- Split View
-
Views
-
Cite
Cite
Zhiyuan Hu, Jiaqi Zhang, Yizhou Du, Kangwei Shi, Guangqian Ren, Babar Iqbal, Zhicong Dai, Jian Li, Guanlin Li, Daolin Du, Substrate availability regulates the suppressive effects of Canada goldenrod invasion on soil respiration, Journal of Plant Ecology, Volume 15, Issue 3, June 2022, Pages 509–523, https://doi.org/10.1093/jpe/rtab073
- Share Icon Share
Abstract
Invasive alien plants not only decrease riparian vegetation diversity but also alter wetland ecosystem carbon processes, especially when they displace the original vegetation. Invasive Canada goldenrod (Solidago canadensis L.) has colonized large areas of disturbed and undisturbed land in southeastern China, yet little is known regarding how it affects soil carbon cycling. To explore the response patterns of soil respiration following S. canadensis invasion and their driving mechanisms, an observational field study and a greenhouse experiment simulating invasion were performed. In the field study, soil respiration was measured weekly from 21th July 2018 to 15th December 2018. In the greenhouse experiment, soil, autotrophic and heterotrophic respiration were measured every 1st and 15th of the month from 15th July 2019 to 15th December 2019. Soil, autotrophic and heterotrophic respiration were measured using a closed-chamber system with the deep gauze collar root exclusion method. Solidago canadensis invasion appeared to decrease the total soil CO2 emissions in both the field study and the greenhouse experiment. The suppressive effects on soil respiration may be attributed to S. canadensis invasion-induced alterations in the quality and quantity of available soil substrate, suggesting that S. canadensis invasion may impact soil carbon cycling via plant-released substrates and by competing for the soil available substrate with native plant and/or soil microbes. These results have substantial implications for estimations of the effects of invasive plants on belowground carbon dynamics and their contribution to the warming world.
摘要
外来植物入侵不仅会降低河边近岸湿地生态系统植被多样性,而且会改变湿地生态系统的地下碳过程。外来入侵植物加拿大一枝黄花(Solidago canadensis L.)已广泛入侵我国东南部地区,但加拿大一枝黄花入侵对入侵地生态系统地下土壤碳循环过程的影响却知之甚少。本研究通过野外原位观测实验和温室模拟入侵实验,探究外来植物加拿大一枝黄花入侵对入侵地土壤呼吸的影响规律及其驱动因素。野 外原位观测实验开展于2018年7月21日至12月15日,期间每周测定样地土壤呼吸。温室模拟入侵实验开展于2019年7月15日至12月15日,期间每月1日与15日上午测定土壤呼吸、自养呼吸和异养呼吸。土壤呼吸、自养呼吸和异养呼吸通过静态箱结合深埋根系隔离法测定。野外原位观测实验和温室模拟入侵实验结果均显示,加拿大一枝黄花的入侵降低了土壤二氧化碳的排放通量。加拿大一枝黄花入侵对土壤呼吸的抑制作用可能归因于其入侵引起的土壤可利用底物质量与数量的变化,表明外来入侵植物加拿大一枝黄花可通过改变植物释放基质以及与本地植物和/或土壤微生物争夺土壤有效基质而影响土壤碳循环。这些研究结果对于评估外来入侵植物对入侵地地下碳动态的影响以及对全球变暖的贡献具有重要意义。
INTRODUCTION
Globally, soils contain approximately 2-fold more carbon (C) than the atmosphere; therefore, soils play a crucial role in the accumulation and sequestering of atmospheric CO2, and in the course and rate of global warming (Johnston and Sibly 2018; Tang et al. 2020a). Soil CO2 emissions via soil respiration are the main source of C loss in the soil, exceeding all other C exchanges between the terrestrial and the atmosphere (Zhang et al. 2018). Soil respiration comprises autotrophic respiration and heterotrophic respiration. Autotrophic respiration stems from the vegetation roots and the associated microbes and mycorrhizae in the rhizosphere, while heterotrophic respiration stems from the decomposition of vegetation detritus and organic matter by microbes and fauna in the soil (Song et al. 2019; Tang et al. 2020a, 2020b). Soil respiration is a key ecosystem function that links aboveground and belowground C processes and plays an essential role in regulating global C cycling (Wang et al. 2017). Numerous studies have demonstrated that alterations to the vegetation community could have a profound impact on soil respiration by changing the type and productivity of vegetation, the quality and quantity of litterfall and the physicochemical and biological characteristics of soil (Chen et al. 2019; Fay et al. 2012; Wang et al. 2015; Yan et al. 2018; Zhang et al. 2016). Owing to their low C release rate, wetland soils are considered as a fundamental C sink, storing 12% to 20% of total terrestrial C and a major contributor to the atmospheric C pool (Chen et al. 2015). Thus, any changes in soil respiration in a wetland ecosystem due to the succession of vegetation induced by alien plant invasions, can lead to significant changes in atmospheric CO2 concentration, which directly contributes to global warming (Chen et al. 2012; Lu et al. 2022; Wang et al. 2017).
Alien plant invasion, a component of global change intentionally or unintentionally induced by humans, poses a significant threat to the wetland vegetation community by dramatically shifting the biodiversity, structure and functioning of the vegetation community, and altering ecosystem element cycling processes (Chen et al. 2012; Li et al. 2022; Wang et al. 2019; Wu et al. 2019; Yang et al. 2016; Yuan et al. 2022; Zhang et al. 2009, 2020b). Carbon cycling in the soil is an ecosystem element process that is significantly impacted following alien plant invasions (Zhang et al. 2016). Alterations to the structure and succession of the vegetation induced by alien plant invasion may affect important soil processes by altering vegetation inputs and soil biotic and abiotic environments (Osborne and Gioria 2018; Wan et al 2018; Wolkovich et al. 2010). Numerous studies have reported the threats of invasive alien plants to plant communities in-depth and proposed non-mutually exclusive hypotheses (Dai et al. 2016, 2020; Zhou and Staver 2019). However, the potential effects of alien plant invasions on climate change via shifts in soil C biogeochemical processes (e.g. soil CO2 emissions) have not been well studied. The available literature demonstrates that the effects of invasive alien plants on soil CO2 emissions vary among ecosystem types, suggesting that changes in the soil C cycle induced by invasive alien plants may be ecosystem-dependent (Chen et al. 2015; Duval et al. 2020; Hsieh et al. 2016; Jones et al. 2019; Potts et al. 2008; Yang et al. 2016; Zhang et al. 2018). Nevertheless, the response of soil respiration to alien plant invasions is poorly understood, limiting understanding of the mechanistic framework underlying soil CO2 emission responses to invasion and posing challenges to accurate prediction of the C cycle in soil and mitigation of its contribution to global warming.
Canada goldenrod (Solidago canadensis L.), which is native to North America, is a rhizomatous herbaceous perennial goldenrod weed that has become one of the most rapidly proliferating invasive alien plants in southeastern China, in regions such as the original riparian habitats of common reed (Phragmites australis (Cav.) Trin. ex Steud) (Zhang and Wan 2017; Yuan et al. 2022). Solidago canadensis invasion threatens native ecosystems, encroaches on the habitats of native plant communities and negatively affects the ecological functions, structure and biodiversity of the invaded ecosystems. Solidago canadensis has many traits, such as their excellent physiological adaptation ability, fast growth rate, high nutrient uptake ability and the complexity components present in their root exudation and litter. These traits are implicated as major factors impacting the soil of invaded ecosystems, as they grant species the ability to directly and/or indirectly modify the abiotic soil environment (e.g. physicochemical properties, resource availability and stoichiometric equilibrium) and biotic soil communities (e.g. microbial communities, nematode communities and food webs) (Ren et al. 2020, 2021; Wang et al. 2018; Wei et al. 2020; Wu et al. 2019; Zhang et al. 2009). However, the mechanistic framework of the invasive effect of S. canadensis on the C flux in the soil remains unclear. Hence, it is necessary to identify how soil respiration varies during the succession of S. canadensis in the invaded ecosystem to understand soil respiration responses to alien S. canadensis invasion and their effects on soil C biogeochemical processes.
