# Predicting Meteorological Variables on Local Level with SARIMA, LSTM and Hybrid Techniques

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Materials and Methods

#### 2.1. Data

#### 2.2. ARIMA Methodology

#### 2.3. Artificial Neural Networks (ANN)–LSTMs

#### 2.4. ARIMA–LSTM Hybrid Methodology

#### 2.5. Benchmarking

## 3. Results

#### 3.1. Temperature and Humidity Forecast

#### 3.2. Wind Speed Forecast

## 4. Discussion—Future Work

## 5. Conclusions

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## References

- Bochenek, B.; Ustrnul, Z. Machine Learning in Weather Prediction and Climate Analyses—Applications and Perspectives. Atmosphere
**2022**, 13, 180. [Google Scholar] [CrossRef] - Georgopoulou, E.; Mirasgedis, S.; Sarafidis, Y.; Hontou, V.; Gakis, N.; Lalas, D.P. Climatic preferences for beach tourism: An empirical study on Greek islands. Theor. Appl. Climatol.
**2019**, 137, 667–691. [Google Scholar] [CrossRef] - Elmira, J.; Remaz, O.D. Intra Urban Air Temperature Distributions in Historic Urban Center. Am. J. Environ. Sci.
**2012**, 8, 503. [Google Scholar] - Scott, D.; Lemieux, C. Weather and Climate Information for Tourism. Procedia Environ. Sci.
**2010**, 1, 146–183. [Google Scholar] [CrossRef] [Green Version] - Zeng, Z.; Wang, Z.; Gui, K.; Yan, X.; Gao, M.; Luo, M.; Geng, H.; Liao, T.; Li, X.; An, J.; et al. Daily Global Solar Radiation in China Estimated From High-Density Meteorological Observations: A Random Forest Model Framework. Earth Space Sci.
**2020**, 7, e2019EA001058. [Google Scholar] [CrossRef] [Green Version] - Shih, H.; Rajendran, S. Comparison of Time Series Methods and Machine Learning Algorithms for Forecasting Taiwan Blood Services Foundation’s Blood Supply. J. Health Eng.
**2019**, 2019, 6123745. [Google Scholar] [CrossRef] - Sillmann, J.; Thorarinsdottir, T.; Keenlyside, N.; Schaller, N.; Alexander, L.V.; Hegerl, G.; Seneviratne, S.I.; Vautard, R.; Zhang, X.; Zwiers, F.W.; et al. Understanding, modeling and predicting weather and climate extremes: Challenges and opportunities. Weather. Clim. Extrem.
**2017**, 18, 65–74. [Google Scholar] [CrossRef] - Wilks, D.S.; Wilby, R.L. The weather generation game: A review of stochastic weather models. Prog. Phys. Geogr. Earth Environ.
**1999**, 23, 329–357. [Google Scholar] [CrossRef] - Shankarnarayan, V.K.; Ramakrishna, H. Comparative study of three stochastic future weather forecast approaches: A case study. Data Sci. Manag.
**2021**, 3, 3–12. [Google Scholar] [CrossRef] - Box, G. Box and Jenkins: Time Series Analysis, Forecasting and Control. In A Very British Affair: Six Britons and the Development of Time Series Analysis during the 20th Century; Mills, T.C., Ed.; Palgrave Macmillan UK: London, UK, 2013; pp. 161–215. [Google Scholar]
- Hyndman, R.J.; Athanasopoulos, G. Forecasting: Principles and Practice; OTexts: Melbourne, Australia, 2014. [Google Scholar]
- Grigonytė, E.; Butkevičiūtė, E. Short-term wind speed forecasting using ARIMA model. Energetika
