Forecasting El Niño and La Niña events using decision tree classifier

Silva, Kárita Almeida; Rolim, Glauco de Souza; Aparecido, Lucas Eduardo de Oliveira

doi:10.1007/s00704-022-03999-5

Cited by 6 publications

(2 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to predict the possibility of a triple La Niña event in the future, based on the three eigenvalue factors derived from PCA, the paper established a time series analysis model and conducted stability tests using the ARIMA model, concluding that La Niña events have a certain periodicity, which means that the possibility of a future La Niña event is high [8].…”

Section: Arimamentioning

confidence: 99%

“…The weights for the 11 indicators affected by La Niña were determined using formulas ( 5) to (8). The specific results are provided in Table 2, which includes indicators such as GDP, hydropower capacity, grain production, total area of fire impact, fire damage, drought-affected area, affected population, death toll, economic losses caused, affected forest area, and area of drought-induced crop failure.…”

Section: Determination Of Weightsmentioning

confidence: 99%

See 1 more Smart Citation

Loss Assessment and Response Strategies Based On ARIMA And TOPSIS Models Under Triple La Niña Events

Liu

2024

HBEM

View full text Add to dashboard Cite

This study aims to develop forecasting and quantitative loss analysis models for potential triple La Niña events in different countries in order to assess and respond to potential La Niña disaster losses. First ,to build the ARIMA model, this study collected data from 2011 to 2020 and selected five indicators including Sea Surface Temperature (SST), Precipitation (PRCP), Temperature(TEMP), Standard Temperature and Pressure(STP) and Sea Level Pressure(SLP) that were more significantly correlated with the La Niña phenomenon, and then conducted correlation analysis using Pearson coefficient. Based on this, the research conducted principal component analysis to obtain three characteristic factors SST, PRCP and TEMP, and based on them, building ARIMA model to predict the possibility of La Niña occurrence in the future. Then, the study used the ADF test to check the smoothness. Second, this study carried out a TOPSIS evaluation to analyse the multiple damages caused by the high temperature and drought brought by the La Niña event in a country, taking into account the entropy weighting method-TOPSIS evaluation model, and determined the ranking of the indicators: La Niña event has the greatest impact on agriculture, followed by ecology and environment. Finally, the results of the score ranking were used to provide solutions. This study predicts the probability of future La Niña events through the development of two models, ARIMA and TOPSIS, assesses and analyses multi-indicator losses, comprehensively evaluates and ranks loss targets, and makes policy recommendations. The results of these studies contribute to the prevention and mitigation of losses caused by La Niña events, ensuring human safety and minimising economic losses.

show abstract

Section: Arimamentioning

confidence: 99%

Section: Determination Of Weightsmentioning

confidence: 99%