Profiling soil volatile organic compounds after N fertilization in a soil grown with Rosmarinus officinalis

The biogenic volatile organic compounds (BVOCs) emissions from cultivated plants-soil systems and N fertilization's role on these emissions are mostly unknown. This study aimed to elucidate the BVOCs emissions from a plant-soil system receiving N fertilization using Rosmarinus officinalis as a model case study. We put forward the hypothesis that N fertilization changes the BVOCs emissions patterns irrespectively the presence of Rosmarinus officinalis and that the emissions are related to N availability and bacterial abundance. To test our hypothesis, we established a microcosm (glass static chambers) study in a fully randomized design with four treatments: soil (S), soil and plant (S + P), soil and fertilizer (S + F), soil, plant, and fertilizer (S + P + F); using NH 4 NO 3 as N source. BVOCs were sampled from the microcosms, using the static headspace sampling technique with a solid-phase microextraction (SPME) device and analyzed by gas chromatography–mass spectrometry (HS-SPME-GC–MS). Sampling occurred 1, 2, 3, 4, 7, 10, 15, 17, 22, 25, and 31 days after applying the NH 4 NO 3 and besides BVOCs, we measured soil NH 4 + and NO 3 − concentration as well as 16S rRNA bacterial abundance. Time drives a significant portion of the variation in the chemical diversity and quantity of the different VOCs found throughout the experiment (PERMANOVA; R 2  = 0.28, p   <   0. 001 ); however, this was dependent on the presence of plant and fertilizer leading to a significant interaction (PERMANOVA; R 2  = 0.33, p   <   0. 001 ). Individual BVOCs emissions were strongly affected by nitrogen fertilization, plant presence and declined during the time. The most dominant BVOCs detected were benzyl alcohol, acetone, toluene, benzaldehyde, and caprolactam. Interestingly, the BVOCs concentration was negatively associated with soil concentration to NH 4 +, NO 3 − and total bacterial abundance.