Big Data Ecosystems for Underground Infrastructure Detection

Wilmersdorfer Strasse, Charlottenburg, Berlin -- "Making the Underground Visible"

Date: 2026-03-25 Research Panel: 7 Ultra-Experts (Copernicus/EO, Drone Sensing, Mobility Data, Social/Crowd Intelligence, Telecom/IoT, Government/Utility, Novel Sources) Project Context: be.liviu.ai Three.js 3D underground model -- seeking cross-pollination from massive external data ecosystems

Concerns

• Data licensing restrictions may prevent commercial use of some government datasets
• Privacy regulations (GDPR) constrain granularity of mobility and social media data
• Drone operations in Berlin require EASA Specific Category authorization
• Some data sources (insurance claims, fleet data) are proprietary and require corporate partnerships
• Temporal alignment: different data sources have different update frequencies
• Coordinate system consistency: Berlin uses ETRS89/UTM 33N (EPSG:25833), Copernicus uses WGS84
• Point cloud and raster data volumes may exceed browser memory for Three.js rendering

Decisions

• Prioritize free/open data sources for initial integration (EGMS, Umweltatlas, Feuerwehr, Stromnetz)
• Use WFS/WMS standards where available for interoperability
• Normalize all spatial data to WGS84 for Three.js, with EPSG:25833 for analysis
• Thermal drone survey identified as highest-impact acquisition to pursue
• infrest Leitungsauskunft identified as most comprehensive single source for utility locations
• DAS fiber optic sensing identified as most innovative frontier technology

Assumptions

• Berlin Copernicus EGMS data has sufficient persistent scatterer density for Charlottenburg (urban area, likely true)
• Sentinel-2 NDVI anomaly detection works in Berlin's temperate climate with moderate vegetation
• Recommunalized BEW (formerly Vattenfall) heating network may be more data-accessible under public ownership
• Three.js can handle the combined data volume with appropriate LOD and tiling strategies
• Web browser memory limits constrain point cloud sizes to approximately 10M points maximum

Traceability

• All web sources cited inline with URLs and accessed 2026-03-25
• Claims marked [UNGROUNDED] where evidence is insufficient
• Expert domains: EO (Expert 1), Drones (Expert 2), Mobility (Expert 3), Social (Expert 4), Telecom/IoT (Expert 5), Government (Expert 6), Novel (Expert 7)
• Research methodology: structured WebSearch across 40+ queries, cross-referenced between experts

Executive Summary: Top 15 Data Sources by Impact/Feasibility

1. Copernicus Earth Observation (All Services, All Products, Berlin-Specific)

1.1 EGMS -- European Ground Motion Service (Sentinel-1 InSAR)

What it is: Free, pan-European mm-precision ground motion measurement derived from Sentinel-1 SAR interferometry. Detects subsidence, uplift, and lateral ground movement.

Data Products:

• Basic: Line-of-sight displacement time series per measurement point
• Calibrated: Referenced to GNSS, consistent across tracks
• Ortho: Decomposed into vertical and east-west components

Specifications:

• Spatial resolution: Individual measurement points (typically 20-50m spacing in urban areas)
• Temporal coverage: Annual updates, time series from 2015+
• Accuracy: mm-level displacement per year
• Format: GeoPackage, CSV, GeoTIFF
• Download: https://egms.land.copernicus.eu/
• License: Free, open access

Berlin/Charlottenburg Relevance:

• Detects subsidence over aging sewer networks (pipe collapse creates settlement bowls)
• Maps ground movement along U-Bahn U7 tunnel corridor (Wilmersdorfer Strasse station)
• Identifies differential settlement between buildings indicating underground voids
• Post-pipe-burst settlement patterns visible in time series
• Construction-induced movement from utility excavations
• [UNGROUNDED] Specific EGMS point density for Charlottenburg not confirmed -- Berlin urban areas typically have high PS-InSAR point density due to buildings/infrastructure acting as persistent scatterers

Three.js Integration:

• Download EGMS Ortho product for Berlin tile
• Parse GeoPackage with ogr2ogr to GeoJSON
• Map measurement points as colored spheres (red=subsidence, blue=uplift) at surface level
• Animate time series as temporal slider
• Overlay with known pipe routes to correlate settlement with infrastructure age

Sources:

• EGMS Portal
• EEA Data Hub
• Detektia EGMS Guide

1.2 Sentinel-1 SAR Coherence Maps

What it is: Interferometric coherence between SAR image pairs indicates surface stability. Low coherence = surface disturbance (excavation, construction, utility work).

Detection Capability:

• Urban areas normally maintain high coherence (stable buildings)
• Drop in coherence between two acquisition dates = ground disturbance
• Can detect: active excavation, road resurfacing, utility trench work, demolition
• Temporal resolution: 6-12 day revisit (Sentinel-1A/B)

Methodology:

• Requires three SAR images: two pre-event (reference coherence), one post-event
• Coherence difference maps highlight disturbed areas
• Pipeline: ESA SNAP toolbox -> coherence estimation -> GeoTIFF export

Berlin Application:

• Historical coherence time series = excavation history map
• Every utility trench, every road cut, every construction site since 2015
• Cross-reference with known utility locations to validate/extend network maps

Sources:

• SentiWiki Applications
• SAR Coherence Change Detection (ISPRS)
• UN-SPIDER Coherence Practice

1.3 Sentinel-2 Multispectral (NDVI Vegetation Anomalies)

What it is: 10m resolution multispectral imagery. Vegetation stress over buried utilities creates detectable NDVI anomalies.

Detection Mechanisms:

• Gas pipe leaks: Methane displaces soil oxygen, stresses vegetation -> NDVI drops below 0.35
• District heating pipes: Elevated soil temperature -> altered growth patterns, early senescence
• Shallow water pipes: Leak moisture creates greener corridors -> NDVI increase
• Recently backfilled trenches: Different soil compaction -> vegetation growth differences for 2-5 years

Specifications:

• Bands: B4 (Red, 665nm), B8 (NIR, 842nm) for NDVI; B8A, B11, B12 for moisture
• Resolution: 10m (B4, B8), 20m (B8A, B11, B12)
• Revisit: 5 days (constellation)
• Format: GeoTIFF, COG
• Access: Copernicus Data Space (https://dataspace.copernicus.eu/) or Sentinel Hub (https://www.sentinel-hub.com/)

Optimal Detection Windows:

• Early spring (NDVI 0.30-0.45): Best contrast for pipe-stress detection
• Post-harvest/late autumn: Exposed soil shows trench scars
• Summer drought: Leak-induced moisture creates obvious green lines

Research Validation:

• Multi-sensor study achieved +/-13m precision with Sentinel-2 for buried pipeline axis detection, improving to +/-0.3m with UAV data
• NDVI threshold of 0.35 separates pipeline-stressed from healthy vegetation
• Source: MDPI Remote Sensing

Three.js Integration:

• Download Sentinel-2 tiles from Sentinel Hub with custom NDVI script
• Generate NDVI difference maps (current vs. baseline)
• Drape as texture on terrain mesh with anomalies highlighted in false color
• Animate seasonal NDVI changes to show persistent linear anomalies

Sources:

• Sentinel Hub NDVI Anomaly Script
• Multi-Component Remote Sensing for Buried Pipelines (MDPI)

1.4 Sentinel-5P / TROPOMI (Atmospheric Methane)

What it is: Global methane (CH4) column measurements from the TROPOMI instrument.

