The study investigated the precipitation variability over Myanmar at the annual and seasonal scales by comparing 12 model outputs from the Coupled Model Intercomparison Project Phase 6 (CMIP6) with gridded observational data provided by the Global Precipitation Climatology Centre (GPCC) from 1970 to 2014. Using Mann–Kendall and Sen's slope estimator, the trend analysis was assessed. Correlation analysis was also used to investigate the relationship of observational and Ensemble means

more »

... on with sea surface temperature (SST) anomalies. Results show a better correlation pattern of ENS with observation precipitation than that of individual selected models during the May-October season than that of the annual scale. Meanwhile, UKESM1-0-LL, NESM3, and HadGEM3-CC31-LL show high correlation with a relatively low root-mean-square difference. A few models roughly capture the spatiotemporal patterns of precipitation during MJJASO over Myanmar. The root mean square errors (RMSEs) of MIROC6, CNRM-ESM2-1, CNRM-CM6, and NESM3 are lower than that of ENS, whereas the RMSEs of CESM2, GFDL-CM4, HadGEM3-CC31-LL, GFDL-ESM4, UKESM1-0-LL, MPI-ESM1-2-HR, MRI-ESM2-0, and IPSL-CM6A-LR are higher than that of ENS, for annual precipitation. Heterogeneous correlation coefficients and slope changes are evident within the country at both annual and seasonal periods. Overall, the ENS showed a long-term increasing annual trend. Most of the model exhibited increasing annual trends while some showed decreasing annual trends. The correlation between the annual series and SST anomalies shows stronger correlation coefficient than that of seasonal. Overall, the correlation analysis of the SST anomalies reveals significant positive and negative relationships with the ENS precipitation. We recommend considering future projections of precipitation changes over Myanmar in future work.