.png){kind=logo}
Unlocking Real-Time Water Quality Data for DC Rivers
How a landmark partnership brought continuous, public-facing water quality monitoring to the Anacostia, Potomac, and Shenandoah
Project Takeaways
Real-Time Sondes Deployed
Major Rivers Observed
Core Partner Organizations
Parameters Monitored
"This project proves what's possible when hardware partners, watershed advocates, and technology teams pool their strengths around a single community goal. Xylem's commitment to donate the sensors, training, and ongoing maintenance removed the single biggest barrier to real-time monitoring in the DC region — and allowed us to focus entirely on building a public-facing platform that watershed residents will actually use. The result is a model we believe can be replicated across urban watersheds nationwide."
Project Deployment Strategy
Working in partnership with Xylem, Anacostia Riverkeeper, Anacostia Watershed Society, Potomac Riverkeeper Network, and the Reservoir Center for Water Solutions, The Commons developed a comprehensive real-time water quality data management and visualization platform for the Anacostia, Potomac, and Shenandoah rivers. The platform ingests continuous sensor data from ten YSI EXO sondes through Xylem's HydroSphere API, processes it through a cloud-based pipeline with industry-standard quality controls, and delivers the results through an intuitive, mobile-responsive Webflow website. The system empowers the more than six million residents of the DC metropolitan area — from casual recreators checking whether it's safe to kayak, to researchers analyzing multi-year trends, to policymakers evaluating watershed restoration outcomes — with the same live data, presented in formats that meet each user where they are.
Xylem Hydrosphere | Telemetered Data Hosting
Centralized API and service that ingests, structures, and serves telemetered water quality data from Xylem YSI EXO sondes.
Heroku | Data Capture and Staging
API Layer supporting data acquisition, quality assurance checks, and routing from every sensor at 15 minutes intervals.
Mongo DB Atlas | Data Archival
Supports data structuring and validation needed to attain long-term hosting and archival across the complete operational lifecycle of the EXO sonde
Webflow | Content Management
Our team chose Webflow for its ease of maintenance and designer-friendly development environment. Webflow's powerful CMS and API enabled real-time site updates and seamless synchronization with Heroku.
Mapbox | Location Experience
Leveraging Mapbox, our team deployed an integrated web mapping experience that displays current monitoring site conditions and allows for easy routing to monitoring site pages.
Project and Environmental Policy Intersection
The Chesapeake Bay Watershed represents one of the most ecologically and economically significant water systems in the United States, serving as home to millions of residents and countless species across six states and the District of Columbia. Despite decades of conservation investment following the passage of the Clean Water Act in 1972, and coordinated recovery efforts under the Chesapeake Bay Program, water quality challenges persist throughout the watershed — particularly in the urban and peri-urban rivers flowing through the nation's capital.
The Anacostia River has historically been among the most polluted urban rivers in the United States, with legacy contamination and active combined sewer overflows (CSOs) that discharge untreated sewage and stormwater during heavy rainfall events. DC Water's Clean Rivers Project — a court-mandated long-term control plan under federal Clean Water Act authority — has invested billions of dollars in tunnel infrastructure to reduce CSO volumes, with the Anacostia River Tunnel already delivering dramatic reductions since coming online. The Potomac and Shenandoah rivers face parallel challenges from urban runoff, agricultural nutrient loading upstream, and emerging contaminants that threaten both ecosystem health and the region's drinking water supply.
Under EPA recreational water quality standards and the Chesapeake Bay Total Maximum Daily Load (TMDL), federal and state agencies are responsible for monitoring compliance — but regulatory monitoring alone provides only a narrow window into real-time conditions. For recreational users, educators, and community advocates, the question is not whether annual averages meet a regulatory threshold; it's whether the river is safe today, at this location. Until now, that question has been answered primarily through volunteer-led weekly grab sampling programs — an indispensable but inherently fragmented approach that misses critical pollution events and leaves the public without timely information.
This gap is precisely where the Xylem partnership changed what was possible. Rather than a traditional vendor-client relationship, Xylem made a multi-year commitment to donate the full suite of YSI EXO sonde hardware, cellular connectivity, solar power systems, sensor calibration supplies, staff training, and ongoing maintenance support across all ten monitoring stations. This contribution — paired with the on-the-water expertise of Anacostia Riverkeeper, Anacostia Watershed Society, and Potomac Riverkeeper Network, who maintain the equipment and validate the data — eliminated the single largest barrier to sustained real-time monitoring: long-term operational cost. The Commons then developed the data management and public dashboard infrastructure, transforming the partnership's sensor network into a living, community-facing resource.
