Mulching in lowland hay meadows drives an adaptive convergence of above- and below-ground traits reducing plasticity and improving biomass: A possible tool for enhancing phytoremediation

Abstract

We aimed to understand the effect of mulching (i.e., cutting and leaving the crushed biomass to decompose in situ) on above- and below-ground plant functional traits and whether this practice may be a potential tool for enhancing the phytoremediation of lowland hay meadows. To this aim, we evaluated at the community level seven years of mulching application in a PCBs and HMs soil-polluted Site of National Interest (SIN Brescia-Caffaro) through the analysis of the floristic composition and the above- and below-ground plant traits. We found that the abandonment of agricultural activities led to a marked increase in the soil organic carbon and pH, and the over-imposed mulching additionally induced a slight increase in soil nutrients. Mulching favored the establishment of a productive plant community characterized by a more conservative-resource strategy, a higher biomass development, and lower plasticity through an adaptative convergence between above- and below-ground organs. In particular, the analysis of the root depth distribution highlighted the key role of roots living in the upper soil layer (10 cm). Mulching did not show a significant effect on plant species known to be effective in terms of PCB phytoremediation. However, the mulching application appears to be a promising tool for enhancing the root web that functions as the backbone for the proliferation of microbes devoted to organic contaminants' degradation and selects a two-fold number of plant species known to be metal-tolerant. However, besides these potential positive effects of the mulching application, favoring species with a higher biomass development, in the long term, may lead to a biodiversity reduction and thus to potential consequences also on the diversity of native species important for the phytoremediation.

Keywords: PCB; fine-root traits; functional traits; global plant spectrum; heavy metals; leaf economics spectrum; phytoremediation; root economics spectrum.

Figure 1

Location of the National Relevance Site (SIN) Brescia-Caffaro in Northern Italy (A) with a detail of the contaminated agricultural areas (B), and of the control (NM; no mulching) and mulching-treated (MU) areas sampled within the study site (C). Crosses indicates the location of the vegetation sampling points, and circles indicate the location of the soil sampling points: pink, soil sampled each 10 cm layer up to 100 cm depth in 2014, light blue, mixed homogeneous soil sampled up to 40 cm in 2014 and 2020.

Figure 2

Comparison of the total biomass (A) and species richness (B) between control (NM; no mulching) and mulching-treated (MU) areas, and relationship between species richness and total biomass (C) in the 10 sampling plots. The lines in figure C indicate the best fitting and its 95% confidence interval obtained through a generalized linear model (GLM). The boxplot indicates the median (line in the middle of the boxes), the interquartile range (boxes) and 1.5 times the interquartile range (whiskers). Results of the Wilcoxon test (A, B) or GLM model (C) are reported in each subfigure; the R² value is the pseudo-R² of the GLM model; ns, not significant (p > 0.05).

Figure 3

Comparison of the community-level above- (A–E) and below-ground traits (F–J) between control (NM; no mulching) and mulching (MU) in the 10 sampling plots. The boxplot indicates the median (line in the middle of the boxes), the interquartile range (boxes), 1.5 times the interquartile range (whiskers) and outliers (circle). Results of the Wilcoxon test are reported in each subfigure; ns, not significant (p > 0.1). AGDW and BGDW, above- and below- ground dry weight, FRP, fine-root percentage, LA, community weighted mean (CWM) of leaf area, LDMC, CWM of leaf dry matter content, LNC, CWM of leaf nitrogen content, MRD, mean root diameter, SLA, CWM of specific leaf area, SRL, specific root length, RL, root length.

Figure 4

Comparison of community-level below-ground traits between control (NM; no mulching) and mulching-treated (MU) areas for each soil depth layer. The boxplot indicates the median (line in the middle of the boxes), the interquartile range (boxes), 1.5 times the interquartile range (whiskers) and outliers (circle). Small letters indicate the results of the Dunn test post-hoc comparisons; ns, not significant (p > 0.05). BGDW, below-ground dry weight (A); RL, root length (B); SRL, specific root length (C); MRD, mean root diameter (D); FRP, fine-root percentage (E).

Figure 5

Comparison of root length (RL) within each class of root diameter at the community-level between control (NM; no mulching) and mulching-treated (MU) areas in the 10 sampled plots. RL is plotted as logarithmic only for facilitating the visualization of smaller values. The boxplot indicates the median (line in the middle of the boxes), the interquartile range (boxes), 1.5 times the interquartile range (whiskers) and outliers (circle). Results of the Wilcoxon test are reported only if significant at p < 0.05 (asterisk) or p < 0.1 (point).

Figure 6

Comparison of soil pH (A), organic carbon (SOC; B) and total nitrogen content (N_tot; C) between control (NM; no mulching) and mulching-treated (MU) areas in the 11 soil plots sampled in the years 2014 and 2020. Small letters indicate the results of the Dunn test post-hoc comparisons.

Figure 7

Redundancy analysis (RDA) ordination of the plant community composition comparing control (NM; no mulching) and mulching-treated (MU) areas in relation to the community-level plant functional traits that were significantly affected by the mulching treatment (see Figures 3 ). The circles represent the 95% confidence interval. Legend: AGDW and BGDW, above- and below- ground dry weight, LDMC, community weighted mean (CWM) of leaf dry matter content, LNC, CWM of leaf nitrogen content, SLA, CWM of specific leaf area, SRL, specific root length. Supporting data are reported in Supplementary Table 1 .