Maintaining the current levels of rice production in the face of future climate changes is an enormous challenge for agriculture. The use of high-resolution projections derived from statistical downscaled global climate models and dynamic downscaled regional climate models can assist in maximizing yield for double-season rice. The thermal indices of secure sowing date (SSd), secure maturation date (SMd), potential growing season (PGS), and growing accumulated temperature ≥ 10 °C (GAT10) were explored in future sub-periods under two greenhouse gas emission scenarios (RCP4.5 and RCP8.5). The trends predicted were advancements in SSd and delays in SMd across major areas of China. A comparison of RCP4.5 and RCP8.5 scenarios indicated that SSd advanced by 1–39 and 1–49 days in 2081–2095 relative to the baseline period of 2001–2015 while SMd was postponed by 11–60 and 11–70 days, respectively. Under RCP4.5, PGS was prolonged by 1–50, 1–60, 21–90, and 31–90 days in 2021–2035, 2041–2055, 2061–2075, and 2081–2095 relative to baseline, respectively. Within these same periods using RCP8.5, PGS advanced 1–50, 31–90, 41–90, and 51–90 days, respectively. A comparison of the relative contributions of SSd and SMd to PGS indicated that higher levels of SMd contributed prominently in the PGS increase. In addition, as the climate continued to warm, GAT10 levels would likely continue in an upward trend and were predicted for 2081–2095 to reach 500–2500 °C·day under RCP4.5 and 1500–3000 °C·day under RCP8.5. Regionally, earlier SSd, later SMd, longer PGS, and larger GAT10 values were projected for southern regions in Guangdong, Guangxi, and Hainan. This type of information is also a potential resource available to farmers to make informed decisions on timely cultivation adjustments and variety selections for specific climatological areas.