Project Planning
The project began with a clear and urgent need: recreational users, researchers, and policymakers across the DC metro area lacked any unified, real-time source of water quality information for the three major rivers flowing through the region. Volunteer-led weekly sampling programs — while scientifically valuable — produced fragmented data with limited temporal resolution, siloed within individual organizations and often requiring technical expertise to interpret. Existing platforms that did surface sensor data suffered from restrictive data retention policies, including HydroSphere's two-year rolling window, which made long-term trend analysis impossible and put valuable historical context at risk of being lost.
Working directly with Xylem, the Reservoir Center for Water Solutions, and the three watershed nonprofit partners, our team executed a comprehensive planning process built around the distinct needs of multiple user groups. Through an initial discovery phase — including stakeholder interviews, survey-based user research with existing platform users, and interviews with new users who had never previously engaged with environmental data tools — we established four primary requirements that would guide the platform's development.
First, the system needed an intuitive mapping interface that allowed a kayaker, a researcher, and a policymaker to each find the information they needed within seconds, translating complex parameters like dissolved oxygen, pH, turbidity, and blue-green algae pigment into clear, color-coded safety indicators. Second, the architecture needed to permanently capture all sensor readings in a durable, queryable cloud database — eliminating the two-year data retention ceiling and enabling analysis across seasons, years, and ultimately decades. Third, the platform needed to meet stringent accessibility and mobile-performance standards, recognizing that many users would be checking conditions on a phone at the riverbank rather than on a desktop. Finally, the back-end needed to be secure, observable, and maintainable by a small team — with clear documentation and handover procedures that would allow the Reservoir Center and its partners to operate the system independently for years to come.
Methodology
The core of this project was the development of an end-to-end data pipeline and public-facing dashboard that could transform continuous sensor telemetry into immediate, trustworthy, and accessible information for watershed communities. Our methodology prioritized user-centered design, rigorous data validation, and a modular cloud architecture that would scale as the partnership's monitoring network expands. The DC Rivers platform contains the following features:
- Interactive Mapbox-powered map displaying the status of all ten monitoring stations with intuitive color-coded indicators (good, impaired, unusual readings, sensor offline)
- Parameter-level filtering for chloride, pH, turbidity, chlorophyll, dissolved oxygen, conductivity, fluorescent dissolved organic matter (fDOM), blue-green algae pigment, and E. coli proxy measurements
- Real-time station detail panels showing current readings, sensor status, and air/water temperature at-a-glance
- Educational "Understanding the Data" pages translating each parameter into plain-language explanations of what it measures, what it tells us, and what causes it to change
- Historical trend visualizations with configurable time windows (7 days, 30 days, 90 days, and longer) for every monitored parameter
- Mobile-first responsive design optimized for on-the-go recreational decision-making
- Integrated external data sources including EPA APIs, USGS flow data, and weather services for environmental context
- QARTOD-compliant automated data quality control applied to every incoming reading
- Automated data archival to permanent cloud storage, removing the two-year HydroSphere retention limitation
We selected Figma as our primary collaborative design environment, conducting iterative workshops with Xylem, Reservoir Center, and watershed partner staff throughout the design phase. Figma enabled our team to move rapidly through low, medium, and high fidelity mockups while keeping all stakeholders aligned on navigation patterns, map behavior, key analytics views, and accessibility decisions. Alongside this, we conducted a full landscape review of comparable water quality, weather, and environmental data platforms — identifying proven UI patterns we could adapt and, equally importantly, common pitfalls we could avoid.
On the back-end, we architected a cloud-native data pipeline built on Heroku. The data ingestion layer continuously polls the HydroSphere API, performs format normalization and validation, applies rate limiting and error handling, and passes each reading through an API layer that handles authentication, data processing (aggregation, calculation, quality checks), and background jobs for data archival, Webflow CMS synchronization, and alerting. Sensor readings and metadata are stored in MongoDB Atlas, while long-term sensor data archives are written to AWS S3 with strict bucket policies, private access, and pre-signed URL distribution. The entire stack is SOC 2 compliant end-to-end, with two-step authentication on all accounts, environment-variable secret management, HTTPS-only connections, and principle-of-least-privilege access control.
Data quality was a non-negotiable foundation of the platform. We implemented a validation process built around QARTOD (Quality Assurance of Real-Time Oceanographic Data) methodology — the industry standard for continuous environmental sensor data. Every incoming reading is evaluated through a series of complementary checks: timing and gap tests confirm data arrive at the expected 15-minute intervals without transmission errors; spike tests identify sudden, unrealistic jumps by comparing each measurement against surrounding values using parameter-specific thresholds drawn from published literature and calibrated to natural river variability; flat-line tests detect stuck sensors; rate-of-change tests flag implausibly rapid shifts; and gross range tests ensure values fall within physically reasonable limits for each parameter. Readings that fail validation are flagged rather than silently discarded, preserving scientific transparency while ensuring the public-facing dashboard only surfaces trustworthy data.