**2016**, 62, 1–2. [Google Scholar] [CrossRef] [Green Version] - Lai, Y.; Dzombak, D.A. Use of the Autoregressive Integrated Moving Average (ARIMA) Model to Forecast Near-Term Regional Temperature and Precipitation. Weather Forecast.
**2020**, 35, 959–976. [Google Scholar] [CrossRef] - Jayawardena, A.; Gurung, A. Noise reduction and prediction of hydrometeorological time series: Dynamical systems approach vs. stochastic approach. J. Hydrol.
**2000**, 228, 242–264. [Google Scholar] [CrossRef] - Paulescu, M.; Paulescu, E.; Gravila, P.; Badescu, V. Time Series Forecasting. In Weather Modeling and Forecasting of PV Systems Operation; Paulescu, M., Paulescu, E., Gravila, P., Badescu, V., Eds.; Springer: London, UK, 2013. [Google Scholar]
- Tylkowski, J.; Hojan, M. Time Decomposition and Short-Term Forecasting of Hydrometeorological Conditions in the South Baltic Coastal Zone of Poland. Geosciences
**2019**, 9, 68. [Google Scholar] [CrossRef] [Green Version] - Kreuzer, D.; Munz, M.; Schlüter, S. Short-term temperature forecasts using a convolutional neural network—An application to different weather stations in Germany. Mach. Learn. Appl.
**2020**, 2, 100007. [Google Scholar] [CrossRef] - Kuhn, M.; Johnson, K. Applied Predictive Modeling, 1st ed.; Springer Science Business Media: New York, NY, USA, 2018; p. 600. [Google Scholar]
- Di, C.; Yang, X.; Wang, X. A Four-Stage Hybrid Model for Hydrological Time Series Forecasting. PLoS ONE
**2014**, 9, e104663. [Google Scholar] [CrossRef] - Jain, A.; Kumar, A.M. Hybrid neural network models for hydrologic time series forecasting. Appl. Soft Comput.
**2007**, 7, 585–592. [Google Scholar] [CrossRef] - Wang, W.; Van Gelder, P.H.; Vrijling, J.; Ma, J. Forecasting daily streamflow using hybrid ANN models. J. Hydrol.
**2006**, 324, 383–399. [Google Scholar] [CrossRef] - Chen, P.; Niu, A.; Liu, D.; Jiang, W.; Ma, B. Time Series Forecasting of Temperatures using SARIMA: An Example from Nanjing. IOP Conf. Ser. Mater. Sci. Eng.
**2018**, 394, 052024. [Google Scholar] [CrossRef] - Ray, S.; Das, S.S.; Mishra, P.; Al Khatib, A.M.G. Time Series SARIMA Modelling and Forecasting of Monthly Rainfall and Temperature in the South Asian Countries. Earth Syst. Environ.
**2021**, 5, 531–546. [Google Scholar] [CrossRef] - Dwivedi, D.K.; Kelaiya, J.; Sharma, G.R. Forecasting monthly rainfall using autoregressive integrated moving average model (ARIMA) and artificial neural network (ANN) model: A case study of Junagadh, Gujarat, India. J. Appl. Nat. Sci.
**2019**, 11, 35–41. [Google Scholar] [CrossRef] - Zhang, G.P. Time series forecasting using a hybrid ARIMA and neural network model. Neurocomputing
**2003**, 50, 159–175. [Google Scholar] [CrossRef] - Tealab, A. Time series forecasting using artificial neural networks methodologies: A systematic review. Future Comput. Inform. J.
