On the Application of Machine Learning Techniques to Regression Problems in Sea Level Studies

Hieronymus, Magnus; Hieronymus, Jenny; Hieronymus, Fredrik

doi:10.1175/jtech-d-19-0033.1

Cited by 18 publications

(16 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is one of the first works that considers both spatial and temporal aspects of the input data; however, predictions are made at a much coarser temporal resolution than that considered in this paper. A higher temporal resolution is considered by Hieronymus et al (2019), which is the most closely related work to ours. Autoregressive neural networks are used to model the sea level time series with the addition of atmospheric forcing reduced by empirical orthogonal function (EOF) decomposition.…”

Section: Introductionmentioning

confidence: 94%

“…The role of the trainable discriminative atmospheric encoder ASE is analysed by replacing it with a reconstructive embedding proposed in Hieronymus et al (2019). A principal component analysis is applied to the atmospheric input to compute a low-dimensional subspace (empirical orthogonal functions, EOFs) that maximizes the data reconstruction.…”

Section: Influence Of the Atmospheric Encodermentioning

confidence: 99%

“…A principal component analysis is applied to the atmospheric input to compute a low-dimensional subspace (empirical orthogonal functions, EOFs) that maximizes the data reconstruction. Following Hieronymus et al (2019), the top three EOFs are used in the subspace construction. The input is projected into this subspace producing a low-dimensional signal that is directly used in the HIDRA regression network.…”

Section: Influence Of the Atmospheric Encodermentioning

confidence: 99%

See 2 more Smart Citations

HIDRA 1.0: deep-learning-based ensemble sea level forecasting in the northern Adriatic

et al. 2021

View full text Add to dashboard Cite

Abstract. Interactions between atmospheric forcing, topographic constraints to air and water flow, and resonant character of the basin make sea level modelling in the Adriatic a challenging problem. In this study we present an ensemble deep-neural-network-based sea level forecasting method HIDRA, which outperforms our set-up of the general ocean circulation model ensemble (NEMO v3.6) for all forecast lead times and at a minuscule fraction of the numerical cost (order of 2×10-6). HIDRA exhibits larger bias but lower RMSE than our set-up of NEMO over most of the residual sea level bins. It introduces a trainable atmospheric spatial encoder and employs fusion of atmospheric and sea level features into a self-contained network which enables discriminative feature learning. HIDRA architecture building blocks are experimentally analysed in detail and compared to alternative approaches. Results show the importance of sea level input for forecast lead times below 24 h and the importance of atmospheric input for longer lead times. The best performance is achieved by considering the input as the total sea level, split into disjoint sets of tidal and residual signals. This enables HIDRA to optimize the prediction fidelity with respect to atmospheric forcing while compensating for the errors in the tidal model. HIDRA is trained and analysed on a 10-year (2006–2016) time series of atmospheric surface fields from a single member of ECMWF atmospheric ensemble. In the testing phase, both HIDRA and NEMO ensemble systems are forced by the ECMWF atmospheric ensemble. Their performance is evaluated on a 1-year (2019) hourly time series from a tide gauge in Koper (Slovenia). Spectral and continuous wavelet analysis of the forecasts at the semi-diurnal frequency (12 h)−1 and at the ground-state basin seiche frequency (21.5 h)−1 is performed. The energy at the basin seiche in the HIDRA forecast is close to that observed, while our set-up of NEMO underestimates it. Analyses of the January 2015 and November 2019 storm surges indicate that HIDRA has learned to mimic the timing and amplitude of basin seiches.

show abstract

Section: Introductionmentioning

confidence: 94%

Section: Influence Of the Atmospheric Encodermentioning

confidence: 99%

Section: Influence Of the Atmospheric Encodermentioning

confidence: 99%

See 1 more Smart Citation

HIDRA 1.0: deep-learning-based ensemble sea level forecasting in the northern Adriatic

et al. 2021

View full text Add to dashboard Cite

show abstract

“…This is one of the first works that considers both spatial and temporal aspects of the input data, however, predictions are made at a much coarser temporal resolution than that considered in this paper. A higher temporal resolution is considered by Hieronymus et al (2019), which is the most closely related work to ours. Autoregressive neural networks are used to model the sea level time-series with addition of atmospheric forcing reduced by empirical orthogonal function (EOF) decomposition.…”

Section: Introductionmentioning

confidence: 94%

“…The role of trainable discriminative atmospheric encoder ASE is analyzed by replacing it with a reconstructive embedding proposed in Hieronymus et al (2019). A principal component analysis is applied to the atmospheric input to compute a lowdimensional subspace (empirical orthogonal functions, EOFs) that maximizes the data reconstruction.…”

Section: Influence Of the Atmospheric Encodermentioning

confidence: 99%

HIDRA 1.0: Deep-Learning-Based Ensemble Sea Level Forecasting in the Northern Adriatic

Žust¹,

Fettich²,

Kristan³

et al. 2020

Preprint

View full text Add to dashboard Cite

Abstract. Complex interactions between atmospheric forcing, topographic constraints to air and water flow, and resonant character of the basin, make sea level modeling in Adriatic a particularly challenging problem. In this study we present an ensemble deep-neural-network-based sea level forecasting method HIDRA, which outperforms our general ocean circulation model ensemble (NEMO v3.6) for all forecast lead times and at a minuscule fraction of the numerical cost (order of 2 × 10−6). HIDRA exhibits larger bias but lower RMSE than NEMO over most of the residual sea level bins. It introduces a trainable atmospheric spatial encoder and employs fusion of atmospheric and sea level features into a self-contained network which enables discriminative feature learning. The HIDRA architecture building blocks are experimentally analyzed in detail and compared to alternative approaches. Results show individual importance of sea level input for accurate forecast lead times below 24 h and of the atmospheric input for longer time leads. The best performance is achieved by considering the input as the total sea level, split into disjoint sets of tidal and residual signals. This enables HIDRA to optimize the prediction fidelity with respect to atmospheric forcing, while compensating for the errors in the tidal model. HIDRA is trained and analysed on a ten-year (2006–2016) timeseries of atmospheric surface fields from a single member of ECMWF atmospheric ensemble. In the testing phase, both HIDRA and NEMO ensemble systems are forced by the ECMWF atmospheric ensemble. Their performance is evaluated on a one-year (2019) hourly time series from tide gauge in Koper (Slovenia). Spectral analysis of the forecasts at semi-diurnal frequency (12 h)−1 and at ground-state basin seiche frequency (21.5 h)−1 is performed by a continuous wavelet transform. The energy at the basin seiche in the HIDRA forecast is close to the observed, while NEMO underestimates it. Analyses of the January 2015 and November 2019 storm surges indicate that HIDRA has learned to mimic timing and amplitude of resonant sea level excitations in the basin.

show abstract

Statistical forecast of the marine surge

Quintana

Drumetz

Leballeur³

et al. 2021

Nat Hazards

View full text Add to dashboard Cite

On the Application of Machine Learning Techniques to Regression Problems in Sea Level Studies

Cited by 18 publications

References 27 publications

HIDRA 1.0: deep-learning-based ensemble sea level forecasting in the northern Adriatic

HIDRA 1.0: deep-learning-based ensemble sea level forecasting in the northern Adriatic

HIDRA 1.0: Deep-Learning-Based Ensemble Sea Level Forecasting in the Northern Adriatic

Statistical forecast of the marine surge

Contact Info

Product

Resources

About