Summary Soil nutrient availability can strongly affect root traits. In tropical forests, phosphorus (P) is often considered the main limiting nutrient for plants. However, support for the P paradigm is limited, and N and cations might also control tropical forests functioning. We used a large‐scale experiment to determine how the factorial addition of nitrogen (N), P and cations affected root productivity and traits related to nutrient acquisition strategies (morphological traits, phosphatase activity, arbuscular mycorrhizal colonisation and nutrient contents) in a primary rainforest growing on low‐fertility soils in Central Amazonia after 1 yr of fertilisation. Multiple root traits and productivity were affected. Phosphorus additions increased annual root productivity and root diameter, but decreased root phosphatase activity. Cation additions increased root productivity at certain times of year, also increasing root diameter and mycorrhizal colonisation. P and cation additions increased their element concentrations in root tissues. No responses were detected with N addition. Here we showed that rock‐derived nutrients determined root functioning in low‐fertility Amazonian soils, demonstrating not only the hypothesised importance of P, but also highlighting the role of cations. The changes in fine root traits and productivity indicated that even slow‐growing tropical rainforests can respond rapidly to changes in resource availability.

Purpose Large parts of the Amazon rainforest grow on weathered soils depleted in phosphorus and rock-derived cations. We tested the hypothesis that in this ecosystem, fine roots stimulate decomposition and nutrient release from leaf litter biochemically by releasing enzymes, and by exuding labile carbon stimulating microbial decomposers. Methods We monitored leaf litter decomposition in a Central Amazon tropical rainforest, where fine roots were either present or excluded, over 188 days and added labile carbon substrates (glucose and citric acid) in a fully factorial design. We tracked litter mass loss, remaining carbon, nitrogen, phosphorus and cation concentrations, extracellular enzyme activity and microbial carbon and nutrient concentrations. Results Fine root presence did not affect litter mass loss but significantly increased the loss of phosphorus and cations from leaf litter. In the presence of fine roots, acid phosphatase activity was 43.2% higher, while neither microbial stoichiometry, nor extracellular enzyme activities targeting carbon- and nitrogen-containing compounds changed. Glucose additions increased phosphorus loss from litter when fine roots were present, and enhanced phosphatase activity in root exclusions. Citric acid additions reduced litter mass loss, microbial biomass nitrogen and phosphorus, regardless of fine root presence or exclusion. Conclusions We conclude that plant roots release significant amounts of acid phosphatases into the litter layer and mobilize phosphorus without affecting litter mass loss. Our results further indicate that added labile carbon inputs (i.e. glucose) can stimulate acid phosphatase production by microbial decomposers, highlighting the potential importance of plant-microbial feedbacks in tropical forest ecosystems.

Purpose The tropical phosphorus cycle and its relation to soil phosphorus (P) availability are a major uncertainty in projections of forest productivity. In highly weathered soils with low P concentrations, plant and microbial communities depend on abiotic and biotic processes to acquire P. We explored the seasonality and relative importance of drivers controlling the fluctuation of common P pools via processes such as litter production and decomposition, and soil phosphatase activity. Methods We analyzed intra-annual variation of tropical soil phosphorus pools using a modified Hedley sequential fractionation scheme. In addition, we measured litterfall, the mobilization of P from litter and soil extracellular phosphatase enzyme activity and tested their relation to fluctuations in P- fractions. Results Our results showed clear patterns of seasonal variability of soil P fractions during the year. We found that modeled P released during litter decomposition was positively related to change in organic P fractions, while net change in organic P fractions was negatively related to phosphatase activities in the top 5 cm. Conclusion We conclude that input of P by litter decomposition and potential soil extracellular phosphatase activity are the two main factors related to seasonal soil P fluctuations, and therefore the P economy in P impoverished soils. Organic soil P followed a clear seasonal pattern, indicating tight cycling of the nutrient, while reinforcing the importance of studying soil P as an integrated dynamic system in a tropical forest context.

