Impaired waters and the Total Maximum Daily Load approach are central drivers to water quality management mandated by the Clean Water Act. Water quality and ecological integrity vary across a gradient of human disturbance, but assessing how ecological integrity is affected by human disturbance is complex and requires robust indicators of ecological health. Indicators are used to quantify stream ecosystem integrity; however, uncertainty and variability of those indicators are poorly understood. We will conduct research that will aid Minnesota regulatory agencies in reducing the uncertainty and variability of indicators of stream ecosystem integrity to allow managers to make decisions based on scientific knowledge and be more defensible than current decisions. We have three objectives: (1) quantify the uncertainty surrounding stream health indicators, (2) evaluate the variability of indicators of stream health at different spatial scales, and (3) deconstruct stream health indices to understand which index metrics contribute most to uncertainty and variability at different spatial scales. Our project will advance management of stream ecological resources by aiding accurate identification of impaired streams, using existing data to understand the history of stream ecosystem integrity, and increasing efficiency with which stream ecosystem integrity is assessed and monitored.