Here, both an observational field study and a greenhouse experiment simulating invasion were performed to explore the response pattern of soil respiration following S. canadensis invasion and the drivers of such alteration processes. Like other invasive alien plant species, S. canadensis invasion can provide more C inputs into the soil such as litterfall, dead roots and root exudates, and further stimulate C processes in the soil. Thus, the present study is based on two hypotheses: (i) S. canadensis invasion will accelerate the total soil respiration as S. canadensis invasion progresses and (ii) S. canadensis invasion will have different effects on autotrophic and heterotrophic respiration during the S. canadensis invasion process.
MATERIALS AND METHODS
Study design
In the field study, the study site was located on a shoreside of a tributary of the Yangtze River (32°14′ N, 119°29′ E) near Zhenjiang City, China. In 2018, the annual air temperature was 17.1 °C and the natural precipitation was 1272.1 mm (Zhenjiang Meteorological Administration, 2019). The study field was characterized by low plant diversity. The original dominant plant was P. australis; however, several of the P. australis-dominated sites were threatened by the invasion of S. canadensis, which occurred in small patches to the east of the study area. Within the study field, three similar and isolated transect lines were selected. According to the different invasion levels (relative density) of S. canadensis, each selected transect line was divided into three study sites. These three study sites included: a low invasion level dominated by P. australis (coverage: 80.22% ± 5.25%), which displayed no or only slight signs of S. canadensis invasion; a moderate invasion level co-dominated by P. australis (coverage: 48.89% ± 4.68%) and S. canadensis (coverage: 40.18% ± 3.70%); and a high invasion level dominated by S. canadensis (coverage: 97.21% ± 0.44%). At each study site, one plot (1 m × 1 m) with over 80% coverage by S. canadensis, P. australis or both was selected. The pH and cation exchange capacity of the soil in the study field were 8.26 ± 0.03 and 10.38 ± 0.05 cmolc kg−1 (means ± standard errors), respectively.
The greenhouse experiment was conducted at a greenhouse on the campus of Jiangsu University (32°12′ N, 119°30′ E) in Zhenjiang City. In 2019, the annual air temperature was 17.1 °C and the natural precipitation was 736.6 mm (Zhenjiang Meteorological Administration, 2020). Seeds of the S. canadensis and P. australis were collected from the aforementioned study field in December 2018. To avoid the S. canadensis invasion effects on the physical, chemical and biological characteristics of soil, the soils were obtained from a natural green field with no S. canadensis invasion on the campus of Jiangsu University. The pH and cation exchange capacity of the soils were 7.52 ± 0.02 and 9.63 ± 0.00 cmolc kg−1 (means ± standard errors), respectively. Soils were sieved and moved into plastic pots (19.5 cm bottom diameter, 24.5 cm top diameter and 26.5 cm height, containing 2.7 kg of air-dried soil each). The S. canadensis and P. australis were cultivated in the soil in April 2019. After two months of cultivation, S. canadensis and P. australis seedlings of similar sizes were carefully transplanted into pots.
To explore the invasion effects of S. canadensis on soil respiration, autotrophic respiration and heterotrophic respiration, different ratios of S. canadensis to P. australis were used as a proxy to represent successive stages of an invasion, using the method of substitution of space for time to simulate the natural S. canadensis invasion process (Zhang et al. 2020a). The S. canadensis invasion treatment was set up at six levels with a total of four seedlings planted in each pot: (i) negative control (bare soil) without plants, (ii) non-invasive stage (0% invasive relative density) with four P. australis seedlings, (iii) early invasive stage (25% invasive relative density) with three P. australis seedlings and one S. canadensis seedling, (iv) intermediate invasive stage (50% invasive relative density) with two P. australis seedlings and two S. canadensis seedlings, (v) dominant invasive stage (75% invasive relative density) with one P. australis seedling and three S. canadensis seedlings and (vi) completely invasive stage (100% invasive relative density) with four S. canadensis seedlings. Treatments were set up in triplicate, placed in the greenhouse under natural light and watered every two days.
Soil, autotrophic and heterotrophic respiration measurements
In the field study, soil respiration was measured in the morning between 08:00 and 10:00 every Saturday from 21th July 2018 to 15th December 2018, using a closed-chamber system with a portable diffusion-type, non-dispersive infrared CO2 sensor (GMP222, Vaisala CARBOCAP, Helsinki, Finland), a handheld controller and logger (MI-70, Vaisala CARBOCAP, Helsinki, Finland) and a polyacrylic chamber (4.4 cm in diameter, 15 cm in height). The soil respiration measurement procedure has been described in detail in previous studies (Li et al. 2017; Noh et al. 2016). CO2 concentrations in the closed chamber were recorded every 5 s for 300 s. The air temperature in the polyacrylic chamber was measured using an air temperature sensor (LKD-597, Lvkedu Technology Co. Ltd, Beijing, China), while the soil temperature and moisture at the 0–10 cm depth were measured using a sensor (TR-6D, Shunkeda Technology Co. Ltd, Beijing, China). Soil respiration was calculated as follows:
where P is the atmospheric pressure, V is the volume of the headspace gas within the closed chamber, A is the chamber base area, R is the gas constant and TAir is the measured air temperature.
The temperature sensitivity of soil respiration (Q10), which is the multiplier by which the soil respiration increases for each 10 °C increase in temperature, was calculated using the following equations (Lloyd and Taylor 1994):
where α and β are regression coefficients and TSoil is the measured soil temperature.
In the greenhouse experiment, the root exclusion method, which inserts deep gauze collars (5.0 cm in diameter, 30 cm in height) in the middle of each pot, was used to partition soil respiration into autotrophic respiration and heterotrophic respiration. This was calculated as the following equation (Suseela et al. 2012):
After two months of cultivation under the treatments, soil respiration and heterotrophic respiration were measured in the morning between 08:00 and 10:00 every 1st and 15th of the month from 15th July 2019 to 15th December 2019 using the aforementioned closed-chamber system outside and inside of deep gauze collars inserted in each pot, respectively. The measurement procedure of soil respiration, heterotrophic respiration and air and soil temperature measurements were performed following the description above.
Vegetation root harvest and measurement
Roots of the vegetation in both the field study and the greenhouse experiment were harvested on 15th December 2018 and 2019, respectively. The root samples were washed, then weighed after oven-drying at 65 °C for 72 h. In the field study, specific root length was calculated using the ratio of root length to dry mass of roots. In the greenhouse experiment, the total root biomass was calculated by summing the dry mass of all roots in each pot.
Soil sample collection and preparation
In the field study, composite soil samples were produced by mixing topsoil samples (0–10 cm depth) collected from ten points in each plot using a soil corer (2.4 cm diameter) on 15th December 2018. In the greenhouse experiment, the same soil sampling procedure using a soil corer (2.0 cm diameter) was performed on the negative control, non-invasion stage, intermediate invasion stage and complete invasion stage treatment pots on 15th December 2019. Composite soil samples were passed through a 2 mm sieve to remove visible plant debris and stones before subdividing the samples for analyses. Each composite sample was divided into two portions, one of which was air-dried for soil physicochemical property analyses, and the other portion was stored at 4 °C until soil microbial extracellular enzymatic activity and microbial biomass analyses were performed. Soil microbial extracellular enzymatic activity analysis was performed within 48 h after sampling, while soil microbial biomass and physicochemical property analyses were performed within two weeks of sample collection.
Soil property measurements
Soil dissolved organic C (DOC) and nitrogen (N, DON) contents were measured following the methods of Li et al. (2020). Soil total C (STC) and N (STN) contents were quantified using an elemental analyzer (vario MACRO; Elementar Analysensysteme GmbH, Langensebold, Germany). Soil total phosphorus (P, STP) and available P (SAP) content were measured following the molybdate colorimetric method (Murphy and Riley 1962; Olsen and Sommers 1982).
