Species selection
The species were selected in the wetlands and grasslands in Nanshuihu National Wetland Park (N24o47’16’’, E113o120’23’’), Guangdong Province, China, based on a field investigation of the dominant alien invasive and native species in 2016. We chose nine alien clonal plant species (i.e., Hydrocotyle verticillata, Alternanthera philoxeroides, Sphagneticola trilobata, Erigeron annuus, Trifolium repens, Eleusine indica, Paspalum dilatatum, Ambrosia artemisiifolia, Amaranthus retroflexus) as the target species, which were dominant in the investigated habitats and co-occurred in subtropical and tropical wetlands or grassland habitats in China (Ma 2014; 2020; Ma and Li 2018). We chose six plant species commonly found and dominant in the wetlands and grasslands that were investigated to construct native communities (Appendix S1: Table S1). To balance the functional diversity in communities, the six native species belonged to the different functional groups (Araliaceae, Amaranthaceae, Compositae, Lamiaceae, Oxalidaceae, Rosaceae). All the selected 15 herbaceous species co-occur in the field (Table S1).
The species used in the experiment had been collected from field sites in Guangdong Province in China. We collected ramets for alien clonal species and seeds for others alien species and all native species (Table S1). The collected ramets were cultivated in a greenhouse at Huazhong Agricultural University (Wuhan, China) to produce enough new clonal fragments for the experiment. The seeds were germinated in potting soil in the greenhouse to produce enough seedlings. Because the time required for germination were known to differ among the species, we sowed them on different dates to ensure that all species were in similar developmental stages at the start of the experiment. We placed all trays with seeds in a greenhouse under natural light conditions, with a temperature between 20 and 28°C. Each ramet and seedling had three leaves and some roots.
Experimental set-up
The experiments were carried out in a greenhouse of Huazhong Agricultural University, Wuhan, China. The greenhouse was open on the sides to allow entrance of insect herbivores and pollinators. However, a fence prevented access to larger animals (e.g., birds and mammals). For each of nine target species, we transplanted one ramet or seedling of an alien target species in the center of each pot (24 cm long × 24 cm wide × 18 cm high). The six native plant species constitute native community planted uniformly surround the alien target species with a hexagonal design. We planted selected 288 square pots filled with a 1:1 mixture of sand (0-0.5 mm) and yellow-brown soil collected from Shizishan Mountain in Wuhan, Hubei Province, China.
The nutrient content of the mixed soil was also determined in this experiment. As the nutrient addition treatment in this experiment was divided into low and high nutrient additions, we needed to ensure that the nutrient content in the pots without nutrient addition treatment was at a lower level. The main method of nutrient addition in this experiment was to use slow-release fertilizer, which mainly contains N, P, and K elements. Based on this, the total N, total P, and total K of the mixed soil were mainly measured. The total N content was 0.25 ± 0.03 g/kg, the total P content was 0.36 ± 0.04 g/kg, and the total K content was 18.24 ± 1.22 g/kg (mean ± SE, n = 10). To test the interactive effects of nutrient availability and enemy suppression on alien plant invasion into resident native communities, we assigned the 32 pots of each alien target species to two levels of nutrient availability (without vs with nutrient) treatments fully crossed with two levels of enemy suppression (without vs with enemy suppression). In other words, per alien species, we had eight pots (i.e., replicates), in each of the four treatment combinations. To create different nutrient availability treatments, the soil was evenly mixed evenly with 6 g water-soluble fertilizer (20% N, 20% P2O5, 20% K2O, g⁄g, Peters Professional, Scotts, Geldermalsen, The Netherlands) as an added nutrient treatment (with nutrient). For the treatment with natural enemies suppression, we sprayed the aboveground parts of the plants with a broad-spectrum insecticide (concentration: 2 ml L− 1 ; main ingredients: chlorpyrifos and fenvalerate; The Dow Chemical Company, Midland, USA) and also added to the soil a solution consisting of a mixture of the same broad-spectrum insecticide (concentration: 2 ml L− 1 ) and two broad-spectrum fungicides with the main ingredients benzimidazole (1.5 g L− 1 ; Nufarm Limited, Contatti, Italy) and copper oxychloride (1.5 g L− 1 ; Dupont agricultural Products, Washington DE, United States) every other week (Wang et al. 2019; Zheng et al. 2015). For the control treatment without enemy suppression, we sprayed the plants with water and added corresponding amounts of water to the soil.
During the experiments, we watered all plants to meet the demand of plant growth. And the pots were randomly positioned and reshuffled every 12 days to avoid effects of possible environment differences. The experiment was conducted from August to October 2015 (70 days). Our experiment lasted for 10 weeks in a greenhouse under natural sunlight. During the experiment, the mean temperature in the greenhouse was set to 27.7°C and the relative humidity to 70.5% (measured by Amprobe TR300, Amprobe, Everett, WA, USA). Light intensity in the greenhouse was 70% of that outside.
Plant harvest and measurements
At the end of experiment, plants were harvested and separated into above-ground and below-ground parts of target species and native community in each pot. All biomass samples were dried at 65°C for 72 hours and then weighed. Moreover, we used biomass proportion as an indicator of plant growth of alien species in the whole community, and it was calculated from the proportion of total mass of the alien target plant to total biomass of the whole community (alien target species plus sum of biomass of the three native species). In addition, we calculated evenness of native community using the Shannon evenness index: J′=H′/ln(S), where H′ is the Shannon–Wiener diversity index: H′ = −Σpiln(pi), where pi is the proportional biomass of each species and S is the number of species in the community (Kardol et al. 2010).
Statistical analyses
To test for differences in biomass production between target alien species and native community in response to nutrient addition and enemy suppression, we fitted a linear mixed-effects model in R 4.1.1 (R Core Team and Team 2021) using the lme function in the nlme package (Pinheiro et al. 2017; Pinheiro et al. 2020). To improve normality and homoscedasticity of the residuals, above-ground biomass production, biomass production and root/shoot ratio of alien target species and the native communities were log-transformed, biomass proportion of the alien target species and native communities in each pot was logit-transformed, and evenness of native community was asin- transformed prior to analyses. The fixed part of the model included nutrient addition (with nutrient vs. without nutrient), enemy (with enemy suppression vs. (without enemy suppression) and all their interactions. To account for non-independence of plants from the same species, and non-independence of species from the same genus, part of models included family, species (nested within genus) as random effects.