Specifications:

• Resolution: 7km x 5.5km (too coarse for street-level)
• Revisit: Daily
• Detects methane plumes from oil/gas facilities, pipelines, urban sources

Berlin Relevance:

• Can identify city-wide methane anomalies but CANNOT resolve individual street leaks
• The Yamal-Europe pipeline (passes through Germany) was monitored by Sentinel-5P + AI, detecting 13 emission events
• Urban areas contribute 35% of detected methane plumes globally
• Verdict: LOW utility for Wilmersdorfer Strasse specifically -- resolution too coarse. Better as city-wide background layer.

Sources:

• ESA Methane Monitoring
• ESA Trio of Sentinels

1.5 Copernicus Emergency Management Service (CEMS)

What it is: On-demand satellite-based flood mapping and historical flood extent data.

Berlin Data Available:

• Historical flood depth maps across Europe (2015-2024) at 20m resolution from JRC
• CEMS has been activated for German flooding events (2021 catastrophic floods mapped)
• Global Flood Awareness System (GloFAS) provides probabilistic flood forecasts

Berlin Application:

• Historical flood extents for Charlottenburg indicate areas where combined sewer overflows occurred
• Correlate flood zones with underground infrastructure failure hotspots
• Climate projection data for infrastructure stress modeling

Access: https://emergency.copernicus.eu/mapping/

Sources:

• CEMS Flood Maps
• JRC Flood Hazard Catalogue

1.6 Copernicus Urban Atlas + Building Height + Imperviousness

Urban Atlas:

• Land use/land cover at 2.5m minimum mapping unit for Berlin FUA
• 2021 status layer available; 2024 in production (850+ FUAs)
• Classes: residential fabric density, industrial, transport, green areas
• Format: Vector (Shapefile/GPKG)
• Access: https://land.copernicus.eu/en/products/urban-atlas

Building Height 2012:

• 10m raster, height from normalized DSM (satellite stereo imagery)
• Covers Berlin FUA
• Source: IRS-P5 stereo + VHR satellite + LiDAR
• Access: https://land.copernicus.eu/en/products/urban-atlas/building-height-2012

Imperviousness:

• Soil sealing percentage at 10m resolution (HRL Imperviousness)
• Critical for: stormwater runoff modeling, infiltration zones, sewer load estimation
• 100% impervious = all water goes to sewer = maximum underground infrastructure stress

Three.js Integration:

• Building heights directly feed 3D building extrusion
• Imperviousness as ground texture (darker = more sealed)
• Urban Atlas classes for land-use coloring

Sources:

• GeoVille Urban Atlas 2024 Production
• GlobalBuildingAtlas (ESSD)

1.7 ERA5-Land Reanalysis (C3S)

What it is: ECMWF reanalysis providing hourly climate variables from 1950 to present at 9km resolution.

Relevant Variables:

• Soil temperature at 4 depth levels (7cm, 28cm, 100cm, 289cm)
• Soil moisture at 4 depth levels
• Precipitation (total, snowfall)
• Surface runoff, sub-surface runoff

Berlin Application:

• Soil temperature at depth = freeze/thaw cycle stress on pipes
• Long-term moisture trends = groundwater-infrastructure interaction
• Extreme precipitation events = combined sewer overflow triggers
• Climate projections for future infrastructure stress

Access: https://cds.climate.copernicus.eu/datasets/reanalysis-era5-land

2. Beyond-Visible-Spectrum Drones

2.1 Thermal/LWIR Drones (8-14 micrometer)

What it is: Drone-mounted uncooled microbolometer cameras detecting thermal radiation from underground heat sources.

Hardware:

• DJI Matrice 350 RTK + Zenmuse H30T (640x512 thermal, 1280x1024 zoom)
• FLIR Vue TZ20 or DJI Zenmuse XT2
• Workswell WIRIS Pro (640x512, 7.5-13.5 micrometer)
• Cost: EUR 5,000-25,000 (camera) + EUR 3,000-15,000 (drone)

Detection Capabilities:

• District heating leaks: BEW Berlin Fernwarme (2,000+ km network) -- thermal plumes visible at 50m altitude. DroneSystems documented 1,345+ confirmed leakages using this method.
• Sewer outfalls: Thermal contrast of warm wastewater entering rivers
• Active power cables: Resistance heating of underground cables
• Building basement mapping: Heat loss patterns reveal basement extent and utility penetrations
• Steam/hot water pipe insulation failure: Even without leaks, degraded insulation creates surface thermal anomalies

Optimal Conditions:

• Pre-dawn winter flights (maximum delta-T between underground heat and cold surface)
• No precipitation 48h prior
• Wind < 5 m/s
• Snow cover ENHANCES detection (melted patches over hot pipes)

Research Validation:

• German dataset (Munich, Karlsruhe) of thermal drone imagery for district heating leak detection published on Zenodo (2019-2021 flights)
• Machine learning classification of thermal anomalies from UAV imagery validated by district heating companies

Three.js Integration:

• Georeference thermal orthomosaic with RTK coordinates
• Drape thermal layer as texture on terrain (red=hot anomalies)
• 3D heatmap extrusion: height = temperature anomaly magnitude
• Overlay with known Fernwarme routes for gap analysis

Sources:

• ScienceDirect: UAV ML Leak Detection
• ResearchGate: UAV Thermal Anomaly Detection DHN
• Zenodo: Thermal Anomaly Dataset Germany
• Impact Aerial: Thermal Leak Detection Guide

2.2 Multispectral/Hyperspectral Drones

What it is: Drones carrying cameras with 5-300+ spectral bands for detecting gas leaks, vegetation stress, and moisture.

Methane Detection (SWIR):

• SWIR cameras with laser diodes tuned to CH4 absorption at 1653.2nm
• Detection: 5,000 ppm*m at up to 13.6m distance
• Real-time methane visualization overlay on flight video
• Tested across pavement, grass, and metallic surfaces

Vegetation Stress:

• cm-level NDVI over pipe routes (vs. 10m Sentinel-2)
• Detects stress within single growing season after pipe installation
• Red-edge indices (705-740nm) more sensitive than broadband NDVI for early stress

Hardware:

• MicaSense RedEdge-P (5-band multispectral, ~EUR 5,000)
• Headwall Nano-Hyperspec (270+ bands, ~EUR 50,000+)
• Quantum Solutions Q.Fly Water (SWIR moisture mapping, ~EUR 15,000+)

Sources:

• Nature: Drone Methane Imaging Performance
• SPH Engineering: Methane Detection PoC
• Spectroscopy Online: Real-Time Methane

2.3 LiDAR Drones

What it is: Airborne laser scanning for cm-precision surface elevation models.

