4.1 Typical System Topology

The monitoring system is organized into three distinct layers: the Field Layer containing all measurement and data acquisition equipment at monitoring points; the Network Layer providing secure, resilient data transport; and the Platform Layer delivering all data management, alarming, and reporting functions. This three-layer architecture provides clear separation of concerns, enabling independent scaling and maintenance of each layer without disrupting the others.

Groundwater well stations (GW-1 through GW-n) connect to the platform via cellular LTE with VPN encryption, while drinking water plant nodes (DW-A, DW-B) use fiber optic connections for higher bandwidth and lower latency. All field devices communicate through a secure VPN gateway before reaching the platform, ensuring that no field device is directly exposed to the internet. The firewall enforces strict ingress/egress rules, and all access is logged for audit purposes.

System Topology Diagram

Figure 4.1: Typical System Topology — Three-Layer Architecture with Field Stations, Network Layer, and Cloud Platform

The topology supports both cloud-hosted and on-premises platform deployments. For cloud deployments, the platform layer runs in a managed cloud environment with geographic redundancy. For on-premises deployments, the platform server is hosted within the customer's network, with the VPN gateway providing secure access for remote field stations. Hybrid configurations — where some data is processed at the edge and only aggregated results are sent to the cloud — are also supported for bandwidth-constrained deployments.

Station Classification and Network Assignment

Station Class Typical Application Network Path Data Interval Alarm Latency Target
GW-1 High-risk groundwater wells (industrial, contamination risk) LTE → VPN → Platform 15 min ≤30 min
GW-2 Standard groundwater monitoring wells LTE → VPN → Platform 60 min ≤2 h
GW-3 Remote baseline wells (low risk) LTE or LoRa → Gateway → Platform 4 h ≤8 h
DW-A Critical drinking water nodes (intake, disinfection) Fiber → Industrial Switch → Platform 1 min (continuous) ≤3 min
DW-B Standard drinking water nodes (distribution, storage) Fiber or Ethernet → Platform 5 min ≤15 min

4.2 Network Architecture & Security

Network architecture for environmental monitoring systems must balance accessibility (operators need to access data from multiple locations) with security (field devices must not be exposed to cyber threats). The recommended architecture uses a defense-in-depth approach with multiple security layers, ensuring that a breach of any single layer does not compromise the entire system.

All field-to-platform communications use TLS 1.2 or higher encryption. Field devices authenticate using device certificates, not shared passwords. The VPN gateway terminates all field device connections and presents them to the platform as a single, controlled interface. Role-based access control (RBAC) ensures that operators can only access data and functions appropriate to their role, and all access attempts are logged with timestamps and user identity.

Security Layer Mechanism Implementation Verification
Physical Locked cabinets, tamper switches, CCTV IP65+ enclosures, tamper seals, vibration sensors Annual physical security audit
Network VPN tunnel, firewall, network segmentation IPSec/OpenVPN, stateful firewall, VLAN separation Penetration test, firewall rule review
Transport TLS encryption, certificate authentication TLS 1.2+, device certificates, MQTT over TLS Certificate expiry monitoring
Application RBAC, audit logging, session management Role-based permissions, immutable audit log, session timeout Access log review, privilege audit
Data Data integrity, backup, retention Checksums, encrypted backups, retention policy enforcement Backup restoration test, data integrity check

4.3 Equipment Wiring Design

Proper wiring design is critical for system reliability and maintainability. Poor wiring — incorrect wire gauges, missing shields, inadequate surge protection, or improper grounding — is responsible for a significant proportion of field failures. The wiring diagram below shows the complete internal wiring of a standard groundwater monitoring station cabinet, including power, signal, and communications connections.

Cabinet Wiring Diagram

Figure 4.2: Groundwater Monitoring Station Cabinet Wiring Diagram — Complete Power, Signal, and Communications Connections with Color-Coded Wire Specifications

The wiring design follows a strict color-coding convention to facilitate troubleshooting and maintenance. Red wires carry positive DC power, black wires carry negative/ground, blue wires carry signal/data, yellow wires are shields, and green/yellow wires are protective earth. All wire sizes are specified based on current carrying capacity and voltage drop calculations, not arbitrary selection. Shield termination — grounding at one end only — is critical for preventing ground loops that introduce noise into sensor signals.

Wiring Design Rules

  1. Wire Sizing: Power wires sized for ≤3% voltage drop at full load; signal wires minimum 22 AWG shielded twisted pair.
  2. Shield Termination: Shield grounded at RTU end only; sensor end left floating to prevent ground loops.
  3. Surge Protection: SPD on all power entries (Type 2 minimum); signal SPD on all external cable entries.
  4. Cable Separation: Power and signal cables routed in separate conduits or with minimum 150 mm separation.
  5. Drip Loops: All cables entering the cabinet must have a drip loop below the entry point to prevent water ingress.
  6. Labeling: All wires labeled at both ends with wire number, source, and destination.
  7. Terminal Blocks: Use screw-type or spring-cage terminal blocks rated for the wire gauge; no wire-to-wire connections inside the cabinet.
  8. Grounding: Chassis ground bar connected to earth ground with minimum 6 AWG conductor; earth resistance <10 Ω verified at commissioning.

4.4 Power Architecture

Power Type Application Key Components Sizing Rule Autonomy Target
Solar + Battery Remote wells without mains PV panel, MPPT controller, LiFePO4 battery Solar: 3× daily load; Battery: 5× daily load ≥5 days (worst-case irradiance)
Mains + UPS Plant nodes, critical stations AC input, UPS, DC PSU UPS: 2× normal load for 2 h minimum ≥2 h (mains failure)
Hybrid (Mains + Solar) Semi-remote stations with unreliable mains Both above, automatic transfer switch Solar sized for mains-off periods ≥3 days (mains failure)
Solar Sizing Warning: Always use worst-case monthly irradiance data for the site location, not annual averages. In many regions, December or January irradiance can be 40–60% below the annual average. Undersized solar systems are the leading cause of winter data gaps in remote monitoring networks.