SDA-for-SDGs Keynote · University of Toronto · 2026

Measuring Global
Water Security.

Sensors, satellites, data, policy, and business models — what's now measurable, and what each unlocks.

Evan Thomas
Professor, CU Boulder · Mortenson Center in Global Engineering & Resilience
CEO, Virridy Inc.
The Challenge

Water Quality Impairment at Scale

4B
People

drink microbially contaminated water

60%
of GDP

threatened by water insecurity

10%
of Emissions

half from water management, half from unmanaged human wastewater

50%
U.S. Rivers

fail Clean Water Act standards

Root cause: The gap persists because it's invisible, under-funded, and operationally orphaned — the same conditions the global WASH sector has been wrestling with for 25 years, with mixed results.
The Challenge · Global E. coli Data
10MObservations
180KSites
40Years
7Databases
Pillar 1 · From Visibility to Accountability
The Lume in Action
Seine River · Paris
Pastoralist Boreholes · Kenya
Field Testing · Boulder Creek
In-Situ Monitoring · Bridge Deployment
Mortenson Center in Global Engineering — Overview
19

Global Mortality Among Children Under 5

Total deaths by cause, 1990–2021. Sources: IHME GBD 2021, UNICEF IGME 2024, WHO Global Health Estimates.

This chart shows the major causes of death in children under 5 from 1990 to 2021. The good news is that nearly every cause has declined substantially — lower respiratory infections dropped from over 10 million to under 3 million deaths. But notice malaria ticked back up recently, and the total is still nearly 5 million children per year. These are largely preventable deaths from causes we know how to address with existing technologies and interventions.
21

Extreme Poverty by Region

People living below $2.15/day (2017 PPP). Sub-Saharan Africa is the only region where absolute numbers are rising. Source: World Bank, 2024. By 2030, 9 in 10 people in extreme poverty will live in Sub-Saharan Africa.

This chart tells one of the most important stories in global health. East Asia, driven largely by China, saw an extraordinary decline in extreme poverty — from over a billion people to about 29 million. South Asia has also made remarkable progress. But Sub-Saharan Africa is the only region where the absolute number of people in extreme poverty is rising. By 2030, nine out of ten people living in extreme poverty will be in Sub-Saharan Africa. This is the geographic concentration that drives most of the disease burden we have been looking at.
22

Per Capita CO2 Emissions

High-income countries have the highest per capita emissions, while low-income countries with the greatest disease burden contribute the least.

Now look at who is causing climate change versus who is bearing the burden of disease. High-income countries have the highest per-capita CO2 emissions, while low-income countries with the greatest disease burden contribute the least. This is the equity dimension of climate and health — the populations least responsible for emissions are the most vulnerable to their health consequences, from heat stress to vector-borne disease expansion to food insecurity.
10

Death Rate from Diarrhoeal Diseases

Annual deaths per 100,000, all ages. Disease burden tracks fecal contamination — Sub-Saharan Africa and South Asia carry it. Source: IHME GBD via Our World in Data.

64

Rwanda — Country Context

Key Indicators

  • Population: 14 million (2024)
  • GDP per capita: ~$800
  • Under-5 mortality: 38 per 1,000 live births
  • Over 80% rely on firewood as primary fuel
  • Most rural households drink untreated water

Study Location

Western Province, Rwanda — 96 sectors cluster-randomized, reaching 101,000 households with water filters and improved cookstoves.