Specifications:

• Accuracy: centimeter-level without GCPs
• Point density: 100-1,000+ pts/m2 (vs. satellite ~1 pt/m2)
• Penetrates vegetation canopy (multiple returns)
• Systems: Riegl miniVUX-3UAV, DJI Zenmuse L2, Livox Avia

Infrastructure Detection:

• Subtle subsidence: Detect 1-2cm settlement bowls over pipe routes from repeat surveys
• Trench scars: Micro-elevation differences along backfilled utility trenches
• Manhole/valve cover mapping: Automatic feature extraction from point cloud
• Drainage grade analysis: Precise slope measurement for sewer flow modeling

Leak Detection via LiDAR Intensity:

• Near-infrared LiDAR pulses reflect differently from wet vs. dry surfaces
• Intensity drops of 1.2 dB (grass) to 2.2 dB (paved) detected at leak sites
• Combines geometric + radiometric data in single flight

Three.js Integration:

• Import classified point cloud as THREE.Points
• Ground mesh generation from filtered DTM points
• Building/vegetation segmentation for 3D scene
• Temporal differencing between flights = deformation map

Sources:

• Using drone-based LiDAR intensity for leak detection (T&F)
• LiDAR Drones Guide 2026
• MDPI: UAV LiDAR Digital Subsidence Model

2.4 Drone-Mounted GPR (Ground Penetrating Radar)

What it is: GPR antenna mounted on UAV, transmitting radar pulses into ground from low altitude.

Key Product: SPH Engineering + Radar Systems Zond Aero 500 NG

• First universal dual-mode (airborne + ground) GPR
• Frequency options: 50-1000 MHz
• High-frequency (1000 MHz): fine detail, shallow (~0.5m)
• Low-frequency (50-300 MHz): deeper penetration (several meters), lower resolution
• Maintains constant antenna height above surface

Detection Capabilities:

• Underground cables, water/sewage pipes, gas pipes, storage tanks
• Pipe diameter estimation from hyperbolic reflections
• Void/cavity detection (sinkhole precursors)
• Soil layer interfaces

Limitations:

• Requires very stable, low-altitude flight (1-3m AGL)
• Signal attenuation in wet clay soils
• Urban RF interference can degrade signal
• Currently lower resolution than ground-based GPR carts

Three.js Integration:

• GPR B-scans as texture planes along survey lines
• 3D radargram volumes from grid surveys
• Detected pipe locations as colored tubes at measured depths
• Integration with DIN-standard depth assumptions for validation

Sources:

• SPH Engineering: Drone GPR
• Measur: Drone GPR Guide
• Zond Aero 500 NG Universal GPR

2.5 Magnetometer Drones

What it is: UAV-mounted magnetometers detecting anomalies in Earth's magnetic field caused by ferrous (iron/steel) underground objects.

Detection Capabilities:

• Cast iron water mains, steel gas pipes, reinforced concrete sewers
• Ferrous valve bodies, fire hydrant connections
• Abandoned/unknown ferrous pipes and tanks
• Detection depth: several meters depending on object mass
• Limitation: Cannot detect non-ferrous pipes (PVC, PE, copper, aluminum)
• SPH Engineering test: detected all ferrous targets EXCEPT 1.5mm wall steel pipe (too thin)

Hardware:

• GEM Systems GSMP-35U (potassium vapor, 0.003 nT sensitivity)
• Geometrics MagArrow (cesium vapor)
• QuSpin QTFM (total-field magnetometer)
• Sub-nanoTesla precision achievable with multirotor UAVs

Three.js Integration:

• Magnetic anomaly map as false-color terrain texture
• 3D dipole modeling to estimate pipe depth/orientation
• Combine with GPR for material identification (ferrous vs. non-ferrous)

Sources:

• SPH Engineering: Drone Magnetometers
• sUAS News: UAV Magnetometer Comparison Test
• PMC: MagNimbus vs MagArrow

2.6 Acoustic/Ultrasonic Drones

What it is: Drone-mounted microphone arrays for acoustic leak detection from air.

Key Product: CRYSound 2626G

• 64-microphone array
• Real-time acoustic visualization during flight
• Detects ultrasonic signatures of pressurized leaks

Applications:

• Water main leaks (pressurized hiss detectable from altitude)
• Gas leak acoustic signatures
• Sewer blockage/overflow sounds

Maturity: Early commercial stage -- fewer validated deployments than thermal or LiDAR

Sources:

• MFE Acoustic Leak Detection

2.7 SWIR Moisture Mapping Drones

What it is: Short-Wave Infrared cameras computing Normalized Difference Moisture Index (NDMI) from drone altitude.

Key Product: Quantum Solutions Q.Fly Water

• Simultaneous NIR + SWIR capture
• Onboard NDMI computation
• Live color-coded moisture maps streamed to controller
• Resolution: up to 400x finer than satellite NDMI
• Detects: surface wetness from underground water leaks, irrigation patterns, seepage

Berlin Application:

• Pavement moisture mapping after rain reveals where water pools = subsurface drainage failure
• Persistent wet spots on dry days = active leak indicators
• Green strip anomalies along pipe routes during drought

Sources:

• Quantum Solutions Q.Fly Water
• DRONELIFE: SWIR Payload

2.8 Drone Regulations -- Berlin

Regulatory Framework:

• EU Regulation 2019/947 (delegated reg 2019/945) -- EASA categories
• Open Category: <25kg, VLOS, max 120m AGL -- for small survey drones
• Specific Category: Required for BVLOS, urban overflights >4kg, or without EC class label
• Authorization: Standard Scenario (STS) declaration, PDRA, or operational authorization
• Remote ID mandatory since 2024 for most operations
• U-Space: May be required for congested urban areas (Berlin city center)

Berlin Specifics:

• Controlled airspace near TXL (now closed) and BER (Schonefeld) -- Charlottenburg is outside BER CTR
• German UAS geographic zones: check LBA AIP, Deutsche Flugsicherung (DFS) droniq app
• Registration: LBA (Luftfahrt-Bundesamt) operator registration required
• Insurance: Mandatory liability insurance for all drone operations

Sources:

• Germany Drone Laws 2025
• EASA Specific Category
• EASA Regulations 2025

3. Mobility Big Data

3.1 Uber/Bolt Movement Data

What it is: Aggregated GPS traces from ride-hailing vehicles providing street-segment speed and travel time data.

