Most if not all coastal states in the US have established numeric thresholds for dissolved oxygen (DO) to protect aquatic life in estuaries. Some are in the process, or have recently completed, revisions of their criteria based on newer science. Often, a toxicological approach has been applied, wherein laboratory exposure tests identify DO values associated with acute or chronic DO effects across a range of taxa and criteria values are based on the requirements of the most sensitive species, which are then assumed to also protect less sensitive species. While recognizing that this approach has validity, the most sensitive species in estuaries may be less likely to suffer DO-related mortality than more tolerant species, whose tolerance likely reflects adaptive response to chronic low DO exposure, which they are unable to avoid. Meanwhile, chronic DO effects on growth, behavior, and trophic interactions may be more pervasive than more acute effects. Given that portions of productive estuarine and coastal waters, even healthy ones, can and do experience periods of low oxygen, environmental policy makers need a science-based approach that is realistic and protects biotic integrity. Pros and cons of existing approaches to addressing temporal and spatial variability in assessment and implementation of DO thresholds will be examined. DO dynamics will be described for several estuaries using a mixture of traditional and continuous DO monitoring, along with proposed approaches to quantify and limit DO impacts. Areas of scientific uncertainty related to establishing and implementing DO thresholds will be identified to guide future monitoring and research.