Network Best Practices
Guidelines for designing and operating effective seismic networks with Grillo sensors.
Network design principles
Coverage vs density trade-off
Consider your goals:
| Goal | Strategy |
|---|---|
| Wide area monitoring | Spread sensors across the region |
| Precise location | Higher sensor density |
| Early warning | Sensors near potential sources |
| Building monitoring | Sensors throughout structure |
Minimum sensors for detection
For earthquake detection (with Events enabled):
| Detection goal | Minimum sensors |
|---|---|
| Basic detection | 3-4 sensors |
| Location accuracy | 6+ sensors |
| Magnitude accuracy | 10+ sensors |
| High precision | 20+ sensors |
Sensor placement strategies
Geographic distribution
For regional networks:
- Space sensors 10-50 km apart depending on goals
- Ensure coverage of populated areas
- Place sensors around potential earthquake sources
- Consider geographic barriers (mountains, rivers)
For urban networks:
- Denser spacing in city centers
- Coverage of critical infrastructure
- Multiple sensors per neighborhood for redundancy
Strategic locations
Prioritize sensors at:
- Emergency response facilities
- Schools and hospitals
- Government buildings
- Transportation hubs
- Critical infrastructure
Avoid problematic locations
- Near major highways (traffic noise)
- Inside industrial facilities (machinery vibration)
- On unstable ground (poor coupling)
- Upper floors of buildings (amplified motion)
Data quality considerations
Site selection checklist
For each sensor location:
- Ground floor or basement level
- Solid foundation (concrete preferred)
- Away from HVAC equipment (>3 meters)
- Away from heavy traffic
- Stable power supply
- Reliable network connectivity
- Accessible for maintenance
Noise sources to avoid
| Source | Minimum distance |
|---|---|
| HVAC units | 3 meters |
| Heavy machinery | 10 meters |
| Major roads | 20 meters |
| Railway lines | 50 meters |
Redundancy and reliability
Network redundancy
Plan for failures:
- No single points of failure
- Overlapping coverage where possible
- Mix of connectivity types (WiFi + cellular)
- Backup power where critical
Connectivity diversity
For robust networks:
- Use Grillo One with Ethernet where possible
- Deploy Grillo Pulse with cellular in remote areas
- Ensure multiple ISPs across the network
- Test failover scenarios
Network organization
Naming conventions
Establish consistent naming:
Station codes:
- Use FDSN-compatible format when possible
- Example: NET.STA.LOC.CHA
- Keep codes meaningful but concise
Network names:
- Geographic: "San Francisco Bay Network"
- Organizational: "University Seismic Network"
- Purpose: "EEW Network - Phase 1"
Documentation
Maintain records of:
- Sensor locations and installation details
- Site conditions and photos
- Network topology diagram
- Maintenance history
- Contact information for each site
Operational best practices
Monitoring
Regular checks:
- Daily: Review sensor status dashboard
- Weekly: Check data quality indicators
- Monthly: Verify all sensors reporting
- Quarterly: Physical site inspections
Maintenance
Scheduled activities:
- Clean sensors annually
- Verify mounting stability
- Update firmware when available
- Replace batteries (Pulse) as needed
- Document all maintenance
Incident response
When sensors go offline:
- Check dashboard for error indicators
- Attempt remote restart if available
- Check network connectivity
- Schedule site visit if needed
- Document issue and resolution
Scaling your network
Starting small
Begin with:
- 3-5 sensors in key locations
- Prove the concept works
- Learn operational requirements
- Refine processes
Growing strategically
Expand by:
- Filling coverage gaps
- Adding density in priority areas
- Responding to user feedback
- Following a documented plan
Enterprise considerations
For large deployments:
- Centralized management
- Automated monitoring
- Tiered support structure
- Regular reporting
Common mistakes to avoid
Planning mistakes
- Insufficient density - Too few sensors for detection goals
- Poor site selection - Noisy locations degrade data quality
- No redundancy - Single points of failure
- Unclear objectives - Not knowing what you want to detect
Operational mistakes
- Neglecting maintenance - Sensors fail without attention
- Ignoring alerts - Missing important signals
- Poor documentation - Can't troubleshoot without records
- No testing - Assumptions about system behavior
Example network designs
Small community network (5 sensors)
Purpose: Community awareness
Coverage: 10 km radius
Sensors:
- 1x Fire station (central)
- 1x School (north)
- 1x Library (south)
- 1x Community center (east)
- 1x Private residence (west)
Urban monitoring network (20 sensors)
Purpose: Early warning
Coverage: City center + suburbs
Sensors:
- 5x Downtown (1 km spacing)
- 5x North district
- 5x South district
- 3x Industrial area
- 2x Port facilities
Research network (50+ sensors)
Purpose: Seismic research
Coverage: Fault zone monitoring
Sensors:
- Dense arrays near fault
- Regional background stations
- Reference stations on bedrock
- Borehole sensors (if applicable)