Infrastructure Proxy:

• Speed anomalies on road segments correlate with: road surface degradation, utility cut patches, active construction
• Temporal speed drops = ongoing excavation/utility work
• Persistent speed reduction vs. baseline = permanent road quality issue (settlement, pothole)

Data Availability:

• Uber Movement: Speed data per road segment per hour (requires 5+ unique trips/hour)
• Covers 98,210 road segments in NYC-scale cities; Berlin coverage [UNGROUNDED -- needs verification]
• Note: Uber Movement may have been deprecated/integrated into Google tools

Sources:

• Uber Movement Street Speeds
• UrbComp 2020: Speed Anomalies

3.2 Google Maps / Waze

Waze for Cities (formerly CCP):

• Real-time construction zone reports from 140M+ users
• Road closure data feed (XML/JSON GeoRSS, updated every 2 minutes)
• Differentiates planned construction vs. emergency utility work
• Partners share road closure data; Waze shares traffic + incident data back
• Requires partnership agreement (free for government/academic)

Google Traffic:

• Historical and real-time speed data per road segment
• Available via Google Maps Platform Roads API
• Construction/incident overlays

Berlin Application:

• Every construction report on Wilmersdorfer Strasse = potential utility work
• Historical closure data = excavation frequency map (higher frequency = more aging infrastructure)
• Cross-reference with utility records to identify which utility was being worked on

Sources:

• Waze for Cities
• Waze Data Feed

3.3 HERE Technologies

Probe Data:

• 44 million connected vehicles contributing live sensor data
• Integrates data from Audi, BMW, Mercedes-Benz vehicles
• Provides: real-time traffic, road conditions, sign detection
• Road Surface Quality: [UNGROUNDED] -- HERE likely has this data from suspension sensors but not confirmed as a publicly available product for Berlin specifically

Floating Car Data:

• GPS position + timestamp + speed + heading per vehicle
• Available through HERE Developer portal with pricing tiers
• https://developer.here.com/documentation/traffic-probe-data

3.4 Strava Metro

What it is: De-identified, aggregated cycling and pedestrian movement data from Strava users.

Infrastructure Insights:

• Route avoidance patterns = poor road surface quality
• Speed drops at specific locations = potholes, utility cuts, uneven surfaces
• Just 10% of roads carry 75% of cycling traffic -- deviations from expected patterns indicate issues
• Free for government and urban planners since 2019
• Partnership with 4,000+ city agencies worldwide

Berlin Application:

• Cycling speed profiles along Wilmersdorfer Strasse and parallel streets
• Seasonal patterns: ice/frost damage visible as spring speed drops
• Route choice changes after utility work = quality of restoration

Access: https://metro.strava.com/ (free for government/academic)

Sources:

• Strava Metro
• T&F: Strava Metro Literature Review

3.5 E-Scooter Fleet Data (Tier, Lime, Voi)

What it is: Accelerometer + GPS data from millions of e-scooter trips.

Road Quality Metrics:

• RMS acceleration, skewness, kurtosis, standard deviation of vibration
• Deep learning models classify road roughness into severity levels
• IMU (3-axis gyroscope + 3-axis accelerometer) data at high sample rates
• ScooterLab: Research testbed for micromobility sensing (NSF-funded)

Berlin Application:

• Berlin has major scooter fleets (Tier HQ is in Berlin)
• Every trip = road surface quality measurement
• Vibration spikes at specific coordinates = utility trench boundaries, manhole covers, settlement
• Temporal degradation tracking: same road segment getting rougher over months = subsidence

Sources:

• PMC: E-Scooter Data Acquisition
• Frontiers: Deep Learning Road Roughness E-Scooter
• ScooterLab (arXiv)

3.6 Berlin VIZ Floating Car Data / Taxi GPS

What it is: GPS traces from Berlin taxi fleet + other FCD sources, operated by VMZ Berlin Betreibergesellschaft.

Data:

• 300+ taxis provide city-wide coverage hourly
• 1,000+ fixed measurement points for traffic speed
• Purchased FCD from multiple providers
• Congestion hotspot analysis published openly

Berlin Application:

• Traffic speed anomalies on Wilmersdorfer Strasse segments
• Historical congestion data correlated with utility work periods
• Speed reduction patterns after infrastructure events

Access: https://viz.berlin.de/en/ (traffic map free, raw data may require agreement)

Sources:

• VIZ Berlin
• VMZ Berlin
• ResearchGate: Berlin Taxi GPS Analysis

3.7 Connected Car Data (Tesla, BMW, Mercedes)

What it is: Fleet-wide suspension/accelerometer data from connected vehicles.

Tesla Specifics:

• Software update 2022.20+: vehicles scan for potholes and rough roads
• Fleet-aggregated road surface map downloaded to individual vehicles
• Suspension auto-adjusts based on crowdsourced roughness data
• This data exists but is proprietary to Tesla

BMW/Mercedes via HERE:

• Contribute sensor data to HERE platform
• Wheel speed sensors + drivetrain data = roughness measurement
• 3-6% connected vehicle penetration provides statistically significant coverage

Berlin Application:

• mm-precision road surface quality from suspension telemetry
• Identifies every utility cut, manhole settlement, frost heave
• Temporal evolution: road degradation rate = underground infrastructure stress indicator

Sources:

• Frontiers: Connected Vehicle Pavement Quality
• Tesla Pothole Scanning
• MDPI: Pavement Quality Connected Vehicles

4. Social Media & Crowd Intelligence

4.1 Twitter/X and Mastodon

Real-Time Incident Detection:

• NLP + ML achieves 88.27% accuracy for traffic-related tweet classification
• Keywords: "Wasserrohrbruch", "Gasgeruch", "Rohrbruch", "Uberschwemmung", "Baustelle" + Charlottenburg/Wilmersdorfer
• Global flood detection dataset: 88M tweets -> 10,000+ flood events in 176 countries

Berlin Application:

• Monitor German-language tweets for infrastructure incidents in Charlottenburg
• Temporal correlation between social media reports and utility failure events
• Sentiment analysis for infrastructure quality perception
• Mastodon instances (berlin.social) for local community reports

Sources:

• Nature: Global Social Media Flood Database
• Springer: Real-Time Traffic Incident Detection from Twitter

4.2 Google Street View / Mapillary -- Time Series Change Detection

Mapillary:

• Crowd-sourced street-level imagery with computer vision
• Detects 100+ object classes including manholes (object--manhole), utility covers, valve markers, utility cabinets
• Semantic segmentation assigns labels to every pixel
• Free API tier available
• Berlin: https://www.mapillary.com/app/?lat=52.5065&lng=13.3088&z=16

Google Street View Time Series:

• CityPulse framework: largest street-view change detection dataset (1,000 coords, 6 cities, 2007-2023)
• ~10 images per location over 16 years
• Deep learning models detect construction, infrastructure changes
• Bentley Systems integration: automated roadway asset detection and inspection from Street View
• Stanford HAI research: AI maps urban change from Street View archives

Berlin Application:

• Historical Street View for Wilmersdorfer Strasse: identify all visible manhole covers, utility cabinets, valve markers, hydrant locations
• Time series: detect when road was excavated and resurfaced (color/texture change)
• Mapillary object detection: automatic inventory of surface-visible infrastructure
• Excavation scars visible for years after backfill (asphalt color difference)

Three.js Integration:

• Mapillary API -> extract detected manhole/utility cover positions -> place markers in 3D model
• Street View imagery as facade textures for buildings
• Time-lapse animation of street surface changes

Sources:

• Mapillary Object Detections
• CityPulse (arXiv)
• Stanford HAI: AI Urban Change

4.3 nebenan.de

What it is: Germany's largest neighborhood social network (3M+ users, 8,000+ neighborhoods).

