38 Years of Vegetation Restoration on Eroded Red Soils

Description

This database contains specific data for the subsequent analysis of environmental factors and microbial properties of red soils under different vegetation types. Vegetation restoration has been proven as an important strategy for recovering soil health. This study hypothesize that the physicochemical and microbial properties of eroded red soils have continued to improve through 38 years of revegetation. To test the hypothesis, soil and litter physicochemical properties, soil microbial abundance, diversity, and community structure, and soil microbial nutrient-cycling functions were evaluated through uneroded and restored eroded red soils under Pinus massoniana, Cinnamomum camphora, and Lespedeza bicolor.

Steps to reproduce

Fig_1: Soil enzyme activities. The spectrophotometric methods were employed to determine the activity of nine soil enzymes. Fig_2: Soil microbial alpha diversity and community structure. Microbial alpha diversity indices (Shannon index and Chao1 index) were calculated using the “vegan” package in R Non-metric multidimensional scaling analysis based on Bray-Curtis dissimilarity matrices among samples were performed using the “vegan” package. Fig_3: Z-score values of relative abundance of nutrient-cycling functions. Fig_5: Mean squared error of core microbial orders and spearman’s correlation matrix of core microbial orders and environmental factors. Random forest models were employed to analyze the relative importance of microbial taxa using the “randomForest” package in R. The importance of variables was assessed based on the percentage increase in mean squared error (IncMSE%), with higher IncMSE% values indicating greater variable importance. Spearman correlation analysis was then conducted to explore associations between these microbial taxa and environmental factors. Fig_S2: Relative abundance of soil microorganisms at the phylum level. Fig_S3: Spearman's correlation matrix for environmental factors and results of the Mantel test. Mantel tests were implemented using the “linkET” package to assess the association between environmental factors and microbial properties. Pairwise similarity matrices of microbial properties and environmental factors were computed using the “psych” package, based on Spearman’s correlation coefficients. Fig_S4: Relative abundance of carbon cycling functions. The relative abundance of each gene was normalized to transcripts per million based on gene length and sequencing depth. Fig_S5: Relative abundance of nitrogen cycling functions. The relative abundance of each gene was normalized to transcripts per million based on gene length and sequencing depth. Fig_S6: Relative abundance of phosphorus cycling functions. The relative abundance of each gene was normalized to transcripts per million based on gene length and sequencing depth.

Institutions

• Nanjing Normal University

  Jiangsu, Nanjing

Categories

Licence