Soil Recovery After Fire: Texture and Porosity Shape Microbial Functions
Description
This file contains the original data underlying the article “Soil Recovery After Fire: Texture and Porosity Shape Microbial Functions.” We tested the hypothesis that soil texture and porosity control microbial survival and the speed of functional recovery after wildfire‑level heating. Intact surface samples from four Chilean soils (arid, semi‑arid, Mediterranean, temperate) were heated for 20 min at 300, 600 or 900 °C, cooled under sterile conditions and incubated 48 h; unheated soils served as controls. Four independent cores were analysed per soil × temperature. The Excel workbook provides: (i) soil‑organic‑matter loss and thermal‑load estimates, (ii) kinetic parameters of β‑glucosidase, urease, peroxidase and phosphatase (V max, K m, V max / K m, Q 10), (iii) microbial‑biomass C, N, P, (iv) concentrations of H₂O₂, •OH and O₂•⁻, (v) relative abundances of 18 bacterial genera (16S rRNA; raw reads in ENA ERP122329) and (vi) principal‑component scores summarising the multivariate response. Column names, units and methods are explained in a README sheet; the companion “Scripts” folder contains R code for figure reproduction and the 1‑D heat‑conduction model. Key findings shown by these data are: clay‑rich, porous soils retained up to 0.68 % SOM and 50–90 % of enzyme efficiency at 900 °C, whereas sandy soils lost almost all SOM (≤ 0.08 %) and suffered > 80 % declines in V max / K m and microbial biomass; H₂O₂ peaked at 3.5 µmol g⁻¹ in semi‑arid soil, indicating strong oxidative stress; thermotolerant genera such as Thermus and Bacillus dominated early recolonisation; PCA separates sandy from clayey soils along a gradient of SOM, biomass and fine particles. Researchers can reuse the dataset to calibrate soil‑heating models, compare enzyme thermodynamics across climates or include the measurements in meta‑analyses of fire effects on soil biology. Released under CC‑BY 4.0; cite as: Merino C. et al. (2025) Soil Recovery After Fire – Supporting Data, Mendeley Data, V1, DOI 10.17632/XXXXX.Y.
Files
Steps to reproduce
Sampling and fire simulation Intact 0–5 cm soil cores (64 × 38 × 60 mm) were taken from four Chilean sites spanning arid, semi‑arid, Mediterranean and temperate climates. After sieving < 2 mm, samples were rewetted to 80 % water‑holding capacity and pre‑incubated 7 d at 20 °C. Four independent cores per soil were heated in a Nabertherm L 9/11 muffle furnace for 20 min at 300, 600 or 900 °C; unheated controls remained at 20 °C. Cores cooled in a class II laminar‑flow hood and incubated 48 h to permit microbial reactivation. Heat‑transfer quantification A one‑dimensional finite‑difference conduction model (R 4.3.2; packages pracma, signal) used measured bulk density and thermal conductivity (KD2‑Pro needle probe) plus literature heat capacities to derive soil‑specific diffusivity (α = λ/ρC). Surface temperature equalled the furnace set‑point, the initial profile 20 °C. Modelled peak temperatures and cumulative thermal load were validated against type‑K thermocouples placed in sacrificial cores. Analytical methods • Soil organic matter: loss‑on‑ignition, 5 g, 4 h @ 550 °C. • Enzyme kinetics: 1 g soil in 100 mL water; substrates pNPG (β‑glucosidase), urea (urease), tetramethyl‑benzidine + H₂O₂ (peroxidase), pNPP (acid/alkaline phosphatase). Seven substrate levels (0–500 µM) read on a BioTek Synergy HTX microplate reader. Michaelis–Menten parameters (V max, K m) obtained by nonlinear least‑squares (nls); catalytic efficiency = V max/K m; Q₁₀ from rate ratios. • Reactive‑oxygen species: H₂O₂ by iodometric titration; •OH by terephthalic‑acid fluorescence (310/425 nm); O₂•⁻ by hydroxylamine‑sulfanilamide colorimetry (530 nm). • Microbial biomass C and N: chloroform fumigation–extraction (TOC‑VCSH analyser; Kjeldahl distillation). Biomass P: fumigation–incubation followed by molybdate‑ascorbate assay. • Bacterial community: DNA extracted with Qiagen PowerSoil Pro, quantified on Qubit 4. The V1‑V2 region of 16S rRNA (primers 27F/338R) was sequenced on an Illumina MiSeq (2 × 250 bp). Reads were processed in QIIME2 2023.5 (DADA2 denoising, SILVA 138 taxonomy). Genus‑level counts are provided here; raw reads are archived in ENA (ERP122329). Statistics and reproducibility Normality (Shapiro–Wilk) and homogeneity (Levene) were tested in R 4.3.2. Repeated‑measures ANOVA within soils and two‑way ANOVA (soil × temperature) were followed by Tukey HSD (α = 0.05). Principal‑component analysis on centred‑scaled variables used the vegan package; graphics were produced with ggplot2. All R scripts—including the heat‑conduction routine—are supplied in the accompanying “Scripts” folder. Data package The Excel workbook in this repository contains one sheet per figure (raw replicate values) plus a README detailing column names, units and detection limits. Following the outlined sampling, furnace treatment, assay protocols and R workflow enables full reproduction of our results and facilitates comparative or modelling studies on post‑fire soil recovery.
Institutions
- Universidad de La Frontera