Infrastructure Value:

• Hyperlocal complaints about water pressure, gas smells, flooding, noise
• Reports visible only to verified neighborhood residents
• Data access: DIFFICULT -- private platform, no public API
• Would require partnership agreement or manual monitoring

Sources:

• nebenan.de Wikipedia

4.4 Ordnungsamt-Online Berlin

What it is: Berlin's official citizen reporting platform for public space issues.

Capabilities:

• Citizens report defects with GPS location + photos
• Categories include: Strassenschaden (road damage), Kanaldeckel (manhole covers), Wasseraustritt (water leaks)
• Reports routed to responsible district office (Charlottenburg-Wilmersdorf)
• App + web interface, anonymous reporting possible
• Automated routing to correct agency

Berlin Application:

• Direct infrastructure failure reports with precise geocoding
• Historical pattern: repeated reports at same location = chronic issue
• Cross-reference with underground utility map to identify failing segments

Access: https://ordnungsamt.berlin.de/ -- public reports viewable at aktuelle Meldungen

Sources:

• Ordnungsamt-Online
• Berlin.de Mobile App
• Charlottenburg-Wilmersdorf Ordnungsamt

5. Telecom & IoT Big Data

5.1 Telecom Infrastructure as Underground Map

Fiber Optic Routes:

• FTTH rollout data = documented conduit locations
• Deutsche Telekom, Vodafone, O2 all have duct networks in Berlin
• Fiber routes follow existing utility corridors

Cell Tower Mapping:

• OpenCelliD: largest open database of cell towers globally
• CellMapper: crowd-sourced coverage mapping
• 5G small cells (250-300m spacing) = dense location network in urban areas
• Signal attenuation patterns COULD theoretically indicate underground metallic structures [UNGROUNDED -- no validated research found for this specific application]

Sources:

• OpenCelliD
• CellMapper

5.2 Distributed Acoustic Sensing (DAS) via Telecom Fiber

What it is: Existing telecom fiber optic cables repurposed as distributed vibration sensors using C-OTDR interrogators.

How it works:

• Interrogator sends laser pulses through existing fiber
• Rayleigh backscattering detects vibrations along entire fiber length
• Meter-level spatial resolution over km distances
• Detects: vehicle traffic, construction, pipe leaks, soil collapse, subway trains

Recent Urban Research:

• Urban underground sensing system built on existing roadside fiber optic infrastructure
• Identifies vehicle driving, construction behavior, road subsurface cavity events
• Real-time, continuous monitoring without new sensor installation

Berlin Application:

• Berlin has extensive telecom fiber networks along Wilmersdorfer Strasse
• DAS could detect: water hammer in pipes, gas flow changes, metro vibrations, construction activity
• Leak acoustic signatures detectable through fiber adjacent to water mains
• Requires partnership with Deutsche Telekom or fiber provider

Three.js Integration:

• Fiber route as linear feature in 3D model
• Color-coded by detected vibration type along its length
• Real-time animation of acoustic events

Sources:

• AP Sensing: DAS Technology
• PMC: DAS for Linear Infrastructure
• Springer: DAS Urban Road Events

5.3 Smart Meter Data (Stromnetz Berlin AMI)

What it is: Advanced Metering Infrastructure providing real-time voltage, current, and power quality data from every connected building.

Infrastructure Detection:

• Voltage notching/dips = underground cable degradation
• Power quality anomalies = transformer or cable faults
• Consumption patterns reveal active vs. inactive service connections
• Phase imbalance patterns indicate cable routing topology

Berlin Context:

• Stromnetz Berlin operates 36,000 km of underground cable (99% underground)
• Smart meter rollout underway (EU mandate)
• Data access: utility-internal, would require partnership

Sources:

• Stromnetz Berlin: Structure
• US DOE: AMI Summary

5.4 LoRaWAN / IoT Sensor Networks

What it is: Low-power wide-area network sensors deployed in manholes, drains, and underground infrastructure.

Deployed Applications:

• Manhole water level sensors (flood early warning)
• Manhole cover displacement detection (theft/unauthorized access)
• Soil moisture monitoring
• Air quality in confined spaces

Berlin:

• The Things Network (TTN) has community gateways in Berlin
• LoRaWAN signal penetrates from beneath heavy manhole covers through dense urban environments
• [UNGROUNDED] Specific Berlin municipal LoRaWAN deployment for infrastructure monitoring not confirmed

Three.js Integration:

• Real-time sensor values displayed as floating labels or color-coded markers
• Water level animations in sewer cross-sections
• Alert visualization when thresholds exceeded

Sources:

• Semtech: Preventing Floods with LoRaWAN
• YZ Systems: Sewer Level Monitoring

5.5 Connected Car Data as IoT Sensors

(See Section 3.7 for detailed treatment)

Summary: Tesla, BMW, Mercedes vehicles act as mobile IoT sensors measuring road surface at mm precision. 3-6% fleet penetration already provides statistical coverage of major roads. Berlin has high connected-vehicle density.

6. Government & Utility Databases

6.1 Berlin Umweltatlas (Environmental Atlas) -- CRITICAL SOURCE

What it is: Comprehensive environmental database for Berlin maintained by Senate Administration. This is a GOLDMINE for underground infrastructure context.

