Agroforestry is globally recognized as a sustainable land management practice that enhances productivity and ecological stability. Among its various models, the cotton tree-crop system is gaining attention in semiarid and tropical regions because it increases productivity, improves soil fertility, diversifies income, and mitigates environmental degradation. However, limited research has been conducted on cotton (Gossypium hirsutum)-based agroforestry systems in Bangladesh. This two-year field study aimed to evaluate the yield, economic performance, and environmental impact of cotton-tree agroforestry systems under an irrigated system managed through conventional farming practices. A single-factor randomized complete block design (RCBD) with four treatments and three replications was used. The treatments included the following: T0 = sole Gossypium hirsutum cropping, T1 = Gossypium hirsutum-Swietenia macrophylla agroforestry system, T2 = Gossypium hirsutum-Mangifera indica agroforestry system, and T3 = Gossypium hirsutum-Psidium guajava agroforestry system. The systems were established in 4-year-old plantations with a 2 m x 2 m spacing agroforestry model. The results revealed that the highest seed cotton yield (4.94 Mg ha-1) was obtained in T0, whereas the lowest yield (3.61 Mg ha-1) was observed in T3. The T2 treatment achieved the highest net profit (US$3967 per hectare), benefit-cost ratio (2.19), and land equivalent ratio (1.67). The T1 treatment had the largest amount of carbon sequestration (42.20 Mg ha-1 yr-1) but the lowest profitability. While the T3 treatment had the lowest carbon sequestration (24.78 Mg ha-1 yr-1), it did not negatively affect the environmental quality. Based on these results, both the Gossypium hirsutum-Mangifera indica (T2) and Gossypium hirsutum-Psidium guajava (T3) agroforestry systems provided greater environmental benefits, higher economic returns, and superior land equivalent ratios than cotton cultivation alone, although the magnitude of these effects varied between systems.