Rwanda field Rwanda community
Let me set the scene. Rwanda is a small, densely populated country in East Africa with about 14 million people and a GDP per capita of roughly 800 dollars. Over 80% of the population relies on firewood for cooking, and most rural households drink untreated surface water. Under-5 mortality is 38 per 1,000 live births, which is dramatically better than it was 20 years ago but still 15 times higher than Norway. Our study was based in Western Province, where we cluster-randomized 96 sectors to reach 101,000 households.
65

Programme Implementation

Child with LifeStraw filter

LifeStraw Family 2.0 water filter in household

EcoZoom cookstove

EcoZoom Dura improved cookstove in use

Sensor monitoring

Electronic sensor monitoring deployment

Here you can see the three main components of the programme. On the left, the LifeStraw Family 2.0 water filter, which is a gravity-fed hollow fiber membrane filter that removes bacteria and parasites. In the middle, the EcoZoom Dura improved cookstove, designed to burn wood more efficiently and reduce smoke. And on the right, the electronic sensors we deployed to objectively monitor whether people were actually using these technologies. That third component turned out to be the most important contribution.
Community health worker training in Rwanda
67

Key Results — Tubeho Neza Trial

Health Outcomes (Children Under 5)

  • 29% reduction in 7-day prevalence of diarrhea
  • 25% reduction in acute respiratory infections
  • 97.5% reduction in fecal contamination of drinking water
  • 38% reduction in cryptosporidium seroconversion

Air Quality — A Cautionary Finding

  • Personal PM2.5 exposure remained unchanged despite improved cookstoves
  • Stove stacking: traditional fire use increased from 24% to 49% over study period

The Adherence Problem

  • Self-reported filter use: 67%
  • Sensor-detected filter use: 37%
  • Self-reported stove use: 84%
  • Sensor-detected stove use: 37%
  • Reported use declined: 75% → 68% → 65% across survey rounds

Economics

  • 5-year programme cost: ~$12 million
  • Estimated 5-year benefit: >$66 million
  • Fuelwood savings: 65,000 tons — enough to reverse regional deforestation

Sources: Kirby, Nagel et al. (2019) PLoS Medicine; Thomas et al. (2018) Lancet Planetary Health; Thomas (2019) The Conversation

Here are the headline results, and they tell a nuanced story. On the water side, we saw a 97.5% reduction in fecal contamination and a 29% reduction in childhood diarrhea. Those are strong results. But on the air quality side, personal PM2.5 exposure was unchanged despite the improved cookstoves, because of stove stacking. Households kept using their traditional fires alongside the new stoves. And the adherence data is perhaps the most important finding: self-reported filter use was 67%, but sensor-detected use was only 37%. People were telling us they were using the filter almost twice as much as they actually were. This gap fundamentally changed how I think about intervention evaluation.
68

Technology & Digital Monitoring

Water Filters

  • LifeStraw Family 2.0 household water filters
  • Significant microbiological effectiveness reducing E. coli contamination
  • Free distribution with carbon waiver for credit generation

Remote Sensing Innovation

  • Electronic sensors remotely transmitting usage data
  • Sensor-reported use was substantially lower than self-reported use
  • Demonstrated critical value of objective digital monitoring
  • Published in ACS Environmental Science & Technology

Carbon Credit Model

  • Pay-for-performance model funded by voluntary carbon credits
  • Health, livelihood, and environmental benefits substantially outweighed costs
  • Fuel savings and averted healthcare costs = largest economic gains
Sensor technology
The technology and monitoring story is really the engineering contribution of this project. The LifeStraw filters were highly effective at removing pathogens when used, which is the efficacy story. But the sensor data revealed a very different effectiveness story. We published these findings in ACS Environmental Science and Technology, and they demonstrated that self-reported data systematically overestimates use. The carbon credit model was also innovative because it created a pay-for-performance financing mechanism. Health and economic benefits substantially outweighed costs, with fuel savings and averted healthcare being the largest gains.
Bar Graph · Cups Filled
Water Pump · Functionality
Sensor slide 1
Sensor slide 2
Sensor slide 3
Sensor slide 4
Sensor slide 5
Sensor slide 6
Sensor slide 7
Sensor slide 8
Sensor slide 9
Sensor slide 10
Sensor slide 11
Sensor slide 12
Sensor slide 13
Sensor slide 14
Sensor slide 15
Sensor slide 16
Sensor slide 17
Sensor slide 18
virridy.com
Our Solution

Global Water Security Solutions

Water sensor technology
💧

Lume Sensors

Three optical modes — TLF (E. coli), Chlorophyll-a (algae), FDOM (organics). 75%+ accuracy, >94% categorical. 1-year battery life.