Relevant Layers:

Groundwater:

• Groundwater levels of Main Aquifer + Panke Valley Aquifer (annual maps since 2001)
• Expected Highest Groundwater Level (EHGL) 2022 -- critical for basement flooding risk
• ~1,000 monitoring stations across Berlin
• Four aquifer layers distinguished down to 150m depth
• Data: WMS layers via Umweltatlas portal
• https://www.berlin.de/umweltatlas/en/water/groundwater-levels/

Geology:

• Geological Map 1:25,000 (historical, 1875-1883 mapping + modern updates)
• Geological Outline: subsurface units, Ice Age deposits, geological cross-sections
• Data to ~2m depth routinely; deeper for specific boreholes
• https://www.berlin.de/umweltatlas/en/soil/geological-map/

Soil:

• Soil type map, soil sealing (impervious surface), contamination sites
• Critical for: utility installation difficulty, settlement risk, corrosion rate
• https://www.berlin.de/umweltatlas/en/soil/

Noise Maps (Strategic Noise Maps 2022):

• Road, rail, aircraft, industry noise at building facade level
• Day and night levels
• Relevant: construction noise impact zones, infrastructure activity indicators
• https://www.berlin.de/umweltatlas/en/traffic-noise/noise-pollution/2022/maps/

Three.js Integration:

• Groundwater level as translucent blue plane at measured depth
• Geological layers as colored cross-section planes
• Soil type as terrain texture
• Noise map as surface overlay

6.2 FIS-Broker (Berlin Geoportal)

What it is: The central geoportal for Berlin's spatially-referenced datasets. Thousands of WMS/WFS layers.

Access Methods:

• Web viewer: https://fbinter.stadt-berlin.de/fb/
• WFS API: query-fis-broker-wfs on GitHub (https://github.com/derhuerst/query-fis-broker-wfs)
• WMS services for map layer visualization

Infrastructure-Relevant Layers (selection):

• StEP Gasversorgung (Strategic Energy Plan -- Gas Supply): main gas pipeline network
• Versorgungsgebiete (supply areas)
• Strassenbefahrung (road survey data)
• Baumkataster (tree registry -- trees interact with underground pipes)
• Bodenbelastungskataster (soil contamination registry)
• Denkmalschutz (historic preservation -- constrains excavation)

Sources:

• FIS-Broker Portal
• GitHub: FIS-Broker Data
• Technologiestiftung: FIS-Broker to QGIS

6.3 infrest -- Leitungsauskunft (Utility Information Portal)

What it is: Germany's largest utility information aggregation service, connecting ALL network operators for construction planning.

Scale:

• 18,600 connected infrastructure operators (ISB)
• 6.5 million+ utility information requests processed
• Covers: water, gas, electricity, telecom, district heating, fiber optic

How it works:

• Submit location query (street address or polygon)
• Receive combined plan excerpts from ALL operators in that area
• Digital delivery via SaaS platform

Berlin Application:

• Single query for Wilmersdorfer Strasse returns ALL utility locations from ALL operators
• Most comprehensive data source for actual pipe/cable positions
• Requires registration + per-query fee (or construction project justification)

Sources:

• infrest Homepage
• infrest Company Profile

6.4 Berliner Wasserbetriebe (BWB)

What it is: Berlin's municipal water and wastewater utility.

Network:

• Drinking water pipe network: 7,816 km
• Wastewater sewers: 9,746 km
• Three main media: sewers (Kanalnetz), drinking water lines (TWL), wastewater pressure lines (ADL)

Data Access:

• Utility information via infrest portal for construction projects
• General network maps may be available through FIS-Broker
• Regelblattverzeichnis (standards catalog) for network design at bwb.de
• Historical monitoring since 1869

Sources:

• BWB: Path of the Water
• BWB: New Building Connection

6.5 NBB Netzgesellschaft / GASAG (Gas)

What it is: NBB operates Berlin's 7,000 km gas distribution network (GASAG Group).

Data Available:

• StEP Gasversorgung map available through FIS-Broker
• Main pipeline network visible in strategic energy plan
• Hydrogen readiness roadmap (3-phase network upgrade)
• Utility information via infrest

Sources:

• NBB Netzgesellschaft
• FIS-Broker StEP Gas

6.6 Stromnetz Berlin (Electricity)

What it is: Berlin's electricity distribution grid operator.

Key Facts:

• 36,000 km of cable (99% underground)
• High voltage (110 kV), medium voltage (10 kV), low voltage (0.4 kV)
• Dedicated Open Data portal: https://www.stromnetz.berlin/en/technology-innovations/open-data-stromnetz-berlin/

Open Data Available:

• Grid topology data on Berlin Open Data portal
• Cable information service for construction planning
• Smart City initiatives with data sharing

Sources:

• Stromnetz Berlin Open Data
• Stromnetz Berlin Facts

6.7 BEW Berliner Energie und Warme (District Heating)

What it is: Formerly Vattenfall Warme, recommunalized in May 2024. Operates largest urban district heating network in Western Europe.

Network:

• 2,000+ km of piping
• 700,000 apartments + 8,000 buildings connected
• Now publicly owned by State of Berlin

Data:

• Network maps may become more accessible under public ownership
• Thermal drone surveys would map actual network topology independent of operator data

Sources:

• BEW Wikipedia
• Berlin.de: State Takes Over Heating

6.8 BVG (U-Bahn Underground)

What it is: Berlin's public transit operator with 155.64 km of underground rail.

Key Data:

• 175 stations, ~80% underground
• Two distinct tunnel profiles: Kleinprofil (U1-U4) and Grossprofil (U5-U9)
• U7 (32 km, longest fully underground line in Germany) runs through Charlottenburg
• Wilmersdorfer Strasse U-Bahn station is on U7

Data Access:

• Station locations available in GTFS feeds and OSM
• Precise tunnel geometry: NOT publicly available (security-sensitive)
• General alignment known from route maps and Wikipedia documentation

Sources:

• Berlin U-Bahn Infrastructure (Wikipedia)

6.9 Berliner Feuerwehr Open Data

What it is: Daily fire brigade operational data published as open data.

Data Structure:

• CSV files on GitHub: github.com/Berliner-Feuerwehr/BF-Open-Data
• Categories: response times, mission keywords, resources
• Spatial resolution: LOR planning rooms (542 areas)
• Includes: fire station locations, service areas (INSPIRE model)
• Updated daily since June 2024

Infrastructure Application:

• Filter missions by keywords: Wasserrohrbruch, Gasaustritt, Kanalisation, Uberschwemmung
• Spatial clustering of water/gas incidents = infrastructure failure hotspots
• Response time patterns = access difficulty (narrow streets, deep infrastructure)

Access:

• Berlin Open Data: Einsatze
• GitHub Repository

6.10 Berlin LOR Demographic Data

What it is: Berlin divided into 542 planning rooms (Planungsraume) with demographic statistics.

Data:

• Population density, age structure, migration, social indicators
• Available as WFS, GeoJSON, Shapefile, RDF (Linked Open Data)
• 3 hierarchy levels: Prognoseraum (58) -> Bezirksregion (143) -> Planungsraum (542)

Infrastructure Application:

• Population density = infrastructure demand load
• Building age (from census) = infrastructure age proxy
• High-density + old buildings = high-stress underground networks

Access:

• WFS: daten.berlin.de LOR datasets
• RDF/LOD: github.com/berlinonline/lod-berlin-lor

7. Novel / Unconventional Sources

7.1 Insurance Claims Data (GDV)

What it is: The German Insurance Association (GDV) aggregates claims data from 460+ member companies.

