Over the past few decades, atmospheric nitrogen (N) deposition increased rapidly, and changed the structure and function of terrestrial ecosystems (Galloway et al. 2008). To describe effects of N deposition on ecosystems, a N saturation theory was developed basing on studies from temperate forests (Aber et al. 1989). One view of this theory proposes that continuous N input may reduce system requirements for N, and finally the system may become N-saturated. In this case, carbon (C), phosphorus (P) or water limitation is expected to occur or be aggravated (Aber et al. 1989). This is a useful theory for predicting dynamics of terrestrial ecosystems in response to increased N deposition (Chen et al. 2018d).
This proposition, however, has yet to specify which nutrient would be the most limited after removing N limitation. In plants, this topic has been studied widely. Recent studies tend to suggest that N deposition may switch N limitation to P limitation (Braun et al. 2010; Crowley et al. 2012; Deng et al. 2017; Gress et al. 2007; Li et al. 2016), although some studies find no effects (Finzi 2009; Weand et al. 2010) or other limitations, such as calcium (Ca) limitation (McNulty et al. 1996). In microbes, however, the discussion is relatively few, and the situation might differ from of plants. For microbes, C is often a more limited element compared to N, P or other nutrients (Soong et al. 2020). Thus, when N limitation is removed after continuous N input, C may be the most limited factor for microbial growth, rather than P or other nutrients. However, this expectation is not consistent with many recent studies, which suggest that N addition may aggravate microbial P limitation (Marklein and Houlton 2012). Since microbes are as important as plants in an ecosystem, more studies are undoubtedly needed to address how microbial resource limitation changes after increased N deposition.
The ecoenzymatic stoichiometry method provides a new tool to study such topic. Compared to traditional methods that measure effects of substrate additions on microbial biomass or respiration as indicators of microbial resource limitation (Traoré et al. 2016), ecoenzymatic stoichiometry has a number of advantages. First, it is much faster, because it measures activities of only four enzymes, including β-D-glucosidase (BG), L-leucine aminopeptidase (LAP), β-N-acetylglucosaminidase (NAG), and acid/alkaline phosphatase (AP). Second, using enzymes as proxy indicators of C, N, and P acquisition, it is much easier for us to understand which nutrient is more limited to an ecosystem (Sinsabaugh 1994). However, most previous studies reported effects of N addition on the activity of the single enzyme (Chen et al. 2016; Jian et al. 2016; Marklein and Houlton 2012), very few studies have assessed responses of ecoenzymatic stoichiometry to N additions (Wang et al. 2015). By collecting published data regarding single enzyme activity in response to N addition, a previous meta-analysis reported that nitrogen deposition may aggravate microbial C limitation (Chen et al. 2018d). However, this study has limitations: firstly, the selected published studies rarely reported C, N, and P acquisition enzymes at the same time, which largely limited the data’s comparability and the conclusion’s generality; secondly, most selected studies were conducted in N-limited systems rather than N-saturated systems, and thus cannot answer the above question regarding changes of resource limitation after a system has been N-saturated. These limitations highlight the importance of more field experiments conducted in N-saturated systems and more experiments measuring enzymes fully for calculating ecoenzymatic stoichiometry.
Topography can modulate many microbial processes, but is often ignored in the previous N-deposition studies. Most N-deposition simulation experiments were conducted in one topography position (mostly in flat ground) or did not distinguish topography positions (Zhang et al. 2021), while few experiments estimated whether and how topography regulates the responses of microbial processes to N deposition (Zhang et al. 2013). In a previous study, divergent responses of soil asymbiotic N2 fixation to N addition has been found between the valley and slope, implying the important effect of topography on ecosystem processes (Wang et al. 2019). However, whether microbial resource limitation status has similar responses needs further exploration. In general, soil N level will be lower in the upslope (due to erosion) than in the downslope or valley (Weintraub et al. 2015), resulting in a higher sensitivity of microbes to the addition of N in the upslope than in the downslope or valley. Nevertheless, the N leaching will be higher in the upslope than in the downslope or valley (Wang et al. 2019), which in turn weakens the N addition effects. These contrary mechanisms make the topography effects uncertain. Lacking knowledge of topography effects has become an important limitation for model prediction regarding responses of terrestrial ecosystem processes to atmospheric N deposition.
Therefore, in this study we conducted a two-year N addition experiment in a subtropical karst forest, where soil microbes have been proven to be N-saturated (Chen et al. 2018b), to test how N deposition changes the status of microbial resource limitation in such a N-saturated situation. In order to investigate whether topography regulates effects of N additions on microbial resource limitation status, the N-addition experiment was set up at two topography positions, i.e., a valley and a slope. To our knowledge, this is the first N-deposition simulation experiment site that considers the effects of topography in the subtropical forest. We hypothesized that 1) nitrogen deposition may aggravate microbial C limitation based on the previous studies; and 2) the response of microbial limitation to nitrogen additions may differ between valley and slope.