Benefit value, height checks and group areas

What changed

• The benefit value is stormwater plus carbon. Air-quality and cooling dollars stay in each tree’s full record, marked not release-eligible.¹
• PM2.5 removal and the cooling index are kept in the full record and not published: the PM2.5 method awaits confirmation of its sources.
• Heights are checked: 80 map markers have a height that is not plausible, and their estimates are withheld; 2,692 have a height to check. Record-to-crown links are checked: 253 records may be on a different tree from their crown, and 5,556 are to check.²
• Height class replaces the life-stage label, and crown replacement time replaces the irreplaceability class. 951,906 map markers are not classed.
• 3,824 map markers inside a Notable Group of Trees area carry the area’s Schedule 10 item, and the 229 areas are a map layer. Trees holding a Council point carry their Schedule 10 entry as the July 2026 schedule lists it.³
• Every source carries its licence; the 4,800 records derived from OpenStreetMap carry the Open Database Licence. Change since 2013 is release-eligible, with its limits.⁴
• The Species filter lists 1,028 species by Latin name under their common names. 11 misspelled or outdated Latin names read their accepted names, and the kauri survey’s records read Kauri.

Benefit value

Up to release 6.2 the benefit value added four services: air quality, cooling, stormwater and carbon. Air quality priced PM2.5 removal at NZ$25 per kg on a removal rate far above published measurements, and cooling was an exploratory index priced in dollars. From release 6.3 the benefit value is stormwater plus carbon, each crown’s with a band set by its valuation confidence (±30% for modelled medium, ±60% for modelled low).¹

Estimated annual benefits, sensitivity scenario. Release 6.2’s stormwater and carbon came to about NZ$76.74 million a year; release 6.3 withholds the estimates of the 80 crowns whose height is not plausible (−NZ$9,352 a year) and values the 103 crowns drawn outside the v5 survey on their drawn outline (−NZ$671 a year).

Map ranks, comparisons among similar trees, Area, property totals and the release totals use the benefit value. Stormwater is priced at NZ$3.50 per cubic metre of avoided runoff (1.5 times in flood-prone areas) and carbon at NZ$50 per tonne CO₂e; neither price is an audited Auckland value. The estimates remain sensitivity scenarios.

PM2.5 removal and the cooling index

Release 6 estimated PM2.5 removal as 0.008 kg per m² of leaf area a year. That rate is far above the removal rates published for urban trees, so its tonnes and its dollars were dropped from the benefit value.⁵

Release 6.3 builds a replacement, a deposition method after Nowak et al. (2013) with Auckland Council’s 2024 PM2.5 monitoring, and an unpriced cooling index (crown area × (paved fraction + 2 × degrees of mean air temperature above 18 °C)). Both are kept in each tree’s full record and are not published: the release publishes them only when every coefficient is confirmed from a cited source and the method’s check against the published range passes, and the method awaits confirmation of its sources. The map, the tree pages and the totals show neither.⁵

Height and record checks

Each crown’s measured height (the highest point of the 2024 LiDAR canopy height model within the crown) is checked. A height is not plausible at 55 m or more, or at 30 m or more with no vegetation returns, or more building than vegetation returns, in the crown’s laser-return record. It is to check over 30 m with a crown under 20 m², with a highest point more than 3 times the crown’s mean vegetation height (20 m or more), or more than 10% above the credible maximum height published for the recorded species or genus.²

Each record linked to a crown is checked against it: the LiDAR height at the recorded point is under a quarter of the crown height (crown 8 m or more); the crown is more than 10% above the credible maximum height for the recorded species or genus (crown 8 m or more); or the growth-form model reads the crown as another growth form than the recorded species’ class, at 0.8 or more. Two or more of these mark the record and crown as possibly different trees; one marks them to check. Of 42,491 linked records, 253 may be on a different tree and 5,556 are to check, on 4,188 map markers. A link a reviewer confirmed stays confirmed. These crowns are counted in totals and left out of comparisons, Species maxima and the property tallest tree.

Tree pages show each check in the release’s words. The map leaves flagged crowns out of the largest, tallest and highest-benefit trees in view, and flagged heights out of the mean height.

Height class and crown replacement time

The height class is the crown’s measured height as a share of the reference height for its growth form (evergreen broadleaf 22 m, deciduous broadleaf 25 m, conifer 35 m, palm or palm-like 14 m), on the growth form the service estimates use, rounded down to 0.1%: very small under 18%, small 18–35%, medium 35–60%, large 60–90%, very large 90% or more. It replaces the life-stage label, which named the same height-only classes in life-stage words; “veteran” is kept for veteran features recorded in the field.

Crown replacement time is the years a newly planted tree would take to grow a crown of this area at the planning rate for its growth form (2.2, 2.6, 2.0 and 0.6 m² a year), in bands that follow the years alone: under 5, 5–15, 15–40, 40–80, 80 years or more. It replaces the irreplaceability class, which placed every crown 25 m or taller in its top class whatever its regrowth time. Tall tree, 25 m or taller, is a separate fact.

Not classed: 917,582 crowns under 5 m tall (height class and regrowth years kept), 80 crowns whose height is not plausible and 34,244 map markers with no crown estimate: 26.6% of the 3,574,821 map markers. Not-classed crowns are left out of comparisons. Tall tree: 204,156 crowns.

