Pacific El Research Articles

Impact of anomalous ocean warming on the abundance and foraging habits of the California Sea Lion (Zalophus californianus) from the San Benito Archipelago in the Mexican Pacific

Abstract The California sea lion (Zalophus californianus; CSL) reflects environmental conditions of the Northeastern Pacific, where resource availability was importantly reduced for different taxa during climatic variations such as the North Pacific heatwave (The Blob 2014 to 2015) or El Niño 2015 to 2016 due to the effect of these anomalies on primary productivity. Our objective was to determine the abundance and foraging habits of CSLs from the San Benito Archipelago (Mexican Pacific) in a period (2013 to 2019) that involved warm anomalies. Annual counts were conducted by sex/age classes; scats were sampled to identify the main prey consumed; and pup fur (maternal foraging indicators) was collected for analysis of δ13C and δ15N. The prey importance index (PII) and trophic level were calculated and the Siber assessment in R was used to estimate isotopic niches. Colony abundance decreased by approximately 50% from 2013 to 2019. Values of δ13C showed significant interannual differences (P < 0.001) with more negative values (−17.2‰ to −16.7‰) from 2015 onwards, indicating offshore foraging habits. Significant interannual differences (P < 0.001) were found for δ15N, with a declining trend (20.4‰ to 19.07‰), and a difference of more than 1‰ between 2013 and 2019, evidencing a probable shift of foraging grounds toward higher latitudes. Based on the PII, Strongylura exilis (30%), Merluccius productus (24%), Ommastrephes bartramii (10%), and Sardinops sagax (10%) were the most important prey, showing significant interannual differences (P < 0.001). CSLs had a mean trophic level of 4.89, with a slight increasing trend toward the end of the period. Dietary plasticity was found in CSLs, possibly caused by environmental changes that modified prey availability and led to variations in foraging grounds, and consequently in the colony abundance over time.

Lag-WALS approach incorporating ENSO-related quantities for altimetric interannual SLA forecasts in the South China Sea

A novel approach using lag weighted-average least squares (Lag-WALS) is proposed to forecast the interannual sea level anomaly (SLA) in the South China Sea (SCS) using lagged equatorial Pacific El Niño–Southern Oscillation (ENSO)-related quantities. Through empirical orthogonal function (EOF) and wavelet coherence method, we first investigated the relationships between sea surface temperature (SST) and SLA (both steric sea level (SSL) and non-steric sea level (NSSL)) in the equatorial Pacific, and then explored their cross-correlations with the interannual SCS SLA. A robust alignment was found between the first spatiotemporal mode of EOF (i.e. EOF1 and first principal component (PC1)) from SLA/SSL and SST across the equatorial Pacific, both of which exhibited a typical ENSO horseshoe spatial pattern in EOF1. Good consistency between the SCS SLA and the SST/SLA/SSL PC1 was revealed, with the SCS SLA lagging behind the SST, SLA, and SSL by several months at most grid locations. In contrast, the NSSL exhibited large disparities with the SST PC1 or the interannual SCS SLA. The lag-WALS model performed better at the SCS boundaries than in the central region, with an average STD/MAE/Bias (RMSE/MAE/Bias) for internal (external) accuracies of 1.01/0.80/–0.002 cm (1.39/1.13/–0.08 cm), respectively. The altimetric-observed SLA seasonal patterns agreed with the Lag-WALS model-forecasted SLA. A similar situation applies to regionally-averaged SLA time series. These results underscore the ability of the Lag-WALS model to accurately forecast the SCS SLA at the interannual scale, which is crucial for early warning of abnormal sea level changes in the SCS.

