Hot or not? connecting rhizosphere hotspots to total soil respiration

Abstract Aims Soil organic carbon (C) efflux is tightly linked to the rhizosphere, where soil microorganisms rapidly decompose organic compounds released from roots. Recently, imaging approaches have greatly improved our understanding of small-scale C-turnover heterogeneity and promoted the term ‘rhizosphere hotspots’ for highly active areas. However, despite often assumed, the effect of these hotspots on total soil C balances is still unknown. We aim to bridge this gap by correlating rhizosphere imaging data to soil respiration on individual plant scale. Methods We grew 17 maize ( Zea mays L. ) plants in rhizoboxes filled with sandy arable soil. After four weeks, the plants were labelled with 14 CO 2 and root exudation was visualized and quantified by 14 C-imaging one day after labeling. The evolved CO 2 was trapped in NaOH and 14 CO 2 as well as total CO 2 was quantified before and after labelling. Enzyme activity (β-glucosidase) was quantified by soil zymography. Results Bulk soil β-glucosidase activitiy negatively correlated to total CO 2 efflux, and was the most important predictor (R 2 = 0.55). Total and rhizosphere specific 14 C-activity were solely correlated to 14 CO 2 efflux (r = 0.51, r = 0.58). A combination of bulk soil β-glucosidase activity, rhizosphere- 14 C activity and root biomass, explained about 75% of variance in CO 2 efflux. Conclusions This indicates that root exudation and enzyme-activity hotspots are suitable predictors for total soil respiration, particularly when combined with root biomass to account for three-dimensional variation, and that hotspots on the rhizosphere scale are directly linked to larger scale C balances.