Effect of long-term K fertilizer application and returning wheat straw to soil on crop yield and soil K under different planting systems in northwestern China

Experiments of long-term K fertilizer application and returning wheat straw to soil in Ningxia irrigation silting soils and Qinghai castanozems in northwestern China were started in 1992. Effect of K fertilizer application and returning wheat straw to soil on crop yield and soil K status in the plough layer under different planting systems was studied. The results showed that K fertilizer and wheat straw could improve crop yield significantly in Ningxia site, with the treatments being in the following order: NPK+StrawNPKNP+StrawNP. The time when the positive effect of K application on yield became significant in Ningxia differed with varieties, however, K application or returning wheat straw to soil did not have any significant effect on wheat yield in Qinghai. Coefficient of variance of wheat yield in each treatment of Ningxia was lower than in Qinghai and was higher than maize, but yield efficiency of K fertilizer application was on the contrary. Except for treatment NPK+Straw, soil K was deficient in each treatment of the two fixed sites, and the consumption of soil K was more serious in the rotation system of Ningxia than in wheat monoculture of Qinghai. K fertilizer and straw could improve content of water-soluble K, non-specifically adsorbed K, non-exchangeable K and total K in contrast to NP, however reduce proportion of mineral K and improve those of others in the two fixed sites, and content and proportion of specifically adsorbed K had no relations with K fertilizer and straw. Compared with the beginning of the experiment, temporal variability characteristics of content and proportion of different forms of K differed between two soil types, which were closely related to K status of original soil and planting system.