| This paper studied the impacts of major agricultural meteorological disasters on grain production in Shaanxi Province. Based on the historical data of natural disasters and the grain yield data in Shaanxi province from 1978 to 2011, Mann-Kendall test and Morlet wavelet were used in this paper to analyze the rates of some major agricultural meteorological disasters in Shaanxi province, including droughts, floods, hail and frost. Then, we used grey relational analysis method to dynamically analyze the correlations between the affected areas and grain yield resulted from these four major agro-meteorological disasters. Results showed that the correlation efficiencies by drought, floods, hail and frost were 0.7806, 0.767, 0.7731, and 0.7355 respectively. From the disaster rate trends, we could see that the drought disaster rates in the 1980s changed slightly, which displayed an upward trend from the early 1990s to 2002, and a downward trend after 2002. The frost showed an upward trend, the hail was in a slight upward trend, and the floods possessed a slight downward trend. M-K alternation test results showed that the alternation in drought disaster took place between 1984 and 2011, the alternation point in frost disaster appeared in 2003, and floods and hail did not had any obvious alternations in 33 years in Shaanxi province. The results of wavelet analysis showed that the drought stricken rate had significant quasi 2～3 a and quasi 2～4 a scale cycles, and the intensities of these cycles were changed over time; The hail stricken rate had significant quasi 2～3 a, quasi 3 a, quasi 3～4 a and quasi 2～6 a scale cycles; The frost stricken rate had significant quasi 2～4 a and quasi 4～8 a scale cycles; The flood stricken rate had significant quasi 3 a, quasi 2～4 a and quasi 2～6 a scale cycles. |