Sources and credits

What ALTO changes

ALTO crops, resamples, reprojects and combines these data. It matches records to LiDAR crowns, draws outlines on imagery and adds its own estimates. Each source record keeps its original identifier and position in the tree’s full record. The organisations below have not reviewed or endorsed ALTO’s estimates.

Species maxima, the largest measured tree of each species, are ALTO’s ranking: species names from Auckland Council’s tree register and Notable Trees schedule, heights and crowns from the 2024 LiDAR survey. Tree pages credit and link each register that lists a tree.

Change since 2013 (map layer, version 1, released 13 Oct 2026) is ALTO’s comparison of the Auckland 2013, 2016–18 and 2024 LiDAR surveys, credited to their licensors under LiDAR surveys. In a blind photo check, 92% of the changes shown agreed with the reference labels (88–95%; 315 changes; labels from machine-learning image-interpretation models, no field check). Its limits are on the release 6.2 methods page.

Email the Better Places Lab (t.welch@auckland.ac.nz) to correct a credit, or to ask for anything to be changed or removed.

Tree records

• Auckland Council tree register (incl. Auckland Transport), retrieved June 2026. Trees recorded by Auckland Council, including street trees owned by Auckland Transport and trees owned by Auckland Council Parks: 50,311 records, retrieved in June 2026. © Auckland Council. The Council’s public service for the register now requires a token, so tree pages show register IDs without a link.
• Notable trees. Auckland Unitary Plan Notable Trees Overlay and Notable Group of Trees Overlay. Plans and Places, Auckland Council. Licensed under CC BY 4.0. Links: Notable Trees Overlay, Notable Group of Trees Overlay, Schedule 10 Notable Trees Schedule. ALTO links a tree to its Schedule 10 entry (from release 6.3, the entry’s item, trees listed, names and address as the July 2026 schedule gives them); Auckland Council determines whether a tree is protected. The map’s Notable Group of Trees areas are Council’s overlay; a tree inside one is not thereby one of the scheduled trees. The overlay is indicative; the Plan is the legal record.
• Kauri survey records. Ruru ObsKauri Tiaki (valid and public records). Bio Info Team, Environmental Services, Auckland Council. Item page, with the Council’s terms of use. ALTO shows each record’s status as the Council publishes it. Not a complete record of kauri or kauri dieback; verify before relying on it.
• OpenStreetMap. Trees mapped by OpenStreetMap contributors, and the addresses, golf courses and island roads used in property summaries. © OpenStreetMap contributors, available under the Open Database Licence (ODbL). ALTO’s tree records derived from OpenStreetMap stay under the ODbL; the method that matches them to ALTO’s trees is in the code repository.
• KYTE ground visits. Photographs, notes and measurements from KYTE contributors, published after a reviewer accepts them. Contributors are not named.
• New Zealand Tree Register. Run by the New Zealand Notable Trees Trust. Where a tree page or a KYTE field mission names the register record for the same tree, it links to that record on the register. ALTO does not copy the register’s text or photographs; photographs on the register belong to their photographers. Register recognition has no legal status.
• iNaturalist observations. Research-grade observations of woody plants in Auckland with open locations and a stated accuracy of 15 m or better, from iNaturalist, read through the iNaturalist API on 17 Oct 2026. A tree page lists the records whose location is inside its crown or within 5–15 m of it. Each record keeps its own licence, shown beside it: ALTO shows the species, month and the observer’s credit for CC BY and CC0 records and links to the others. ALTO copies no photographs, and a record never changes the species ALTO shows for a tree.

LiDAR surveys

• Auckland 2024 LiDAR, Parts 1 and 2 (point clouds, 1 m DEM and DSM), flown by Woolpert from 30 April to 4 November 2024. Sourced from the LINZ Data Service and licensed by Regional Software Holdings Limited (Part 1 DEM and DSM, Part 2 DEM) and Auckland Council (Part 2 DSM) for re-use under the Creative Commons Attribution 4.0 International licence. Layers 121990, 121992, 122580, 122581.
• Auckland North LiDAR 2016–18 (flown by Aerial Surveys) and Auckland South LiDAR 2016–17 (flown by AAM NZ). Sourced from the LINZ Data Service and licensed by Auckland Council for re-use under the Creative Commons Attribution 4.0 International licence. Layers 106410, 105089, 104318, 104382.
• Auckland LiDAR 2013, flown by NZ Aerial Mapping and Aerial Surveying Ltd. Copyright in the underlying dataset from which this work has been derived is owned by Auckland Council. Sourced from the LINZ Data Service, layers 53405 and 53406.

LINZ’s guidance: attributing elevation or aerial imagery data. The map’s short credit reads “Sourced from LINZ. CC BY 4.0”.

Aerial photographs

Tree pages read New Zealand Imagery from https://registry.opendata.aws/nz-imagery when they open; the list of surveys below was accessed on 12 October 2026. Each survey is licensed by the licensor named below for re-use under CC BY 4.0. Toitū Te Whenua LINZ hosts and processes the archive. Each photograph on a tree page names its own survey and the date it was accessed.

The 2024–25 survey is also used to separate leaves from other surfaces when crowns are drawn, and to describe growth form.

Scanned historical surveys, 1939–2004: aerial photographs from film, scanned and orthorectified, read from the same archive. The list below was accessed on 17 October 2026 from the archive’s STAC catalogue. Each survey is licensed by the licensors named below for re-use under CC BY 4.0; the producer is as LINZ names it. A tree page shows the surveys whose capture area covers the tree.

Street-level imagery

Where a street tree has a suitable photograph on Mapillary, its tree page links to it, with the date it was taken. Images: Mapillary contributors, licensed under CC BY-SA 4.0; each image’s page on Mapillary names its contributor. ALTO keeps only the image’s ID, date and distance from the tree, chosen from Mapillary’s public coverage, and does not copy or show the photographs.

Basemaps

• Esri World Imagery (“Hybrid”, “Satellite”, the tree page’s map and the aerial figures on the panel and the tree page): Powered by Esri. Source: Esri, Vantor, Earthstar Geographics, and the GIS User Community. In Auckland the imagery is Auckland Council’s. Used under Esri’s terms of use.
• Esri World Boundaries and Places (the labels over “Hybrid”): Powered by Esri. Esri, HERE, Garmin, © OpenStreetMap contributors, and the GIS user community.
• Esri World Light Gray Canvas (“Light”, the default): Powered by Esri. Esri, HERE, Garmin, © OpenStreetMap contributors, and the GIS user community.
• OpenStreetMap (“Streets”): © OpenStreetMap contributors. Tiles from the OpenStreetMap Foundation under its tile usage policy.

Land, boundaries and context

• NZ Primary Land Parcels (layer 50823) and NZ Building Outlines (layer 101290): contains data sourced from the LINZ Data Service licensed for reuse under CC BY 4.0. Parcels are not for defining legal boundaries.
• Public Conservation Land: Crown Copyright: Department of Conservation Te Papa Atawhai 2026. CC BY 4.0. Taken from the Department of Conservation’s own service (DOC Public Conservation Land) on 2 October 2026; it may be out of date.
• Local board boundaries and land areas: Stats NZ – Tatauranga Aotearoa, Territorial Authority Local Board 2026, CC BY 4.0.
• Auckland Council, all CC BY 4.0:
  • Unitary Plan base zones and designations: Plans and Places, Auckland Council.
  • Park extents: Auckland Council.
  • Impervious surfaces 2017 and land cover 2017: Lynker Analytics and Healthy Waters, Auckland Council.
  • Stormwater assets: Healthy Waters, Auckland Council.
  • Flood plains: Stormwater Hydraulic Modelling Team, Auckland Council.
  • Flood prone areas and overland flow paths: WSP; Auckland Council: Healthy Waters, Geospatial.
  • Predicted air temperature, November 2021 to March 2022: Arup New Zealand Limited, Lynker Analytics, Statistics NZ, Ministry for the Environment, Auckland Council, Contains modified Copernicus Climate Change Service information 2024.

