The question is very general, and its answer depends on many different factors. But overall, most of the studies showed that the time series models (e.g., SARIMA) and the machine learning methods (e.g., ANN and ANFIS) had reliable skills for precipitation forecasting. For instance:
Article Rainfall forecasting by technological machine learning models
Article Prediction of daily precipitation using wavelet—neural networks
Article Precipitation Forecast Using Artificial Neural Networks in S...
Article REPLY to Discussion of “Precipitation Forecasting Using Wave...
Article Monthly Precipitation Forecasting with a Neuro-Fuzzy Model
Article Application of seasonal time series model in the precipitati...
Thank you very much all colleagues Am I asking my question more clearly? If we want to have a one year accurate forecast of precipitation, the following questions are raised: Which station parameter data should be used? What are the parameters of the upper atmosphere stations?
Yen, M., Liu, D., Hsin, Y. et al. Application of the deep learning for the prediction of rainfall in Southern Taiwan. Sci Rep 9, 12774 (2019). https://doi.org/10.1038/s41598-019-49242-6
Application of the deep learning for the prediction of rainfall in Southern Taiwan,
Thank you for your suggestion Time series have been done and it seems to be a short-term and cross-sectional method. Due to the fact that we are looking for the best method, other colleagues have suggested the WRFmodel. Which time series method do you suggest?
I think you can use probably the GFS Forecast Model: Precipitation as commonly used by the Meteorologists. I do believe also, that through technological innovation, the model is kept on changing. You can use whatever available on the market.
Dear Prof. Zainab Mohammadi, you might find it interesting to read the following external post.
I can guess that using non-traditional data science tools combined with AI (tensor prog and deep learning), they collect huge amounts of weather data, so it is possible to obtain new unexpected results.
AI developed by Google that can predict rain in a range of 6 hours, in less than 10 minutes [1].
From the meteorological point of view only a synoptic meteorologist can predict rainfall. I know that this is not what you mean, though: -)) Predictions which are mathematically manipulated or made numerically for good reason are simulations, not real predictions.