Soil microbial biomass C (MBC) and N (MBN) contents were measured following the chloroform fumigation extraction method (Brookes et al. 1985; Vance et al. 1987). A total of six C-, N- and P-acquisition microbial extracellular enzymatic activities were measured following DeForest (2009), including three C-acquisition enzymes (β-1,4-glucosidase (BG), β-D-1,4-cellobiohydrolase (CBH) and β-1,4-xylosidase (BX)), two N-acquisition enzymes (β-1,4-N-acetylglucosaminidase (NAG) and L-leucine aminopeptidase (LAP)) and one P-acquisition enzyme (alkaline phosphatase (AP)). The detailed soil microbial extracellular enzymatic activity measurement procedures have been described previously (Li et al. 2018). All enzyme reactions were incubated in the dark at 25 °C for 4 h, except LAP, which was incubated for 5 h, before measurement at 355 nm excitation wavelength and 460 nm emission wavelength using a Multimode Microplate Reader (infinite M1000PRO; Tecan, Männedorf, Switzerland).
The soil microbial nutrient limitation and nutrient status were assessed based on the vector length (VL), vector angle (VA) and threshold elemental ratio (TER) of C-, N- and P-acquisition microbial extracellular enzymatic activities. Vector length represents soil microbial C limitation, with longer vector length indicating stronger microbial C limitation and vector angle represents soil microbial N and P limitation, with a vector angle larger or smaller than 45° indicating microbial P limitation and N limitation, respectively (Moorhead et al. 2013). Threshold elemental ratio represents the elemental ratio at which metabolic control of microbial metabolic limitation switches between C limitation and nutrient limitation (Sinsabaugh et al. 2009). Vector length, vector angle and threshold elemental ratio were calculated as the following equations:
where MBC:N is the ratio of MBC to MBN, and n0 is the dimensionless normalization constant representing the intercept of ln(AG + BG + BX) vs. ln(NAG + LAP).
Statistical analyses
For the field study, a one-way analysis of variance (ANOVA) followed by Fisher's least significant difference test was performed to evaluate the effects of S. canadensis invasion on soil respiration, Q10 and soil properties. Spearman correlation tests were performed to determine the relationships between soil respiration and Q10 with soil properties.
For the greenhouse experiment, ANOVA followed by Fisher's least significant difference test was performed to evaluate the effect of S. canadensis invasion on soil respiration, autotrophic respiration, heterotrophic respiration, total root biomass and soil properties. Spearman correlation tests were performed to determine the relationships between soil respiration, autotrophic respiration and heterotrophic respiration with total root biomass and soil properties. Partial least squares path modeling (PLS-PM) was performed to evaluate possible pathways by which variables affect soil respiration, autotrophic respiration and heterotrophic respiration among treatments.
ANOVA and Spearman correlation test were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA), and PLS-PM was performed using Amos in IBM SPSS version 24.0 (SPSS Inc., Chicago, IL, USA).
RESULTS
Soil properties and total root biomass
In the field study, soil properties differed among study sites. Soil dissolved organic C was 9.75% and 16.11% higher at the high S. canadensis invasion level site than at the moderate and low S. canadensis invasion level sites, respectively (P < 0.01). Solidago canadensis invasion induced significant changes in soil available P, with the highest value observed at the moderate invasion site and the lowest value observed at the low invasion site (P < 0.05). Compared to the low invasion site, the high and moderate invasion sites displayed changes in soil total C of −10.00% and 13.40%, and in soil total N of −7.75% and 16.26%, respectively. Soil total P was significantly increased by 22.55% and 3.36% at high and moderate invasion sites, respectively (P < 0.01), resulting in changes in soil resource ratios of soil total C to total P of −5.90% and −9.71%, and in soil total N to total P by −3.56% and 31.82% at high and moderate invasion sites, respectively, compared to the low invasion site (Table 1). The highest specific root length was observed at the low invasion site, and the lowest specific root length was observed at the moderate invasion site (Supplementary Fig. S1a).
Properties . | F . | P . | Study sites . | . | . |
---|---|---|---|---|---|
. | . | . | LI . | MI . | HI . |
DOC (×10–1 mg C g−1 soil) | 36.00 | ** | 1.95 ± 0.01c | 2.09 ± 0.05b | 2.32 ± 0.02a |
SAP (×10–2 mg P g−1 soil) | 7.98 | * | 1.78 ± 0.12b | 2.87 ± 0.26a | 2.29 ± 0.17ab |
STC (mg C g−1 soil) | 0.57 | ns | 12.47 ± 1.19 | 13.71 ± 3.05 | 10.80 ± 0.74 |
STN (mg N g−1 soil) | 0.76 | ns | 1.03 ± 0.03 | 1.12 ± 0.24 | 0.87 ± 0.06 |
STP (×10–1 mg P g−1 soil) | 7.9 | * | 6.70 ± 0.21b | 6.92 ± 0.31b | 8.21 ± 0.33a |
STC:N | 0.25 | ns | 11.96 ± 0.74 | 12.23 ± 0.26 | 12.41 ± 0.16 |
STC:P | 2.18 | ns | 18.55 ± 1.31 | 19.64 ± 3.87 | 13.11 ± 0.36 |
STN:P | 2.79 | ns | 1.55 ± 0.03 | 1.61 ± 0.31 | 1.06 ± 0.03 |
MBC (×10–1 mg C g−1 soil) | 0.32 | ns | 8.02 ± 4.00 | 4.43 ± 2.39 | 5.74 ± 3.11 |
MBN (×10–2 mg N g−1 soil) | 1.77 | ns | 2.21 ± 0.73 | 1.07 ± 0.55 | 2.67 ± 0.54 |
MBC:N | 0.89 | ns | 32.63 ± 14.24 | 40.66 ± 4.60 | 20.96 ± 10.40 |
Properties . | F . | P . | Study sites . | . | . |
---|---|---|---|---|---|
. | . | . | LI . | MI . | HI . |
DOC (×10–1 mg C g−1 soil) | 36.00 | ** | 1.95 ± 0.01c | 2.09 ± 0.05b | 2.32 ± 0.02a |
SAP (×10–2 mg P g−1 soil) | 7.98 | * | 1.78 ± 0.12b | 2.87 ± 0.26a | 2.29 ± 0.17ab |
STC (mg C g−1 soil) | 0.57 | ns | 12.47 ± 1.19 | 13.71 ± 3.05 | 10.80 ± 0.74 |
STN (mg N g−1 soil) | 0.76 | ns | 1.03 ± 0.03 | 1.12 ± 0.24 | 0.87 ± 0.06 |
STP (×10–1 mg P g−1 soil) | 7.9 | * | 6.70 ± 0.21b | 6.92 ± 0.31b | 8.21 ± 0.33a |
STC:N | 0.25 | ns | 11.96 ± 0.74 | 12.23 ± 0.26 | 12.41 ± 0.16 |
STC:P | 2.18 | ns | 18.55 ± 1.31 | 19.64 ± 3.87 | 13.11 ± 0.36 |
STN:P | 2.79 | ns | 1.55 ± 0.03 | 1.61 ± 0.31 | 1.06 ± 0.03 |
MBC (×10–1 mg C g−1 soil) | 0.32 | ns | 8.02 ± 4.00 | 4.43 ± 2.39 | 5.74 ± 3.11 |
MBN (×10–2 mg N g−1 soil) | 1.77 | ns | 2.21 ± 0.73 | 1.07 ± 0.55 | 2.67 ± 0.54 |
MBC:N | 0.89 | ns | 32.63 ± 14.24 | 40.66 ± 4.60 | 20.96 ± 10.40 |
Abbreviations: LI = low invasion level, MI = moderate invasion level, HI = high invasion level, DOC = soil dissolved organic carbon (C), SAP = soil available phosphorus (P), STC = soil total C, STN = soil total nitrogen (N), STP = soil total P, STC:N = ratio of STC to STN, STC:P = ratio of STC to STP, STN:P = ratio of STN to STP, MBC = soil microbial biomass C, MBN = soil microbial biomass N, MBC:N = ratio of MBC to MBN. Different letters denote significant differences (P < 0.05) among study sites. Different letters denote significant differences (P < 0.05) among study sites. ns = not significant at the level of P > 0.05; *significant at the level of P < 0.05; and **significant at the level of P < 0.01.
