Data Management and Analytics Platform for Water Quality Monitoring Data - Actionable Decision-Making

Presentation Description: The abundance of environmental and water resources related data has highlighted a gap in data management and accessibility within the water and wastewater industry. Spreadsheets provide flexibility in analytics (e.g., charts, pivot tables, trend analysis), but are cumbersome for complex data management, prone to error, and are not suitable for large datasets. Databases, such as Microsoft Access or SQL Server, are designed to handle larger data volumes, but generally require custom development for graphical visualizations and analytics. Third party software, such as PowerBI and ArcGIS Online, can offer a solution, but finding a single product that combines geospatial analysis with the management of very large time-series data is challenging. To address this, CDM Smith and its digital company, Trinnex, have developed a cloud-based solution for geospatial and time-series data management and analysis and are applying this data platform to a large dataset collected by Los Angeles Sanitation and Environment (LASAN).
LASAN is responsible for managing critical water quality data collected for various monitoring programs and watersheds, including monitoring for NPDES Permits and TMDLs. A key component of the watershed protection strategy involves continuous monitoring (at 15-minute interval) of water quality and meteorologic data across four urban lakes. The dataset, spanning back to 2014, contains over 40 million individual records including various water quality (e.g., temperature, dissolved oxygen, phycocyanin, chlorophyll-a, conductivity, turbidity) and meteorologic (e.g., air temperature, rainfall, humidity) parameters. To store and visualize these data, we utilize dataCAST - a data management and analytics platform designed for storing and displaying large volumes of sampling and timeseries data. New data are automatically imported on an hourly schedule from continuous data loggers then displayed in timeseries, statistical, or categorically grouped figures. Additionally, the data logger sites can be mapped, and data subsets can be exported for analysis in other software. LASAN uses dataCAST to monitor real-time water quality trends, such as tracking algae concentrations or diurnal dissolved oxygen patterns, and to identify data logger maintenance needs or malfunctions. Future developments include automated reporting to assist with in-house (and potentially regulatory) reports and/or data-driven water quality forecasts powered by machine learning models.