Abstract: Intercropping has been shown to improve P availability in red soil, but a quantitative mechanistic analysis of the relationships between soil biological activity and P availability has not been conducted in field experiments. The aim of this study was to determine the specific effect of soil biological activity on P availability in red soil. In this study, maize/soybean intercropping and maize monocropping with four P application levels 0 kg(P₂O₅)∙hm⁻², 60 kg(P₂O₅)∙hm⁻², 90 kg(P₂O₅)∙hm⁻², and 120 kg(P₂O₅)∙hm⁻², labeled as P0, P60, P90, and P120, respectively were used to analyze P availability and its relationship with major soil factors during a 4 year field experiment, which compared the values in the first and last years (2017 and 2020). Regression, redundancy, and structural equation model analyses were conducted to determine these relationships. The results showed that the land equivalent ratio of dry matter mass (LER) and land equivalent ratio of P uptake (LERp) were both greater than 1, indicating that intercropping increases biomass production and P uptake of crops. LER significantly (P<0.05) decreased with an increase in P application level in 2020, but LERp did not. This indicates that the advantage of P absorption in intercropping was not affected by P level, and low P levels still had the ability to maintain this advantage. Compared with monocropping, intercropping significantly (P<0.05) increased the soil available P content in maize rhizosphere; the increasing amplitude was from 24% to 103%, and the role of planting pattern changed from a significant level (P<0.05) in 2017 to an extremely significant level (P<0.01) in 2020. This indicates the effects of intercropping and P application on maize rhizosphere P availability in red soil with low fertility over multiple planting years. Regression analysis showed that the slope of the linear relationship between both Olsen-P and microbial biomass P (MBP) with maize P absorption in 2020 was lower than that in 2017 under monoculture (varied from 4.73 to 4.42 in Olsen-P and 19.68 to 7.16 in MBP), whereas the slope increased under intercropping (varied from 4.12 to 4.44 in Olsen-P and 13.72 to 17.78 in MBP). Redundancy analysis showed that the increase in soil available P and microbial biomass carbon (MBC) accounted for P absorption varied from 37.6% and 10.0% in monoculture to 33.3% and 13.8% in intercropping, respectively. This result shows that intercropping decreased the effect of available P on P uptake and increased the role of MBC. Through structural equation model analysis, increasing the activity of alkaline phosphatase was found to directly increase the content of soil available P, whereas acid phosphatase could increase soil available P by increasing MBP. These two enzymes are speculated to affect P availability in rhizosphere soil via different mechanisms. Therefore, under low P application levels, intercropping can enhance P availability in red soil by increasing soil microbial biomass and phosphatase activity in the maize rhizosphere, thus maintaining the dry matter mass and phosphorus uptake of maize. The quantitative contribution of microbial processes should be further studied.