Native soil microbes buffer savanna trees against nutrient limitation but are drought sensitive
Data files
Mar 26, 2025 version files 64.68 KB
-
Gx2data.csv
49.66 KB
-
Gx2HarvestedSeedlings.csv
6.01 KB
-
README.md
9.01 KB
Abstract
Belowground microbial communities are vital to ecosystem nutrient cycling, plant health, and resource acquisition, yet belowground plant-soil interactions in savannas remain understudied, especially in their responses to environmental stressors like drought and nutrient limitation. Here, we evaluate if native soil microbiomes have positive or negative effects on tree growth and if these effects are dependent on the level of resource availability. We grew 6 tree species from Kruger National Park, South Africa, for 8 weeks under fully factorial soil inoculant, water stress, and nitrogen limitation treatments (i.e., sterile/inoculated soils, droughted/non-droughted water supply, and low/high rate of nitrogen supply). In all resource treatments, inoculated plants grew significantly more than sterile plants. Under low nitrogen, trees increased investment in nitrogen-fixing nodules and mycorrhizal associations, leading to increased mass gain. Soil inoculant was most beneficial in non-droughted water conditions, indicating that microbial symbiont effects decreased under drought.
Synthesis: Belowground microbial symbionts improved savanna tree growth in limited resource environments and could be critical for plant growth in the field. However, drought substantially affected tree growth and the benefit of native soil microbes to tree growth, indicating that extreme droughts could create lasting consequences for both aboveground tree growth and belowground beneficial microbial communities.
https://doi.org/10.5061/dryad.h44j0zpx9
Description of the data and file structure
At the time of transplanting, 10 individuals of each tree species were harvested for initial size and biomass estimates. We recorded starting biomass, stem height, basal diameter, diameter at 10 cm height, root diameter, root length, wet and dry leaf stem mass, and fine root (0-2 mm) and coarse root (> 2 mm) dry mass. We also recorded stem height and basal diameter of all transplanted trees. Throughout the experiment, soil moisture was measured at three depths (2 cm, 5 cm, 10 cm) for all trees every two weeks, both before and after watering, to ensure that soil moisture was consistent across the same watering treatments (with probes sterilized with 70% ethanol after every pot). Soil moisture remained consistent within or
treatments with slight variation across the three depths measured over the course of the experiment, ensuring there were no substantial water availability differences across species or treatment, with the
treatment receiving 40% less water than the
treatment (Supp. Fig. 1). Seedlings were harvested after 8 weeks, when we recorded stem height, basal diameter, diameter at 10 cm, root diameter, root length, and wet and dry leaf, stem, coarse root, fine root, and nodule biomass. To calculate mass gain, we subtracted final total dry weights from the averaged total dry weight of the 10 seedlings harvested at the start of the experiment. Allometric equations were not used since all seedlings were small and similar in size at transplant, immediately after germination.
Additionally, fine root samples were collected from each seedling and stored at 4°C in 95% ethanol for mycorrhizal colonization analyses. Ten intersections of mycorrhizal colonization were taken per species per treatment (2 roots per individual tree). Roots were pooled by species and treatment, soaked in deionized water, and rinsed three times to remove ethanol. Roots were cut into 1-cm sections, cooked at 70°C in 10% KOH for 2-6 hours, acidified with 1% HCl, and stained with a 0.05% trypan blue solution for 15 minutes. Roots were de-stained in a lactic acid glycerol solution overnight (Wurzburger and Wright 2015). Root sections were mounted on microscope slides and the number of mycorrhizal structures (arbuscules, vesicles, and hyphae) were quantified with a compound microscope using a random-intercept method (McGonigle et al. 1990). Mycorrhizal colonization was calculated as the percentage of mycorrhizal density colonized with arbuscules, vesicles, and hyphae by fine-root length (Wurzburger and Wright 2015). Neither nodulation nor mycorrhizal colonization was observed in ‘sterile’ treatments, whereas both were abundant in ‘inoculation’ treatments, which confirms that treatments worked as intended.