The Different Effects of Two Types of El Niño on Eastern China’s Spring Precipitation During the Decaying Stages

El Niño is one of the most significant global climatic phenomena affecting the East Asian atmospheric circulation and climate. This study uses multi-source datasets, including observations and analyses, and statistical methods to investigate the variations and potential causes of boreal spring precipitation anomalies in eastern China under different El Niño sea surface temperature conditions, namely, the Eastern Pacific and Central Pacific (EP and CP) El Niño cases. The findings reveal that, particularly along the Yangtze–Huaihe valley, spring precipitation markedly increases in most regions of eastern China during the EP El Niño decaying stages. Conversely, during the CP El Niño decaying stages, precipitation anomalies are weak, with occurrences of weak negative anomalies in the same regions. Further analyses reveal that during the decaying spring of different El Niño cases, differences in the location and strength of the Northwest Pacific (NWP) abnormal anticyclone, which is associated with the central–eastern Pacific warm sea surface temperature anomaly (SSTA), result in distinct anomalous precipitation responses in eastern China. The SSTA center of the EP El Niño is more easterly and stronger. In the meantime, NWP abnormal anticyclones are more easterly and have a broader range, facilitating water vapor transport over eastern China. By contrast, the CP El Niño SSTA center is westward and relatively weaker, leading to a relatively weak, westward, and narrower anomalous NWP anticyclone that causes less significant water vapor transport anomalies in eastern China. This paper highlights the diverse impacts of El Niño diversity on regional atmospheric circulation and precipitation, providing valuable scientific references for studying regional climate change in East Asia.

Spatiotemporal pattern of ENSO-induced modulation on landscape fires over Pacific Rim from 2001‒2020

While El Niño-Southern Oscillation (ENSO) modulates landscape fire activities across the Pacific Rim, the detailed patterns and mode-specific variations of these effects remain poorly understood. This study aims to address that gap by systematically assessing the spatiotemporal variations in burned areas from 2001‒2020 during different modes of ENSO, namely, Eastern Pacific (EP) El Niño, EP La Niña, Central Pacific (CP) El Niño and CP La Niña. We evaluate the modulation patterns using classical correlation and cross-spectrum analyses, focusing on five hotspot regions: Temperate North America (TENA), Southern Hemisphere South America (SHSA), Northern Hemisphere South America (NHSA), Equatorial Asia (EQAS) and Australia and New Zealand (AUST). El Niño and La Niña exhibit asymmetrical opposite effects on burned areas, as expected. The findings reveal spatial non-uniformity in the impact of ENSO on landscape fires, with La Niña enhancing burned areas in TENA and SHSA, whereas El Niño predominantly affects AUST, EQAS and NHSA. CP and EP ENSO events differ in intensity and their large-scale circulation patterns. These differences cause notable variations in the regional burned area. Water vapour redistribution emerges as the key driver. The study emphasises the heterogeneity in ENSO modes in regulating landscape fires, providing insights for future landscape fire risk assessment and prevention efforts under climate change.

Synchronous decadal climate variability in the tropical Central Pacific and tropical South Atlantic

Pantropical climate interactions across ocean basins operate on a wide range of timescales and can improve the accuracy of climate predictions. Here, we show in observations that Central Pacific (CP) El Niño-like sea surface temperature (SST) anomalies have coevolved with tropical South Atlantic SST anomalies on a quasi-decadal (~10-year) timescale over the past seven decades. During the austral autumn–winter season, decadal warm SSTs in the tropical CP effectively induce tropical SST cooling in the South Atlantic, mainly by strengthening the South Atlantic subtropical anticyclone via an extratropical atmospheric wave teleconnection in the southern hemisphere. Partially coupled pacemaker simulations corroborate the observational findings, indicating that tropical CP decadal SSTs play a primary pacing role, while Atlantic feedback is of secondary importance throughout the study period. Our results suggest that the tropical CP could be an important source of decadal predictability for tropical South Atlantic SST and the surrounding climate.

