The impact of climate change on food production in China has become the focus of public attention. Affected by global warming, wide range of continuous drought has become the most serious threat to agricultural production. In recent years, seasonal drought in the southern China became more and more serious, which has seriously harmed the development of the society and economy in our country. Therefore, how to adapt to seasonal drought has become a common concerned topic of the whole society. As the subject of micro decisionmaking in agricultural production sector, farmers play an important role in choosing adaptive behaviors to seasonal drought, so study on farmers' behaviors and their influence factors is the precondition and foundation to scientifically reduce the disaster. Based on the survey data of 755 households in 41 counties of Jiangxi Province, we analyzed farmers ' adaptation behaviors to seasonal drought and the influencing factors from theoretical and empirical perspectives of view. The statistical results show that 68.6% of the farmers will take corresponding behaviors when facing the seasonal drought, that replanting (seedling), switching to other crops, and adjusting the sowing and harvest dates are the most preferred adaptive measures of farmers, while the capital and labor force are the most important limiting factors for farmers to choose adaptive measures. By using Probit regression model, Poisson regression model and Tobit regression model, this paper empirically analyzed the impact factors of farmers' adaptation behaviors to seasonal drought and their relations with the scale and intensity of the behaviors. The results show that social capital, drought occurrence frequency, and the availability of disaster prevention information have positive effects on framers ' willingness of adopting behaviors and the scale and intensity of the behaviors, while the distance to the nearest highway has negatively effect on the framers' willingness of adopting behaviors and the scale and intensity of the behaviors.