ENSO‐GTC: ENSO Deep Learning Forecast Model With a Global Spatial‐Temporal Teleconnection Coupler

Mu, Bin; Qin, Bo; Yuan, Shijin

doi:10.1029/2022ms003132

Cited by 9 publications

(6 citation statements)

References 68 publications

(85 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In prior work, we have demonstrated that the Nonhomogeneous Hidden Markov Model (NHMM) applied to gridded, monthly SST data can objectively identify 5 ENSO flavors (whose spatial patterns and time of occurrence are similar to those of the 5 types of events typically identified), their temporal dynamics 24 , and their implications for global precipitation 23 . Further its 1-18 month ahead prediction skill for the benchmark Niño 3.4 index is comparable to that of other popular machine learning models 19,[25][26][27] . Given its interpretability and relative accuracy, we use it to explore how the temporal dynamics of ENSO may have changed relative to changing global temperature.…”

Section: Introductionsupporting

confidence: 56%

The influence of global warming on ENSO dynamics

Zhang,

Yan,

Nayak

et al. 2024

Preprint

View full text Add to dashboard Cite

How the dynamics of the El Niño-Southern Oscillation (ENSO) are being modified under a changing climate continues to be an open question with significant implications for global climate impacts. We present the first statistical attribution of how the spatio-temporal dynamics of ENSO have changed due to global warming over the last century using a non-homogeneous Hidden Markov Model of gridded tropical Pacific Sea Surface Temperature (SST) with smoothed global temperature as a covariate for the associated latent states. We identify 5 latent states whose space and time patterns correspond to classical El Niño and La Niña (LN), mild La Niña and Central Pacific El Niño (CP), and neutral ENSO patterns. The monthly probability of occurrence, and the transition probabilities of these states have changed significantly, and global temperature changes are identified as strong determinants of these changes. For instance, the probability of observing LN and CP has changed by 0.300 and 0.767 over the last 100 years. The estimated sensitivity of these probabilities to global temperature is (0.248±0.002)/℃ and (0.722±0.008)/℃, translating to 97% and 98% of the change, respectively, given the 1.22℃ warming. Similarly, the sensitivity of La Niña persistence to global temperature change is estimated as (28.9±0.2) month/℃, and this explains 98% of the variance in the increased persistence of La Niña patterns that are associated with recent recurrent droughts and floods in parts of the world.

show abstract

Section: Introductionsupporting

confidence: 56%

The influence of global warming on ENSO dynamics

Zhang,

Yan,

Nayak

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Recently, deep learning (DL) methods have been widely applied to ENSO forecasts (Fang et al., 2022; Geng & Wang, 2021; Guo et al., 2020; Ham et al., 2019, 2021; Hassanibesheli et al., 2022; He et al., 2019; Mu et al., 2021, 2022; Nooteboom et al., 2018; Saha & Nanjundiah, 2020; Wang et al., 2021; Yan et al., 2020; Zhou & Zhang, 2023; Zhou et al., 2023). Among various DL models, the convolutional neural network (CNN)‐based models have achieved significant success in ENSO prediction.…”

Section: Introductionmentioning

confidence: 99%

CNN‐Based ENSO Forecasts With a Focus on SSTA Zonal Pattern and Physical Interpretation

Sun,

Chen,

et al. 2023

Geophysical Research Letters

View full text Add to dashboard Cite

Deep learning (DL) has achieved notable success in El Niño‐Southern Oscillation (ENSO) forecasts. Most DL‐based models focused on forecasting ENSO indices while the zonal distribution of sea surface temperature anomalies (SSTA) over the equatorial Pacific was overlooked. To provide accurate predictions for the SSTA zonal pattern, this study developed a model through leveraging the merits of the cosine distance in constructing the convolutional neural network. This model can skillfully predict the SSTA zonal pattern over the equatorial Pacific 1 year in advance, remarkably outperforming current dynamical models. Moreover, the physical interpretation of the model prediction reveals that the sources for ENSO predictability at different lead times are distinct. For the 10‐month‐lead predictions, the precursors in the north Pacific, south Pacific and tropical Atlantic play critical roles in determining the model behaviors; while for the 16‐month‐lead predictions, the initial signals in the tropical Pacific associated with the discharge‐recharge cycle are essential.

show abstract

“…ANNs have shown potential in capturing non-linear relationships intrinsic to ENSO (Ham et al 2019;Zhao and Sun 2022;Zhang et al 2022). Some studies employing recurrent neural networks and Convolutional Neural Networks (CNN) have achieved modest success in lead times ranging from 6 to 12 months and beyond, often outperforming traditional models (Mu et al 2021(Mu et al , 2022Liu et al 2023;Patil et al 2023;Wang and Huang 2023;Chen et al 2023a, b). Compared to traditional ANN and CNN, the combination of Autoencoders (AE) (Saha et al, 2020;Ibebuchi and Richman 2024) and Long Short-Term Memory Networks (LSTM) represent a viable novel alternative for ENSO prediction.…”

Section: Introductionmentioning

confidence: 99%

Deep learning with autoencoders and LSTM for ENSO forecasting

Ibebuchi,

Richman

2024

Clim Dyn

View full text Add to dashboard Cite

El Niño Southern Oscillation (ENSO) is the prominent recurrent climatic pattern in the tropical Pacific Ocean with global impacts on regional climates. This study utilizes deep learning to predict the Niño 3.4 index by encoding non-linear sea surface temperature patterns in the tropical Pacific using an autoencoder neural network. The resulting encoded patterns identify crucial centers of action in the Pacific that serve as predictors of the ENSO mode. These patterns are utilized as predictors for forecasting the Niño 3.4 index with a lead time of at least 6 months using the Long Short-Term Memory (LSTM) deep learning model. The analysis uncovers multiple non-linear dipole patterns in the tropical Pacific, with anomalies that are both regionalized and latitudinally oriented that should support a single inter-tropical convergence zone for modeling efforts. Leveraging these encoded patterns as predictors, the LSTM - trained on monthly data from 1950 to 2007 and tested from 2008 to 2022 - shows fidelity in predicting the Niño 3.4 index. The encoded patterns captured the annual cycle of ENSO with a 0.94 correlation between the actual and predicted Niño 3.4 index for lag 12 and 0.91 for lags 6 and 18. Additionally, the 6-month lag predictions excel in detecting extreme ENSO events, achieving an 85% hit rate, outperforming the 70% hit rate at lag 12 and 55% hit rate at lag 18. The prediction accuracy peaks from November to March, with correlations ranging from 0.94 to 0.96. The average correlations in the boreal spring were as large as 0.84, indicating the method has the capability to decrease the spring predictability barrier.

show abstract

ENSO‐GTC: ENSO Deep Learning Forecast Model With a Global Spatial‐Temporal Teleconnection Coupler

Abstract: El Niño-Southern Oscillation (ENSO) is one of the most dominant phenomena that modulate the global climate changes and varieties (Holton & Dmowska, 1989), the annual amplitudes of which usually derive significant impacts on social economy and human sustainability (

Cited by 9 publications

References 68 publications

The influence of global warming on ENSO dynamics

The influence of global warming on ENSO dynamics

CNN‐Based ENSO Forecasts With a Focus on SSTA Zonal Pattern and Physical Interpretation

Deep learning with autoencoders and LSTM for ENSO forecasting

Contact Info

Product

Resources

About