Responses of the Southern Ocean mixed layer depth to the eastern and central Pacific El Niño events during austral winter

Responses of the Southern Ocean mixed layer depth to the eastern and central Pacific El Niño events during austral winter

Distinct Features of Tropical Cyclone Landfall over East Asia during Various Types of El Niño

Abstract Numerous studies focus on the impacts of ENSO diversity on tropical cyclone (TC) activities in the western North Pacific (WNP). In recent years, there is a growing threat of landfalling and northward-moving TCs in East Asia, accompanying an increase in central Pacific (CP) El Niño. Here, we aim to discover variations in landfalling TCs during various types of CP El Niño (CP-I and CP-II El Niño). It is found that significant changes in landfalling and going northward TCs over East Asia north 20°N are modulated by CP-I El Niño. During CP-I El Niño, TCs tend to landfall more often over the mainland of China with longer duration, moving distance, and stronger power dissipation index (PDI) after landfall and increased TC-induced rainfall, due to favorable conditions (beneficial steering flow, weak vertical wind shear, increased specific humidity, increased soil moisture, and temperature), especially significant over the northeastern part. The situation over the mainland of China is reversed during eastern Pacific (EP) El Niño and CP-II El Niño, with a significant decrease in the characteristics with corresponding unfavorable environments. Over the Korean Peninsula and Japan, the frequency of TC landfalls, as well as the duration and the moving distance after landfall, exhibits greater levels during CP-I and CP-II El Niño than during EP El Niño due to favorable steering flow, and thus, TC-induced rainfall enhances correspondingly. Regarding the PDI over the Korean Peninsula and Japan, it remains relatively consistent across all El Niño types. However, a notable increase in the PDI during EP El Niño could be attributed to the higher intensity of TCs prior to landfall.

Long-term variability of the Kuroshio since 1788 based on coral records

The Kuroshio transports warm water poleward from low- to mid-high latitude regions and strongly affects ocean-atmosphere-land interactions. Previous studies have shown that the Kuroshio has strengthened at its origin during recent decades. However, changes in the Kuroshio on a centennial timescale are unclear due to the lack of observations before the Industrial Revolution. In this study, a precisely dated and monthly resolved Porites coral Sr/Ca record from south of Taiwan island, where coral growth is influenced by the Kuroshio, was used to generate a continuous reconstruction of Kuroshio transport (KT) during the period from 1788 to 2013. The data show a consistent decline in KT since 1788, with the rate of decline increasing since the 1950s, probably due to rapid oceanic warming. The southward shift of the bifurcation latitude of the North Equatorial Current, which is related to accelerated global oceanic warming, might have led to the decrease in KT. Natural variability in the phase transition of the Pacific Decadal Oscillation and El Niño-Southern Oscillation might also influence the long-term changes in KT, particularly before the 1950s.

Pronounced spatial disparity of projected heatwave changes linked to heat domes and land-atmosphere coupling

Heatwaves are projected to substantially increase at a global scale, exacerbating worldwide heat-related risks in the future. However, understanding future heterogeneous heatwave changes and their origins remains challenging. By analyzing the output of various climate models from the Coupled Model Intercomparison Project Phase 6, we found pronounced spatial disparity of projected heatwave increases in the Northern Hemisphere, even outstretching seven-fold inter-regional differences in extreme heatwave occurrences, attributed primarily to future changes in heat-dome-like circulations and soil moisture–temperature coupling. Specifically, we found that by the end of the 21st century, the modulations of combined Pacific El Niño and positive Pacific Meridional Mode on magnified heat-dome-like circulations would be translated into summertime hotspots over western Asia and western North America. Amplified soil moisture–temperature couplings then further aggravate the heatwave intensity over these two hotspots. This study provides support for formulating impact-based mitigation strategies and efficiently addressing the potential future risks of heatwaves.

Why does there occur spring predictability barrier for eastern Pacific El Niño but summer predictability barrier for central Pacific El Niño?

Why does there occur spring predictability barrier for eastern Pacific El Niño but summer predictability barrier for central Pacific El Niño?

