Hydrology involves the study of processes describing the movement of water on the earth’s surface, from the land surface to the subsurface and from the land surface to the atmosphere. These processes occur on a range of time scales, from hours to days to weeks (i.e. flash floods) to longer decadal and inter-decadal time frames involving climate variability (i.e. droughts). Quantifying the uncertainty in hydrologic predictions and considering this uncertainty in subsequent decision-making is becoming increasingly important for both research and operational modeling purposes. The main sources of uncertainty include the data (e.g. spatial and temporal representation of input and output timeseries), the perceptual and numerical model (e.g. watershed flowpaths and residence times, process descriptions and mathematical implementation) and the modeling approach (e.g. the method chosen for parameter estimation). This presentation provides an overview of recent work centered on reducing the uncertainty in hydrologic predictions, primarily through the use of common optimization techniques and the integration of alternative observations or “soft” data in model development and calibration.
Copyright. All Rights Reserved.