In release 6.2, 441,472 crowns were in the top irreplaceability class; 154,139 of them were there by height with a regrowth time under 80 years, and 258,818 of the 441,472 take 80 years or more. Regrowth years now use the growth form of the published service estimates.

Group areas and Schedule 10 entries

Council’s Notable Group of Trees Overlay (Auckland Unitary Plan; layer retrieved 4 Jun 2026) has 229 group areas. 3,824 map markers lie inside one: 3,266 crowns, 473 earlier detections and 85 records. 2,980 of those crowns had no link to Schedule 10 (334 on private land in 75 areas). Each carries the area’s Schedule 10 item and name, and its tree page reads “Inside a Notable Group of Trees area”, with the note that whether the tree is one of the scheduled trees has not been established. The flag does not link a tree to Schedule 10: 3,570 map markers are linked to Schedule 10, as in release 6.2. ALTO checks the Notable Trees schedule only; Auckland Council determines whether a tree is protected. The 229 areas are a map layer under Layers, dashed.³

The July 2026 Schedule 10 is read entry by entry (item, trees listed, common and botanical names, address, legal description), and each tree holding a Council point carries its entry: 2,480 entries, which together list about 4,787 scheduled trees. Where an entry lists 2 or more trees and ALTO maps their Council points on fewer crowns, the tree page says so; for 96 items every point lies on one crown. For example, item 866 lists 2 pōhutukawa, and both Council points lie on one crown. A 40-row sample of the parse was checked against the schedule.

A reviewer’s note can now travel through the corrections ledger: tree T162DYHP, where Council’s point for item 1186 lies, carries one, and the field visit shown on it reads “A field visit found a tree at the Council point.” Corrections on tree pages read in tree numbers and plain names.

Licences

Every source in the release’s catalogue carries its licence, licensor, producer and attribution in the provider’s own form; Sources and credits lists them, and the release metadata and the full-record schema carry the same notice. The 4,800 records derived from OpenStreetMap stay under the Open Database Licence 1.0: their tree pages say so, ALTO’s noncommercial terms do not apply to them, and exports that contain them carry its share-alike condition. Change since 2013 is release-eligible from release 6.3, with its limits as release 6.2’s methods page states them.⁴

Species names

• The Species filter lists 1,028 species across the 56,258 map markers with a recorded species: each common name with its Latin names beneath it (Cherry: Prunus sp., Prunus serrulata, Prunus serrula and Prunus × yedoensis). A common name recorded without a Latin name is its own entry. The filter reads each marker’s species key.
• 11 misspelled or outdated Latin names read their accepted names (3,546 records), checked against the iNaturalist taxonomy (accessed 17 Oct 2026); for example, London plane is Platanus × hispanica. The name as supplied stays in each record and on the tree page.
• The 1,205 kauri survey records read Kauri; the survey result codes their names carried are kept in their own field with the survey’s credit and caveat.
• i-Tree reference heights are stored in metres (i-Tree supplies feet; kauri 165 ft is 50.3 m). Pūriri (952 records) reads evergreen broadleaf and cabbage tree (720 records) palm or palm-like; i-Tree’s classes are kept beside them.

What did not change

• Tree numbers, positions and crown outlines: release 6.2’s. The map shows 3,574,821 map markers and 4,200,336 tree numbers; 1,447 crown outlines, nearly all earlier crowns outside the v5 survey, are cleaned to their polygon parts, with the same area.
• Growth form, Council positions, the near-canopy layer and the Change since 2013 file are release 6.2’s.
• Every value release 6.3 retires stays in each tree’s full record: the life-stage label, the irreplaceability class, the air-quality and cooling dollars and the release 6 PM2.5 rate.
• Trunk diameter: from tool 0.2.15 its label follows the New Zealand convention of 1.4 m above the ground (on a slope, from mid-slope), where earlier pages said 1.3 m above the uphill ground; ALTO’s trunk diameters are estimates from the crown, so no number changed.

Deferred

• PM2.5 removal and the cooling index: published once the method’s sources are confirmed and its check passes.
• Plantation forest: crowns in exotic forest are classed by regrowth time like every other crown; release 6.3 does not overlay a land-cover map (LCDB).
• Trunk diameter calibration and ranges: they need reviewed girth measurements; KYTE’s girth missions collect them.
• Crowns at the 400 m² merge limit: not re-segmented; they keep the note “at or near the 400 m² segmentation limit”.
• The benefit class of a recorded species, and accepted names from national name lists: the benefit estimates keep the growth-form class.
• For a multi-tree Schedule 10 entry, the other crowns of the parcel: no “may be one of the trees” line yet.
• Lighter map tiles, a packed full-record archive and a place index.

Limits

• The height checks rest on the 2024 LiDAR survey and published maximum heights; a real tall tree can be flagged where its returns were misclassified.
• The record check’s third rule reads the growth-form model, which can itself be wrong (a kauri read as broadleaf), so one rule alone marks a record only to check.
• The group-area flag places a marker inside Council’s indicative overlay; it does not establish that the tree is one of the scheduled trees. The overlay may have changed since it was retrieved.
• Height class and crown replacement time use ALTO planning defaults for four growth forms; species within a growth form vary. Trunk diameter and root zones keep the release 5.2 growth-form class.
• Benefit estimates are not independently validated; they remain sensitivity scenarios.