Uncertainty Estimation with Deep Learning for Rainfall-Runoff Modelling

Klotz, Daniel; Kratzert, Frederik; Gauch, Martin; Sampson, Alden Keefe; Klambauer, Günter; Hochreiter, Sepp; Nearing, Grey

doi:10.31223/x5js4t

Cited by 16 publications

(20 citation statements)

References 53 publications

(36 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This Catchment Attributes and Meteorological Large Sample data set (CAMELS; Newman et al, 2015;Addor et al, 2017) consists of daily meteorological and daily discharge data from 671 catchments in CONUS ranging in size from 4 km 2 to 25,000 km 2 that have largely natural flows and long streamflow gauge records . Again, to be consistent with previous studies (Kratzert et al, 2019c(Kratzert et al, , 2021Klotz et al, 2021;Gauch et al, 2021b;Newman et al, 2017;Frame et al, 2021), we used the 531 of 671 CAMELS catchments that were chosen for model benchmarking by Newman et al (2017), who removed basins with (i) large discrepancies between different methods of calculating catchment area, and (ii) areas larger than 2,000 km 2 . CAMELS includes daily discharge data from the USGS Water Information System, which are used as training and evaluation target data.…”

Section: Datamentioning

confidence: 99%

Technical Note: Data assimilation and autoregression for using near-real-time streamflow observations in long short-term memory networks

Nearing

Klotz

Sampson³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Abstract. Ingesting near-real-time observation data is a critical component of many operational hydrological forecasting systems. In this paper we compare two strategies for ingesting near-real-time streamflow observations into Long Short-Term Memory (LSTM) rainfall-runoff models: autoregression (a forward method) and variational data assimilation. Autoregression is both more accurate and more computationally efficient than data assimilation. Autoregression is sensitive to missing data, however an appropriate (and simple) training strategy mitigates this problem.

show abstract

Section: Datamentioning

confidence: 99%

Technical Note: Data assimilation and autoregression for using near-real-time streamflow observations in long short-term memory networks

Nearing

Klotz

Sampson³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Our modelling approach uses deep learning (DL) techniques, which have produced accurate predictions on a wide variety of tasks, including rainfall-runoff modelling (Huntingford et al, 2019), and represent a fruitful area of further exploration for hydrologists and Earth scientists (Reichstein et al, 2019). For a more complete picture on the uses of DL techniques in hydrology, an interested reader is referred to Shen (2018), Beven (2020), Nearing et al (2020), and Kratzert et al (2018).…”

Section: Introductionmentioning

confidence: 99%

“…However, one architecture explicitly designed for time series simulation, the long short-term memory (LSTM) network (Hochreiter et al, 2001;Hochreiter, 1991), has recently demonstrated credible performance for modelling hydrological signatures across the continental United States (CONUS) (Kratzert et al, 2018(Kratzert et al, , 2019Duan et al, 2020;Feng et al, 2020;Gauch et al, 2021b;Fang et al, 2018Fang et al, , 2020. More recent work has begun not only to explore the accuracy of forecasts but also to use LSTMs to (i) provide estimates of uncertainty (Klotz et al, 2020), (ii) explore the ability of the LSTM to integrate prior physical knowledge into DL model architectures (Hoedt et al, 2021;Jiang et al, 2020), and (iii) to use LSTMs to produce predictions at multiple timescales from a single model (Gauch et al, 2021a).…”

Section: Introductionmentioning

confidence: 99%

“…This paper provides an evaluation of LSTM model ability across a large sample of GB catchments. We explore the association between catchment characteristics and the differences in model performances and present a data-driven benchmark that reflects the null-hypothesis of what information is present in a large-sample dataset (Nearing et al, 2020). Future modelling efforts may seek to assess whether hydrological theories encoded in conceptual and process-based models may contain more information than the benchmarks provided here .…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Benchmarking data-driven rainfall–runoff models in Great Britain: a comparison of long short-term memory (LSTM)-based models with four lumped conceptual models

Lees

Buechel

Anderson

et al. 2021

Hydrol. Earth Syst. Sci.

