• The color-changing of the orangutan’s face is now automated by fetching from the overall_score in the database

• Created backend login with api connectors and data simulations
• Implemented metrics and index calculations based on WHO and EU targets (aligned with Green City Accord Indicators Guidebook (2024))
• Implemented Database Infrastrucure in Supabase

• New concept using a video to project animal faces
• Those faces can change color (red/orange/green) based on the index of Aarhus Green City Index (WIP, not yet implemented)
• Put the 3D concept in the legacy folder

• A website concept with a 3D object of a low-poly tree
• Featuring window tracking with phone

The Aarhus Green City Index (AGCI) is a comprehensive environmental monitoring system that collects, processes, and visualizes sustainability metrics for the city of Aarhus, Denmark. The system generates an environmental health score across five key dimensions:

• Air Quality
• Water Management
• Nature & Biodiversity
• Waste & Circular Economy
• Noise Pollution

The results are visualized through an interactive installation featuring animal faces that change colors (red/orange/green) based on the environmental scores.

agci/
├── backend/
│   ├── collectors/              # Data collectors for each environmental dimension
│   ├── pipeline/                # Data processing and index calculation
│   ├── storage/                 # Database interaction (Supabase)
│   ├── installation/            # Physical installation controllers
│   └── tests/                   # Test scripts
├── data/processed/              # Storage for processed index data
├── legacy/                      # Previous version of the frontend (3D tree)
└── [frontend files]             # Current frontend (HTML, CSS, JavaScript)

The system uses Supabase as its backend database with the following tables:

1. raw_metrics:

   • id (uuid): Primary key
   • dimension (text): Environmental dimension (air, water, nature, waste, noise)
   • metric_name (text): Metric identifier
   • value (float): Raw measured value
   • unit (text): Measurement unit
   • source (text): Data source (API or Simulated)
   • collection_timestamp (datetime): Collection time

2. normalized_scores:

   • id (uuid): Primary key
   • dimension (text): Environmental dimension
   • metric_name (text): Metric identifier
   • raw_value (float): Original raw value
   • normalized_score (float): Score from 0-100
   • calculation_method (text): Method used for normalization
   • date (date): Calculation date

3. dimension_scores:

   • id (uuid): Primary key
   • date (date): Calculation date
   • dimension (text): Environmental dimension
   • score (float): Normalized dimension score

4. green_city_index:

   • id (uuid): Primary key
   • date (date): Calculation date
   • overall_score (float): Overall index score
   • air_score (float): Air quality score
   • water_score (float): Water management score
   • nature_score (float): Nature & biodiversity score
   • waste_score (float): Waste & circular economy score
   • noise_score (float): Noise pollution score
   • target_score (float): Target score for comparison

The system collects data for each environmental dimension using specialized collector classes:

Class: AirQualityCollector

Data Source: Real-time air quality data from Open-Meteo API (https://air-quality-api.open-meteo.com/v1/air-quality)

Metrics:

• PM2.5 particulate matter concentration (μg/m³)
• PM10 particulate matter concentration (μg/m³)
• NO2 nitrogen dioxide concentration (μg/m³)

Normalization:

• PM2.5: 100 points at 5μg/m³ (WHO guideline), 0 points at 25μg/m³ (EU limit)
• PM10: 100 points at 15μg/m³ (WHO guideline), 0 points at 40μg/m³ (EU limit)
• NO2: 100 points at 10μg/m³ (WHO guideline), 0 points at 40μg/m³ (EU limit)

Class: WaterManagementSimulator

Data Source: Simulated data based on Danish water consumption patterns

Metrics:

• Water consumption (liters per capita per day)
• Infrastructure Leakage Index (ILI)
• Treatment compliance with EU directives (%)

Normalization:

• Water consumption: 100 points at 100L/capita/day, 0 points at 200L/capita/day
• ILI: 100 points at 1.0 (no leakage), 0 points at 6.0
• Treatment compliance: 100 points at 100% compliance, 0 points below 50% compliance

Class: NatureBiodiversitySimulator

Data Source: Simulated data with seasonal variations

Metrics:

• Protected area percentage (%)
• Tree canopy coverage (%)
• Bird species change percentage (%)

Normalization:

• Protected area: 100 points at 10%+, 0 points at 0%
• Tree canopy: 100 points at 30%+, 0 points at 5% or below
• Bird species change: 100 points at +20%, 0 points at -20%

Class: WasteSimulator

Data Source: Simulated waste data with seasonal and holiday patterns

Metrics:

• Waste generation per capita (tonnes/year)
• Recycling rate (%)
• Landfill rate (%)

Normalization:

• Waste per capita: 100 points at 0.2 tonnes/year, 0 points at 0.7 tonnes/year
• Recycling rate: Direct percentage (100% recycling = 100 points)
• Landfill rate: 100 points at 0%, 0 points at 60%+

Class: NoiseSimulator

Data Source: Simulated noise data with daily and seasonal patterns

Metrics:

• Population exposed to Lden ≥ 55 dB (%)
• Population exposed to Lnight ≥ 50 dB (%)
• Population with high sleep disturbance (%)

Normalization:

• Lden exposure: 100 points at 0%, 0 points at 40%+
• Lnight exposure: 100 points at 0%, 0 points at 40%+
• Sleep disturbance: 100 points at 0%, 0 points at 25%+

The Green City Index is calculated in three steps:

1. Raw Data Collection: Each collector generates or retrieves metrics for its environmental dimension

2. Normalization: Raw metrics are converted to normalized scores (0-100) based on their respective thresholds

3. Index Calculation:

   • Each dimension score is the average of its normalized metrics
   • The overall index is a weighted average of the five dimension scores
   • Default weights are 20% per dimension

1. Database Type Mismatch: There's a type mismatch when inserting into the green_city_index table. The error occurs because floating-point values are being sent to an integer column:

   Failed to store index: {'code': '22P02', 'details': None, 'hint': None, 'message': 'invalid input syntax for type integer: "100.0"'}
   

2. Missing Dimension Scores: The code attempts to store individual dimension scores, but there's no evidence in the logs that the store_dimension_score() method is being called correctly.

The frontend visualizes the Green City Index through:

1. Animal Faces: Digital projections that change expressions and colors based on environmental scores
2. Color System:
   • Green (70-100): Excellent environmental conditions
   • Orange (40-69): Moderate environmental conditions
   • Red (0-39): Poor environmental conditions
3. Background Effects: Dynamic sky effects that change based on time of day and index values

• Python 3.8+
• Supabase account with project set up
• Environment variables for Supabase access:

  • SUPABASE_URL

  • SUPABASE_KEY

  • in the root of the directory create .env file with these two environment variables

1. Clone the repository
2. Install dependencies: pip install -r backend/requirements.txt
3. Configure Supabase credentials as environment variables
4. Run initial setup: python -m backend.storage.setup_database
5. Execute data collection: python -m backend.pipeline.green_city_index

• Test Supabase connection: python -m backend.tests.test_supabase_connection
• Test simulations: python -m backend.tests.test_simulation

1. Database Connection Errors:

   • Verify Supabase credentials are correctly set
   • Check if Supabase service is running

2. Missing Data:

   • Verify that all collectors are running properly

3. Type Mismatch in Database:

   • Ensure Supabase table schemas match expected types:
     • Score columns should be float8 or numeric not integer
   • Adjust normalization functions if needed