Drought soil legacy overrides maternal effects on plant growth

Abstract

Maternal effects (i.e. trans-generational plasticity) and soil legacies generated by drought and plant diversity can affect plant performance and alter nutrient cycling and plant community dynamics. However, the relative importance and combined effects of these factors on plant growth dynamics remain poorly understood.We used soil and seeds from an existing plant diversity and drought manipulation field experiment in temperate grassland to test maternal, soil drought and diversity legacy effects, and their interactions, on offspring plant performance of two grassland species (Alopecurus pratensis and Holcus lanatus) under contrasting glasshouse conditions.Our results showed that drought soil legacy effects eclipsed maternal effects on plant biomass. Drought soil legacy effects were attributed to changes in both abiotic (i.e. nutrient availability) and biotic soil properties (i.e. microbial carbon and enzyme activity), as well as plant root and shoot atom ¹⁵N excess. Further, plant tissue nutrient concentrations and soil microbial C:N responses to drought legacies varied between the two plant species and soils from high and low plant diversity treatments. However, these diversity effects did not affect plant root or shoot biomass.These findings demonstrate that while maternal effects resulting from drought occur in grasslands, their impacts on plant performance are likely minor relative to drought legacy effects on soil abiotic and biotic properties. This suggests that soil drought legacy effects could become increasingly important drivers of plant community dynamics and ecosystem functioning as extreme weather events become more frequent and intense with climate change. A plain language summary is available for this article.

Keywords: 15N; above‐ground–below‐ground interactions; climate change; climate extremes; drought shelters; extracellular soil enzymes; mycorrhizae; plant diversity.

Figure 1

Schematic depiction of the experimental design. In the field, seeds from one subordinate [Alopecurus pratensis] and one dominant [Holcus lunatus] grass species and soils were collected from both ambient and drought subplots within both low and high plant diversity plots (n = 5 for both high and low diversity plots, yielding a total of 20 subplots from which soils and seeds were collected). Next, seeds from both species from both ambient and drought origin were planted into pots containing ambient soils and drought legacy soils taken from the same field plot (one individual per pot). During the glasshouse experiment, half of the pots from each treatment combination were well‐watered and half were droughted for 4 weeks

Figure 2

Shoot biomass averaged across all plants (Alopecurus pratensis and Holcus lanatus) from seeds of ambient and drought maternal origin grown in ambient and drought legacy soils. Points accompanied by the same lower case letter do not differ at p < 0.05 (Tukey's HSD). Data are means plus one standard error (n = 138). ANOVA results are presented in Supporting Information Tables S4a,b

Figure 3

Shoot nitrogen (%N) content of plants (panel a); shoot atom %¹⁵N excess (panel b); root %N content (panel c); and root atom %¹⁵N excess (panel d) of plants (averaged across Alopecurus pratensis and Holcus lanatus) grown in soils from ambient and drought legacies. Within each panel that contains more than two bars, bars topped with the same lower case letter do not differ at p < 0.05 (Tukey's HSD). Data are means ± 1 SE (shoot %N and shoot atom %¹⁵N excess, root %N: n = 138, root atom ¹⁵N excess: n = 136). ANOVA results are presented in Supporting Information Tables S5a,b

Figure 4

Microbial C (carbon) (panel a) of soils from ambient and drought legacy; microbial carbon‐to‐nitrogen (C:N) ratios (panel b) of soils from ambient and drought legacy, low and high diversity legacy soils that were well‐watered and droughted; amino acid deaminase (DEA) (panel c) and urease (URE) (panel d) measurements of soils from low and high diversity legacy, and ambient and drought legacy. Data are averaged across the two study species (Alopecurus pratensis, Holcus lanatus). Within each panel, points accompanied by the same lower case letter do not differ at p < 0.05 (Tukey's HSD). Data are means ± 1 SE (n = 138). ANOVA results are presented in Supporting Information Tables S6 and 7Sa,b

Figure 5

Total inorganic nitrogen (TIN) measurements (panel a) from Alopecurus pratensis and Holcus lanatus grown in low and high diversity legacy, ambient and drought legacy soils; shoot %C (panel b) of Alopecurus pratensis and Holcus lanatus from ambient and drought maternal origin grown in soils from ambient and drought legacy; mycorrhizal vesicle % colonization (panel c) of Alopecurus pratensis and Holcus lanatus grown in soils from ambient and drought legacy, low and high diversity legacy soils and shoot carbon‐to‐nitrogen (C:N) ratios (panel d) of Alopecurus pratensis and Holcus lanatus grown in soils from ambient and drought legacy, low and high diversity legacy soils. Within each panel, points accompanied by the same lower case letter do not differ at p < 0.05 (Tukey's HSD). Data are means ± 1 SE (n = 138 except total inorganic N: n = 137). ANOVA results are presented in Supporting Information Tables S4–S6a,b