Pacific Interannual and Multidecadal Variability Recorded in δ18O of South American Summer Monsoon Precipitation

AbstractThe South American summer monsoon (SASM) generates important hydroclimatic impacts in (sub‐)tropical South America and isotopic tracers recorded in paleoclimatic archives allow for assessing its long‐term response to Pacific variability prior to modern observations. Stable oxygen isotopes in precipitation integrate hydroclimatic changes during the SASM mature phase from December to February (DJF) in response to the Interdecadal Pacific Oscillation (IPO) and El Niño—Southern Oscillation (ENSO), respectively. Here, results from the isotope‐enabled Community Atmosphere Model v.5 are compared with highly resolved and precisely dated isotopic records from speleothems, tree rings, lake and ice cores during the industrial era (1880–2000 CE) and validated against observations from the International Atomic Energy Agency (IAEA) network. Pacific sea surface temperatures (SSTs) are coupled to the isotopic composition of SASM precipitation through perturbations in the Walker circulation associated with low‐ (IPO) and high‐frequency (ENSO) variability, impacting convective activity over tropical South America and the tropical Atlantic. Changes in convection over this monsoon entrance region ultimately control the downstream oxygen isotopic composition of precipitation recorded in paleoclimate archives. Overall, model results, paleoclimate records and IAEA data agree on the isotopic response to Pacific SST forcing. These results highlight the potential for long isotopic paleoclimate records to reconstruct Pacific climate variability on both high‐ and low‐frequency timescales. Furthermore, the isolation of the IPO signal in a diverse set of isotopic archives invites the reinterpretation of other paleoclimate proxies for identifying this historically overlooked forcing.

Enhanced interaction between ENSO and the South Atlantic subtropical dipole over the past four decades

AbstractThis study investigates the relationship between sea surface temperature (SST) anomalies in the subtropical Atlantic Ocean, as represented by the Southern Atlantic subtropical dipole (SASD), and SST anomalies in the tropical Pacific Ocean, identified by the El Niño‐Southern Oscillation (ENSO). Contrary to the previously held notion of a weak relationship between SASD and ENSO as suggested by earlier literature, our analysis reveals a substantial inverse correlation between the two. This correlation exhibits significant multi‐decadal variability, which has notably intensified over the most recent two decades compared with the preceding two decades. This intensification in the SASD–ENSO inverse correlation may be attributed to the shift in ENSO regime from predominance of eastern Pacific El Niño to central Pacific El Niño events around the turn of the century. This transition triggers wavetrains that propagate along different paths, consequently influencing the South Atlantic subtropical high and inducing alterations in anomalous SST patterns in the subtropical Atlantic Ocean. These findings advance our comprehension of the interactions between South Atlantic and Pacific SST variations, which strongly influence rainfall patterns, particularly in South America and southern Africa. Understanding such teleconnection holds promise for improving sub‐seasonal to seasonal precipitation predictions in these regions.

Holocene extreme flood events in the middle reaches of the Lancang‒Mekong River basin recorded by a high-altitude lake in southwestern China

The Lancang‒Mekong River (LMR) basin is a region particularly susceptible to floods induced by extreme precipitation. However, owing to limited historical flood records, it is difficult to understand the processes and mechanisms of extreme floods in the LMR basin. In the present study, sediments from a high-altitude lake (Lake Baima, 4700 m above sea level) located in the upper-middle reaches of the LMR are used to reconstruct climate and hydrology variability since the last deglaciation based on a range of climate proxies and methods, including loss-on-ignition, grain size, X-ray diffraction, and X-ray fluorescence. Through the investigation of modern and historical floods in the LMR basin, we combine the climate proxies with hydrological frequency curves to reconstruct the paleoflood sequence during the Holocene. Our results suggest that the temperature and precipitation in southwestern China increased during the Holocene compared to the last deglaciation, indicating a response to changes in Northern Hemisphere summer solar insolation. In addition, Holocene sediments of Lake Baima are primarily composed of brown background layers and numerous light yellow/gray event layers. These event layers are further identified as flood events based on sedimentary facies, redness, and main exogenous elements content and their rate of change. The results show that extreme flood events in the LMR basin occurred more frequently during the early and late Holocene, with seven once-in-1000-year floods during the two intervals; while there were low-frequency flood periods during the mid-Holocene. The occurrence of Holocene floods may be related to external forcing, such as the abrupt changes in total solar irradiance (TSI) and volcanic activity, as well as the internal variability of the wider climate system including the Pacific Decadal Oscillation (PDO) and El Niño‒Southern Oscillation (ENSO). When the TSI was low, the Intertropical Convergence Zone (ITCZ) in the Northern Hemisphere shifted northward, and the Western Pacific Subtropical High (WPSH) moved eastward, accompanied with a negative phase of the PDO and La Niña-like phase of the ENSO. The air-sea interactions increased sea surface temperatures, leading to high precipitation and increasing flood risk in the LMR basin.

