The effects of meteorological conditions on asthma in Haikou, a tropical city in China, are still unclear. This study aimed to determine the relationships between meteorological factors and the number of asthma hospital visits in Haikou. A Poisson generalized additive model combined with a distributed lag nonlinear model is used to model the nonlinear exposure–response relationship between the daily mean temperature and asthma hospital visits. The daily mean pressure and air quality are used as covariates and simultaneously control the mixed effects of holiday effects, weekend effects, and long-term trends. The results indicate that there is a significant statistical relationship between the daily mean temperature and asthma hospital visits, which shows an inverted J-shaped relationship. When the daily mean temperature is below the reference value (29.3 °C), the number of asthma patients increases considerably, and there is a marked lag in the prevalence of asthma. The longest lag is 9 days, and the most pronounced impact of the daily mean temperature on the number of asthma hospital visits can be found when the lag time is 1–4 days. When the daily mean temperature is 10 °C, the cumulative effect of the relative risk of asthma is 2.204, an increase of 120.4% (95% CI 1.294–3.755). If the daily mean temperature is below the 2.5th percentile value (14.8 °C), the relative risk significantly increases by more than 5.3% (95% CI 1.000–1.110), and the longest lasting impact time is 5 days. This indicates that increases in asthma hospital visits in Haikou, China, are significantly correlated with low-temperature weather. We suggest that preventive measures for asthma should take low-temperature weather into account. Additionally, we also found that extremely high temperatures have a certain impact on the increase in asthma hospital visits, but that the correlation is not significant.