Scale:

• 25+ million claims processed in 2024 (69,000+ per day in P&C alone)
• Risk statistics: fire, water damage, storms/hail, burglary
• Hochwasser-Check: risk assessment for 22.4 million German addresses
• 270,000 residential buildings in high-risk flood zones
• Only 46% of German households have flood insurance

Infrastructure Application:

• Water damage claims density = pipe failure frequency map
• Claims cost patterns: higher claims in areas with older infrastructure
• GDV Zonierungssystem (ZUERS): flood risk zones per address
• Data access: aggregated statistics publicly available; detailed claims data requires GDV partnership

Sources:

• GDV: Natural Hazards 2022
• GDV Homepage

7.2 Real Estate Platforms (ImmoScout24, WG-Gesucht)

What it is: Property listings containing implicit infrastructure data.

Infrastructure Signals:

• "Keller feucht" (damp basement) = high groundwater or sewer infiltration
• "Keller uberflutet" (flooded basement) = combined sewer overflow area
• Building age (Baujahr) from listing = infrastructure generation
• "Sanierung" (renovation) mentions = recent infrastructure upgrade
• Energy certificate data = building connection quality

Data Access:

• Web scraping possible (Apify scraper exists for ImmoScout24 with 50+ data points per listing)
• Terms of service may restrict automated collection
• Manual sampling for Wilmersdorfer Strasse area feasible

7.3 Energy Consumption Patterns

What it is: District heating and gas consumption data as infrastructure topology proxy.

Detection Method:

• High heating consumption + no gas connection = district heating customer (pipe route confirmation)
• Sudden consumption drops = disconnection or pipe failure
• Consumption clusters = network branch structure
• Energy Atlas Berlin: consumption and renewable potential for ~550,000 buildings

Access: Berlin Energy Atlas (partial open data); detailed consumption data is utility-confidential

7.4 Delivery Fleet Data (DHL, Amazon, Lieferando)

What it is: GPS + stop pattern data from millions of delivery trips.

Infrastructure Value:

• Stop duration anomalies = access difficulties (road works, blocked paths)
• Route diversions = construction/excavation zones
• DHL operates 27,000 e-vans globally; significant Berlin fleet
• 7 stops/hour (van), 10 stops/hour (bike) -- deviation = road condition issue
• Out-of-route miles (3-10% of total) often caused by infrastructure-related detours

Data Access:

• Proprietary fleet data -- requires corporate partnership
• [UNGROUNDED] DHL smart city data sharing interest not confirmed
• Amazon has not published urban infrastructure datasets

Sources:

• DHL: Last-Mile Logistics

7.5 Citizen Science Air Quality (Sensor.Community)

What it is: Global network of 14,000+ citizen-operated air quality sensors, originated in Germany (formerly Luftdaten).

What it Measures:

• Particulate matter (PM2.5, PM10)
• Temperature, humidity, atmospheric pressure
• Noise levels (some sensors)
• Does NOT currently measure methane -- would need additional CH4 sensors

Berlin Application:

• Dense sensor network in Berlin (Germany is largest contributor)
• Temperature anomalies near ground level could indicate district heating leaks
• Humidity spikes could correlate with water leak events
• Noise patterns near construction sites

Access: Open API, real-time data at https://sensor.community/

Sources:

• Sensor.Community

7.6 Sound Monitoring / Acoustic Fingerprinting

Berlin Noise Infrastructure:

• Strategic Noise Maps calculated every 5 years (latest: 2022)
• Covers road, transit, airport, industrial noise at facade level
• Available through Umweltatlas as WMS layers

Research Frontier: Urban Acoustic Fingerprinting

• Low-cost distributed acoustic sensor networks for real-time sound monitoring
• Deep learning classification of sound sources (construction, traffic, infrastructure events)
• Stadtlarm Project: field study of urban noise monitoring in German cities
• Potential: continuous acoustic monitoring could detect water hammer, pump failures, sewer flow changes

Sources:

• Berlin Strategic Noise Maps
• Stadtlarm Project (ResearchGate)
• MDPI: Low-Cost Urban Sound Monitoring

7.7 OpenStreetMap Infrastructure + Open Infrastructure Map

What it is: Community-contributed underground infrastructure data.

OSM Tags for Underground:

• man_made=pipeline (type: gas, water, oil, sewage)
• tunnel=yes (for covered/underground waterways)
• location=underground attribute
• inlet=* for storm drain inlets
• waterway=pressurised for underground pressurized water

Open Infrastructure Map:

• https://openinframap.org -- renders OSM electricity, telecoms, oil, gas infrastructure
• Global coverage, varies in completeness by region

Berlin Application:

• OSM Berlin has extensive utility mapping by local contributors
• Combined with official data for gap analysis
• Historical changesets = mapping activity timeline

Sources:

• Open Infrastructure Map
• OSM Pipeline Tag

Integration Architecture: How All Feeds Connect to Three.js Model

                    +--------------------------------------------------+
                    |           Three.js 3D Underground Model            |
                    |         (be.liviu.ai - Wilmersdorfer Str.)         |
                    +--------------------------------------------------+
                              |              |              |
                    +---------+    +---------+    +---------+
                    | TERRAIN |    | NETWORK |    | EVENTS  |
                    | LAYERS  |    | LAYERS  |    | LAYERS  |
                    +---------+    +---------+    +---------+
                         |              |              |
    +--------------------+----+  +------+------+  +----+--------------------+
    |                         |  |             |  |                         |
    v                         v  v             v  v                         v
+--------+  +---------+  +--------+  +--------+  +--------+  +---------+
|EGMS    |  |Umwelt-  |  |infrest |  |FIS-    |  |Feuer-  |  |Ordnungs-|
|InSAR   |  |atlas    |  |Utility |  |Broker  |  |wehr    |  |amt      |
|Points  |  |Ground-  |  |Plans   |  |WFS     |  |Incidents| |Reports  |
|        |  |water    |  |        |  |Layers  |  |        |  |         |
+--------+  +---------+  +--------+  +--------+  +--------+  +---------+
|Sentinel|  |Geology  |  |BWB     |  |Stromnetz| |Twitter/ |  |Mapillary|
|2 NDVI  |  |Soil     |  |Water   |  |Open    |  |Mastodon |  |CV       |
|        |  |         |  |Sewer   |  |Data    |  |NLP     |  |Detections|
+--------+  +---------+  +--------+  +--------+  +--------+  +---------+
|S-1 SAR |  |ERA5     |  |NBB/    |  |BVG     |  |Waze    |  |Street   |
|Coherence|  |Climate  |  |GASAG   |  |U-Bahn  |  |Constr. |  |View     |
|        |  |         |  |Gas     |  |        |  |Reports |  |TimeSeries|
+--------+  +---------+  +--------+  +--------+  +--------+  +---------+
|Copernicus| |Berlin  |  |BEW     |  |LOR     |  |Strava  |  |Sensor   |
|Urban    |  |Noise   |  |Heating |  |Demo-   |  |Metro   |  |Community|
|Atlas    |  |Maps    |  |        |  |graphics|  |Cycling |  |Air/Noise|
+--------+  +---------+  +--------+  +--------+  +--------+  +---------+

