Research, management, and modeling for water quality and resources, ecological processes, agricultural and urban planning, wildlife habitat, and much more depend on accurate water level and discharge measurement. Greater and more complex water resource needs for parties with competing interests demand more detailed models and effective science communication than are possible with the traditional scalar measurements. The scientific community recognizes the need for new methods that provide context for manual and programmatic extraction of a broader range information to make measurements and predictions, drive insights and facilitate communication between stakeholders and the public. Methods have been developed to accurately measure water level in ground-based imagery, but it is critical to improve those methods and create standardized tools, workflows, and software for use by the scientific community, academia, and industry. The purpose of this research was to find ways to develop and analyze tools and methods to measure and/or predict water level and streamflow in images with less artificial conditioning of the water scene than was previously possible. Three research studies were identified to accomplish this: 1) A site study to determine whether it is possible to predict water level and streamflow accurately to fill data gaps in United States Geological Survey historical data from a time-series of images with no artificial calibration or waterline target in the scene, 2) creation of a free, open-source software suitable for use by hydrologists and ecologiststo measure water level in images with a pixel to world coordinate calibration and waterline search target installed in a stream or river, and 3) creation of a new calibration target that improved the bow-tie calibration target and measurement surface used in #2 by allowing for calibration in every image and that occluded less of the scene allowing for improved contextual analysis.