Full Report
Water and wastewater systems, like other industrial and manufacturing facilities use pumps, valves, motors,drives, relays, analyzers, process sensors, engineering workstations, operator displays, and communication networks. Because these systems depend on electronic communications to monitor and control physical processes, control system cyber incidents occur whenever electronically communicated process information becomes corrupted, delayed, unavailable, or untrustworthy. Network […]
Analysis Summary
# Best Practices: Integrity of Process Information in Water and Wastewater Systems
## Overview
These practices address a critical gap in traditional cybersecurity: the transition from focusing solely on **network security** to ensuring the **integrity and trustworthiness of process information**. In water and wastewater systems, physical safety depends on the accuracy of data from sensors, pumps, and valves. These guidelines focus on preventing incidents where process information is corrupted, delayed, or untrustworthy, regardless of whether a network breach has occurred.
## Key Recommendations
### Immediate Actions
1. **Audit Process Data Points:** Identify all critical physical process sensors (pressure, flow, chemical levels) and actuators (pumps, valves) that lack manual verification methods.
2. **Establish "Ground Truth" Baselines:** Verify that operator displays (HMIs) accurately reflect physical reality by performing field checks on analog gauges vs. digital readouts.
3. **Implement Alarm Rationalization:** Review control system alarms to ensure delays or "frozen" data (stale data) trigger immediate alerts to operators.
### Short-term Improvements (1-3 months)
1. **Cross-Check Sensor Logic:** Implement logic within PLCs or engineering workstations that compares related data points (e.g., if a pump is "on," flow sensors should show movement) to detect "corrupted" but seemingly valid data.
2. **Evaluate Sensor-to-Controller Integrity:** Review the physical and electronic paths of process sensors to identify points where signals can be manipulated before they reach the network.
3. **Update Incident Response Plans:** Include scenarios where the network is "clean" but process data is untrustworthy (e.g., a sensor failure or out-of-band manipulation).
### Long-term Strategy (3+ months)
1. **Adopt Process-Informed Engineering:** Integrate cybersecurity into the physical engineering design phase, ensuring systems remain safe even if electronic communication is compromised.
2. **Deploy Hardware-Level Monitoring:** Invest in solutions that monitor the physics of the process (Level 0/Level 1 devices) rather than just Level 2/3 network traffic.
3. **Redundancy of Sensing Technologies:** Deploy diverse sensing technologies (e.g., using both ultrasonic and pressure-based level sensing) to make it harder for a single corrupted data stream to cause a safety incident.
## Implementation Guidance
### For Small Organizations
- **Focus on Manual Overrides:** Ensure that staff are trained to operate valves and pumps manually if electronic displays appear "frozen" or untrustworthy.
- **Physical Inspections:** Increase the frequency of physical site visits to verify sensor readings.
### For Medium Organizations
- **PLC Logic Hardening:** Program "sanity checks" into PLCs to detect impossible physical states (e.g., a tank filling faster than the pump capacity allows).
- **Basic Network Segmentation:** Separate the process control traffic from general business traffic to reduce latency and data "delays."
### For Large Enterprises
- **Digital Twin Integration:** Use digital twins to run real-time simulations of the physical process; if the physical system deviates from the model, trigger an integrity alert.
- **Supply Chain Integrity:** Require vendors of sensors and analyzers to provide documentation on the integrity and "trustworthiness" of the data their devices generate.
## Configuration Examples
*While the article emphasizes a shift away from just "configurations" toward "integrity," the following logic-based configuration is recommended:*
**Example: Stale Data Detection (PLC Logic)**
pascal
// Pseudocode for detecting a "Frozen" Sensor
IF (Current_Sensor_Value == Previous_Sensor_Value) AND (System_Status == ACTIVE) THEN
Increment_Stale_Counter;
IF (Stale_Counter > Max_Threshold) THEN
Trigger_Integrity_Alarm("Sensor Data Unchanging - Potential Corruption");
## Compliance Alignment
- **NIST SP 800-82:** Guide to Industrial Control Systems (ICS) Security.
- **ISA/IEC 62443:** Security for Industrial Automation and Control Systems (specifically Data Integrity requirements).
- **AWWA G430:** Security Practices for Operations and Management.
- **Cyber-Informed Engineering (CIE):** DOE framework for functional resilience.
## Common Pitfalls to Avoid
- **"Network Blindness":** Assuming that because the firewall shows no intrusions, the process data is accurate.
- **Ignoring Level 0/1:** Focusing all security budget on IT-centric tools (SIEM/IDS) while ignoring the sensors and actuators (Level 0/1).
- **Trusting "Legal" Values:** Failing to realize that a value can be within a "normal" range but still be physically impossible or incorrect for the current system state.
## Resources
- **ControlGlobal (Industrial Control Insights):** hxxps://controlglobal[.]com/55388935
- **Applied Control Solutions (ACS):** hxxp://www[.]realtimeacs[.]com
- **CISA Water and Wastewater Sector Resources:** hxxps://www[.]cisa[.]gov/topics/critical-infrastructure-sectors/water-and-wastewater-systems