Green energy carbon
🌱

Water for Carbon

Drinking water treatment, precision irrigation, and watershed restoration. Gold Standard, Verra & Regen certified. 29-49% diarrhea reduction.

Global network
🌍

Global Portfolio

10 active projects across 12 countries. IoT-verified impact — no self-reporting. 5.6x cost-benefit ratio across health, livelihood, and environment.

virridycarbon.com
Portfolio

High Integrity Water Projects Across Africa

thelume.ai
Product

Meet the Lume

Continuous water quality monitoring — for the cost of a single grab sample.

Three interchangeable optical modes: TLF (280/350nm, E. coli), Cl-A (470/680nm, algae/HABs), FDOM (365/480nm, dissolved organics). Plus turbidity and temperature.

Sampling: 30s–24h intervals. Cellular + satellite connectivity. 1-year battery on hourly sampling. Solar or wall charging. No calibration or maintenance required.

Key Applications

  • Drinking water source protection
  • Recreational water & beach advisories
  • Wastewater discharge & CSO monitoring
  • Agricultural return flow monitoring
  • Reservoir and intake monitoring

$200/month — includes device, connectivity, cloud dashboard, API, and firmware updates.

thelume.ai

Field Testing the Lume

Researchers deploying the Lume sensor in natural stream environments for real-time E. coli monitoring and validation studies.

thelume.ai
Virridy

Seine River, Paris

Virridy’s Lume sensors monitoring water quality for recreational swimming safety along the Seine River.

thelume.ai
Technology

Hardware & Analytics

Sensor Capabilities

  • TLF: E. coli & microbial contamination (280/350 nm)
  • Cl-A: algal biomass & bloom detection (470/680 nm)
  • FDOM: dissolved organics & nutrient loading (365/480 nm)
  • Integrated turbidity, temperature, GPS
  • No regular calibration or cleaning required

Operations

  • Sampling: 30 sec to 24 hours (remote config)
  • Battery: up to 1 year; solar or wall charging
  • Cellular & satellite connectivity
  • Single integrated unit — no external power or telemetry
  • ML quantification with protected dashboard & API

Economics

  • $200/month vs. competitors at $5K–$30K
  • 700+ data points/month/site
  • Hand-removable cover for tool-free field maintenance
  • Captures transient contamination events that weekly sampling misses
Lume v1.2 exploded technical view with labeled components
thelume.ai/research
Intelligence

Analytical Features

Lume sensor deployed in mountain stream

Adaptive Contamination Detection

The Lume uses machine learning to learn normal conditions and trigger contamination alerts based on learned patterns rather than fixed thresholds.
US Patent 11,506,606 B2 — Bedell, Fankhauser, Sharpe, Wilson & Thomas

Machine learning code

Automated System-State Classification

Time-series data from water infrastructure sensors are analyzed to classify system states and support operational decisions without manual inspection or rule-based logic.
US Patent 11,507,861 B2 — Wilson, Coyle, Thomas & Croshere

Natural waterfall

Natural Waters Application

Gradient-boosted decision tree models achieving 75%+ accuracy across 0–1,000 CFU/100mL, >94% categorical accuracy with site calibration, and 7% MAPE (log-transformed).

thelume.ai/research
Performance

Drinking Water Classification

The Lume has been validated for drinking water monitoring across chlorinated and unchlorinated supplies. Binary classification at regulatory thresholds of 1 and 10 CFU/100 mL yields 91–92% overall accuracy with Cohen's kappa of 0.82–0.84.

