In order to explore a more reasonable index weighting method in climatic evaluation of different flue-cured tobacco growing areas,climatic data of 12 typical tobacco growing areas in Yunnan were collected,evaluated by single and comprehensive weighting methods and the results of weighting and evaluation were compared.The results showed that:(1) The results of descriptive statistics for climatic factors at 12 tobacco growing areas showed that the average temperature in the field period was between 19.23 and 20.58 degrees celsius,which was appropriate.The average rainfall in the mature period was at a high level of 290.45 mm,and the sunshine hours in the field were sufficient.(2) The highest weighting coefficients in three single weighting methods were 0.173 for the precipitation in the period of vigorous growth index in AHP,0.143 for the average temperature in the period of vigorous growth index in standard deviation method and 0.195 for the average temperature in the period of vigorous growth index in entropy method.And the R^2 mean was 2.36.(3) The highest and lowest weighting coefficients were the average temperature in the period of vigorous index (0.150,0.152 and 0.149) and the sunshine hours for whole year index (0.107,0.107 and 0.108) for all the three comprehensive weighting methods.And the R^2 mean was 0.44.(4) The comprehensive weighting method based on the sum of deviation squares had larger gaps among evaluation scores,and the comprehensive weighting method based on game theory had better correlations among evaluation results,while the comprehensive weighting method based on unit constraints had higher evaluation scores.After combining the three comprehensive weighting methods,the weight difference among climatic indices was less,and the evaluation results had better convergence.Comprehensive weighting method was suitable for the index weighting research in climatic evaluation of flue-cured tobacco growing areas.
Chinese Tobacco Science
flue-cured tobacco growing areas
single weighting method
comprehensive weighting method