Interdecadal change in the relationship between the South China late rainy season rainfall and equatorial Pacific SSTs

Abstract The South China (SC) late rainy season rainfall (SCLR) is important to people’s livelihoods and properties of this densely populated and economically developed region. Based on observation and reanalysis datasets, this study identifies that the relationship between the SCLR and equatorial Pacific SSTs (EPSSTs) experiences an interdecadal change around the late 1970s. The SCLR is negatively correlated with the eastern EPSSTs during 1948–1977 (the previous epoch) but positively correlated with the central EPSSTs during 1978–2022, especially during the last three decades 1993–2022 (the post epoch). This is due to the interdecadal change in El Niño flavors and EPSSTs-tropical Atlantic SSTs (TASSTs) relationship. With the increasing frequency of central Pacific El Niños in the post epoch, the El Niño-related warm SST anomaly center shifts from the eastern equatorial Pacific (EEP) to the central equatorial Pacific (CEP). Correspondingly, the anomalous cyclone exerted by the CEP warming reaches SC and causes rainfall surplus there (westward propagation pathway). In contrast, the EEP warming in the previous epoch can’t influence the SCLR through the westward propagation pathway, but through shifting the westerly jet southward via heating the tropical troposphere. The southward shifted westerly jet crashes the Qinghai-Tibet Plateau, triggering cyclonic circulations across the northern East Asia and North Pacific, which in turn cause anticyclonic circulations over the western SC and reduce SC rainfall (eastward propagation pathway). On the other hand, the CEP warming in the post epoch barely influence the SCLR through the eastward propagation pathway because of the enhanced negative relationship between EPSSTs and TASSTs.

The Influence of ENSO Diversity on Future Atlantic Tropical Cyclone Activity

Abstract El Niño–Southern Oscillation (ENSO) influences seasonal Atlantic tropical cyclone (TC) activity by impacting environmental conditions important for TC genesis. However, the influence of future climate change on the teleconnection between ENSO and Atlantic TCs is uncertain, as climate change is expected to impact both ENSO and the mean climate state. We used the Weather Research and Forecasting Model on a tropical channel domain to simulate 5-member ensembles of Atlantic TC seasons in historical and future climates under different ENSO conditions. Experiments were forced with idealized sea surface temperature configurations based on the Community Earth System Model (CESM) Large Ensemble representing: a monthly varying climatology, eastern Pacific El Niño, central Pacific El Niño, and La Niña. The historical simulations produced fewer Atlantic TCs during eastern Pacific El Niño compared to central Pacific El Niño, consistent with observations and other modeling studies. For each ENSO state, the future simulations produced a similar teleconnection with Atlantic TCs as in the historical simulations. Specifically, La Niña continues to enhance Atlantic TC activity, and El Niño continues to suppress Atlantic TCs, with greater suppression during eastern Pacific El Niño compared to central Pacific El Niño. In addition, we found a decrease in the Atlantic TC frequency in the future relative to historical regardless of ENSO state, which was associated with a future increase in northern tropical Atlantic vertical wind shear and a future decrease in the zonal tropical Pacific sea surface temperature (SST) gradient, corresponding to a more El Niño–like mean climate state. Our results indicate that ENSO will remain useful for seasonal Atlantic TC prediction in the future.

