Application of a neural network model to forecasting of El Niño and La Niña

Lubkov, A.S.; Voskresenskaya, E. N.; Marchukova, O.V.

doi:10.1088/1755-1315/386/1/012040

Cited by 6 publications

(6 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Other similar works include a standard neural network applied by Lubkov et al [42] to forecast El Niño and La Niña based on a set of global climatic indices of atmosphere-ocean system oscillations, with a lead time of 3 months. Petersik and Dijkstra [43] applied Gaussian density neural network and quantile regression neural network to forecast ONI, for lead times of up to 21 months.…”

Section: Related Workmentioning

confidence: 99%

Forecasting El Niño and La Niña Using Spatially and Temporally Structured Predictors and a Convolutional Neural Network

Hashemi

2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

El Niño-Southern Oscillation (ENSO) is characterized by large-scale fluctuations of sea surface temperature in the central and eastern tropical Pacific accompanied by changes in the atmospheric circulation. ENSO events are of two main types: El Niño and La Niña. Oceanic Niño index (ONI) determines the five consecutive 3-month running mean of sea surface temperature (SST) anomalies, in the Niño 3.4 region (5°S-5°N, 170°W-120°W). El Niño is a phenomenon in the equatorial Pacific Ocean characterized by a value of greater than 0.5°C for ONI. La Niña is a phenomenon in the equatorial Pacific Ocean characterized by a value of less than-0.5°C for ONI. The lingering of either of these two phenomena could induce severe droughts, while either of them following the other could cause massive floods. In both cases, deaths and substantial pecuniary loss is unavoidable, making their forecast of great significance. This study takes over the challenge of forecasting these two phenomena with one year lead time which has proven difficult in the literature. This research's contribution is restructuring ENSO events' predictors, including SST, sea level pressure, surface wind speed, and wind stress in a spatially and temporally meaningful way and designing a convolutional neural network (CNN) that takes advantage of this structure to forecast ENSO events of different types (i.e. Central Pacific and Eastern Pacific) in the next year. Not only a high precision in forecasts was achieved, but also it was shown that the proposed model has the potential to achieve higher recalls if a larger number of samples from the positive class would become available.

show abstract

Section: Related Workmentioning

confidence: 99%

Forecasting El Niño and La Niña Using Spatially and Temporally Structured Predictors and a Convolutional Neural Network

Hashemi

2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

show abstract

“…The adaptation of the model was carried out by analogy with the paper [61] and included the stages of preprocessing, modeling and postprocessing. At the stage of preliminary data processing, a search was made for the correlation relationships between the frequency of intense cyclones and the values of the indices of large-scale signals in the previous 2, 4 and 6 months, after that the input parameters having the greatest relationship with the prognostic parameter were selected for the model operational work.…”

Section: Neural Network Modelmentioning

confidence: 99%

Intense Cyclones in the Black Sea Region: Change, Variability, Predictability and Manifestations in the Storm Activity

et al. 2020

View full text Add to dashboard Cite

Cyclonic activity in the midlatitudes is a form of general atmospheric circulation, and the most intense cyclones are the cause of hydrometeorological anomalies that lead to economic damage, casualties and human losses. This paper examines the features of variability of intense cyclonic activity in the Black Sea region and the examples of their regional manifestations in the storm types. Based on 6-hourly NCEP/NCAR reanalysis data on 1000 hPa geopotential height fields with 2° × 2° spatial resolution and using the methodology by M.Yu. Bardin, objective data were obtained for the identification and estimation of the frequency of deep cyclones (reaching 0.75 and 0.95 quantiles by intensity and depth—intense and extreme cyclones, respectively) for the Black Sea region during the period 1951–2017. Additionally, a specific methodology of more precise cyclone identification based on spherical spline interpolation was successfully applied, and then the two methodologies were compared. The key point of the study is the following: In the background of negative significant linear trends and interdecadal variability (period of about 35 years), typical scales of their interannual variability on the periods of about 2.5–3.5 and 6–8 years were identified. These periods coincide with the time scales of the North Atlantic Oscillation and El Nino–Southern Oscillation, providing an outlook for further study of the patterns of their connection. Besides, seasonal forecasts of frequency of intense cyclones in the Black Sea region were successfully modeled using an artificial neural network technique. Finally, the case studies of regional manifestations of deep cyclones in the types of storms in the northern Black Sea coast revealed substantial differences in the location of deep centers of cyclones and storm tracks associated with the large-scale pressure fields.

show abstract

“…For instance, [21] found that the application of a neural network to create a nonlinear PCA approach was successful in extracting periodic modes in the tropical Pacific. References [22][23][24][25] used convolutional neural networks to forecast the time amplitude and type of ENSO. Labe and Barnes [26] applied ANN to predict the onset of slowdowns in decadal warming trends of global mean surface temperature.…”

Section: Introductionmentioning

confidence: 99%

Non-linear modes of global sea surface temperature variability and their relationships with global precipitation and temperature

Ibebuchi,

Richman

2024

Environ. Res. Lett.

View full text Add to dashboard Cite

Sea surface temperature (SST) modes are comprised of variability that arises from inherently nonlinear processes. Historically, a limitation arises from applying linear statistics to define these modes. Accurate depiction of the complex, non-linear nature of SST modes of variability necessitates the specification of a model capable of producing nonlinear patterns. In this study, we apply an artificial neural network algorithm integrated with autoencoders to analyze the seasonal non-linear global SST modes allowing for improved characterization of the modes and their large-scale temperature and precipitation teleconnections. Our results show that during boreal summer, SST cooling over the central to eastern tropical Pacific co-occurs with the Arctic amplification. In recent decades, the negative SST trend in the central to eastern tropical Pacific, combined with the positive trend in the western tropical Pacific is linked to an increase in the amplitude of SST modes associated with the Arctic warming, resulting in warmer temperatures over large portions of the global land, particularly over Greenland. In boreal winter, El Niño Southern Oscillation (ENSO) is the prominent global SST mode. The distinct spatiotemporal patterns of ENSO modes are associated with unique effects on regional land temperature and precipitation. The central Pacific El Niño is more associated with the combination of warm and dry conditions over western Australia, and the northern part of South America. Conversely, the central to eastern El Niño is more associated with the combination of warm and dry conditions over parts of Southern Africa, and the northern part of South America. The spatiotemporal patterns and trends in the amplitude of the analyzed non-linear global SST modes alongside their regional influences on temperature and precipitation are discussed. The broader impact of this study is on the potential of neural networks in effectively delineating non-linear global SST modes and their associations with regional climates.

show abstract

Application of a neural network model to forecasting of El Niño and La Niña

Cited by 6 publications

References 10 publications

Forecasting El Niño and La Niña Using Spatially and Temporally Structured Predictors and a Convolutional Neural Network

Forecasting El Niño and La Niña Using Spatially and Temporally Structured Predictors and a Convolutional Neural Network

Intense Cyclones in the Black Sea Region: Change, Variability, Predictability and Manifestations in the Storm Activity

Non-linear modes of global sea surface temperature variability and their relationships with global precipitation and temperature

Contact Info

Product

Resources

About