Abstract: Mathematical modeling is an important approach for investigating future trends in population size and age structure and for supporting evidence-based population policies. Based on data from China's national censuses and statistical yearbooks from 1990 to 2020, a time-series model is employed to estimate age-specific fertility and mortality rates. Subsequently, an age-structured population model is developed to project future trends in population size, age structure, and population ageing from 2021 to 2070. The average age and life expectancy are also calculated. The results indicate that the average childbearing age is projected to increase gradually, while the crude birth rate declines from 8.58‰ to 6‰ and the annual number of newborns decreases from 12 million to 6 million. The total population is projected to peak at 1.424 billion in 2026 and then decline to 1.013 billion by 2070. The proportions of children (0-14 years) and the elderly (≥60 years) are projected to change from 18.96% and 18.79% to 10.12% and 41.05%, respectively, while the proportion of the oldest-old population (aged ≥80 years) is expected to increase from 13.76% to 34.03% within the elderly population. Additionally, both the average age and life expectancy show an upward trend over the next 50 years, rising from 39 and 79 years to 51 and 85 years, respectively. These findings suggest a continuing decline in births and total population, a polarized age structure, and an accelerated process of population ageing.