Properties . | F . | P . | Study sites . | . | . |
---|---|---|---|---|---|
. | . | . | LI . | MI . | HI . |
DOC (×10–1 mg C g−1 soil) | 36.00 | ** | 1.95 ± 0.01c | 2.09 ± 0.05b | 2.32 ± 0.02a |
SAP (×10–2 mg P g−1 soil) | 7.98 | * | 1.78 ± 0.12b | 2.87 ± 0.26a | 2.29 ± 0.17ab |
STC (mg C g−1 soil) | 0.57 | ns | 12.47 ± 1.19 | 13.71 ± 3.05 | 10.80 ± 0.74 |
STN (mg N g−1 soil) | 0.76 | ns | 1.03 ± 0.03 | 1.12 ± 0.24 | 0.87 ± 0.06 |
STP (×10–1 mg P g−1 soil) | 7.9 | * | 6.70 ± 0.21b | 6.92 ± 0.31b | 8.21 ± 0.33a |
STC:N | 0.25 | ns | 11.96 ± 0.74 | 12.23 ± 0.26 | 12.41 ± 0.16 |
STC:P | 2.18 | ns | 18.55 ± 1.31 | 19.64 ± 3.87 | 13.11 ± 0.36 |
STN:P | 2.79 | ns | 1.55 ± 0.03 | 1.61 ± 0.31 | 1.06 ± 0.03 |
MBC (×10–1 mg C g−1 soil) | 0.32 | ns | 8.02 ± 4.00 | 4.43 ± 2.39 | 5.74 ± 3.11 |
MBN (×10–2 mg N g−1 soil) | 1.77 | ns | 2.21 ± 0.73 | 1.07 ± 0.55 | 2.67 ± 0.54 |
MBC:N | 0.89 | ns | 32.63 ± 14.24 | 40.66 ± 4.60 | 20.96 ± 10.40 |
Properties . | F . | P . | Study sites . | . | . |
---|---|---|---|---|---|
. | . | . | LI . | MI . | HI . |
DOC (×10–1 mg C g−1 soil) | 36.00 | ** | 1.95 ± 0.01c | 2.09 ± 0.05b | 2.32 ± 0.02a |
SAP (×10–2 mg P g−1 soil) | 7.98 | * | 1.78 ± 0.12b | 2.87 ± 0.26a | 2.29 ± 0.17ab |
STC (mg C g−1 soil) | 0.57 | ns | 12.47 ± 1.19 | 13.71 ± 3.05 | 10.80 ± 0.74 |
STN (mg N g−1 soil) | 0.76 | ns | 1.03 ± 0.03 | 1.12 ± 0.24 | 0.87 ± 0.06 |
STP (×10–1 mg P g−1 soil) | 7.9 | * | 6.70 ± 0.21b | 6.92 ± 0.31b | 8.21 ± 0.33a |
STC:N | 0.25 | ns | 11.96 ± 0.74 | 12.23 ± 0.26 | 12.41 ± 0.16 |
STC:P | 2.18 | ns | 18.55 ± 1.31 | 19.64 ± 3.87 | 13.11 ± 0.36 |
STN:P | 2.79 | ns | 1.55 ± 0.03 | 1.61 ± 0.31 | 1.06 ± 0.03 |
MBC (×10–1 mg C g−1 soil) | 0.32 | ns | 8.02 ± 4.00 | 4.43 ± 2.39 | 5.74 ± 3.11 |
MBN (×10–2 mg N g−1 soil) | 1.77 | ns | 2.21 ± 0.73 | 1.07 ± 0.55 | 2.67 ± 0.54 |
MBC:N | 0.89 | ns | 32.63 ± 14.24 | 40.66 ± 4.60 | 20.96 ± 10.40 |
Abbreviations: LI = low invasion level, MI = moderate invasion level, HI = high invasion level, DOC = soil dissolved organic carbon (C), SAP = soil available phosphorus (P), STC = soil total C, STN = soil total nitrogen (N), STP = soil total P, STC:N = ratio of STC to STN, STC:P = ratio of STC to STP, STN:P = ratio of STN to STP, MBC = soil microbial biomass C, MBN = soil microbial biomass N, MBC:N = ratio of MBC to MBN. Different letters denote significant differences (P < 0.05) among study sites. Different letters denote significant differences (P < 0.05) among study sites. ns = not significant at the level of P > 0.05; *significant at the level of P < 0.05; and **significant at the level of P < 0.01.
In the greenhouse experiment, compared to the non-invasion stage treatment, the complete invasion stage and intermediate invasion stage S. canadensis treatments displayed changes in dissolved organic C of 13.15% and −0.91%, decreases in soil dissolved organic N of 13.73% and 17.68%, and increases in the soil available P of 9.22% and 0.12%, respectively. Solidago canadensis invasion induced a decrease in soil microbial biomass properties; however, this effect was not statistically significant. Additionally, S. canadensis invasion decreased the ratio of C-acquisition to P-acquisition microbial extracellular enzymatic activity (EEAC:P, P < 0.05), and the ratio of N-acquisition to P-acquisition microbial extracellular enzymatic activity (EEAN:P, P > 0.05). The vector lengths ranged from 0.39 to 0.96. The vector length was 58.96% and 14.66% higher for the non-invasion treatment than for the intermediate and complete invasion treatments, respectively (Table 2). The highest total root biomass was observed at the non-invasion treatment, and the lowest total root biomass was observed at the intermediate invasion treatment (Supplementary Fig. S1b).
