Investigating a papakāinga question

Research question: Which parts of a teaching whenua deserve closer investigation for possible papakāinga development, and what spatial questions remain unresolved?

Useful for: whenua Māori owners, trusts, whānau project teams, planners and GIS practitioners preparing evidence for discussion.

This workflow is spatial screening. It does not decide where housing should go and does not replace engineering, planning, geotechnical, infrastructure or whenua decision-making. Its purpose is to make the spatial evidence visible before more detailed work begins.

What you will produce

• a source register
• a QGIS project in NZTM2000
• terrain and hazard context
• a simple access and services layer
• three synthetic candidate areas
• a comparison table that does not hide the reasoning inside one score
• an A3 discussion map
• a list of questions requiring non-GIS investigation

Working structure

Create three simple project tables rather than downloading a synthetic pack:

Table	Use
layer_register	Record each source layer, publisher, date, purpose and notes.
candidate_comparison	Compare candidate areas without hiding the reasoning inside a single score.
questions_register	Keep unresolved planning, access, servicing, whenua and technical questions visible.

Keep original authoritative source downloads unchanged in 01_source. Use NZGD2000 / New Zealand Transverse Mercator 2000 (EPSG:2193) as the working project CRS for normal land-area, distance and terrain work.

Source and checking

For every contextual layer record the provider, dataset name, source reference, date/version, what it was created to represent and a limitation relevant to the papakāinga question. Use Provenance and source checking.

Keep evidence and interpretation separate. A flood intersection is a geometric result under a published scenario. A nearby service point is not proof of capacity. A cadastral parcel is not automatically the same geometry as a Māori Land Court block.

Source data

Use a synthetic teaching whenua polygon. For public contextual data use current publisher sources where available.

Layer	Useful source	What it contributes
cadastral parcels	LINZ Data Service	current cadastral framework; not a substitute for title or Māori Land Court records
Māori land record context	Pātaka Whenua	current Māori Land Court electronic record for land within court jurisdiction
aerial imagery	LINZ aerial imagery	visible occupation, vegetation, roads and recent landscape context
elevation	LINZ 1 m DEM/available LiDAR products	slope and terrain screening
hazards	relevant council or other authoritative hazard publisher	published scenario/extent, with date and scenario retained
road/address context	LINZ and council sources	location and access context
planning/services	relevant council or network provider	only where publicly released and current enough for the purpose

For LiDAR-derived terrain, keep horizontal and vertical reference systems visible. LINZ national LiDAR deliverables use NZTM2000 and NZVD2016; older local datasets may use different vertical datums.

Project setup

Create:

papakainga-investigation/
  01_source/
  02_working/
  03_project/
  04_outputs/

Create the three working tables in a spreadsheet, CSV or GeoPackage according to the project. Keep authoritative source downloads separately in 01_source.

Step 1: establish the whenua and cadastral context

Add the teaching whenua and current LINZ parcels. Style parcels as thin outlines so they provide context without becoming the dominant interpretation of the whenua.

If the real-world starting point is a Māori land block, search Pātaka Whenua and record the block name, district and relevant identifiers separately from LINZ parcel IDs. Read block, parcel and title before joining anything.

Do not force a one-to-one relationship where the sources show one-to-many or many-to-one relationships.

Step 2: build terrain evidence

Add a DEM. If it is already in NZTM2000, clip it to a buffer around the study area rather than processing an entire regional raster.

In QGIS use Raster → Analysis → Slope or the equivalent Processing algorithm. Save the derived raster into 02_working with the source and date in its metadata or filename.

Use broad slope classes for discussion rather than pretending the class thresholds determine buildability. A useful teaching scheme is:

• 0–5°

• 5–10°

• 10–15°

• 15°

These are comparison classes only. Engineering thresholds come from the project, terrain and relevant professional assessment.

Step 3: add hazard layers properly

Add published flood, coastal, landslide or other hazard information relevant to the study area. Put the publisher, scenario and date in the layer name where possible, for example:

Council flood model, 1% AEP + climate scenario, published 2025

Do not merge several hazard scenarios into a single red polygon called hazard.

If a candidate intersects a modelled extent, the GIS result is that the geometries intersect under that published scenario. It does not by itself state the depth, consequence, consent outcome or engineering response.

Step 4: map access as evidence, not assumption

Keep these separate where they exist:

• legal/cadastral road parcel
• formed road
• visible track
• entrance/gate
• practical route observed on site

The distinction matters. A legal road can be unformed and a visible track can cross land without establishing a legal right of access.

Step 5: add services carefully

Use public service data only where it is actually published. If the location of a water, wastewater, electricity or telecommunications asset is approximate, record that.

Where no reliable spatial layer exists, create a question such as confirm nearest feasible power connection with provider instead of inventing a line.

Step 6: create candidate areas

Create a GeoPackage polygon layer candidate_areas with fields:

candidate_id
scenario
area_ha
terrain_note
hazard_note
access_note
services_note
other_note
status

Draw three synthetic candidates. The purpose is to practise comparison, not to create a suitability model.

Use Vector → Geoprocessing Tools → Intersection or Processing equivalents to identify hazard intersections. Use distance tools for indicative distances to roads or service points where the source supports that use.

Step 7: compare transparently

Create a candidate_comparison table and keep evidence and interpretation in separate columns.

A useful final table looks like this:

Candidate	Terrain evidence	Hazard evidence	Access evidence	Services evidence	Next question
A	mostly 0–10°	outside selected flood extent	near formed road	public power point nearby	confirm site servicing
B	mixed	partial intersection	near formed road	similar to A	understand flood depth/scenario
C	steeper	outside selected flood extent	longer route	farther from public service point	confirm practical access and terrain

Do not calculate 82/100 suitable. The table is more useful because people can see what drove the comparison.

Step 8: make the discussion map

Create an A3 landscape layout with:

• teaching whenua
• candidate areas
• slope or hillshade
• selected hazard layer
• formed roads/access
• service context where reliable
• north arrow, scale and coordinate reference
• source/date note
• prominent statement that candidates are synthetic teaching data

Use a second inset map if the main map needs wider access or catchment context.

Quality checks

Before export, check:

1. Project CRS is EPSG:2193.
2. DEM vertical datum is known.
3. Hazard layer includes publisher/scenario/date.
4. Parcel geometry has not been presented as Māori Land Court block geometry.
5. Visible tracks are not labelled as legal access.
6. Candidate areas are clearly synthetic.
7. The map does not imply a final development decision.
8. Source links are recorded in the source register.

What the result means

The result identifies spatial advantages, constraints and unknowns worth investigating. It is evidence for a better conversation about the whenua.

It does not establish legal access, engineering feasibility, planning approval, title interests, infrastructure capacity or a preferred papakāinga location.

Next useful research

Continue with Papakāinga and housing GIS, Māori land data, Property and cadastre, LiDAR and elevation, and Investigating whenua.

Last reviewed: 4 September 2026