Soil microbial communities, soil, environmental and forest structural properties across old-growth forest and naturally regenerating or actively restored selectively logged tropical forest in Sabah, Borneo, March 2018 v2

Soil samples and forest structural and environmental data were collected at nine sites during one campaign in March 2018 . Three sampling sites were chosen in each of three forest types of old-growth, naturally regenerating and actively restored logged forest. Old-growth forest was located in the Danum Valley Conservation Area, and logged forest sites were situated within the adjacent Innoprise Face Foundation Rainforest Rehabilitation Project (INFAPRO) area. A geospatial transect design was employed for soil sampling and measurement of microclimate variables at each site to assess variation in microbial communities over cm to m scales. This comprised three transects radiating out from one centre point positioned at 120 ° to one another. Sampling points were located at the centre point, then at increasing distances of 10 cm, 30 cm, 90 cm, 2.7 m, 8.4 m and 24.3 m along each transect relative to the centre. The circumferences of all tree stems with diameter at breast height (DBH) >5 cm within a 2.5 m buffer of all transects were recorded at each site for site-level calculation of stem density and basal area. Soil physicochemical properties were analysed at the Sabah Forest Research Centre, Sepilok. Organic layer soil microbial community data were generated by 16S and ITS amplicon sequencing at UKCEH Wallingford, and Phospholipid Fatty Acid (PLFA) analysis at Lancaster University in the UK. Fungal functional guild classifications were assigned to fungal sequences using the FUNGuild annotation tool. Alpha diversity indices for microbial groups and relative abundances of fungal guilds were generated after rarefying by minimum read counts for each group . Rates of spatial turnover in soil microbial taxa for different microbial groups were calculated using the distance decay approach via linear modelling.