Properties . | F . | P . | Treatments . | . | . | . |
---|---|---|---|---|---|---|
. | . | . | NC . | NI . | II . | CI . |
DOC (×10–1 mg C g−1 soil) | 0.47 | ns | 1.36 ± 0.08 | 1.49 ± 0.04 | 1.48 ± 0.24 | 1.69 ± 0.31 |
DON (×10–2 mg N g−1 soil) | 3.67 | ns | 1.51 ± 0.24 | 1.19 ± 0.05 | 0.98 ± 0.05 | 1.03 ± 0.03 |
SAP (×10–2 mg P g−1 soil) | 1.81 | ns | 1.67 ± 0.07 | 1.65 ± 0.05 | 1.65 ± 0.04 | 1.80 ± 0.06 |
STC (mg C g−1 soil) | 0.54 | ns | 3.77 ± 0.17 | 3.76 ± 0.09 | 3.69 ± 0.17 | 3.56 ± 0.06 |
STN (×10–1 mg N g−1 soil) | 0.65 | ns | 1.83 ± 0.03 | 2.20 ± 0.23 | 1.88 ± 0.19 | 2.10 ± 0.18 |
STP (×10–1 mg P g−1 soil) | 0.71 | ns | 6.56 ± 0.09 | 6.69 ± 0.02 | 6.70 ± 0.15 | 6.72 ± 0.02 |
STC:N | 1.25 | ns | 21.58 ± 3.12 | 17.43 ± 1.51 | 19.89 ± 1.76 | 16.58 ± 1.34 |
STC:P | 0.62 | ns | 5.76 ± 0.31 | 5.63 ± 0.14 | 5.52 ± 0.35 | 5.30 ± 0.07 |
STN:P (×10–1) | 0.53 | ns | 2.81 ± 0.52 | 3.29 ± 0.34 | 2.81 ± 0.26 | 3.24 ± 0.26 |
MBC (mg C g−1 soil) | 0.16 | ns | 2.08 ± 031 | 2.19 ± 0.66 | 2.08 ± 0.25 | 1.81 ± 0.18 |
MBN (×10–3 mg N g−1 soil) | 1.13 | ns | 1.67 ± 0.48 | 2.64 ± 0.94 | 1.82 ± 0.23 | 1.24 ± 0.23 |
MBC:N (×103) | 0.11 | ns | 1.48 ± 0.40 | 1.59 ± 1.15 | 1.15 ± 0.10 | 1.66 ± 0.53 |
EEAC (μg h−1 g−1 soil) | 1.4 | ns | 2.84 ± 1.25 | 2.57 ± 0.56 | 1.11 ± 0.38 | 3.54 ± 0.99 |
EEAN (μg h−1 g−1 soil) | 0.76 | ns | 9.00 ± 5.10 | 5.62 ± 1.96 | 2.89 ± 0.38 | 5.91 ± 1.76 |
EEAP (μg h−1 g−1 soil) | 0.82 | ns | 8.17 ± 1.76 | 3.99 ± 0.71 | 8.69 ± 3.30 | 10.80 ± 5.03 |
EEAC:N (×10–1) | 0.22 | ns | 6.57 ± 4.71 | 6.43 ± 3.32 | 3.66 ± 0.95 | 6.12 ± 0.32 |
EEAC:P (×10–1) | 4.24 | * | 3.22 ± 0.79ab | 6.36 ± 0.45a | 1.32 ± 0.06b | 5.05 ± 1.93a |
EEAN:P | 0.83 | ns | 1.37 ± 0.90 | 1.62 ± 0.73 | 0.42 ± 0.11 | 0.80 ± 0.28 |
TER (×104) | 0.59 | ns | 1.15 ± 0.83 | 0.73 ± 0.27 | 5.47 ± 0.10 | 1.33 ± 0.34 |
VL (×10–1) | 0.81 | ns | 7.77 ± 4.40 | 9.55 ± 2.56 | 3.91 ± 0.88 | 8.15 ± 1.46 |
VA (o) | 1.03 | ns | 49.34 ± 16.70 | 39.11 ± 12.73 | 67.85 ± 5.36 | 53.74 ± 9.11 |
Properties . | F . | P . | Treatments . | . | . | . |
---|---|---|---|---|---|---|
. | . | . | NC . | NI . | II . | CI . |
DOC (×10–1 mg C g−1 soil) | 0.47 | ns | 1.36 ± 0.08 | 1.49 ± 0.04 | 1.48 ± 0.24 | 1.69 ± 0.31 |
DON (×10–2 mg N g−1 soil) | 3.67 | ns | 1.51 ± 0.24 | 1.19 ± 0.05 | 0.98 ± 0.05 | 1.03 ± 0.03 |
SAP (×10–2 mg P g−1 soil) | 1.81 | ns | 1.67 ± 0.07 | 1.65 ± 0.05 | 1.65 ± 0.04 | 1.80 ± 0.06 |
STC (mg C g−1 soil) | 0.54 | ns | 3.77 ± 0.17 | 3.76 ± 0.09 | 3.69 ± 0.17 | 3.56 ± 0.06 |
STN (×10–1 mg N g−1 soil) | 0.65 | ns | 1.83 ± 0.03 | 2.20 ± 0.23 | 1.88 ± 0.19 | 2.10 ± 0.18 |
STP (×10–1 mg P g−1 soil) | 0.71 | ns | 6.56 ± 0.09 | 6.69 ± 0.02 | 6.70 ± 0.15 | 6.72 ± 0.02 |
STC:N | 1.25 | ns | 21.58 ± 3.12 | 17.43 ± 1.51 | 19.89 ± 1.76 | 16.58 ± 1.34 |
STC:P | 0.62 | ns | 5.76 ± 0.31 | 5.63 ± 0.14 | 5.52 ± 0.35 | 5.30 ± 0.07 |
STN:P (×10–1) | 0.53 | ns | 2.81 ± 0.52 | 3.29 ± 0.34 | 2.81 ± 0.26 | 3.24 ± 0.26 |
MBC (mg C g−1 soil) | 0.16 | ns | 2.08 ± 031 | 2.19 ± 0.66 | 2.08 ± 0.25 | 1.81 ± 0.18 |
MBN (×10–3 mg N g−1 soil) | 1.13 | ns | 1.67 ± 0.48 | 2.64 ± 0.94 | 1.82 ± 0.23 | 1.24 ± 0.23 |
MBC:N (×103) | 0.11 | ns | 1.48 ± 0.40 | 1.59 ± 1.15 | 1.15 ± 0.10 | 1.66 ± 0.53 |
EEAC (μg h−1 g−1 soil) | 1.4 | ns | 2.84 ± 1.25 | 2.57 ± 0.56 | 1.11 ± 0.38 | 3.54 ± 0.99 |
EEAN (μg h−1 g−1 soil) | 0.76 | ns | 9.00 ± 5.10 | 5.62 ± 1.96 | 2.89 ± 0.38 | 5.91 ± 1.76 |
EEAP (μg h−1 g−1 soil) | 0.82 | ns | 8.17 ± 1.76 | 3.99 ± 0.71 | 8.69 ± 3.30 | 10.80 ± 5.03 |
EEAC:N (×10–1) | 0.22 | ns | 6.57 ± 4.71 | 6.43 ± 3.32 | 3.66 ± 0.95 | 6.12 ± 0.32 |
EEAC:P (×10–1) | 4.24 | * | 3.22 ± 0.79ab | 6.36 ± 0.45a | 1.32 ± 0.06b | 5.05 ± 1.93a |
EEAN:P | 0.83 | ns | 1.37 ± 0.90 | 1.62 ± 0.73 | 0.42 ± 0.11 | 0.80 ± 0.28 |
TER (×104) | 0.59 | ns | 1.15 ± 0.83 | 0.73 ± 0.27 | 5.47 ± 0.10 | 1.33 ± 0.34 |
VL (×10–1) | 0.81 | ns | 7.77 ± 4.40 | 9.55 ± 2.56 | 3.91 ± 0.88 | 8.15 ± 1.46 |
VA (o) | 1.03 | ns | 49.34 ± 16.70 | 39.11 ± 12.73 | 67.85 ± 5.36 | 53.74 ± 9.11 |
Abbreviations: NC = negative control, NI = non-invasion stage, II = intermediate invasion stage, CI = complete invasion stage, DON = soil dissolved organic N, EEAC = soil microbial extracellular C-acquisition enzyme activity, EEAN = soil microbial extracellular N-acquisition enzyme activity, EEAP = soil microbial extracellular P-acquisition enzyme activity, EEAC:N = the ratio of soil microbial extracellular C-acquisition enzymes to N-acquisition enzymes, EEAC:P = the ratio of soil microbial extracellular C-acquisition enzymes to P-acquisition enzymes, EEAN:P = the ratio of soil microbial extracellular N-acquisition enzymes to P-acquisition enzymes, TER = threshold elemental ratio, VL = vector length, VA = vector angle.