The front-end experience is delivered through Webflow, chosen for its combination of visual design flexibility, CMS capabilities for educational content and blog posts, and CDN-backed performance with built-in DDoS protection. Custom JavaScript modules connect the Webflow presentation layer to the live REST API, and Mapbox GL JS powers the primary interactive map. Chart.js handles all time-series visualizations, automatically simplifying on smaller screens to ensure readability on mobile devices. Role-based CMS access controls allow Xylem, Reservoir Center, and partner organization staff to publish blog updates, success stories, and educational content directly, while sensitive configuration remains protected.
.png)
Throughout development, our team produced comprehensive documentation and standard operating procedures covering the full technology stack — including runbooks for build and deployment, monitoring and alerting, database backups, and API key rotation. A four-to-five-week hands-on knowledge transfer period was built into the project plan, ensuring that Reservoir Center and partner staff could confidently operate, troubleshoot, and evolve the platform independently after handover.
Results and Outcomes
The result of this collaborative effort is a public-facing water quality platform that makes the DC region one of the most transparently monitored urban watersheds in the United States. Ten real-time stations across three major rivers now stream continuous data to a permanent cloud repository, surface that data through a mobile-friendly map and dashboard experience designed for both technical and non-technical audiences, and do so under a partnership model specifically structured for long-term sustainability. The platform demonstrates that when hardware manufacturers, community watershed organizations, technology providers, and convening institutions align around a shared public benefit, they can deliver infrastructure that neither party could have built alone.
A partnership model built for the long haul
The most significant outcome of this project is not the dashboard itself — it's the partnership structure that sustains it. Xylem's multi-year commitment to provide the full YSI EXO hardware suite, calibration supplies, cellular connectivity, solar power systems, training, and maintenance funding transforms what is typically the highest ongoing cost center of a monitoring program (equipment and upkeep) into a donated contribution. That donation is matched by the on-the-water expertise of Anacostia Riverkeeper, Anacostia Watershed Society, and Potomac Riverkeeper Network, whose staff physically maintain the sondes, perform calibration, and validate the sensor data against periodic grab samples. The Commons contributes the data infrastructure and public dashboard; the Reservoir Center for Water Solutions serves as the convening institution and hosts the ongoing community engagement phase.
This four-way structure — hardware partner, watershed stewards, technology partner, and convening institution — produces a platform that is genuinely greater than the sum of its parts. The watershed nonprofits gain access to continuous data streams and a professionally designed public dashboard they would not otherwise have the capacity to build. Xylem's hardware reaches the recreational, research, and advocacy communities most likely to benefit from it. The Reservoir Center and The Commons produce a reference architecture that other urban watershed collaborations can study and replicate. And the public gains a single, trusted source of real-time river health information for the waterways that run through their city.
Critically, the partnership was designed with explicit sustainability commitments rather than one-time project handoffs. Xylem's five-year commitment to calibration supplies and maintenance funding, combined with modest ongoing hosting costs covered by the Reservoir Center, means the platform has a clear operational runway from day one — and a partnership framework that can be extended or expanded as the network grows.
Permanent, accessible data for the next decade of decisions
The second major outcome of this project is the creation of a durable, queryable, public archive of high-resolution water quality data for the Anacostia, Potomac, and Shenandoah rivers. HydroSphere's two-year data retention policy — while perfectly reasonable for an operational monitoring tool — meant that historical context older than twenty-four months was at constant risk of being lost. Our cloud-based architecture eliminates that ceiling entirely. Every validated reading from every sonde is now archived in MongoDB Atlas and AWS S3, under strict access controls, for the indefinite long term.
The downstream value of that archive is substantial. Researchers can analyze multi-year patterns in dissolved oxygen, chlorophyll, and fDOM to evaluate the effectiveness of DC Water's Clean Rivers Project infrastructure investments. Educators can use the data in classroom curricula across DC, Maryland, Virginia, and West Virginia. Policymakers can evaluate the real-world impact of TMDL implementation. And recreational users gain not just a snapshot of current conditions but the ability to see how a river's health has changed over weeks, months, and years — a level of transparency that transforms public engagement with watershed protection from abstract advocacy into concrete, visible progress.
Combined with the educational infrastructure built into the platform — plain-language parameter explanations, mobile-optimized navigation, and integration with external data sources from EPA, USGS, and weather services — the result is a resource that meets users wherever they are in their relationship with the rivers. A kayaker checking safety conditions on their phone at Buzzard Point, a middle school science teacher in Arlington pulling up a chlorophyll time series for their ecosystems unit, and a policymaker reviewing multi-year trends for a committee hearing now all have access to the same trustworthy data, presented in formats appropriate to their needs. That is the vision of democratized environmental data this partnership set out to build — and a model we believe belongs in urban watersheds across the country.
Support the Digital Services Program
.svg)
.svg)