The interaction between rape nitrogen content and water status and influencing factor of lighting in spectrum analysis were developed. In order to avoid the effect of lighting fluctuation, the correction models which based on variation ratio of spectral reflectance were established in each spectral characteristic wavelength of rape nitrogen content and moisture content. The method of principal component analysis (PCA) was used to establish the nitrogen spectral predictive model under different moisture content levels in order to avoid the interaction between rape nitrogen and water status. The combination model was used to predict the nitrogen content of rape. The result showed that the correlation coefficient between the estimated values and the real values is 0.92, the root mean square error is 0.53. The precision of combination model is higher than the previous model which used PCA method and 6 wavelength variables. 