Properties . | F . | P . | Treatments . | . | . | . |
---|---|---|---|---|---|---|
. | . | . | NC . | NI . | II . | CI . |
DOC (×10–1 mg C g−1 soil) | 0.47 | ns | 1.36 ± 0.08 | 1.49 ± 0.04 | 1.48 ± 0.24 | 1.69 ± 0.31 |
DON (×10–2 mg N g−1 soil) | 3.67 | ns | 1.51 ± 0.24 | 1.19 ± 0.05 | 0.98 ± 0.05 | 1.03 ± 0.03 |
SAP (×10–2 mg P g−1 soil) | 1.81 | ns | 1.67 ± 0.07 | 1.65 ± 0.05 | 1.65 ± 0.04 | 1.80 ± 0.06 |
STC (mg C g−1 soil) | 0.54 | ns | 3.77 ± 0.17 | 3.76 ± 0.09 | 3.69 ± 0.17 | 3.56 ± 0.06 |
STN (×10–1 mg N g−1 soil) | 0.65 | ns | 1.83 ± 0.03 | 2.20 ± 0.23 | 1.88 ± 0.19 | 2.10 ± 0.18 |
STP (×10–1 mg P g−1 soil) | 0.71 | ns | 6.56 ± 0.09 | 6.69 ± 0.02 | 6.70 ± 0.15 | 6.72 ± 0.02 |
STC:N | 1.25 | ns | 21.58 ± 3.12 | 17.43 ± 1.51 | 19.89 ± 1.76 | 16.58 ± 1.34 |
STC:P | 0.62 | ns | 5.76 ± 0.31 | 5.63 ± 0.14 | 5.52 ± 0.35 | 5.30 ± 0.07 |
STN:P (×10–1) | 0.53 | ns | 2.81 ± 0.52 | 3.29 ± 0.34 | 2.81 ± 0.26 | 3.24 ± 0.26 |
MBC (mg C g−1 soil) | 0.16 | ns | 2.08 ± 031 | 2.19 ± 0.66 | 2.08 ± 0.25 | 1.81 ± 0.18 |
MBN (×10–3 mg N g−1 soil) | 1.13 | ns | 1.67 ± 0.48 | 2.64 ± 0.94 | 1.82 ± 0.23 | 1.24 ± 0.23 |
MBC:N (×103) | 0.11 | ns | 1.48 ± 0.40 | 1.59 ± 1.15 | 1.15 ± 0.10 | 1.66 ± 0.53 |
EEAC (μg h−1 g−1 soil) | 1.4 | ns | 2.84 ± 1.25 | 2.57 ± 0.56 | 1.11 ± 0.38 | 3.54 ± 0.99 |
EEAN (μg h−1 g−1 soil) | 0.76 | ns | 9.00 ± 5.10 | 5.62 ± 1.96 | 2.89 ± 0.38 | 5.91 ± 1.76 |
EEAP (μg h−1 g−1 soil) | 0.82 | ns | 8.17 ± 1.76 | 3.99 ± 0.71 | 8.69 ± 3.30 | 10.80 ± 5.03 |
EEAC:N (×10–1) | 0.22 | ns | 6.57 ± 4.71 | 6.43 ± 3.32 | 3.66 ± 0.95 | 6.12 ± 0.32 |
EEAC:P (×10–1) | 4.24 | * | 3.22 ± 0.79ab | 6.36 ± 0.45a | 1.32 ± 0.06b | 5.05 ± 1.93a |
EEAN:P | 0.83 | ns | 1.37 ± 0.90 | 1.62 ± 0.73 | 0.42 ± 0.11 | 0.80 ± 0.28 |
TER (×104) | 0.59 | ns | 1.15 ± 0.83 | 0.73 ± 0.27 | 5.47 ± 0.10 | 1.33 ± 0.34 |
VL (×10–1) | 0.81 | ns | 7.77 ± 4.40 | 9.55 ± 2.56 | 3.91 ± 0.88 | 8.15 ± 1.46 |
VA (o) | 1.03 | ns | 49.34 ± 16.70 | 39.11 ± 12.73 | 67.85 ± 5.36 | 53.74 ± 9.11 |
Properties . | F . | P . | Treatments . | . | . | . |
---|---|---|---|---|---|---|
. | . | . | NC . | NI . | II . | CI . |
DOC (×10–1 mg C g−1 soil) | 0.47 | ns | 1.36 ± 0.08 | 1.49 ± 0.04 | 1.48 ± 0.24 | 1.69 ± 0.31 |
DON (×10–2 mg N g−1 soil) | 3.67 | ns | 1.51 ± 0.24 | 1.19 ± 0.05 | 0.98 ± 0.05 | 1.03 ± 0.03 |
SAP (×10–2 mg P g−1 soil) | 1.81 | ns | 1.67 ± 0.07 | 1.65 ± 0.05 | 1.65 ± 0.04 | 1.80 ± 0.06 |
STC (mg C g−1 soil) | 0.54 | ns | 3.77 ± 0.17 | 3.76 ± 0.09 | 3.69 ± 0.17 | 3.56 ± 0.06 |
STN (×10–1 mg N g−1 soil) | 0.65 | ns | 1.83 ± 0.03 | 2.20 ± 0.23 | 1.88 ± 0.19 | 2.10 ± 0.18 |
STP (×10–1 mg P g−1 soil) | 0.71 | ns | 6.56 ± 0.09 | 6.69 ± 0.02 | 6.70 ± 0.15 | 6.72 ± 0.02 |
STC:N | 1.25 | ns | 21.58 ± 3.12 | 17.43 ± 1.51 | 19.89 ± 1.76 | 16.58 ± 1.34 |
STC:P | 0.62 | ns | 5.76 ± 0.31 | 5.63 ± 0.14 | 5.52 ± 0.35 | 5.30 ± 0.07 |
STN:P (×10–1) | 0.53 | ns | 2.81 ± 0.52 | 3.29 ± 0.34 | 2.81 ± 0.26 | 3.24 ± 0.26 |
MBC (mg C g−1 soil) | 0.16 | ns | 2.08 ± 031 | 2.19 ± 0.66 | 2.08 ± 0.25 | 1.81 ± 0.18 |
MBN (×10–3 mg N g−1 soil) | 1.13 | ns | 1.67 ± 0.48 | 2.64 ± 0.94 | 1.82 ± 0.23 | 1.24 ± 0.23 |
MBC:N (×103) | 0.11 | ns | 1.48 ± 0.40 | 1.59 ± 1.15 | 1.15 ± 0.10 | 1.66 ± 0.53 |
EEAC (μg h−1 g−1 soil) | 1.4 | ns | 2.84 ± 1.25 | 2.57 ± 0.56 | 1.11 ± 0.38 | 3.54 ± 0.99 |
EEAN (μg h−1 g−1 soil) | 0.76 | ns | 9.00 ± 5.10 | 5.62 ± 1.96 | 2.89 ± 0.38 | 5.91 ± 1.76 |
EEAP (μg h−1 g−1 soil) | 0.82 | ns | 8.17 ± 1.76 | 3.99 ± 0.71 | 8.69 ± 3.30 | 10.80 ± 5.03 |
EEAC:N (×10–1) | 0.22 | ns | 6.57 ± 4.71 | 6.43 ± 3.32 | 3.66 ± 0.95 | 6.12 ± 0.32 |
EEAC:P (×10–1) | 4.24 | * | 3.22 ± 0.79ab | 6.36 ± 0.45a | 1.32 ± 0.06b | 5.05 ± 1.93a |
EEAN:P | 0.83 | ns | 1.37 ± 0.90 | 1.62 ± 0.73 | 0.42 ± 0.11 | 0.80 ± 0.28 |
TER (×104) | 0.59 | ns | 1.15 ± 0.83 | 0.73 ± 0.27 | 5.47 ± 0.10 | 1.33 ± 0.34 |
VL (×10–1) | 0.81 | ns | 7.77 ± 4.40 | 9.55 ± 2.56 | 3.91 ± 0.88 | 8.15 ± 1.46 |
VA (o) | 1.03 | ns | 49.34 ± 16.70 | 39.11 ± 12.73 | 67.85 ± 5.36 | 53.74 ± 9.11 |
Abbreviations: NC = negative control, NI = non-invasion stage, II = intermediate invasion stage, CI = complete invasion stage, DON = soil dissolved organic N, EEAC = soil microbial extracellular C-acquisition enzyme activity, EEAN = soil microbial extracellular N-acquisition enzyme activity, EEAP = soil microbial extracellular P-acquisition enzyme activity, EEAC:N = the ratio of soil microbial extracellular C-acquisition enzymes to N-acquisition enzymes, EEAC:P = the ratio of soil microbial extracellular C-acquisition enzymes to P-acquisition enzymes, EEAN:P = the ratio of soil microbial extracellular N-acquisition enzymes to P-acquisition enzymes, TER = threshold elemental ratio, VL = vector length, VA = vector angle.
