Objectives Soil fertility is the fundamental resource for sustainable development of rice production. We researched the variation feature of red paddy soil fertility under long-term fertilization, specified the correlation between soil fertility index (SFI) and rice yield, in order to provide a useful tool for the quantitative improvement and evaluation of soil fertility.
Methods This study was based on a long-term experiment of red paddy soil started in 1982. The SFIs of different treatments were calculated by Fuzzy method. The fitting equation between SFI and rice yield was established and verified in a county scale.
Results After 30 years of fertilization, the SFI increased slowly and linearly with the fertilization years in all the treatments except no fertilizer control, and the annual increases of SFI were in range of 0.0035–0.0136. The annual average SFI of five manure containing treatments were all higher than that of the NPK treatment, and those in PKM and NPKM were 70.6% and 68.4% higher than that in CK. There was a significant linear positive correlation between the average SFI and the annual average absolute and relative rice yields in 1984, 1988, 1994, 1998, 2004, 2008, and 2012 (P < 0.05), and the correlation coefficients were as high as 0.708, 0.683, respectively. The verification results based on SFI in the county scale showed the high consistency between the predicted and the measured absolute and relative rice yields, with the RRMSE both less than 25%.
Conclusions Input of organic fertilizer effectively improves soil fertility in red paddy soil. The rice yield is significantly correlated with SFI in hilly red soil area, and SFI and rice yield still have some potential to improve. The rice yield in county scale could be predicted and evaluated by the relational model between SFI and rice yield.
DownLoad: