BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//talks.osgeo.org//foss4g-europe-2026//talk//RKPEN9
BEGIN:VTIMEZONE
TZID:EET
BEGIN:STANDARD
DTSTART:20001029T050000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:EET
TZOFFSETFROM:+0300
TZOFFSETTO:+0200
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:EEST
TZOFFSETFROM:+0200
TZOFFSETTO:+0300
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-foss4g-europe-2026-RKPEN9@talks.osgeo.org
DTSTART;TZID=EET:20260630T120000
DTEND;TZID=EET:20260630T123000
DESCRIPTION:The subdivision control check (udstykningskontrol\, UKS) is the
  process of verifying that a land subdivision complies with planning and l
 and-use regulations. In Denmark\, chartered land surveyors are legally obl
 igated to complete this check for every cadastral case submitted to the Da
 nish Geodata Agency (Geodatastyrelsen). The UKS requires the surveyor to m
 anually gather data from 11 different legal themes\, ranging from protecte
 d nature areas and coastal protection lines to soil contamination\, road b
 uilding lines\, and local plans (resulting in querying 17 different geospa
 tial data themes)\, before verifying whether a proposed cadastral change i
 s legally permissible. Each theme is governed by its own sector legislatio
 n\, and the surveyor must cross-reference open government datasets from mu
 ltiple national portals including data themes such as nature\, environment
 al\, planning\, cadastral\, coastal zones\, heritage sites\, agricultural 
 etc. In current practice this process is time-consuming\, fragmented\, and
  prone to human error [Hosseini et al.\, 2025b]\, yet it remains a mandato
 ry prerequisite before any cadastral case can be registered.\n\nResearch Q
 uestion\nThis paper presents a web-based\, AI-enabled PostGIS engine that 
 automates the UKS workflow. The system accepts WFS links\, performs dynami
 c geospatial analysis\, and structures its output specifically to enable a
  generative AI model to interpret the geographic properties of each GIS an
 alysis. The central research question is: how can AI be utilised as a tool
  to streamline the subdivision control check\, and to what degree can a lo
 cally hosted\, open-source LLM produce legally grounded\, evidence-based a
 nswers when provided with deterministic geospatial results as context? Not
 ably\, while the cadastral use case drives the design\, the core contribut
 ion is a generalisable architecture: the combination of WFS ingestion\, Po
 stGIS analysis\, and AI interpretation can be applied to any regulatory co
 mpliance workflow where spatial evidence must be matched against legal req
 uirements.\n\nSystem Architecture and Open-Source Stack\nThe system is bui
 lt entirely on free and open-source components. The web application is dev
 eloped in Next.js\, which exposes the APIs that connect the user interface
  to the geospatial and AI backend. The data layer is managed through Supab
 ase\, which provides three databases built on PostgreSQL: a primary databa
 se storing case information and parcel geometries\; a results database hol
 ding PostGIS outputs in structured JSON\; a vector database for the CAG em
 beddings.\nAt the analytical core is PostGIS\, which performs all 17 geosp
 atial analyses deterministically against the parcel geometry retrieved fro
 m the Danish cadastral register. The system accepts WFS endpoints\, reads 
 GetFeature responses\, and constructs a bounding box envelope around the s
 elected parcel. This envelope is used to query each WFS service\, and the 
 returned features are parsed and stored. Spatial operations include within
 -polygon tests\, line intersections and distance calculations. The outputs
  are structured to serve as precise inputs for the AI interpretation phase
 \, since the quality of the LLM response is only as good as the spatial ev
 idence it receives.\nFor natural language interpretation\, the system uses
  Ollama\, an open-source platform for running LLMs locally\, serving the M
 eta Llama 3.1 8B model [Ollama\, u.d.a]. The relevant legal texts are embe
 dded using the Nomic-embed-text model and stored in the vector database. T
 his constitutes a Cache-Augmented Generation (CAG) architecture [Chan et a
 l.\, 2025]: rather than expecting the model to recall Danish land law from
  its training data\, the system caches the legislation and injects it as c
 ontext at inference time. This constrains the model to a closed legal know
 ledge space\, which reduces the risk of hallucination.\n\nPipeline Phases 
 and Case Demonstration\nThe processing pipeline consists of four phases. P
 hase 1 accepts a parcel identifier via the web interface\, retrieves the c
 adastral geometry\, and initialises the case. Phase 2 runs the orchestrato
 r\, which queries all WFS endpoints in parallel and populates the database
 . Phase 3 executes the PostGIS analyses\, producing a structured result re
 cord per theme with a preliminary decision flag\, spatial evidence\, and a
 n agent log. Phase 4 passes these results alongside the embedded legislati
 ve context to the LLM\, which produces a completed draft of the UKS form. 
 The paper includes a case-oriented walkthrough demonstrating the system on
  a real cadastral parcel\, showing the analysis outputs for each of the 17
  themes and the corresponding AI-generated interpretations. Crucially\, th
 e geospatial results are themselves meaningful and verifiable independentl
 y of the AI layer: in many themes\, the spatial finding is already the ans
 wer\, and the AI provides the legal framing and documentation around it.\n
 \nResults\nThe system was evaluated against real cadastral cases and the g
 enerated UKS drafts were compared to manually prepared versions. The PostG
 IS layer correctly identified overlaps and distances across all 17 themes.
  The LLM layer produced coherent\, legislation-referenced responses in the
  majority of test cases\, with output quality closely tied to the specific
 ity of the spatial evidence provided. Beyond the cadastral domain\, the ar
 chitecture is directly applicable to other land-use compliance workflows w
 here spatial data must be checked against regulatory thresholds. Obvious e
 xamples include wind turbine siting (setback distances to dwellings and na
 ture areas)\, solar farm permitting\, and environmental impact screening.\
 n\nRelevance for FOSS4G\nAll geospatial data originates from Danish nation
 al open data infrastructures operating under INSPIRE-compliant WFS standar
 ds. The full stack\, Next.js\, PostGIS\, Supabase\, Ollama\, Llama 3.1\, a
 nd Nomic-embed-text\, is open source. The architecture is generalisable to
  any jurisdiction exposing land-restriction data through WFS services. The
  system code and data schema will be made publicly available under an open
 -source licence and available on GitHub. The study contributes to a broade
 r discussion on responsible LLM integration into professional legal-techni
 cal workflows [Hosseini et al.\, 2025a]\, specifically the role of determi
 nistic spatial evidence as a grounding mechanism that makes AI output trac
 eable and verifiable.\n\nConclusion\nThe presented system demonstrates tha
 t a web-based open-source GIS and LLM pipeline can automate complex\, legi
 slation-bound cadastral workflows in a robust and practically useful way. 
 Human oversight is preserved throughout\, as the surveyor reviews and appr
 oves all outputs.
DTSTAMP:20260605T043250Z
LOCATION:A01
SUMMARY:Automating the Subdivision Control Check: An Open-Source GIS and LL
 M Pipeline for Cadastral Case Preparation - Lasse Hedegaard Hansen\, Nickl
 as Nordhaug
URL:https://talks.osgeo.org/foss4g-europe-2026/talk/RKPEN9/
END:VEVENT
END:VCALENDAR