The ecological stoichiometry theory provides a framework to understand organism fitness and population dynamics based on the stoichiometric mismatch between organisms and their resources. Recent studies have revealed that different soil animals occupy distinct multidimensional stoichiometric niches (MSNs), which likely determine their specific stoichiometric mismatches and population responses facing resource changes. The goals of the present study are to examine how long-term forest plantations affect the multidimensional elemental contents of litter and detritivores and the population size of detritivores that occupy distinct MSNs.

We evaluated the contents of ten elements of two detritivore taxa (lumbricid earthworms and julid millipedes) and their litter resources, quantified their MSNs and the multidimensional stoichiometric mismatches, and examined how such mismatch patterns influence the density and total biomass of detritivores across three forest types spanning from natural forests (oak forest) to plantations (pine and larch forests).

Sixty-year pine plantations changed the multidimensional elemental contents of litter but did not influence the elemental contents of the two detritivore taxa. Earthworms and millipedes exhibited distinct patterns of MSNs and stoichiometric mismatches, but they both experienced severer stoichiometric mismatches in pine plantations than in oak forests and larch plantations. Such stoichiometric mismatches led to lower density and biomass of both earthworms and millipedes in pine plantations. In other words, under conditions of low litter quality and severe stoichiometric mismatches in pine plantations, detritivores maintained their body elemental contents but decreased their population biomass.

Our study illustrates the success of using the multidimensional stoichiometric framework to understand the impact of forest plantations on animal population dynamics, which may serve as a useful tool in addressing ecosystem responses to global environmental changes.

On the Dongling mountain of Beijing city, China, we established five sampling plots (50 m × 50 m) in different valleys for each of three forest types: oak, larch, and pine, dominated by Quercus liaotungensis Koidz., Larix principis rupprechtii Mayr., and Pinus tabulaeformis Carr., respectively. To measure the total dry biomass and density of earthworms and millipedes, in each plot we collected three subsamples (30 cm in diameter) comprising litter (L layer), organic matter (F and H layer), and top 10 cm soil (Ah layer).  These three subsamples were pooled into one aggregate sample for each plot (n = 15 in total) and taken to the laboratory to extract earthworms and millipedes by sieving and hand-picking. Then, the samples were heated for 24 h in a Tullgren apparatus equipped with a 40 W bulb to collect smaller earthworms and millipedes.

Animals were freeze-dried for 72 h and individual body mass (dry weight) were measured using an electronic microbalance (MYA 5.4Y, RADWAG Wagi Elektroniczne, Poland) to the nearest microgram (±0.1µg). Adult individuals of earthworms and millipedes (body mass higher than 5 mg and 2 mg, respectively) were selected for elemental analysis. Specimens larger than 10 mg were ground individually to a homogenous powder using a bead mill homogenizer (Bead Ruptor 12, Omni International, USA).

Approximately 2 mg powder of each ground sample was used to quantify C and N contents using a vario EL cube CHNOS Elemental Analyzer (Elementar Analysensysteme GmbH, Germany). Another ~2 mg powder of the ground sample was digested with nitric acid (2 ml and 70% by weight) for quantifying the contents of Ca, Cu, K, Mg, Mn, Na, P, and Zn using an iCAP 6301 ICP-OES Spectrometer (Thermo Fisher, USA). For specimens smaller than 10 mg, we used their whole body to quantify C/N or the other eight elements, except for six large individuals of earthworms and eight large individuals of millipedes, which we cut into two halves prior to quantifying C/N and the other elements. From each plot, we further took another five leaf litter subsamples (0.5 m × 0.5 m, in total n = 75) for litter biomass and elemental content analysis. We also measured the ten elemental contents of the five litter samples from each plot.  For each litter sample, after freeze-dried for 72 h, we ground approximately 5 g leaf litters (including microbes living on them) and then used ~ 8 mg and ~ 50 mg powders to measure C and N and the other eight elements, respectively.