Most leaf functional trait studies in the Amazon basin do not consider ontogenetic variations (leaf age), which may influence ecosystem productivity throughout the year. When leaf age is taken into account, it is generally considered discontinuous, and leaves are classified into age categories based on qualitative observations. Here, we quantified age-dependent changes in leaf functional traits such as the maximum carboxylation rate of ribulose-1,5-biphosphate carboxylase/oxygenase (Rubisco) (Vcmax), stomatal control (Cgs%), leaf dry mass per area and leaf macronutrient concentrations for nine naturally growing Amazon tropical trees with variable phenological strategies. Leaf ages were assessed by monthly censuses of branch-level leaf demography; we also performed leaf trait measurements accounting for leaf chronological age based on days elapsed since the first inclusion in the leaf demography, not predetermined age classes. At the tree community scale, a nonlinear relationship between Vcmax and leaf age existed: young, developing leaves showed the lowest mean photosynthetic capacity, increasing to a maximum at 45 days and then decreasing gradually with age in both continuous and categorical age group analyses. Maturation times among species and phenological habits differed substantially, from 8 ± 30 to 238 ± 30 days, and the rate of decline of Vcmax varied from −0.003 to −0.065 μmol CO2 m−2 s−1 day−1. Stomatal control increased significantly in young leaves but remained constant after peaking. Mass-based phosphorus and potassium concentrations displayed negative relationships with leaf age, whereas nitrogen did not vary temporally. Differences in life strategies, leaf nutrient concentrations and phenological types, not the leaf age effect alone, may thus be important factors for understanding observed photosynthesis seasonality in Amazonian forests. Furthermore, assigning leaf age categories in diverse tree communities may not be the recommended method for studying carbon uptake seasonality in the Amazon, since the relationship between Vcmax and leaf age could not be confirmed for all trees.

<p>In large parts of the Amazon rainforest low soil phosphorus availability may prevent the stimulation of forest growth in response to elevated atmospheric CO<sub>2</sub> (eCO<sub>2</sub>). One strategy of plants could be to increase the relative allocation of the extra C belowground to their root systems to enhance nutrient acquisition and alleviate the potential phosphorus limitation, but little is known about the responses of tropical lowland forest species. We hypothesized that in tropical understory plants will trigger a first a fast upregulation of fine root phosphatase activities, followed by changes in fine root productivity and adaptions of morphological parameters, such as specific root length (SRL), specific root area (SRA) and root tissue density (RTD) to enhance phosphorus mobilization, increase its availability and exploit a larger soil and litter volume.</p><p>We tested our hypothesis in the first CO<sub>2</sub> enrichment experiment in Central Amazonia at a low soil phosphorus site, increasing CO<sub>2</sub> levels by 200 ppm relative to CO<sub>2</sub> ambient (aCO<sub>2</sub>) concentrations using open top chambers (OTC) in the forest understory. We monitored potential root phosphatase activity, root productivity, and morphological traits in the soil with ingrowth cores (0-15 cm) and in the litter layer, as well as root biomass stocks in 0-5 and 5-10 cm of depth.</p><p>In contrast to our hypothesis, we observed a reduction in fine root productivity (<1mm diameter), from 0.038 ± 0.01 mg cm<sup>2</sup> day<sup>-1 </sup>under aCO<sub>2</sub> to 0.013 ± 0.004 mg cm<sup>2 </sup>day<sup>-1</sup> after 12 months of eCO<sub>2.</sub> On the other hand, the fine root biomass stock (<2mm diameter) increased at 5-10 cm from 0.86 ± 0.18 at aCO<sub>2</sub> to 1.74 ± 0.65 mg<sup>-1</sup> cm<sup>2</sup> with eCO<sub>2</sub>, but there was no effect of eCO<sub>2</sub> on fine root biomass in the litter layer. However, roots growing in the litter layer significantly increased their SRL and showed a strong tendency of higher SRA in response to eCO<sub>2 </sub>(SRL: 4.66 ± 1.08 and 9.58 ± 2.12 cm mg<sup>-1</sup>; SRA:  0.63 ± 0.18 and 1.0 ± 0.25 cm<sup>2</sup> mg<sup>-1</sup> with aCO<sub>2</sub> and eCO<sub>2</sub>, respectively), but we did not observe changes in root morphological parameters in the soil, only a tendency towards decreasing RTD. Moreover, we found a strong trend towards an increase in potential root phosphatase activity with eCO<sub>2 </sub>in the litter by 20.0 % (aCO<sub>2</sub>: 66.16 ± 10.4; eCO<sub>2</sub>: 79.39 ± 20.8 nmol mg<sup>-1 </sup>dry root h<sup>-1</sup>) and soil by 45.61% (aCO<sub>2</sub>: 97.42 ± 30.76; eCO<sub>2</sub>:141.86 ± 34.04 nmol mg<sup>-1 </sup>dry root h<sup>-1</sup>).</p><p>Our initial results suggest that understory plants intensified the investment in fine root dynamics in litter layer as response to eCO<sub>2</sub> (e.g., increase in SRL and potential root phosphatase activity) Furthermore, with a potential increase in root phosphatases exudation (litter and soil) in the first year with eCO<sub>2</sub>, our results reinforce the importance of this mechanism to mobilize inorganic P. Our results provide an initial understanding of nutrient mechanisms acquisition under eCO<sub>2</sub> in a tropical forest, which can be incorporated into ecosystem models to allow more reliable predictions of forest productivity under eCO<sub>2</sub>.</p>