The Interdecadal Changes in the Spatial Structure of the South Pacific Oscillation and Its Implication for ENSO Diversity

Abstract The South Pacific Oscillation (SPO), characterized by a north–south dipole-like pattern of sea level pressure anomalies, is one of the key factors in understanding tropical–extratropical interactions in the South Pacific. We show that in boreal summer (June–August), the center of the northern lobe sea level pressure anomalies in the SPO is shifted to the east gradually after the 1960–70s. This study focuses on the relationship between the boreal summer SPO and following winter El Niño–Southern Oscillation (ENSO) diversity before and after the eastward shift of the SPO’s subtropical lobe. The eastward shift of the SPO’s subtropical lobe altered both the seasonal footprint mechanism and the trade wind charging mechanism associated with the SPO and thus profoundly influenced the ENSO diversity. It is revealed that when the northern lobe of the SPO shifts to the west of its average location, it tends to strengthen the eastern Pacific (EP) El Niño mainly via the seasonal footprint mechanism. But after the SPO’s northern lobe shifts to the east of its average location, it tends to promote the development of central Pacific (CP) El Niño mainly via the trade wind charging mechanism. The changes in the spatial structure of convection over the tropical Pacific and Indian Oceans may be one of the possible causes for the eastward shift in the SPO’s northern lobe. The findings in the present study have implications for a better understanding of ENSO diversity. Significance Statement Previous studies have demonstrated that the South Pacific Oscillation (SPO), as an important El Niño–Southern Oscillation (ENSO) precursor in the South Pacific, has the potential to provide an enhancement of the prediction of specific ENSO flavor. However, the historical variation in the SPO’s spatial structure and related changes in the relationship with the diversity of ENSO are still unclear. In this paper, we show that the subtropical lobe of the boreal summer (June–August) SPO is shifted to the east gradually after the 1960–70s. The changes in the spatial structure have also altered both the seasonal footprint mechanism and the trade wind charging mechanism which play important roles in the developmental processes of different types of ENSO. Our work highlights the importance of the interdecadal changes in the spatial structure of the SPO in understanding the relationship between the SPO and ENSO diversity.

What Determines the East Asian Winter Temperature during El Niño?—Role of the Early Onset El Niño and Tropical Indian Ocean Warming

Abstract Atmospheric teleconnections from the Pacific El Niño are key to determining the East Asian winter climate. Using the database for policy decision-making for future climate change (d4PDF) large-ensemble simulations, the present study investigates a mechanism for the warm and cold East Asian winters during El Niño with a focus on atmospheric teleconnections triggered by anomalous sea surface temperature (SST) patterns in the tropical Indo-Pacific. Our results show that the western Pacific (WP) teleconnection pattern plays a primary role in the warm winters in East Asia. The WP pattern tends to appear in years when both an early El Niño and the positive phase of the Indian Ocean dipole (IOD) mode develop in boreal autumn. In those years, the tropical Indian Ocean (TIO) strongly warms in the following winter, forming a distinct zonal contrast in precipitation anomalies over the tropical Indo-Pacific through a reduced Walker circulation. The Rossby wave source anomalies indicate that the WP pattern is associated with the weakened Indo-Pacific Walker circulation. By contrast, the WP pattern does not dominate in the cold winters due to the absence of strong TIO warming. The present study proposes a mechanism that promotes the excitation of the WP pattern through the upper-troposphere divergence in East Asia associated with the Walker circulation modulated by the tropical Indo-Pacific interbasin interaction. Significance Statement The East Asian winter temperature variability is controlled not only by the strong atmospheric internal variability in the midlatitudes and high latitudes but also by remote forcing from the tropical ocean. Our study investigates how El Niño exerts diverse impacts on the East Asian winter temperature, depending on where atmospheric convection intensifies in the tropical Pacific Ocean and the Indian Ocean. Our results show that an intense warming of the tropical Indian Ocean and the early development of El Niño are the major factors for warm winters in East Asia. Given that a precursor of the intense Indian Ocean warming appears in boreal autumn, our findings should contribute to the improvement of seasonal prediction for the East Asian winter climate.

