Uncertainty of Hydraulic Habitat Models Due to ADCP-Derived Inputs

Boat-mounted Acoustic Doppler Current Profilers (ADCP), originally designed to measure total discharge in a river, incidentally collect a high density of hydraulic measurements that can be post-processed to generate certain inputs for 1-dimensional habitat models (e.g., PHABSIM). These inputs include: 1) the discharge component of the stage-discharge rating curves, 2) bed elevation profiles for the wetted portion of the cross-section, and 3) water column velocity profiles.  Data collection efficiency is improved over traditional field methods; however, converting ADCP data to a format suitable for model input relies on significantly more post-processing and the application of complex algorithms, which frequently confound an intuitive assessment of their accuracy. Nevertheless, it is important to consider the error associated with these data and post-processing techniques, and subsequently, to understand the implications for the sensitivity of the model and the wisdom of the decisions they support. Using example data from the Suwannee River, Florida, this presentation will identify sources of uncertainty in ADCP data and quantify the variance of input parameters generated from these data.  Finally, the System for Environmental Flow Analysis (SEFA), a PHABSIM-type model, will be used to explore sensitivity of the Area Weighted Suitability (AWS) statistic (a.k.a. Weighted Useable Area, or WUA).