**2018**, 3, 334–340. [Google Scholar] [CrossRef] - Hajirahimi, Z.; Khashei, M. Hybrid structures in time series modeling and forecasting: A review. Eng. Appl. Artif. Intell.
**2019**, 86, 83–106. [Google Scholar] [CrossRef] - Saba, T.; Rehman, A.; AlGhamdi, J.S. Weather forecasting based on hybrid neural model. Appl. Water Sci.
**2017**, 7, 3869–3874. [Google Scholar] [CrossRef] [Green Version] - Deng, Y.; Fan, H.; Wu, S. A hybrid ARIMA-LSTM model optimized by BP in the forecast of outpatient visits. J. Ambient Intell. Humaniz. Comput.
**2020**, 1–11. [Google Scholar] [CrossRef] - Mieczkowski, Z. The tourism climatic index: A method of evaluating world climates for tourism. Can. Geogr. Le Géographe Can.
**1985**, 29, 220–233. [Google Scholar] [CrossRef] - Rutty, M.; Scott, D.; Matthews, L.; Burrowes, R.; Trotman, A.; Mahon, R.; Charles, A. An Inter-Comparison of the Holiday Climate Index (HCI:Beach) and the Tourism Climate Index (TCI) to Explain Canadian Tourism Arrivals to the Caribbean. Atmosphere
**2020**, 11, 412. [Google Scholar] [CrossRef] [Green Version] - Hyndman, R.J.; Khandakar, Y. Automatic Time Series Forecasting: The forecast Package for R. J. Stat. Softw.
**2008**, 27, 1–22. [Google Scholar] [CrossRef] [Green Version] - Hyndman, R.J.; Athanasopoulos, G.; Bergmeir, C.; Caceres, G.; Chhay, L.; O’Hara-Wild, M.; Petropoulos, F.; Razbash, S.; Wang, E.; Yasmeen, F.; et al. Forecast: Forecasting Functions for Time Series and Linear Models. 2021. Available online: https://cran.r-project.org/web/packages/forecast/forecast.pdf (accessed on 20 May 2022).
- Petnehazi, G. Recurrent Neural Networks for Time Series Forecasting. arXiv
**2018**, arXiv:1901.00069. [Google Scholar] - Lewis, N.D.C. Neural Networks for Time Series Forecasting with R: An Intuitive Step by Step Blueprint for Beginners; AusCov: Scotts Valley, CA, USA, 2017. [Google Scholar]
- Hu, C.; Wu, Q.; Li, H.; Jian, S.; Li, N.; Lou, Z. Deep Learning with a Long Short-Term Memory Networks Approach for Rainfall-Runoff Simulation. Water
**2018**, 10, 1543. [Google Scholar] [CrossRef] [Green Version] - Elsaraiti, M.; Merabet, A. A Comparative Analysis of the ARIMA and LSTM Predictive Models and Their Effectiveness for Predicting Wind Speed. Energies
**2021**, 14, 6782. [Google Scholar] [CrossRef] - Ruder, S. An Overview of Gradient Descent Optimization Algorithms. arXiv
**2016**, arXiv:1609.04747, 4747. [Google Scholar] - Kingma, D.P.; Ba, J. Adam: A Method for Stochastic Optimization. arXiv
**2015**, arXiv:1412.6980. [Google Scholar] - Pappenberger, F.; Ramos, M.H.; Cloke, H.L.; Wetterhall, F.; Alfieri, L.; Bogner, K.; Mueller, A.; Salamon, P. How do I know if my forecasts are better? Using benchmarks in hydrological ensemble prediction. J. Hydrol.
**2015**, 522, 697–713. [Google Scholar] [CrossRef] [Green Version] - Dave, E.; Leonardo, A.; Jeanice, M.; Hanafiah, N. Forecasting Indonesia Exports using a Hybrid Model ARIMA-LSTM. Procedia Comput. Sci.
**2021**, 179, 480–487. [Google Scholar] [CrossRef]

