LAUSR

Abstract Rhizosphere pH is the consequence of numerous, complex processes and is a key factor of biogeochemistry at the soil-root interface. This work aimed at evaluating the impact of nutrient root absorption on rhizosphere pH using a mechanistic model formalizing major nutrient cation and anion transport, under the influence of concentration, electric potential gradients and of convection. The electrical neutrality of the system was maintained through a H+ root efflux or influx (or equivalent opposite OH− fluxes). The model outputs were compared to the rhizosphere pH of maize (Zea mays L.), ryegrass (Lolium perenne L.) and Alpine pennycress (Noccaea caerulescens (J. & C. Presl) F. K. Mey.) grown in three different soils, determined with planar optodes. A global sensitivity analysis was carried out in order to highlight the dominating factors. The composition of 37 soil solutions from the literature was used to predict rhizosphere pH. The model most of the time predicts the alkalinisation of the rhizosphere by up to +3.3 pH units, given that in aerobic soils, nitrate is the most absorbed nutrient and its internalization is equilibrated by H+ root absorption (or equivalent OH− excretion). This alkalinisation is close to that observed in the rhizosphere of ryegrass and Alpine pennycress. The rhizosphere pH predictions are very sensitive to the initial composition of the soil solution and quite insensitive to root absorption kinetics and soil buffer power for the various nutrients. Because it favours nitrate diffusion, water content is positively correlated with rhizosphere alkalinisation. Moreover, the lower the root density, the lower the effect of the root uptake on both ion concentrations and on rhizosphere pH. Convection has little influence on rhizosphere pH. This study confirms the prominent role of mineral nutrition in the determination of rhizosphere pH. The model therefore constitutes a tool for understanding and predicting rhizosphere pH.