Application of high-resolution, remotely sensed data for transient storage modeling parameter estimation

This paper presents a method that uses high‐resolution multispectral and thermal infrared imagery from airborne remote sensing for estimating two model parameters within the two‐zone in‐stream temperature and solute (TZTS) model. Previous TZTS modeling efforts have provided accurate in‐stream temperature predictions; however, model parameter ranges resulting from the multiobjective calibrations were quite large. In addition to the data types previously required to populate and calibrate the TZTS model, high‐resolution, remotely sensed thermal infrared (TIR) and near‐infrared, red, and green (multispectral) band imagery were collected to help estimate two previously calibrated parameters: (1) average total channel width (BTOT) and (2) the fraction of the channel comprising surface transient storage zones (β). Multispectral imagery in combination with the TIR imagery provided high‐resolution estimates ofBTOT. In‐stream temperature distributions provided by the TIR imagery enabled the calculation of temperature thresholds at which main channel temperatures could be delineated from surface transient storage, permitting the estimation ofβ. It was found that an increase in the resolution and frequency at which BTOT and β were physically estimated resulted in similar objective functions in the main channel and transient storage zones, but the uncertainty associated with the estimated parameters decreased. Key Points High‐resolution imagery can be used to estimate TZTS parameters Estimating parameters from imagery decreases uncertainty in other parameters Increasing spatial resolution of key parameters decreases uncertainty