**Figure 2.**Comparison of SARIMA, LSTM and hybrid using daily averaged MAE with real data for 1-day prediction horizon for 230 days of the test set. With dashed lines the average error of all the dataset is shown. Units of MAE are in Celsius in this temperature example.

**Figure 3.**Comparison of SARIMA, LSTM and hybrid using daily averaged MAE with real data for 2-day prediction horizon for 230 days of the test set. With dashed lines the average error of all the dataset is shown. The boxes indicate the predictions that we plot separately to observe their 48-h behaviour. Units of MAE are in Celsius in this temperature example.

**Figure 4.**Comparison of SARIMA, LSTM and hybrid with real data on temperature example for 2-day prediction horizon in January.

**Figure 5.**Comparison of SARIMA, LSTM and hybrid with real data on temperature example for 2-day prediction horizon during March; stable parabolic temperature profile.

**Figure 6.**Comparison of SARIMA, LSTM and hybrid with real data on temperature example for 2-day prediction horizon in March; stable temperature example where hybrid is not achieving a correction on the SARIMA.

**Figure 7.**Comparison of SARIMA, LSTM and hybrid with real data on temperature example for 2-day prediction horizon in April; example with very big MAE.

**Figure 8.**Comparison of SARIMA, LSTM and hybrid with real data on temperature example for 2-day prediction horizon in June; example with very big MAE.

**Figure 9.**Comparison of SARIMA, LSTM and hybrid using daily averaged MAE with real data for 2-day prediction horizon on humidity prediction for 230 days of the test set. With dashed lines the average error of all the dataset is shown. Units of MAE are in percentage (%) in this humidity example.

**Figure 10.**Comparison of SARIMA, LSTM and hybrid with real data on humidity example for 2-day prediction horizon in June; example with very big MAE.

**Figure 11.**Comparison of SARIMA, LSTM and hybrid using daily averaged MAE with real data for 2-day prediction horizon on wind speed prediction for 230 days of the test set. With dashed lines the average error of all the dataset is shown. Units of MAE are in m/s in this wind speed example.

**Figure 12.**Comparison of SARIMA, LSTM and hybrid with real data for 2-day prediction horizon during December; wind speed example; low wind speed during previous days.

**Figure 13.**Comparison of SARIMA, LSTM and hybrid with real data for 2-day prediction horizon during May; wind speed example; high wind speed during previous days.

**Table 1.**MAE errors for the three methods for 1- and 2-day prediction of wind speed (m/s) for the three predicted variables, averaged over each day and then over the test set. MAE units are the same as the predicted variable.

Prediction Variable | Methods | MAE 1-Day | MAE 2-Day |
---|---|---|---|

Temperature (Celsius) | SARIMA (1, 0, 2) (2, 1, 2) | 1.58 | 1.86 |

LSTM (Adam 40 units each of two hidden layers) | 2.12 | 2.14 | |

Hybrid | 1.56 | 1.85 | |

Climatological benchmark | 2.25 | 2.26 | |

Persistence benchmark | 2.44 | 2.44 | |

Humidity (%) | SARIMA (5, 1, 0) (2, 0, 0) | 10.62% | 12.33% |

LSTM (Adam 40 units each of two hidden layers) | 9.54% | 10.01% | |

Hybrid | 10.30% | 12.01% | |

Climatological benchmark | 11.15% | 11.16% | |

Persistence benchmark | 14.14% | 14.16% | |

Wind Speed (m/s) | SARIMA (0, 1, 5) (0, 0, 2) | 2.46 | 2.78 |

LSTM (Adam 60 units each of two hidden layers) | 2.73 | 2.79 | |

Hybrid | 2.41 | 2.70 | |

Climatological benchmark | 2.87 | 2.88 | |

Persistence benchmark | 3.67 | 3.68 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Parasyris, A.; Alexandrakis, G.; Kozyrakis, G.V.; Spanoudaki, K.; Kampanis, N.A.
Predicting Meteorological Variables on Local Level with SARIMA, LSTM and Hybrid Techniques. *Atmosphere* **2022**, *13*, 878.
https://doi.org/10.3390/atmos13060878

**AMA Style**

Parasyris A, Alexandrakis G, Kozyrakis GV, Spanoudaki K, Kampanis NA.
Predicting Meteorological Variables on Local Level with SARIMA, LSTM and Hybrid Techniques. *Atmosphere*. 2022; 13(6):878.
https://doi.org/10.3390/atmos13060878

**Chicago/Turabian Style**

Parasyris, Antonios, George Alexandrakis, Georgios V. Kozyrakis, Katerina Spanoudaki, and Nikolaos A. Kampanis.
2022. "Predicting Meteorological Variables on Local Level with SARIMA, LSTM and Hybrid Techniques" *Atmosphere* 13, no. 6: 878.
https://doi.org/10.3390/atmos13060878