Confusion matrices for binary classification of water quality using sensor predictions versus laboratory-observed E. coli concentrations at two regulatory thresholds.

thelume.ai/research
Performance

Chlorinated Supply Monitoring

The Lume detects chlorine residual presence in treated water supplies with 85% accuracy, distinguishing pre- and post-chlorinated samples.

Left: Predicted vs observed E. coli on log axes. Right: Binary classification of chlorine residual presence (accuracy 0.85, kappa 0.70).

thelume.ai/research
Performance

Natural Waters — Local Model

The Lume algorithm has been extensively validated against Colilert E. coli in freshwater systems. Over 75% of predictions fall within the analytical uncertainty bounds of the Colilert reference method, with 7% MAPE in log-transformed space.

Left: Boulder Creek test dataset. Right: Categorical classification into three management-relevant bins (<10, 10–100, >100 MPN/100 mL). Balanced accuracy 95%, Cohen's kappa 0.84.

thelume.ai/research
Performance

Natural Waters — Global Model

Temporally structured cross-validation across the global dataset. RMSE ranged from 0.55 (training) to 0.63 log units (test), with MAPE below 22% across both splits.

Left: Global dataset cross-validation. Right: Seine River, Paris — binary classification achieving 96.8% accuracy and 94% balanced accuracy using three TLF sensors.

thelume.ai
Development

Lume Development

⚙

Hardware Engineering

Internal and contracted expertise in IoT, signals, embedded systems, hardware design.

🧪

Analytical Science

Internal and contracted expertise in watershed AI/ML and front-end product development.

🏭

Manufacturing

Contract manufacturer established in China. 200 units produced, 60 sold as of March 2026.

🚚

Distribution

Internal through 2026; contracted thereafter.

📈

Sales & Applications

Extensive internal sales and research expertise.

🛡

Risk Management

Lume 1.1 traction will support positive margin on COGS in 2026 before further NRE investment.

thelume.ai
Market

Microbial Sensor Global Market

$5.57B

Global water quality sensor market in 2024, projected to reach $12.9B by 2033 (CAGR ~9%).

200K

Cl-A sensors sold in US in past 10 years. Annual market ~$100M. Microbial is the next frontier.

SensorDescriptionSetupEst. CostAccuracy
Virridy Lume Tryptophan sensor; ML model analysis Single, fully integrated IoT sensor $200/month/site 75%+ accuracy, >94% categorical
Proteus Sonde Multiparameter sonde Requires data logger, site-specific calibration $14K - $24K+ +/- 10 CFU/100mL
Chelsea UViLux Tryptophan, CDOM, BTEX, BOD Not specified ~$5,000 0.01 QSU sensitivity
YSI (Xylem) Chlorophyll only, no tryptophan Multiparameter sonde $4,985+ N/A for TLF
In Situ Inc. FDOM, CDOM, Cl-A, no tryptophan Multiparameter sonde + telemetry ~$10,000 N/A for TLF
virridycarbon.com
Business Model

Pricing Strategy

Competition sells for $5K - $30K one-time hardware cost. Virridy focuses on recurring revenue.

  • $557K Lume revenue YTD March 2026 (target exceeded); targeting $1M+ for full year
  • Target 50%+ margin on COGS
  • Convert from one-time sales to recurring subscriptions in 2026
  • Minimum 10+ units, 12-month contract
  • Includes: device lease, connectivity, cloud dashboard, API, fleet monitoring, firmware updates, onboarding support

Key Figures

$200
/month/unit
$1,200
COGS per unit

12-month minimum contract = 2x COGS

Hand-Off · What Comes Next

What's working overseas,
ready to come home.

Visibility, government commitment, capacity — each pillar of Vessel's roadmap has international precedent. The next eighty minutes are about which lessons fit here.

Laura, over to you.
Vessel Convening · May 5, 2026
CU Boulder · Mortenson Center in Global Engineering & Resilience
CU Boulder Mortenson Center in Global Engineering & Resilience
← → navigate · O overview · F fullscreen · N notes view