A Data-Based Mechanistic Modelling Approach to Real-Time Flood Forecasting

Beven, Keith

doi:10.1142/9781848162716_0016

Cited by 8 publications

(5 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lake WL forecasting could also be carried out using mechanistic models, focused on the causality of inputoutput relationships (Baker et al, 2018). It has been frequently used for rainfall-runoff process, streamflow and river water level modeling (Lees, 2000;Romanowicz et al, 2008;Young et al, 2014;Smith et al, 2014). However, ML models have several advantages over the mechanistic models.…”

Section: Discussionmentioning

confidence: 99%

Advanced Hybrid Machine Learning Algorithms for Multistep Lake Water Level Forecasting

Fijani¹,

Khosravi²,

Barzegar³

et al. 2021

Preprint

View full text Add to dashboard Cite

Random Tree (RT) and Iterative Classifier Optimizer (ICO) based on Alternating Model Tree (AMT) regressor machine learning (ML) algorithms coupled with Bagging (BA) or Additive Regression (AR) hybrid algorithms were applied to forecasting multistep ahead (up to three months) Lake Superior and Lake Michigan water level (WL). Partial autocorrelation (PACF) of each lake’s WL time series estimated the most important lag times — up to five months in both lakes — as potential inputs. The WL time series data was partitioned into training (from 1918 to 1988) and testing (from 1989 to 2018) for model building and evaluation, respectively. Developed algorithms were validated through statistically and visually based metric using testing data. Although both hybrid ensemble algorithms improved individual ML algorithms’ performance, the BA algorithm outperformed the AR algorithm. As a novel model in forecasting problems, the ICO algorithm was shown to have great potential in generating robust multistep lake WL forecasts.

show abstract

Section: Discussionmentioning

confidence: 99%

Advanced Hybrid Machine Learning Algorithms for Multistep Lake Water Level Forecasting

Fijani¹,

Khosravi²,

Barzegar³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Techniques such as the Kalman filter, or stochastic autoregres-5 sive modelling, can be used with the advantage that an estimate of the variance of the forecast can also be updated at the same time (see for example, Sene et al, 2014;Young et al, 2014;Smith et al, 2012Smith et al, , 2013a. No explicit account of potential epistemic uncertainties is made in this approach; the aim is only to improve the forecast and minimize the forecast variance at the required lead time as new data become available 10 for assimilation.…”

Section: Discussionmentioning

confidence: 99%

Epistemic uncertainties and natural hazard risk assessment – Part 2: Different natural hazard areas

Beven

Almeida

Aspinall

et al. 2016

Preprint

View full text Add to dashboard Cite

This paper discusses how epistemic uncertainties are considered in a number of different natural hazard areas including floods, landslides and debris flows, dam safety, droughts, earthquakes, tsunamis, volcanic ash clouds and pyroclastic flows, and wind storms. In each case it is common practice to treat most uncertainties in the form of 5 aleatory probability distributions but this may lead to an underestimation of the resulting uncertainties in assessing the hazard, consequences and risk. It is suggested that such analyses might be usefully extended by looking at different scenarios of assumptions about sources of epistemic uncertainty, with a view to reducing the element of surprise in future hazard occurrences. Since every analysis is necessarily conditional 10 on the assumptions made about the nature of sources of epistemic uncertainty it is also important to follow the guidelines for good practice suggested in the companion Part 1 by setting out those assumptions in a condition tree.

show abstract

“…A different strategy is to assume that all uncertainties can be treated statistically and use a data assimilation approach to correct for over or under-prediction as the event proceeds. Techniques such as the Kalman filter, or stochastic autoregressive modelling, can be used with the advantage that an estimate of the variance of the forecast can also be updated at the same time (see, for example, Sene et al, 2014;Young et al, 2014;Smith et al, , 2013a. No explicit account of potential epistemic uncertainties is normally made in this approach; the aim is only to improve the forecast and minimize the forecast variance at the required lead time as new data become available for assimilation.…”

Section: Uncertainty Quantification In Real-time Flood Managementmentioning

confidence: 99%

Epistemic uncertainties and natural hazard risk assessment – Part 1: A review of different natural hazard areas

Beven

Almeida

Aspinall

et al. 2018

Nat. Hazards Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. This paper discusses how epistemic uncertainties are currently considered in the most widely occurring natural hazard areas, including floods, landslides and debris flows, dam safety, droughts, earthquakes, tsunamis, volcanic ash clouds and pyroclastic flows, and wind storms. Our aim is to provide an overview of the types of epistemic uncertainty in the analysis of these natural hazards and to discuss how they have been treated so far to bring out some commonalities and differences. The breadth of our study makes it difficult to go into great detail on each aspect covered here; hence the focus lies on providing an overview and on citing key literature. We find that in current probabilistic approaches to the problem, uncertainties are all too often treated as if, at some fundamental level, they are aleatory in nature. This can be a tempting choice when knowledge of more complex structures is difficult to determine but not acknowledging the epistemic nature of many sources of uncertainty will compromise any risk analysis. We do not imply that probabilistic uncertainty estimation necessarily ignores the epistemic nature of uncertainties in natural hazards; expert elicitation for example can be set within a probabilistic framework to do just that. However, we suggest that the use of simple aleatory distributional models, common in current practice, will underestimate the potential variability in assessing hazards, consequences, and risks. A commonality across all approaches is that every analysis is necessarily conditional on the assumptions made about the nature of the sources of epistemic uncertainty. It is therefore important to record the assumptions made and to evaluate their impact on the uncertainty estimate. Additional guidelines for good practice based on this review are suggested in the companion paper (Part 2).

show abstract

A Data-Based Mechanistic Modelling Approach to Real-Time Flood Forecasting

Cited by 8 publications

References 52 publications

Advanced Hybrid Machine Learning Algorithms for Multistep Lake Water Level Forecasting

Advanced Hybrid Machine Learning Algorithms for Multistep Lake Water Level Forecasting

Epistemic uncertainties and natural hazard risk assessment – Part 2: Different natural hazard areas

Epistemic uncertainties and natural hazard risk assessment – Part 1: A review of different natural hazard areas

Contact Info

Product

Resources

About