Deforestation for agriculture increases microbial carbon use efficiency in subarctic soils

This study investigated how and  through which pathways deforestation and conversion to agricultural land (i.e. grassland, cropland) alters the microbial carbon use efficiency (CUE) in subarctic soils to allow the development of mitigation strategies to alleviate C losses. We assessed CUE using ¹⁸O-labelled water in a paired-plot approach on soils collected from 19 farms across the subarctic region of Yukon, Canada, comprising 14 pairs of forest-to-grassland conversion and 15 pairs of forest-to-cropland conversion. Microbial CUE significantly increased following conversion to grassland and cropland. Land-use conversion resulted in a lower estimated abundance of fungi, while the archaeal abundance increased, as assessed by qPCR. Interestingly, structural equation modelling revealed that increases in CUE were mediated by a rise in soil pH and a decrease in soil C:N ratio rather than by shifts in microbial community composition, i.e. the ratio of fungi, bacteria and archaea. Our findings indicate a direct control of abiotic factors on microbial CUE via improved nutrient availability and facilitated conditions for microbial growth.

The R code was developed under R v3.6.3 and adapted to work under version R v.4.1.2.

The repository includes the following files:

• general_soil_parameters_per_site.csv - general soil data assessed on pooled reference forest plot (n=19)
• general_soil_parameters_per_plot.csv - general soil data assessed on pooled replicated field samples (n=48)
• sample_data.csv - data measured for each laboratory sample (n=147)

• Land-use change effects on 18O-CUE.Rproj - Rproject (load project to work on provided scripts and data)
• load_data_script.R - loads required data
• Multivariate_normality_script.R - tests for multivariate normaility in dataset
• PCA_script.R - calculates PC1 and 2 of clay mineralogy data to reduce dimensions
• map_Yukon_script.R - create Figure 1
• plot_density_script.R - create Figure 2
• linear_mixed-effects_models_script.R - calculates response ratios
• plot_boxplots_script.R - plot boxplots per land use including compact letter display indicating significant differences, create Figure 3 + 4
• correlogram_script.R - correlation analysis to identify drivers of CUE, create Figure 6
• plot_correlations_script.R - plot drivers of CUE, create Figure 5 + 7
• SEM_script.R - development of structural equation model, create Figure 8