Technology Showcase (Alpha) — This platform demonstrates AI-driven water quality predictions. Not a substitute for official monitoring. Full disclaimer
About PathoWatch

Water Quality Intelligence,
Powered by Data

PathoWatch is a research platform from DataPond, an Israeli water-tech company building AI tools for real-time bio-contamination monitoring of surface waters.

Our Mission

Cleaner, Safer Water Through Data

Microbial contamination is a public health issue affecting millions, but lab testing is slow and expensive. We're building the predictive layer that closes the gap.

Continuous Monitoring

Daily ingestion from EPA, NASA, USGS, and weather APIs - over 24,000+ active US monitoring sites covered.

AI Predictions

Machine learning models trained on hundreds of thousands of historical lab records, aligned with EPA contamination thresholds.

Coverage Gap

13,500+ US sites measure water parameters but lack microbial testing. We bring affordable, predictive coverage to all of them.

DataPond

About the Company

DataPond is an Israeli water-tech company developing AI-powered solutions for real-time microbial bio-contamination detection in surface water. Our mission is to ensure cleaner, safer water by enabling affordable, scalable, and continuous contamination monitoring - improving public health and regulatory compliance.

DataPond's current focus is bio-contamination detection in surface water - rivers, lakes, and coastal beaches. PathoWatch is the public-facing research demonstration of this capability across the United States.

Visit data-pond.co
Methodology

How We Measure Quality

PathoWatch predicts whether a site is above or below the EPA bacterial contamination threshold. Every site maps to a single binary decision - safe or high risk - using the same threshold our model was trained on.

E. coli
>235 CFU/100mL
Freshwater recreational
Source: EPA 2012 RWQC
Fecal Coliform
>500 CFU/100mL
Surface water
Source: EPA legacy
Enterococcus
>130 CFU/100mL
Marine recreational
Source: EPA 2012 BAV

Binary classification

Every site is classified as safe or high risk using the EPA bacterial threshold. This is the same decision rule our ML models were trained and validated against - we don't invent intermediate categories.

What we measure

Our models correctly flag most contamination events while keeping false alarms low. Performance varies by region, sampling density, and seasonality. Sites with sparse recent data carry higher uncertainty - we flag these clearly.

Data sources

We combine EPA Water Quality Portal lab records, NASA VIIRS satellite reflectance for coastal beaches, NASA POWER daily weather, and USGS real-time streamflow. All inputs are refreshed automatically each day.

Technology Showcase (Alpha)

PathoWatch is a technology demonstration built by DataPond to showcase AI-driven water quality prediction capabilities. It is designed for water professionals, researchers, and potential partners evaluating our technology.

This platform is not a certified monitoring system and should not be used as a substitute for official water quality testing, public health advisories, or recreational safety decisions.

Predictions are generated by machine learning models trained on historical data and carry inherent uncertainty. Model performance varies by region, data availability, and environmental conditions. For official water quality information, always consult your local environmental or public health authority.

Want to Get in Touch?

Reach out to discuss our beta program, partnerships, or how PathoWatch can support your water quality program.