Litter quality outweighs climate in driving grassland root decomposition

Abstract

Introduction: Root decomposition plays a critical role in nutrient cycling and carbon storage in grassland ecosystems, yet its global drivers remain poorly understood.

Methods: The study synthesized global data on root decomposition in grasslands to assess the relative importance of climate and litter quality, and to quantify the effects of environmental and biotic factors using a comprehensive meta-analysis.

Results: Results indicated that, at the global scale, litter quality exerted a stronger influence on root decomposition than climatic variables. Random forest analysis identified the ratio of acid-unhydrolyzable residue to nitrogen (AUR:N) and AUR as the most important predictors of mass loss, both of which were significantly and negatively correlated with mass loss. The meta-analysis further demonstrated that both environmental and biotic factors significantly affected root decomposition. Among environmental factors, nitrogen addition (+4.49%), phosphorus addition (+16.26%), warming (+9.80%), increased precipitation (+5.95%), and elevated CO₂ (+14.03%) were found to promote root decomposition, while reduced precipitation (-15.60%) had the negative effect. With respect to biotic factors, grazing (+7.51%) significantly increased decomposition, whereas vegetated soil (-27.84%), increased plant species richness (-4.99%), increased root litter richness (-5.93%), home-field decomposition (-4.34%), and soil biota exclusion (-10.40%) decreased it.

Discussion: These findings highlight the dominant role of litter quality over climate in regulating root decomposition at a global scale, and underscore the sensitivity of belowground processes to environmental and biotic disturbances in grassland ecosystems.

Keywords: climate; grassland; litter quality; meta-analysis; root decomposition.

Figure 1

Global distribution of study sites used in the Random Forest analysis. All sites represent control groups either from natural ecosystems or from field experiments, reflecting conditions without experimental manipulation. The size of the triangles represents the number of observations at each site.

Figure 2

Relative importance of (A) geoclimatic factors only (latitude, longitude, elevation, mean annual temperature (MAT), and mean annual precipitation (MAP)) and (B) initial root litter chemistry only (acid-unhydrolyzable residue (AUR), C, N, P, AUR:N, and C:N) in explaining variation in root litter decomposition, based on Random Forest analysis. **P < 0.01; *P < 0.05; ns, not significant.

Figure 3

Individual importance of predictor variables in explaining the residuals of mass loss for (A) all data, (B) data with decomposition time ≤ 12 months, and (C) data with decomposition time > 12 months.

Figure 4

Relative importance of geoclimatic factors (latitude, longitude, elevation, mean annual temperature (MAT), and mean annual precipitation (MAP)) and initial root litter chemistry (acid-unhydrolyzable residue (AUR), C, N, P, AUR: N, and C: N) in explaining variation in root litter decomposition, based on Random Forest analysis. (A) All data combined; (B) decomposition within 0–12 months; (C) decomposition after 12 months. **P <0.01; *P <0.05.

Figure 5

Correlations between root mass loss and acid-unhydrolyzable residue (AUR) (A), AUR:N ratio (B), and nitrogen (N) content (C). ***P <0.001.

Figure 6

The weighted response ratio (RR₊₊) for the effects of twelve environmental and biotic factors on root litter decomposition (A), and the frequency distributions of the natural logarithm of the response ratio (log_eRR) for each individual factor: nitrogen addition (B), phosphorus addition (C), warming (D), increased precipitation (E), reduced precipitation (F), elevated CO₂ (G), grazing (H), vegetated soil (I), elevated plant richness (J), elevated litter richness (K), home-field decomposition (L), and soil biota exclusion (M). The solid curves represent Gaussian distributions fitted to the frequency data. The x-axis denotes log_eRR, and the y-axis denotes frequency.

Figure 7

The weighted response ratio (RR₊₊) for the effects of eleven individual environmental and biotic factors on root litter decomposition, grouped by decomposition duration. Each panel (a–k) shows the temporal dynamics of treatment effects for nitrogen addition (A), phosphorus addition (B), warming (C), increased precipitation (D), reduced precipitation (E), grazing (F), vegetated soil (G), elevated plant richness (H), elevated litter richness (I), home-field decomposition (J), and soil biota exclusion (K). Error bars represent bias-corrected bootstrap (64999) confidence intervals. The vertical dashed line in orange indicates log_eRR = 0.