Based on the grain yields and plant K uptakes of the 122 K fertilizer field experiments conducted on rape as reference, three different extractants (i.e., 1 mol L^-1 neutral NH₄OAc, 1 mol L^-1 boiling HNO₃, 0.2 mol L^-1 NaBPh₄) were used to extract soil K in the experiment soils, and the data were sorted by extractants and by available K and slowly available K. Then correlation analysis was done of the data to define which one was the accurate determination method for soil available K and corresponding indices for soil K abundance and deficiency were established. It was found that the three extractant varied in extraction capacity and displayed an order of NH₄OAc-K > Boiling HNO₃-K > NaBPh₄-K. The soil Ks extracted with NH₄OAc and NaBPh₄ were higher in bio-availability and NaBPh₄ the highest. For winter rape fields in the Yangtze River valley, the index of “Low”, “Medium”, “High” and “Very High” for soil NH₄OAc-K and NaBPh₄-K was <45 mg kg^-1, 45~115 mg kg^-1, 115~190 mg kg^-1 and >190 mg kg^-1; and <125 mg kg^-1, 125~415 mg kg^-1, 415~755 mg kg^-1 and >755 mg kg^-1, respectively. Suumarization of the results showed that the relative yield and relative plant uptake was highly significantly correlated with soil K content extracted with the NaBPh₄ and NH₄OAc method. The soil K critical indices established using the methods and through the field experiments can be used as reference for soil test based fertilizing recommendation for winter rape production in the valley. Moreover, it is essential to pay more attention to rational K fertilization in the region.