DRONE LAYERS (on-demand acquisition):
+--------+  +---------+  +--------+  +---------+  +--------+
|Thermal |  |LiDAR    |  |GPR     |  |Magneto- |  |SWIR    |
|LWIR    |  |Point    |  |Radar-  |  |meter    |  |Moisture|
|Ortho-  |  |Cloud    |  |grams   |  |Anomaly  |  |Map     |
|mosaic  |  |DTM/DSM  |  |        |  |Map      |  |        |
+--------+  +---------+  +--------+  +---------+  +--------+

Data Flow Pipeline

1. INGEST     -> GeoJSON/GeoTIFF/CSV/WFS -> Normalize to WGS84
2. TRANSFORM  -> Clip to AOI (Wilmersdorfer Str. bounding box)
                 Reproject to EPSG:25833 (ETRS89 / UTM 33N -- Berlin standard)
                 Convert depths to model Z-coordinates
3. STORE      -> PostGIS database (spatial queries)
                 GeoJSON files for Three.js direct load
                 COG (Cloud Optimized GeoTIFF) for raster layers
4. SERVE      -> Express.js API endpoints
                 Tile server for raster layers (TiTiler or similar)
                 WebSocket for real-time event feeds
5. RENDER     -> Three.js scene graph:
                 - THREE.Points for EGMS/InSAR measurement points
                 - THREE.Mesh for terrain, buildings, geological layers
                 - THREE.Line/TubeGeometry for pipe networks
                 - THREE.PlaneGeometry with DataTexture for raster overlays
                 - THREE.Sprite for event markers (incidents, reports)
                 - Temporal slider for time-series data

Three.js Specific Libraries

Data Licensing & Privacy Matrix

Quick Wins (Free, Available Now, Integrate in Days-Weeks)

1. EGMS InSAR Ground Motion -- Download Berlin tile from Copernicus, parse GeoPackage, visualize settlement points as colored spheres in Three.js. Immediate insight into which streets are subsiding and where underground infrastructure may be failing.

1. Berlin Umweltatlas Groundwater + Geology -- WMS layers directly loadable. Groundwater level as translucent blue plane in 3D model. Geology as cross-section. Both explain WHY infrastructure at specific locations behaves as it does.

1. Berliner Feuerwehr Open Data -- Download CSV from GitHub, filter by Charlottenburg-Wilmersdorf + water/gas/sewer keywords, map incidents as 3D event markers with temporal slider.

1. Stromnetz Berlin Open Data -- Grid topology from Berlin Open Data portal. 36,000 km underground cable = massive dataset for the 3D model.

1. Sentinel-2 NDVI Anomaly -- Run Sentinel Hub custom script for NDVI anomaly detection over Charlottenburg. Free processing. Drape result as terrain texture.

1. Mapillary Manhole Detection -- API query for detected manholes/utility covers along Wilmersdorfer Strasse. Free tier. Place as surface markers in 3D model.

1. Ordnungsamt-Online -- Monitor current reports for Charlottenburg infrastructure issues. Manual integration but immediate value.

1. Berlin LOR Demographics + FIS-Broker Layers -- WFS endpoints ready to consume. Population density, tree registry, soil contamination -- all contextual layers.

1. OpenStreetMap + Open Infrastructure Map -- Pipeline and utility data from OSM contributors. Immediate GeoJSON export.

1. Berlin Noise Maps -- WMS from Umweltatlas. Overlay on 3D model as ambient context layer.

Partnership Required (Valuable but Needs Agreements)

1. infrest Leitungsauskunft -- Registration + per-query fee, but returns ALL utility data from ALL operators for any location. Worth every cent for comprehensive underground mapping.

1. Waze for Cities -- Free partnership program, but requires application and approval. Returns real-time construction/utility work data in JSON feed.

1. Strava Metro -- Free for government/academic. Application process. Returns cycling movement patterns revealing road surface quality.

1. BEW Berlin District Heating Maps -- Recommunalized utility may be more open to data sharing under public ownership. Network topology for 2,000+ km of pipes.

1. BWB Sewer/Water Plans -- Available per-project through infrest, but comprehensive dataset would require formal partnership.

1. VMZ Berlin / VIZ Floating Car Data -- Traffic speed data available, raw FCD may require data agreement.

1. GDV Insurance Claims -- Aggregated public stats free; granular claims data by location requires insurance industry partnership.

1. E-Scooter Fleet Data (Tier) -- Tier HQ is in Berlin. Research partnership for anonymized accelerometer data would provide city-wide road surface quality map.

1. DAS via Deutsche Telekom -- Fiber optic sensing requires telecom partnership. Research pilot would be groundbreaking for urban infrastructure monitoring.

Research-Grade (Frontier, Needs Pilot Projects)

1. Drone-Mounted GPR -- Zond Aero 500 NG is commercially available but urban utility survey from drone is cutting-edge. Would need EASA Specific Category authorization, controlled survey design, and ground-truth validation.

1. Hyperspectral Methane Detection Drone -- SWIR cameras with CH4 lasers exist but have not been validated for urban gas network surveying at street scale.

1. DAS Fiber Optic Urban Sensing -- Demonstrated in research but no Berlin deployment for infrastructure monitoring. Joint research project with Telekom and TU Berlin would be novel.

1. Connected Vehicle Road Quality Aggregation -- Tesla/BMW data exists but no public access. Research agreement with OEM or via HERE Technologies would unlock mm-precision road surface data.

1. Deep Learning on Street View Time Series -- CityPulse framework proven in research. Applying to Berlin Street View archives to reconstruct excavation history since 2007 would be novel and publishable.

1. Acoustic Leak Detection Drones -- CRYSound 2626G exists but urban deployment over active streets faces regulatory and noise-floor challenges.

1. LoRaWAN Manhole Sensor Network -- Technology proven globally, but deploying a purpose-built sensor network for Wilmersdorfer Strasse would be a pilot project.

1. Insurance Claims Spatial Analysis -- Correlating water damage claims density with pipe age/material would be academically novel. Requires GDV data partnership.

Appendix: Data Product Quick Reference

Research conducted 2026-03-25 by 7-expert panel. All URLs verified at time of research. Claims marked [UNGROUNDED] require additional verification before implementation.