The reliable Subseasonal‐to‐Seasonal (S2S) forecast of precipitation, particularly extreme precipitation, is critical for disaster prevention and mitigation, which however remains a great challenge for mission agencies and research communities. In this study, a deep learning method based on U‐Net with additional atmospheric factor forecasts included is proposed to improve S2S quantitative forecasts of summer precipitation over Southern China. The weighted loss function integrated by mean square error and threat score is introduced to capture extreme precipitation more precisely. Generally, the U‐Net model shows promising results in both general statistics and extreme events. Predictor importance analyses show that the U‐Net forecast skills at the 1‐week lead time mainly arise from synchronous precipitation forecasts, but the contributions made by atmospheric factor forecasts rise rapidly with increasing lead times. Therefore, the channel combining numerical weather prediction model and deep learning framework is demonstrated promising in S2S precipitation forecasts.

The extended range temperature prediction is of great importance for public health, energy and agriculture. The two machine learning methods, namely, the neural networks and natural gradient boosting (NGBoost), are applied to improve the prediction skills of the 2-m maximum air temperature with lead times of 1–35 days over East Asia based on the Environmental Modeling Center, Global Ensemble Forecast System (EMC-GEFS), under the Subseasonal Experiment (SubX) of the National Centers for Environmental Prediction (NCEP). The ensemble model output statistics (EMOS) method is conducted as the benchmark for comparison. The results show that all the post-processing methods can efficiently reduce the prediction biases and uncertainties, especially in the lead week 1–2. The two machine learning methods outperform EMOS by approximately 0.2 in terms of the continuous ranked probability score (CRPS) overall. The neural networks and NGBoost behave as the best models in more than 90% of the study area over the validation period. In our study, CRPS, which is not a common loss function in machine learning, is introduced to make probabilistic forecasting possible for traditional neural networks. Moreover, we extend the NGBoost model to atmospheric sciences of probabilistic temperature forecasting which obtains satisfying performances.

In this study, we investigated the interdecadal variability in monsoon rainfall in the Myanmar region. The gauge-based gridded rainfall dataset of the Global Precipitation Climatology Centre (GPCC) and Climatic Research Unit version TS4.0 (CRU TS4.0) were used (1950–2019) to investigate the interdecadal variability in summer monsoon rainfall using empirical orthogonal function (EOF), singular value decomposition (SVD), and correlation approaches. The results reveal relatively negative rainfall anomalies during the 1980s, 1990s, and 2000s, whereas strong positive rainfall anomalies were identified for the 1970s and 2010s. The dominant spatial variability mode showed a dipole pattern with a total variance of 47%. The power spectra of the principal component (PC) from EOF revealed a significant peak during decadal timescales (20–30 years). The Myanmar summer monsoon rainfall positively correlated with Atlantic multidecadal oscillation (AMO) and negatively correlated with Pacific decadal oscillation (PDO). The results reveal that extreme monsoon rainfall (flood) events occurred during the negative phase of the PDO and below-average rainfall (drought) occurred during the positive phase of the PDO. The cold phase (warm phase) of AMO was generally associated with negative (positive) decadal monsoon rainfall. The first SVD mode indicated the Myanmar rainfall pattern associated with the cold and warm phase of the PDO and AMO, suggesting that enhanced rainfall for about 53% of the square covariance fraction was related to heavy rain over the study region except for the central and eastern parts. The second SVD mode demonstrated warm sea surface temperature (SST) in the eastern equatorial Pacific (El Niño pattern) and cold SST in the North Atlantic Ocean, implying a rainfall deficit of about 33% of the square covariance fraction, which could be associated with dry El Niño conditions (drought). The third SVD revealed that cold SSTs in the central and eastern equatorial Pacific (La Niña pattern) caused enhance rainfall with a 6.7% square covariance fraction related to flood conditions. Thus, the extra-subtropical phenomena may affect the average summer monsoon trends over Myanmar by enhancing the cross-equatorial moisture trajectories into the North Atlantic Ocean.

Myanmar is located in a tropical region where temperature rises very fast and hence is highly vulnerable to climate change. The high variability of the air temperature poses potential risks to the local community. Thus, the current study uses 42 synoptic meteorological stations to assess the spatiotemporal changes in air temperature over Myanmar during 1971–2013. The nonparametric sequential Mann-Kendall (SqMK), linear regression, empirical orthogonal function (EOF), Principal Component Analysis (PCA), and composite analysis were used to assess the long-term trends in maximum (Tmax) and minimum (Tmin) temperature series and their possible mechanism over the study region. The results indicate that the trend of Tmax has significantly increased at the rates of 90% in summer season, while the Tmin revealed a substantial positive trend in winter season time series with the magnitude of 30%, respectively. Moreover, during a rapid change of climate (1995–2013) we observed an air temperature increase of 0.7 °C. The spatial distributions of EOF revealed relatively warmer temperatures over the whole region except the south in the summer; however, a similar pattern can be seen for the rainy season and winter, implying warming in the central part and cooling in the northern and southern parts. Furthermore, the Indian Ocean Dipole (IOD) influence on air temperature over Myanmar is more prevalent than that of the El Niño Southern Oscillation (ENSO). The result implies that the positive phase of the IOD and negative phase of the Southern Oscillation Index (SOI; El Niño) events led to the higher temperature, resulting in intense climatic extremes (i.e., droughts and heatwaves) over the target region. Therefore, this study’s findings can help policymakers and decision-makers improve economic growth, agricultural production, ecology, water resource management, and preserving the natural habitat in the target region.