Files and variables
File: Gx2data.csv
Description: Gx2data.csv is the csv file with harvested seedling data after 2 months of growth according to experimental treatment. There are 6 species (40 plants per species) that vary across inoculation treatment, nitrogen treatment, and water treatment. Below is the column by column description:
Variables
- SpeciesID: 6 letter species ID code
- Species: full scientific species name
- PlantID: plant ID for the experiment (each species was given a corresponding number between 1 through 40 to keep track over the course of the experiment)
- Pot: Pot number for the experiment (between 1 and 240)
- InocTrt: inoculation treatment, options are Sterile or Inoculated
- STrt: inoculation treatment, options are S- (sterile) or S+ (inoculated)
- NTrt: nitrogen treatment, options are N- (low nitrogen) or N+ (high nitrogen)
- WaterTrt: water treatment, options are drought or watered
- WTrt: water treatment, options are W- (drought) or W+ (watered)
- Sowed: date seed was sowed
- Transplanted: date seed was transplanted for experiment start
- DaysGrownTH: number of days grown since transplant (unit: days)
- Transp_sdiam: stem diameter of the seedling at date of transplant (unit: cm)
- Transp_stemht: stem height of seedling at date of transplant (unit: cm)
- Harvested: date seedling was harvested at experiment end
- Sdiam: stem diameter at harvest (unit: cm)
- Diam10cm: diameter at 10 cm at harvest (unit: cm)
- StemHt: stem height at harvest (unit: cm)
- RootLn: root length of tap root at harvest (unit: cm)
- Rdiam: root diameter at harvest (unit: cm)
- Stem_Bag: bag weight of stem bag (unit: grams)
- Stem_Wet: weight of bag + stem when wet (unit: grams)
- Stem_Dry: weight of bag + stem when dry (unit: grams)
- Leaf_Bag: bag weight of leaf bag (unit: grams)
- Leaf_Wet: weight of bag + leaf when wet (unit: grams)
- Leaf_Dry: weight of bag + leaf when dry (unit: grams)
- Coarse_Bag: bag weight of coarse root bag (unit: grams)
- Coarse_Wet: weight of bag + coarse root when wet (unit: grams)
- Coarse_Dry: weight of bag + coarse root when dry (unit: grams)
- FR_Bag: bag weight of fine root bag (unit: grams)
- FR_Wet: weight of bag + fine root when wet (unit: grams)
- FR_Dry: weight of bag + fine root when dry (unit: grams)
- Nod_Bag: bag weight of root nodule bag (unit: grams)
- Nod_Wet: weight of bag + nodules when wet (unit: grams)
- Nod_Dry: weight of bag + nodules when dry (unit: grams)
- Nod_Num: counted number of nodules on root system at harvest (unit: number of nodules)
- NodStatus: Nodulating or non-nodulating, if species is known to have the capability of nodulating
- MycCol: mycorrhizal colonization, degree of mycorrhizal colonization based on sectioning and staining (unit: % mycorrhizal colonization)
File: Gx2HarvestedSeedlings.csv
Description: Gx2HarvestedSeedling.csv is the csv file with harvested seedling data after germination to understand pre-treatment growth and starting conditions. There are 6 species (10 plants per species) and usually seeds were germinated and grown for about 2 weeks before transplant. These were data taken of seedlings that were not transplanted but that provided data on initial biomass data for the experiment. Roots were not split into coarse and fine roots for this dataset because seedlings were too young to have substantial root systems. Below is the column by column description:
Variables
- Species: 6 letter species ID code
- PlantID: full scientific species name
- Sowed: date seed was planted to germinate
- Harvested: date seedling was harvested for data (also corresponding transplant date for non-harvested seedlings)
- DaysGrown: how many days the seed was grown (unit: days)
- RootDiam: root diameter of seedling at harvest (unit: cm)
- BasalDiam: stem basal diameter at harvest (unit: cm)
- RootLen: root length at harvest (unit: cm)
- StemHt: stem height at harvest (unit: cm)
- Stem_Bag: bag weight of stem bag (unit: grams)
- Stem_Wet: weight of bag + stem when wet (unit: grams)
- Stem_Dry: weight of bag + stem when dry (unit: grams)
- Leaf_Bag: bag weight of leaf bag (unit: grams)
- Leaf_Wet: weight of bag + leaf when wet (unit: grams)
- Leaf_Dry: weight of bag + leaf when dry (unit: grams)
- Root_Bag: bag weight of root bag (unit: grams)
- Root_Wet: weight of bag + root when wet (unit: grams)
- Root_Dry: weight of bag + root when dry (unit: grams)
Code/software
All statistical analyses were conducted in R (R Core Team 2021) using linear mixed-effects models (LMMs) to test the effects of soil and resource treatments. Variables of interest were tree mass gain, root-to-shoot ratio, mass fraction of stem, leaf, coarse root, fine root, nodule biomass, and mycorrhizal colonization in interaction with soil treatment, nitrogen treatment, water treatment, and nodulation status with tree species as random effects. Model selection was done via AIC, with the best model selected as the simplest model within DAIC < 2 of the overall minimum AIC. Linear mixed-effects models were run in lme4 (Bates et al. 2015) and figures in ggplot2 (Wickham 2016). All pairwise comparisons were calculated using the emmeans and cld functions from emmeans and multcomp packages, respectively (Lenth et al. 2018, Hothorn et al. 2008).
Access information
Other publicly accessible locations of the data:
- None
Data was derived from the following sources:
- Greenhouse experiment