Evaluation of Maximum a Posteriori Estimation as Data Assimilation Method for Forecasting Infiltration-Inflow Affected Urban Runoff with Radar Rainfall Input
| Autor: | Peter Steen Mikkelsen, Troels Sander Poulsen, Morten Grum, Nadia Schou Vorndran Lund, Roland Löwe, Jonas Wied Pedersen, Morten Borup |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2016 |
| Předmět: |
lcsh:Hydraulic engineering
Auto-Calibration Flow Forecasting 0208 environmental biotechnology Geography Planning and Development Flow (psychology) 02 engineering and technology Aquatic Science linear reservoir models Biochemistry Data Assimilation Linear Resevoir Models Standard deviation Data assimilation lcsh:Water supply for domestic and industrial purposes Goodness of fit Real-time Control System lcsh:TC1-978 Statistics Maximum a posteriori estimation real-time control flow forecasting data assimilation auto-calibration Maximum a Posteriori estimation urban drainage systems Urban Drainage Systems Water Science and Technology Urban runoff lcsh:TD201-500 Infiltration/Inflow 020801 environmental engineering Environmental science Real-time Control Maximum a Posteriori Estimation |
| Zdroj: | Wied Pedersen, J, Lund, N S V, Borup, M, Löwe, R, Poulsen, T S, Mikkelsen, P S & Grum, M 2016, ' Evaluation of Maximum a Posteriori Estimation as Data Assimilation Method for Forecasting Infiltration-Inflow Affected Urban Runoff with Radar Rainfall Input ', Water, vol. 8, no. 9 . https://doi.org/10.3390/w8090381 Water, Vol 8, Iss 9, p 381 (2016) Water; Volume 8; Issue 9; Pages: 381 |
| Popis: | High quality on-line flow forecasts are useful for real-time operation of urban drainage systems and wastewater treatment plants. This requires computationally efficient models, which are continuously updated with observed data to provide good initial conditions for the forecasts. This paper presents a way of updating conceptual rainfall-runoff models using Maximum a Posteriori estimation to determine the most likely parameter constellation at the current point in time. This is done by combining information from prior parameter distributions and the model goodness of fit over a predefined period of time that precedes the forecast. The method is illustrated for an urban catchment, where flow forecasts of 0–4 h are generated by applying a lumped linear reservoir model with three cascading reservoirs. Radar rainfall observations are used as input to the model. The effects of different prior standard deviations and lengths of the auto-calibration period on the resulting flow forecast performance are evaluated. We were able to demonstrate that, if properly tuned, the method leads to a significant increase in forecasting performance compared to a model without continuous auto-calibration. Delayed responses and erratic behaviour in the parameter variations are, however, observed and the choice of prior distributions and length of auto-calibration period is not straightforward. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|