Impact of soil sample pre-treatment on microbial carbon use efficiency and associated parameters

The ¹⁸O-labelling method is a powerful tool for studying soil microbial carbon use efficiency (CUE), as the label itself does not affect the soil microbial carbon metabolism. Beside parameters of soil microbial activity, the DNA extracted within this method can be used to gain information on soil microbial metagenome, e.g. community composition. Amongst others, the ¹⁸O-CUE method can support studies on the impact of land use change on the soil microbial metabolism and thus soil carbon dynamics.

As soil sample handling and length of incubation prior to analysis have been shown to affect measured parameters of soil microbial activity (e.g. Birch-effect) and community, different sample preparation measures are recommended in the literature depending on the focus of the study. However, so far the sensitivity of CUE and associated parameters to sample pre-treatment is unknown. We therefore tested i) how different sample pre-treatments (freezing, drying, and fresh) affect the parameters of soil microbial metabolism and community measured within the ¹⁸O-CUE method. In order to determine the potential to use the ¹⁸O-CUE in land use change studies we tested ii) if mentioned parameters of the two compared systems forest and cropland were - if applicable -affected in a similar way.

Based on five different paired sites (cropland and forest each), we evaluated the effects on the CUE and associated parameters (respiration, soil microbial biomass C, DNA extracted, abundance of fungi, bacteria and archaea) of six pre-treatments for soil samples via the ¹⁸O-CUE method: (i) direct analysis of field-fresh soil samples, analysis after pre-incubation of (ii) field-fresh, (iii) air-dried, (iv) oven-dried, (v) frozen at -20°C and (vi) in-situ frozen soil samples (dry ice and subsequently liquid N₂).

Among all pre-treatments, the pre- incubation of 14 days as such had the strongest effect on metabolic parameters. Furthermore, while individual parameters (respiration, microbial biomass C) were influenced by the pre-treatment the ¹⁸O-CUE was relatively insensitive. We therefore conclude that not only fresh but also archived, dried soil samples can be used to obtain representative CUE values. Drying soil samples and rewetting led to increased fungal abundance in the forest soil, while this was not the case for croplands.