BEAR (Bayesian error analysis with reordering)

Identifying the time series of data error during the model calibration

% BEAR Algorithm Overview:

1. During the estimation of model parameters, a random sequence of output data errors is generated based on the probability distribution functions (PDF) of the output data error (prior knowledge).
2. The order of this sequence is updated using residual information.
3. By adjusting their orders, a new sequence of output data errors is obtained.

% REFERENCE:

Wu, X., Marshall, L., Sharma, A., 2022. Incorporating multiple observational uncertainties in water quality model calibration. Hydrological Processes 36, e14452. https://doi.org/10.1002/hyp.14452

Wu, X., Marshall, L., Sharma, A., 2021. Quantifying input error in hydrologic modeling using the Bayesian error analysis with reordering (BEAR) approach. Journal of Hydrology 598, 126202. https://doi.org/10.1016/j.jhydrol.2021.126202

Wu, X., Marshall, L., Sharma, A., 2022. Quantifying input uncertainty in the calibration of water quality models: reordering errors via the secant method. Hydrology and Earth System Sciences 26, 1203–1221. https://doi.org/10.5194/hess-26-1203-2022