Relationships between soil nutrient status and nutrient-related leaf traits in Brazilian cerrado and seasonal forest communities

Abstract

Aims

In Brazil, cerrado and seasonal forest occur in close proximity but on soils with very different chemistry and texture. We aimed to compare species from these two vegetation types in terms of leaf N and P concentrations (of green and senesced leaves) and proportional nutrient resorption, quantifying the relationships among these traits, with other key leaf traits, and with soil properties.

Methods

We collected topsoil at 100 25 m² sample plots in south-eastern Brazil and measured leaf traits of 89 woody species occurring therein, expressing them as community-weighted means. Soil nutrient status was indexed using eight standard variables.

Results

In terms of properties such as pH, clay content, and cation exchange capacity, cerrado soils were deemed as being less “fertile” than forest soils, yet cerrado and forest sites did not differ in soil total N or available P (which themselves were negatively correlated). On average, forest species showed higher proportional P resorption but lower N resorption. Leaves with higher nutrient concentrations were less scleromorphic.

Conclusion

In Brazilian cerrado and forests, variation in green- and senesced-leaf nutrients was better aligned with generalised measures of soil fertility than with total N or available P and showed far more clear patterns than nutrient resorption efficiencies.

Appendix

Table 3

Table 3 Species and number of individuals sampled in cerrado and seasonal forest in the Vaçununga State Park, south-eastern Brazil

Table 4 Leaf functional traits and nutrient concentration of species sampled in cerrado and seasonal forest in the Vaçununga State Park, south-eastern Brazil

Table 5 Mean, standard deviation and range of each soil feature in the cerrado and in the seasonal forest. Clay (g.kg⁻¹), cation exchange capacity (CEC, mmol.kg⁻¹), organic matter (OM, mmol.kg⁻¹), pH, and sum of bases (SB, mmol.kg⁻¹), available phosphorus (P, mg.kg⁻¹), total nitrogen (N, mg.kg⁻¹), Aluminium (Al, mmol.kg⁻¹), Cations K⁺ (K, mmol.kg⁻¹), Ca²⁺ (Ca, mmol.kg⁻¹), Mg⁺ (Mg, mmol.kg⁻¹), as well as potential acidity (H.Al, mmol.kg⁻¹), percentage of base saturation (V, %), percentage of aluminium saturation (m, %), silt, and sand proportions (1:1000)

Table 6 Loadings of each soil variable on the first two principal components. Clay, cation exchange capacity (CEC), organic matter (OM), pH, and sum of bases (SB) contributed more strongly to PC1. Available phosphorus (P), total nitrogen (N), and aluminium contributed more strongly to PC2

Table 7 Average soil aluminium concentration of species sampled in cerrado and seasonal forest in the Vaçununga State Park, south-eastern Brazil. We calculated the mean Al concentration across all sites where the species occurred within each vegetation type. We expected cerrado species, or at least some families, to be more associated with high Al sites and forest species to be more associated with low Al sites. Unexpectedly, forest species can occupy, on average, sites with higher Al concentration and there is no pattern of aggregation, in terms of species relatedness, towards high or low Al sites in both vegetation types. Species occurring in both vegetation types, highlighted in bold, tend to be plastic: whereas some were more associated with high Al sites in the forest, others were more associated with high Al sites in the cerrado. Perhaps the soil Al concentration is not a good predictor of either cerrado or forest species preferences or floristic distinctiveness in regards do high or low Al soils. Perhaps what really matters is the difference in species tolerance against the toxic forms of aluminium that are predominant under the more acid soils of cerrado. Although cerrado soils had higher Al concentrations than the forest (Table 5), some forest plots had surprising high Al concentrations. Some of the high Al species in the forest were rare species (Table 3) occurring incidentally in these uncommon high Al plots