Forest Age Rivals Climate to Explain Reproductive Allocation Patterns in Forest Ecosystems Globally

Abstract

Forest allocation of net primary productivity (NPP) to reproduction (carbon required for flowers, fruits, and seeds) is poorly quantified globally, despite its critical role in forest regeneration and a well-supported trade-off with allocation to growth. Here, we present the first global synthesis of a biometric proxy for forest reproductive allocation (RA) across environmental and stand age gradients from a compiled dataset of 824 observations across 393 sites. We find that ecosystem-scale RA increases ~60% from boreal to tropical forests. Climate shows important non-linear relationships with RA, but is not the sole predictor. Forest age effects are comparable to climate in magnitude (MAT: ß = 0.24, p = 0.021; old growth forest: ß = 0.22, p < 0.001), while metrics of soil fertility show small but significant relationships with RA (soil pH: ß = 0.07, p = 0.001; soil N: ß = -0.07, p = 0.001). These results provide strong evidence that ecosystem-scale RA is mediated by climate, forest age, and soil conditions, and is not a globally fixed fraction of positive NPP as assumed by most vegetation and ecosystem models. Our dataset and findings can be used by modellers to improve predictions of forest regeneration and carbon cycling.

Keywords: climate; ecosystem modelling; forest age; forest ecosystems; forest regeneration; reproductive allocation; soil fertility.

FIGURE 1

(a) R/NPP is strongly correlated with a litterfall proxy for RA, R/(R + L), at forested sites (n = 66, R = reproductive litterfall, L = leaf litterfall, NPP = net primary productivity, all fluxes in Mg ha⁻¹yr ⁻¹), adapted from Hanbury‐Brown et al. (2022); point colours indicate Whittaker biomes as shown in (b). (b) Observations of R/(R + L) are distributed across Whittaker biome space. Point size indicates annual average R/(R + L), overlapping points appear darker. (c) Observations of R/(R + L) are distributed across the forested (green) continents.

FIGURE 2

Boxplots of RA proxy R/(R + L) across (a) biomes (n = 824) and (b) plant types defined by biome and dominant leaf morphology (n = 716; sites with undetermined or mixed dominant leaf morphology excluded) and (c) biome and forest age classification (n = 824). Panels show results from ANOVA and Tukey's HSD post hoc comparisons (indicated with letters) on Box‐Cox transformed data (raw data shown in plot). Boxplots show median, quartiles, and 1.5*IQR whiskers.

FIGURE 3

Standardised coefficient estimates (points) and 95% confidence intervals (lines) from final models predicting a) RA proxy (R/(R+ L)), (b) Reproductive litterfall (R) and (c) Leaf litterfall (L). MAT = mean annual temperature (°C), MAP = mean annual precipitation (mm/year), Forest age (young) is the reference level for the categorical variable Forest age, asterisks indicate statistical significance (*p < 0.05, **p < 0.01, ***p < 0.001).

FIGURE 4

Predicted (a) RA proxy (R/ (R + L)), (b) Reproductive litterfall (R), and (c) Leaf litterfall (L) from our final model (Figure 3, Table 1) in mid‐aged forests with mean soil texture, mean soil pH, and mean soil N, as a function of mean annual temperature (MAT) and mean annual precipitation (MAP). Convex hulls are defined by observations, plotted in grey (point size reflects sampling duration in years).

FIGURE 5

Predicted effects of climate terms (mean annual temperature (MAT), mean annual precipitation (MAP), and their quadratic terms and interaction; panels a‐c) and forest age (panels d‐f) on RA proxy (R/(R + L)), reproductive litterfall (R), and leaf litterfall (L) over mean annual temperature (MAT). Forest age is not included in the final model of leaf flux due to statistical non‐significance (panel f).