Methods, models and reference values

• i-Tree (USDA Forest Service and cooperators): species growth form and mature size from the i-Tree Database. The benefit values are an ALTO benefit scenario: simplified equations adapted from published i-Tree Eco components, with ALTO’s own unit values; not run in i-Tree Eco. i-Tree is provided by the i-Tree Cooperative (USDA Forest Service, Davey, Arbor Day Foundation, UCFS, ISA, Casey Trees); i-Tree® is a registered trademark. The Cooperative has not reviewed or endorsed ALTO. Nowak, D. J. (2024). Understanding i-Tree: 2023 summary of programs and methods. General Technical Report NRS-200-2023. USDA Forest Service, Northern Research Station.
• DINOv2 (Meta; Apache License 2.0) provides image features of the 2024 aerial photographs for the growth-form model. Oquab, M. et al. (2024). DINOv2: Learning Robust Visual Features without Supervision. Transactions on Machine Learning Research.
• LightGBM (Microsoft; MIT licence) is the growth-form classifier. Ke, G. et al. (2017). LightGBM: A Highly Efficient Gradient Boosting Decision Tree. Advances in Neural Information Processing Systems 30.
• DeepForest (Weecology, University of Florida; MIT licence) and torchvision (BSD 3-Clause) are the base of the palm detector. Weinstein, B. G. et al. (2020). DeepForest: A Python package for RGB deep learning tree crown delineation. Methods in Ecology and Evolution 11: 1743–1751.
• iNaturalist research-grade observations, obtained through GBIF.org, are an independent check of palm detection. iNaturalist contributors, iNaturalist (2026). iNaturalist Research-grade Observations. Occurrence dataset https://doi.org/10.15468/ab3s5x, accessed via GBIF.org on 2026-10-02. Each observation is under its observer’s licence.
• Pl@ntNet suggests species from KYTE leaf, flower and fruit photographs, shown as unconfirmed leads. The image-based plant species identification service used, is based on the Pl@ntNet recognition API, regularly updated and accessible through the site https://my.plantnet.org/
• NZ Plant Conservation Network species pages are linked as references for the form of native trees in ALTO’s 3D models.
• iNaturalist taxonomy. From release 6.3, 11 misspelled or outdated Latin names in the tree records are shown under their accepted names, checked against the iNaturalist taxonomy (accessed 17 Oct 2026); each record keeps the name as supplied. This use is of names only.
• Published maximum tree heights. From release 6.3, ALTO’s height and record checks compare a crown with the tallest height published for its species or genus, from the New Zealand Plant Conservation Network, Flora of China (eFloras), Monumental Trees, the Gymnosperm Database, PlantNET (Royal Botanic Gardens and Domain Trust, Sydney), the New Zealand Tree Register, the World Agroforestry Agroforestree Database, VicFlora, FloraBase and other floras and registers. Each check names its source and links it; ALTO quotes the height only.

Software and fonts

MapLibre GL JS 4.7.1 (BSD 3-Clause) and PMTiles 3.0.7 (BSD 3-Clause), loaded from unpkg.com · three.js (MIT) · the Inter typeface by The Inter Project Authors (SIL Open Font License 1.1) · in KYTE, Leaflet 1.9.4 (BSD 2-Clause) and exifr 7.1.3 (MIT) · for building the map tiles, tippecanoe (BSD 2-Clause) and go-pmtiles (BSD 3-Clause).

ALTO

Developed by the Better Places Lab (Dr Timothy F. Welch), Te Pare School of Architecture and Planning, Waipapa Taumata Rau | University of Auckland. Code: PolyForm Noncommercial 1.0.0. Documents: CC BY-NC 4.0. Third-party data keep the licences above.