Wireless sensor networks support decision-making in diverse environmental contexts. Adoption of these has increased dramatically due to technological advances that have value while lowering cost. However, real-time information only allows for reactive management. As most interventions take time, predictions across enable better planning and decision making. Prediction engines large water level discharge do exist. they shortcomings their accessibility, automaticity, data requirements. We...

Real-time and model-predictive control promises to make urban drainage systems (UDS) adaptive, coordinated, dynamically optimal. Though early implementations are promising, existing algorithms have drawbacks in computational expense, trust, system-level coordination, labor cost. Linear feedback has distinct advantages interpretation, coordination. However, current methods for building linear controllers require calibrated software models. Here we present an automated method generating...

A sudden surge of data has created new challenges in water management, spanning quality control, assimilation, and analysis. Few approaches are available to integrate growing volumes into interpretable results. Process-based hydrologic models have not been designed consume large amounts data. Alternatively, machine learning tools can automate analysis forecasting, but their lack interpretability reliance on very sets limits the discovery insights may impact trust. To that end, we present a...

A sudden surge of data has created new challenges in water management, spanning quality control, assimilation, and analysis. Few approaches are available to integrate growing volumes into interpretable results. Process-based hydrologic models have not been designed consume large amounts data. Alternatively, machine learning tools can automate analysis forecasting, but their lack interpretability limits the discovery insights may impact trust. To that end, we present a approach, which seeks...