Soil, autotrophic and heterotrophic respiration
In the field study, soil respiration differed significantly among study sites, with the highest value at the low invasion site and the lowest value at the moderate invasion site (P < 0.05, Fig. 1a). Q10 was 21.90% and 15.06% higher at the low invasion site than at the high and moderate invasion sites, respectively (P < 0.01, Fig. 2a).
In the greenhouse experiment, soil respiration, autotrophic respiration and heterotrophic respiration were significantly affected by S. canadensis invasion (P < 0.01, each). Similar to the field study, the highest soil respiration was observed in the non-invasion pots, and the lowest soil respiration was observed in intermediate invasion S. canadensis and P. australis co-dominated pots. Compared to the non-invasion treatment, S. canadensis invasion reduced soil respiration by 12.64% to 31.85% (apart from negative control pots) (Fig. 1b). However, the variation trends of autotrophic and heterotrophic respiration differed to that of soil respiration. With the S. canadensis invasion process (from the non-invasion to complete invasion stage), the autotrophic respiration initially was reduced and then increased with the lowest value observed in the complete invasion treatment pots (Fig. 1c). Meanwhile, heterotrophic respiration was continuously decreasing, with the lowest value observed in the complete invasion treatment pots (Fig. 1d). Compared to the non-invasion treatment, S. canadensis invasion caused autotrophic respiration to vary by −42.96% to 20.83% and reduced heterotrophic respiration by 10.75% to 30.47% (apart from negative control pots). The highest Q10 was observed in the early invasion stage treatment, and the lowest was observed in the negative control treatment (Fig. 2b).
Relationships between soil, autotrophic and heterotrophic respiration with total root biomass and soil properties
In the field study, soil respiration was significantly correlated with soil temperature (R2 = 0.79, P < 0.01), dissolved organic C (R2 = 0.80, P < 0.01), soil total N (R2 = −0.72, P < 0.05), the ratio of soil total C to total P (R2 = −0.73, P < 0.05) and the ratio of soil total N to total P (R2 = −0.80, P < 0.01). Solidago canadensis invasion changed the trend of the soil temperature–respiration curve (Supplementary Fig. S2). Q10 was significantly correlated with soil moisture (R2 = 0.68, P < 0.05), dissolved organic C (R2 = −0.80, P < 0.01), soil available P (R2 = −0.67, P < 0.05), soil total N (R2 = 0.78, P < 0.05), the ratio of soil total C to total P (R2 = 0.77, P < 0.05) and the ratio of soil total N to total P (R2 = 0.80, P < 0.01, Fig. 3a).
In the greenhouse experiment, soil respiration revealed significant positive correlations with soil temperature (R2 = 0.16, P < 0.05), soil total P (R2 = −0.67, P < 0.05), EEAC:P (R2 = 0.75, P < 0.05), vector length (R2 = 0.85, P < 0.01) and total root biomass (R2 = 0.82, P < 0.01). Autotrophic respiration revealed significant positive correlations with soil available P (R2 = 0.71, P < 0.05), microbial biomass N (R2 = −0.86, P < 0.01), threshold elemental ratios (R2 = 0.73, P < 0.05) and total root biomass (R2 = 0.57, P < 0.05), while heterotrophic respiration significantly correlated with soil temperature (R2 = 0.19, P < 0.05), microbial biomass N (R2 = 0.67, P < 0.05) and total root biomass (R2 = 0.54, P < 0.05, Fig. 3b). The PLS-PM analysis demonstrated that the alterations in soil properties, total root biomass and soil microbial characteristics induced by S. canadensis invasion ultimately affected soil respiration. The soil available P (0.99) and total root biomass (0.72) were influential factors for soil respiration. Meanwhile, microbial metabolic C limitation and total root biomass were the direct drivers of soil respiration variation among all test treatments (Fig. 4).
DISCUSSION
The response of soil respiration to Solidago canadensis invasion
Solidago canadensis invasion was originally hypothesized to accelerate soil respiration due to its ability to modify the soil environment and soil biota communities to provide more C inputs into the soil as litterfall, dead roots and root exudates, thereby potentially stimulating the C processes in the soil (Wolkovich et al. 2010; Wu et al. 2019; Zhang et al. 2009, 2016, 2018). In contrast to this prediction, the results of both the field study and the greenhouse experiment demonstrated that S. canadensis invasion reduced the total soil respiration during the S. canadensis invasion process (Fig. 1). On the contrary, an accelerative effect of S. canadensis invasion on soil respiration was observed in an annual grassland by Zhang et al. (2016, 2018), due to S. canadensis invasion-induced alterations in the quality and quantity of C inputs into the soil. It has also been reported that soil CO2 emission was strongly influenced by vegetation traits (e.g. aboveground biomass) and life span (Chen et al. 2015). In addition, S. canadensis is characterized by higher aboveground biomass than native annual herbaceous plants in the grassland, but its biomass was lower than that of P. australis in the present study (Hu 2020; Zhang et al. 2016, 2018). Thus, differences in invaded ecosystem and native vegetation types may account for the divergence in soil respiration responses to S. canadensis invasion.
It was observed that soil respiration demonstrated significant correlations with soil microclimatic conditions (e.g. soil temperature, soil moisture) and substrate availability (e.g. dissolved organic C, total P and resource stoichiometry reflected by the ratios of resources) in both the field study and the greenhouse experiment (Fig. 3). This confirmed that soil respiration is controlled by factors including soil microclimate conditions and substrate availability (Song et al. 2019; Zhou et al. 2013). The observed positive correlations between soil respiration and soil microclimate conditions suggested that S. canadensis invasion altered soil respiration via changes in soil microclimate (Fig. 3; Supplementary Fig. S2). However, the plant-mediated effects on soil microclimate in this study were not as significant as those observed in previous studies (Supplementary Fig. S3, Jones et al. 2019; Potts et al. 2008; Wolkovich et al. 2010). Hence, variation in soil microclimate may not be a major factor inducing the alterations in soil respiration in the present study. Meanwhile, the Q10 value consistently decreased as S. canadensis invasion progressed (Fig. 2), which is inconsistent with the finding that a greater Q10 value was observed in S. canadensis invaded sites than in a native annual herb site in a grassland ecosystem (Zhang et al. 2018). According to Arrhenius kinetics, the Q10 is related to the molecular structure of the organic reactant and decreases with the complexity of the molecular structure (Xiao et al. 2020). Thus, the inconsistent findings may be attributed to differences between S. canadensis and P. australis in morphological traits, life span and components in root exudation and litter, which may further affect soil C cycling due to shifts in soil substrate input. Instead of soil microclimate, changes in the available substrate in the soil induced by S. canadensis invasion may have been causing the decrease in soil respiration.
The responses of autotrophic and heterotrophic respiration to Solidago canadensis invasion
Interestingly, autotrophic respiration was reduced during the initial stage of S. canadensis invasion and promoted from the dominant stage of invasion, showing a similar variation trend to that of soil respiration (Fig. 1b and c). However, heterotrophic respiration continued to decline as the invasion progressed (Fig. 1d). The different variation trends of autotrophic and heterotrophic respiration resulting from S. canadensis invasion confirmed the second hypothesis. Meanwhile, the PLS-PM results revealed that both total root biomass and soil microbial characteristics had direct effects on soil respiration (Fig. 4), which is consistent with the already-established knowledge that plant functional traits can affect soil respiration and C cycling via plant root reparation, plant-released substrates and microbial symbionts associated with the root system (Tang et al. 2020a, 2020b). Combined with the different responses of autotrophic and heterotrophic respiration to S. canadensis invasion, the suppressive effect of S. canadensis invasion on soil respiration can also be revealed from the following two aspects discussed below.
