of dry spells: single dry days (<1 mm rain) and consecutive dry periods (3–20 days). Proxy for drought, fire risk, and pollen seasons.
• Extreme Rain (eci_rain) — Seven daily rainfall intensity thresholds (≥10 mm up to ≥150 mm). Tracks heavy rainfall events associated with flooding and waterborne disease.
• Extreme Wet (eci_wet) — Wet days (≥1 mm) plus consecutive wet spells (3–15 days). Indicator of prolonged damp conditions linked to mould and respiratory illness.
• Extreme Wind (eci_wind) — Days with average wind speed exceeding 5, 7.5, 10, 15, or 20 m/s. Linked to power outages, infrastructure damage, and transport disruption.
• Extreme Fire (eci_fire) — Seasonal Fire Weather Index (FWI) based on temperature, humidity, wind and rainfall. Pilot indicator of wildfire danger.

Climate metric codes

(Data Explorer → Climate Model Projections sidebar)

The most common climate variables shown on the map. The full CMIP6 / climate-index set — degree days, extreme percentiles, radiation and wind indices — is in the technical documentation.

Drinking Water Indicators

(Drinking Water tab)

• Campylobacteriosis — Most common bacterial gastrointestinal illness in New Zealand. Often linked to untreated drinking water after heavy rainfall (5-year moving average rate per 100,000).
• Cryptosporidiosis — Protozoan parasite commonly associated with contaminated surface water following heavy rain or flooding.
• Giardiasis — Protozoan waterborne illness; notifications frequently spike after flooding events.

Heating Fuel

(Heating Fuel tab)

• Household Heating Fuel Types (air_heating) — Proportion of dwellings using wood, coal, electricity, gas, etc. (Census 2013, 2018, 2023). Solid-fuel burning is the main source of winter PM₂.₅ in many NZ towns.

Vehicle Fleet Statistics

(Vehicle Fleet Stats tab)

• Vehicle Age Profile (air_fleetage) — Average age and age-band distribution of the light vehicle fleet by Territorial Authority (2025 snapshot). Older fleets emit more PM₂.₅ and NO₂.
• Number of Motor Vehicles by Type (air_fleettype) — Count of light vehicles, heavy vehicles, motorcycles, etc., by Territorial Authority. Proxy for transport-related air pollution hazard.

Hospital Admissions

(Hospital Admissions tab)

• Temperature-Related Hospital Admission Risk Curves — Exposure– response curves showing how the relative risk (see Relative Risk above) of hospital admission changes across the distribution of a climate variable, for different demographic groups.
• Indicator — The climate variable driving the curve (Temperature, Precipitation, etc.).
• Measure — The health-system metric being modelled (e.g. Hospital Admissions, ED Presentations).
• Outcome — The population grouping that splits the curves (e.g. Age Brackets, Income, Disability status).
• Condition — The specific disease or condition.

Source Principal Economics (August 2025 draft), Vulnerability and Adaptation Assessment – Technical documentation: Indicators, prepared for the Ministry of Health.