Decoupling of nitrogen, phosphorus, and carbon release from fine and coarse roots during 7 years of decomposition

Belowground litter decomposition represents an important source of the limiting nutrients nitrogen (N) and phosphorus (P) to forest soils, but roots also immobilize these nutrients during the decomposition process.

Despite clear implications for soil fertility, the rates and drivers of nutrient immobilization and release (as the percent of increase and decrease of the initial pool) from root litter remain poorly understood, especially in coarse roots (> 2 mm diameter).

To address this gap, we conducted a seven-year field decomposition experiment using roots from three species, across five diameter classes (< 1 mm, 1–2 mm, 2–5 mm, 5–10 mm, and 10–20 mm) in a temperate forest.

Nitrogen dynamics were largely decoupled with P and carbon (C) over the course of the experiment, and both varied by species and root diameter. Roots released P to the surrounding soil within the first year of decomposition. In contrast, roots immobilized N for much longer, with the coarsest roots remaining a net N sink after 7 years. Long-term N release was jointly controlled by initial nutrient and C quality, whereas P release and decomposition rate were better predicted by initial C quality. Initial root nutrients well predicted the difference between long-term N versus P release.

Synthesis. Our results highlight the fact that N and P dynamics should be considered separately when modelling nutrient release during root decomposition and suggest that the functional diversity of belowground biomass may have considerable afterlife effects on the relative availability of N and P in soil. We conclude that root litter, especially coarse root litter, represents an underappreciated N sink in forest soils.

Methods
Root material for incubation from three to five mature individuals of each species was excavated in late May 2012 from the 0–20 cm soil layer, and immediately taken back to the lab for processing. The roots were cleaned with tap water, and dead roots were removed and then divided into five diameter classes: < 1 mm, 1–2 mm, 2–5 mm, 5–10 mm, 10–20 mm, and then mixed for each diameter class and each species to reduce the influence of intra-species variability of roots traits on decomposition. A sample of 2–5 g air-dried roots was placed into each 0.1 mm mesh litter-bag, and the masses were later corrected to oven-dried (65°C) mass using species- and diameter-specific correction factors. The correction factor, i.e., the ratio of the oven-dried (65 °C) mass to the air-dried mass, was determined with another root sample of 5 g for each of the five diameters for the three species. Although the litter-bag method may not accurately estimate root decomposition rate by changing the relationships among root, soil and rhizosphere decomposer community as well as environmental aspects in the short-term, this method is helpful to study the decomposition rate and its relationship to traits by tree species or diameter, especially in natural forests. There were 720 bags in total: 60 bags for each of the < 1 mm, 1–2 mm, and 2–5 mm diameter classes, and 30 bags each of the 5–10 mm and 10–20 mm diameter classes in a species. Litter-bags by species and root diameters were buried on June 3rd, 2012 to 10 cm depth underneath five replicate trees of each species, with species corresponding to the species of root contained in the litterbag. We did not incubate the roots underneath the same individual tree because that digging root introduced a large disturbance to the tree and the soil around the tree. After burying the bags, the litter layer was restored as much as possible to minimize disturbance.

Litterbags were retrieved on eight occasions during the first two years of incubation to effectively capture dynamics during the early stages of decomposition. To characterize later stage dynamics, we retrieved one set of litterbags each in the 3rd, 5th and 7th years of incubation. This resulted in a total of 11 litterbag collections, representing incubation times of 1, 2, 3, 4, 5, 11, 13, 15, 27, 49, and 87 months. In total, 666 of 720 litter bags were retrieved, and the rest were lost or destroyed by wild boar (Sus scrofa L.). Upon removal from the field, all roots were cleaned, oven-dried it at 65 °C, weighed, and analysed for C, N and P concentrations.

Additional root samples of each species and diameter class were collected on May 29th, 2021 to measure the initial root quality (the same as the tree individuals for root material incubation). We characterized each sample for concentrations of C, N, P, Ca, Mg, Si, soluble sugars, starch, hemicellulose, cellulose, lignin, and total phenolics. Additional root samples of each species and diameter class were collected on May 29th, 2021 to measure the initial root quality (the same as the tree individuals for root material incubation). We characterized each sample for concentrations of C, N, P, Ca, Mg, Si, soluble sugars, starch, hemicellulose, cellulose, lignin, and total phenolics.