<p>This study investigates the interdecadal variability of long rains over Tanzania and its linkage to large-scale ocean-atmosphere interactions. High-resolution observational datasets (CRU-TS4.08 and ERSSTv5), NCEP/NCAR reanalysis and Coupled Model Intercomparison Project Phase 6 (CMIP6) historical and future projections were utilised. The Empirical Orthogonal Function (EOF), composite analysis, and regression techniques were applied to explore dominant rainfall modes and their drivers. Seven high-performing CMIP6 models were selected using standard statistical metrics to project future precipitation under SSP2-4.5 and SSP5-8.5 scenarios using an ensemble mean approach. EOF results revealed three leading modes of rainfall variability with a basin-wide monopole (EOF1), a meridional dipole (EOF2) tied to Intertropical Convergence Zone shifts, and a canonical interdecadal mode (EOF3) marked by alternating wet and dry phases between northeastern and southwestern Tanzania. These decadal phases were strongly associated with sea surface temperature (SST) anomalies in the tropical Pacific, indicating a tri-basin control involving El Niño-Southern Oscillation (ENSO), Pacific Decadal Oscillation (PDO) and Interdecadal Pacific Oscillation (IPO). Circulation anomalies indicated that wet decadal phases were characterized by enhanced ascent, low-level cyclonic inflow, and upper-level divergence over Tanzania, while dry phases exhibited the opposite patterns. Running correlation and regression analyses confirmed that the tri-basin indices strongly modulate rainfall variability via Indo-Pacific wave trains. Future ensemble projections suggest that interdecadal rainfall modes, particularly in EOF2 are likely to persist under both SSP2-4.5 and SSP5-8.5 scenarios, maintaining strong teleconnections with global SST patterns. The findings offer valuable improved long-range rainfall prediction and support climate adaptation strategies in Tanzania.</p>