Coccolithophore Assemblage Response to Indian Monsoon–ENSO Teleconnections During the Last Century

AbstractWe present a new, annually resolved coccolithophore relative abundance record (1889–1993 CE) from a laminated marine sediment core obtained from the northeastern Arabian Sea to document assemblage variations during the last century. Our analysis revealed that ∼80% of the assemblage is composed of the coccolithophore species Gephyrocapsa oceanica, Emiliania huxleyi, and Florisphaera profunda. Hierarchical clustering of the coccolithophore data delineates time intervals (1889–1909 CE, 1910–1959 CE, 1960–1993 CE) comparable to previously reported interdecadal changes of the El Niño–Southern Oscillation (ENSO)–Indian monsoon variance. We find that F. profunda (Fp%) is positively correlated with central Pacific ENSO (Niño 4 and Niño 3.4). Frequency analyses of Fp% match ENSO quasiperiodicity reported from instrumental data and proxy records and support the existence of previously described ENSO–Indian monsoon teleconnections. Furthermore, the increased variance of Fp% and associated decrease in marine primary productivity starting ∼1960 CE appears to be driven by decreased upward nutrient transport linked to increased surface water stratification due to surface ocean warming. With the predicted increases in frequency and strength of central Pacific El Niño as well as surface ocean warming and upper ocean stratification in the Indian Ocean over the coming century, our results imply further impacts on coccolithophores in the northeastern Arabian Sea that may have cascading effects on the marine food web and global biogeochemical cycles.

Comparison of machine learning models in forecasting different ENSO types

Accurate forecasting of the El Niño Southern Oscillation (ENSO) plays a critical role in mitigating the impacts of extreme weather conditions linked to ENSO variability on ecosystems. This study evaluates the performance of six machine learning models in forecasting two ENSO types: the Central Pacific El Niño (Niño 4 index) and the East Central Pacific El Niño (Niño 3.4 index). The models analyzed include the Feed Forward Neural Network (FFNN), Long Short-term Memory (LSTM) neural network, eXtreme Gradient Boosting Regressor, K-Nearest Neighbors Regressor, Gradient Boosting Regressor, and Support Vector Regressor, using the ENSO index lagged by six months as the predictor. The models were trained on the monthly ENSO indices from 1870 to 1992 and tested from 1993 to 2023. We also assess the relative predictability of the two ENSO types. Events were defined as when the ENSO index exceeded ±0.4. Our evaluation during the testing period reveals that for the analyzed models, the deep neural network models (LSTM and FFNN) demonstrated superior performance in forecasting ENSO at a 6-month lead time. Furthermore, all models achieved impressive all-season correlations ranging from 0.93 to 0.97 and threat score for the ENSO phases between 0.71 to 0.88 for Niño 3.4 events, and 0.72 to 0.93 for Niño 4 events. The predictability of the two ENSO types depended on the model and strength of the ENSO event. Considering both ENSO phases, La Niña events were forecasted with a higher accuracy relative to El Niño events, and all models, besides the deep learning models, notably fell short in capturing the extreme 2015/2016 Central Pacific El Niño event. These results highlight the potential of machine learning models, particularly the deep learning approaches, for skillful ENSO forecasting, by leveraging its historical data.

Deep ocean warming-induced El Niño changes

The deep ocean, a vast thermal reservoir, absorbs excess heat under greenhouse warming, which ultimately regulates the Earth’s surface climate. Even if CO2 emissions are successfully reduced, the stored heat will gradually be released, resulting in a particular pattern of ocean warming. Here, we show that deep ocean warming will lead to El Niño-like ocean warming and resultant increased precipitation in the tropical eastern Pacific with southward shift of the intertropical convergence zone. Consequently, the El Niño-Southern Oscillation shifts eastward, intensifying Eastern Pacific El Niño events. In particular, the deep ocean warming could increase convective extreme El Niño events by 40 to 80% relative to the current climate. Our findings suggest that anthropogenic greenhouse warming will have a prolonged impact on El Niño variability through delayed deep ocean warming, even if CO2 stabilization is achieved.