In this study, S. canadensis invasion increased dissolved organic C and available P and decreased dissolved organic N in the soil (Tables 1 and 2). As terrestrial nutrient cycles are highly coupled, the changes in available soil substrate may induce an imbalance in nutrient stoichiometry in soils (Xiao et al. 2020). Additionally, the available soil substrate may directly impact the growth, C use efficiency and resource allocation of plants (Duval et al. 2020; Zhang et al. 2009). Consequently, the C allocation strategy may be modified to cope with the competition between invasive plants and native plants, alongside plant and soil microbes. Meanwhile, the competition-induced differences in C allocation belowground between S. canadensis and P. australis would directly impact the root biomass and the autotrophic respiration via plant root respiration (Zhou et al. 2013). Evidence for this was provided by the observation of the same variation trends in total root biomass and autotrophic respiration during the S. canadensis invasion process, and the lowest autotrophic respiration and total root biomass being observed at the middle stage of invasion (moderate invasion and intermediate invasive stage treatments) (Supplementary Fig. S1b). Besides, autotrophic respiration demonstrated a significant relationship with total root biomass, soil available P and the threshold elemental ratio (Fig. 3b). Thus, changes in soil substrate availability could be a possible cause of the suppressive effect of S. canadensis invasion on soil respiration, due to reduced autotrophic respiration (Wang et al. 2015).
On the other hand, it has been proposed that vegetation functional traits can affect the soil microbial community (e.g. its activity and functioning) via plant-released substrates from root exudation and litter (Wang et al. 2017; Yang et al. 2016). Based on the Metabolic Theory of Ecology and the Ecological Stoichiometry Theory, soil extracellular ecoenzymatic stoichiometry was developed to assess the energy and nutrient limitations in soil microbial metabolism (Sinsabaugh et al. 2008, 2009). Soil microbial extracellular enzymatic stoichiometry links substrate availability with microbial nutrient acquisition strategies, which are affected by microbial demand, as microbes obtain and/or compete with plants for available substrate during the decomposition of soil organic matter (He et al. 2020; Moorhead et al. 2013). In this study, S. canadensis invasion reduced the soil microbial demand on C, as reflected by the decrease in C-related microbial characteristics (microbial C limitation reflected by vector length, the ratio of C-acquisition to P-acquisition and C-acquisition to N-acquisition microbial extracellular enzymatic activity) (Table 2), negatively affecting soil C turnover. Previous studies have reported that S. canadensis can change the specific soil microbe functional groups and inhibit the growth and activity of many microbes by releasing several allelopathic compounds (e.g. α-pinene, limonene and germacrene) into the soil (Wang et al. 2018; Zhang et al. 2009). This is consistent with the findings of lower microbial biomass under S. canadensis invasion, and the high phenol and flavone concentrations in S. canadensis leaves, which was observed in a previous companion study in the same field site (Supplementary Table S1; data from Hu 2020). These antagonistic substrates were difficult for soil microbes to decompose and led to a lower soil C turnover, which consequently contributed to the suppression of soil respiration. Thus, the changes in the quality of released substrates could be another possible cause of the suppressive effect of S. canadensis invasion on soil respiration.
Implication of Solidago canadensis invasion effects on soil respiration
Overall, vegetation community succession may alter not only soil microclimate conditions (e.g. hydrology state, thermodynamic state), but also the quality and quantity of soil C substrate inputs and the composition and structure of soil microbial communities (Chen et al. 2015; Duval et al. 2020; He et al. 2020). In the natural riparian habitat, one of the most ecologically sensitive areas, these alterations may further affect the soil C cycling processes by impacting the oxidation, production and transport of CO2 (Hu et al. 2021). Studies focused on the effect of S. canadensis invasion on soil respiration in an originally P. australis dominated area were limited. However, the available literature suggests that Spartina alterniflora invasion similarly affects soil microbial respiration in a tidal wetland originally dominated by P. australis (Chen et al. 2012; Yang et al. 2016), due to the different substrate inputs released from Spartina alterniflora. Given that soil respiration is a composite of multiple soil ecological processes, the consumption of substrates released from plants and the corresponding mechanisms were observed in the present study, as the induced changes in substrate inputs by S. canadensis invasion may have a greater effect on soil, autotrophic and heterotrophic respiration than the induced changes in soil microclimate conditions (Zhou et al. 2013). Despite the remarkable observed suppression of soil CO2 emissions under S. canadensis invasion, the soil C pool (soil total C) did not prospectively increase following the progression of the S. canadensis invasion (Tables 1 and 2). Several factors may have contributed to these findings. One of the possible reasons may be the established fact that the soil C pools are usually much larger than the amount of C released via soil CO2 emissions (Jones et al. 2019). Another possible reason may be the C assimilation and accumulation patterns in S. canadensis, as a significant increase in soil dissolved organic C (Tables 1 and 2) and a higher leaf C content was found in S. canadensis than in P. australis (Supplementary Table S1; data from Hu 2020).
As S. canadensis continues to expand their distribution in the invaded ecosystem, soil CO2 emission changes following S. canadensis invasion may not yet have reached a steady state. Substrates derived from invasive plants may have a continuous priming effect on soil CO2 emissions. Tang et al. (2020a, 2020b) studied the global patterns of soil, autotrophic and heterotrophic respiration in a meta-analysis and suggested that both vegetation and edaphic effects on soil CO2 emissions were always coupled with climate factors on a large scale. Thus, additional studies of long-term observations on the variations following S. canadensis invasion may be needed to increase the understanding of mechanisms underlying the S. canadensis invasion effects on belowground C dynamics and its contribution to climate warming.
CONCLUSIONS
This study recorded the variations in soil CO2 emissions following S. canadensis invasion in both an observational field study and a greenhouse experiment simulating invasion. Overall, it was found that S. canadensis invasion decreased the total soil CO2 emissions in both the field and greenhouse experiments. Specifically, soil and autotrophic respiration continuously decreased during the initial stage of S. canadensis invasion and then increased during the dominant stage of invasion. However, heterotrophic respiration continued to decline during the entire invasion process. Changes in the quality and quantity of available substrate in the soil, induced by S. canadensis invasion, may have a suppressive effect on soil respiration, suggesting that S. canadensis invasion may impact soil C cycling via plant-released substrates and may compete for the available substrate with native plants and/or soil microbes. The observed suppression of soil CO2 emissions under S. canadensis invasion in a wetland originally dominated by P. australis indicated that large-scale S. canadensis invasion may affect C exchanges between the soil and the atmosphere. Meanwhile, in the context of global warming, the present study has essential implications for the assessment of the effects of invasive alien plants on the belowground C dynamics and their contribution to the warming climate in wetland systems.
Supplementary Material
Supplementary material is available at Journal of Plant Ecology online.
Table S1: Leaf characteristic of Common reed (Phragmites australis (Cav.) Trin. ex Steud) and Canada goldenrod (Solidago canadensis L.) among study sites in the field study, presented as means ± standard errors.
Figure S1: Variations in (a) specific root length among study sites in the field study (n = 3) and (b) total root biomass among treatments in the greenhouse experiment (n = 12).
Figure S2: Correlation between soil respiration and soil temperature (a) among all study sites (gray dot, n = 180), (b) in the low invasion level (LI) site (purple dot, n = 60), (c) in the moderate invasion level (MI) site (yellow dot, n = 60) and (d) in the high invasion level (HI) site (green dot, n = 60) in the field study.
Figure S3: Variations in (a) soil temperature and (b) soil moisture among study sites in the field study (n = 60) as well as (c) soil temperature and (d) soil moisture among treatments in the greenhouse experiment (n = 33).
Acknowledgements
We are grateful to Mr. Jie Dong, Weikang Xia, Xijia Zhang and Jiangquan Wang for their help with the experiments during the study period. We are also grateful to the Associate Editor and anonymous referee for providing valuable comments.
Funding
State Key Research Development Program of China (2017YFC1200100); the National Natural Science Foundation of China (31800342, 31770446, 32071521); the China Postdoctoral Science Foundation (2019M651720); the Talent Project from the ‘Double-Entrepreneurial Plan’ in Jiangsu Province; and the Jiangsu University Foundation, the Postgraduate Research and Practice Innovation Program of Jiangsu Province (SJCX19_0568).
Conflict of interest statement: The authors declare that they have no conflict of interest.
REFERENCES
Author notes
These authors contributed equally to this work.