118

View full text Add to dashboard Cite

Abstract. Long short-term memory (LSTM) models are recurrent neural networks from the field of deep learning (DL) which have shown promise for time series modelling, especially in conditions when data are abundant. Previous studies have demonstrated the applicability of LSTM-based models for rainfall–runoff modelling; however, LSTMs have not been tested on catchments in Great Britain (GB). Moreover, opportunities exist to use spatial and seasonal patterns in model performances to improve our understanding of hydrological processes and to examine the advantages and disadvantages of LSTM-based models for hydrological simulation. By training two LSTM architectures across a large sample of 669 catchments in GB, we demonstrate that the LSTM and the Entity Aware LSTM (EA LSTM) models simulate discharge with median Nash–Sutcliffe efficiency (NSE) scores of 0.88 and 0.86 respectively. We find that the LSTM-based models outperform a suite of benchmark conceptual models, suggesting an opportunity to use additional data to refine conceptual models. In summary, the LSTM-based models show the largest performance improvements in the north-east of Scotland and in south-east of England. The south-east of England remained difficult to model, however, in part due to the inability of the LSTMs configured in this study to learn groundwater processes, human abstractions and complex percolation properties from the hydro-meteorological variables typically employed for hydrological modelling.

show abstract

“…Precipitation may serve as a surrogate for groundwater recharge; temperature and relative humidity include the relationship with evapotranspiration and at the same time provide the network with information on seasonality due to the usually distinct annual cycle. As an additional synthetic input parameter, a sinusoidal signal fitted to the temperature curve (T sin ) can provide the model with noise-free information on seasonality, which often allows for significantly improved predictions to be made (Kong-A-Siou et al, 2014). Without doubt, the most important input parameter out of these is P , since groundwater recharge usually has the greatest influence on groundwater dynamics.…”

Section: Input Parametersmentioning

confidence: 99%

Groundwater level forecasting with artificial neural networks: a comparison of long short-term memory (LSTM), convolutional neural networks (CNNs), and non-linear autoregressive networks with exogenous input (NARX)

Wünsch

Liesch

Broda

2021

Hydrol. Earth Syst. Sci.

159

View full text Add to dashboard Cite

Abstract. It is now well established to use shallow artificial neural networks (ANNs) to obtain accurate and reliable groundwater level forecasts, which are an important tool for sustainable groundwater management. However, we observe an increasing shift from conventional shallow ANNs to state-of-the-art deep-learning (DL) techniques, but a direct comparison of the performance is often lacking. Although they have already clearly proven their suitability, shallow recurrent networks frequently seem to be excluded from the study design due to the euphoria about new DL techniques and its successes in various disciplines. Therefore, we aim to provide an overview on the predictive ability in terms of groundwater levels of shallow conventional recurrent ANNs, namely non-linear autoregressive networks with exogenous input (NARX) and popular state-of-the-art DL techniques such as long short-term memory (LSTM) and convolutional neural networks (CNNs). We compare the performance on both sequence-to-value (seq2val) and sequence-to-sequence (seq2seq) forecasting on a 4-year period while using only few, widely available and easy to measure meteorological input parameters, which makes our approach widely applicable. Further, we also investigate the data dependency in terms of time series length of the different ANN architectures. For seq2val forecasts, NARX models on average perform best; however, CNNs are much faster and only slightly worse in terms of accuracy. For seq2seq forecasts, mostly NARX outperform both DL models and even almost reach the speed of CNNs. However, NARX are the least robust against initialization effects, which nevertheless can be handled easily using ensemble forecasting. We showed that shallow neural networks, such as NARX, should not be neglected in comparison to DL techniques especially when only small amounts of training data are available, where they can clearly outperform LSTMs and CNNs; however, LSTMs and CNNs might perform substantially better with a larger dataset, where DL really can demonstrate its strengths, which is rarely available in the groundwater domain though.

show abstract

Uncertainty Estimation with Deep Learning for Rainfall-Runoff Modelling

Cited by 16 publications

References 53 publications

Technical Note: Data assimilation and autoregression for using near-real-time streamflow observations in long short-term memory networks

Technical Note: Data assimilation and autoregression for using near-real-time streamflow observations in long short-term memory networks

Benchmarking data-driven rainfall–runoff models in Great Britain: a comparison of long short-term memory (LSTM)-based models with four lumped conceptual models

Groundwater level forecasting with artificial neural networks: a comparison of long short-term memory (LSTM), convolutional neural networks (CNNs), and non-linear autoregressive networks with exogenous input (NARX)

Contact Info

Product

Resources

About