Over the past few decades, environmental changes caused by anthropogenic activity, such as climate warming and nitrogen (N) deposition, have played a large role in the disturbance of biological diversity and ecosystem functions (Klein et al. 2004; Chen et al. 2013). However, the effects of N deposition on soil microbial communities and their underlying mechanisms within the alpine area are yet to be fully understood. In many tropical and temperate ecosystems, N deposition at rate not more than 50 kg ha−1 year−1 stimulates the nitrification and denitrification of soil microbial communities by elevating the substrate contents of the two processes (Erickson et al. 2001; Baer and Blair 2008). N deposition at rate > 50 kg ha−1 year−1 often continues to stimulate the microbial nitrification process (Lehtovirta-Morley et al. 2011; Zhang et al. 2012), yet depresses the microbial denitrification process (Gao et al. 2015). However, it is unclear whether the N deposition at rate not more than 50 kg ha−1 year−1 will also stimulate microbial nitrification and denitrification in alpine ecosystems.

In this study, we conducted an 8-year (2007–2015) N addition experiment to simulate N deposition in a meadow ecosystem on the Qinghai–Tibetan Plateau, China (N 37° 37′, E 101° 19′). The Qinghai–Tibetan Plateau has served as the roof of the world, the water tower of Asia, and the third pole of the Earth for 65 million years (Zheng et al. 1979; Wu and Yin 2002; Sun et al. 2012). The current N deposition rate at the Qinghai–Tibetan Plateau is recorded at approximately 10 kg ha−1 year−1, with an increasing trend predicted for the next few decades (Lu and Tian 2007). The vegetation type is a typical Kobresia humilis meadow. The soils developed in the Kobresia meadow are Mat–Gryic Cambisol (Fang et al. 2012). Four N addition rates were applied (0, 10, 20, and 40 kg ha−1 year−1), and thus, the actual total N deposition rates (anthropogenically added rate + naturally deposited rate) were approximately 10, 20, 30, and 50 kg ha−1 year−1, respectively. Two types of nitrogenous compounds, NH4Cl and KNO3, were added for each of the latter three rates. Thus, there were seven treatments, with three replicates (in three blocks) for each treatment. In 2015, the abundances of two nitrifying (AOA-amoA and AOB-amoA) and four denitrifying genes (nirS, nirK, narG, and nosZ) of soil microbial communities were quantified using real-time PCR analysis. The N2O emission rate was quantified in situ via gas chromatography (Wang and Wang 2003). In addition, plant community biomass and soil physicochemical indices (soil water content, pH, dissolved organic carbon content, NO3−-N content, and NH4+-N content) were also measured (see Method S1).

The abundances of the two nitrifying and four denitrifying genes initially increased and subsequently decreased as the N deposition rate increased, with maximum values associated with 30 kg ha−1 year−1 (Fig. 1a–f). The N2O emission rate exhibited a similar trend as that of the gene abundances, especially for the treatments of KNO3 addition (Fig. 1g). Specifically, an increase in the N deposition rate from 10 to 30 kg ha−1 year−1 resulted in a rise in the gene abundances and N2O emission rate due to elevated soil N content. This is consistent with tropical and temperate ecosystems (Xie et al. 2018; Zhang et al. 2018). However, further increases in the N deposition rate (from 30 to 50 kg ha−1 year−1) led to a decline in the index values, differing from results in the literature (Frey et al. 2004; Tian et al. 2014; Zeng et al. 2016).

Fig. 1
figure 1

Relationship between community size of a, b nitrifying and c–f denitrifying microorganisms, g N2O emission rate, and nitrogen deposition rate. Error bars represent one standard error

Full size image

This unexpected decline in the indices was not attributed to aboveground plant biomass, soil water content, and dissolved organic carbon content, as these variables responded non-significantly to the N deposition rate, as revealed by a two-way analysis of variance (Fig. 2a, c–d). Furthermore, as the N deposition rate increased from 30 to 50 kg ha−1 year−1, no consistent changes were observed for soil pH under the addition of NH4Cl and KNO3 (Fig. 2b). Specifically, soil pH levels decreased with the addition of NH4Cl, yet remained constant with the addition of KNO3. Thus, soil pH was not an influencing factor for the observed decline in the seven indices. Similarly, soil NO3−-N content remained unchanged under the addition of both NH4Cl and KNO3 (Fig. 2e). In contrast, soil NH4+-N content increased consistently with the rise in nitrogen deposition under the addition of both NH4Cl and KNO3 (Fig. 2f) and is thus identified as a likely cause of the reduced index values.

Fig. 2
figure 2

Effects of N deposition rate on a soil water content, b soil pH, c dissolved organic carbon content, d aboveground plant biomass, e soil NO3−-N content, and f NH4+-N content. Error bars represent one standard error. For clarity, the statistical results with P > 0.05 are not shown

Full size image

Excessive soil NH4+ concentrations are able to change the ion balance in plants through their uptake of cations and thus are toxic to plants (Britto and Kronzucker 2002; Esteban et al. 2016). Moreover, soil microorganisms are considered to be less resistant to environmental pressures compared to higher plants (Zhang et al. 2015; Zhang et al. 2016). Hence, the excessive NH4+ concentration under the nitrogen deposition rate of 50 kg ha−1 year−1 is likely to be toxic to soil microorganisms, leading to the observed decline in the seven indices (Fig. 1). Note the addition of NO3−-N (in the form of KNO3) from 30 to 50 kg ha−1 year−1 was associated with a slight reduction in soil NO3−-N content, while NH4+-N content slightly increased (Fig. 2e, f). This implies that a large amount of NO3−-N was transformed into NH4+-N, possibly driven by plants and/or microorganisms (Esteban et al. 2016). Both the transformation of NO3−-N to NH4+-N and the toxicity of excessive NH4+ for microorganisms require further exploration to fully understand the mechanisms.

In summary, N deposition at rate not more than 50 kg ha−1 year−1 may either stimulate nitrifying and denitrifying microorganisms through increases in soil N content or depress these microorganisms via NH4+ toxicity. In some tropical and temperate ecosystems, many microorganisms are resistant to NH4+ toxicity, thus exhibiting a net positive response to low deposition rates (Ning et al. 2015; Tian et al. 2018). In contrast, microorganisms found in alpine ecosystems do not present such a resistance to NH4+ toxicity, highlighting the vulnerability of microorganisms in such unique ecosystems. Moreover, for other types of ecosystems, N deposition is observed to have a cumulative effect (Dupre et al. 2010). Thus, it is also important to identify whether N deposition rates at rate not more than 30 kg ha−1 year−1 can cause a net negative effect on microorganisms for longer treatments. A further key point is to determine whether the relationship observed on the Qinghai–Tibetan Plateau between the community size of nitrifying and denitrifying microorganisms and nitrogen deposition at rate not more than 50 kg ha−1 year−1 also applies to other alpine ecosystems in the world.