Heavy rainfall in peak growing season had larger effects on soil nitrogen flux and pool than in the late season in a semiarid grassland

Highlights

• N₂O flux and soil total N responses to heavy rainfall depended on seasonal timing.

• Mid-season heavy rainfall increased N₂O flux by 65% due to higher denitrification.

• Enhanced N₂O emissions coincided with 17% of reduction in soil total N.

• Late-season heavy rainfall did not change N₂O emissions and soil total N contents.

Abstract

Increasing heavy rainfalls can strongly affect ecosystem nitrogen (N) cycling processes and thereby alter soil N fluxes and pools. However, the effects of heavy rainfalls on soil N fluxes and pools are poorly understood, particularly with regards to high rainfall timing under field conditions. We conducted a 3-year (2014–2016) manipulative experiment in which heavy rainfall was imposed in middle (plant peak growing stage) or late (plant senescent growth stage) growing season in a semiarid grassland of Inner Mongolia, China to explore the responses of N₂O fluxes and soil total N contents and the underlying microbial mechanisms. Mid-season heavy rainfall promoted soil N₂O emissions by 65% on average across the three years, attributable mainly to increases in denitrifying nirK and nirS abundances induced large denitrification at higher soil water contents. However, archaeal and bacterial amoA and narG genes did not change significantly due probably to counteracting effects of increased soil water content (positive) and soil pH (negative). Mid-season heavy rainfall led to 17% reduction in soil total N by the end of the last year of the study (2016), partly due to the enhanced accumulated N₂O emissions over the three years. In contrast, late-season heavy rainfall did not change N₂O emissions and soil total N contents even though soil water content, soil pH and nirK and nirS abundance were significantly increased, perhaps due to limitation by low temperature. Timing of the heavy rainfall events during the plant growing season strongly influenced their impacts on soil N fluxes and pools and heavy rainfalls in the peak stage of plant growth may potentially cause a positive feedback to global warming and exacerbate N limitation in terrestrial ecosystems.

Introduction

One of the characteristics of global climate change is the increase in frequency, intensity, and extent of heavy rainfalls (Donat et al., 2016, Fischer and Knutti, 2016; Jian et al., 2020), which can have profound impacts on ecosystem nitrogen (N) cycling processes, such as nitrification, denitrification, and resultant nitrous oxide (N₂O) emission (Greaver et al., 2016). Soil N₂O flux variation would further influence the size of soil N pool (Fowler et al., 2013). N₂O is the third most important anthropogenic greenhouse gas with a global warming potential 265 times that of carbon dioxide over a 100-year lifespan (Stocker, 2014). Soil N content was highly related to multiple ecosystem structure and functions, such as carbon uptake and microbial community (Yang et al., 2021, Widdig et al., 2020). As such, understanding the effects of heavy rainfalls on ecosystem N₂O flux and soil N pool is particularly important for assessing global warming potential and ecosystem functions.

Ecosystem N cycling processes are mainly driven by microorganisms. For example, each process in nitrification and denitrification is performed by a specialized group of nitrifiers and denitrifiers (e.g., ammonium‐oxidizing and nitrite‐reducing microorganisms) which contain specific functional genes (e.g., amoA and nirK/nirS) encoding corresponding oxido-reductase (e.g., ammonia monooxygenase and Cu-containing/haem-containing cd1 nitrite reductase) (Kuypers et al., 2018). Changes in precipitation are expected to fundamentally affect N₂O flux and/or soil N pool through directly altering the abundance and structure of nitrifiers and denitrifiers (Chen et al., 2017, Chen et al., 2017, Sun et al., 2018, Zhang et al., 2017, Zhang et al., 2013), in addition to indirectly changing soil conditions, such as soil moisture, aeration, and redox potential (Chen et al., 2017, Chen et al., 2017, Patil et al., 2010, Sun et al., 2018). However, the microbial mechanisms of N₂O fluxes responses to precipitation variation are not well understood due to limited data of soil functional genes (gene markers for nitrifiers and denitrifiers) currently available (Li et al., 2020b, Li et al., 2020b).

Recently, several meta-analysis studies suggested that increased precipitation promoted terrestrial N₂O emissions, but unchanged soil total N content based on current field manipulative studies at a global scale (Li et al., 2020b, Li et al., 2020b, Yan et al., 2018). Note that most of these manipulative studies focused on the scenario that every precipitation event proportionately increased in magnitude over the growing season or whole year, rather than the scenario that abrupt heavy rainfall events within few days. Nevertheless, the two precipitation scenarios may have different ecological impacts (Reyer et al., 2013). Chronic increases in precipitation at a seasonal or yearly scale would produce better soil moisture conditions for nitrifiers and denitrifiers activity (Yang et al., 2018, Zhang et al., 2018). However, heavy rainfall-induced inundation may lead to soil water saturation and inhibit microbial activity, especially for aerobic nitrifiers. Additionally, chronic increases in precipitation may promote N mineralization and N availability, providing substrate for nitrification and denitrification processes, thus ultimately accelerating N₂O emissions (Liu et al., 2017). In contrast, dramatic heavy rainfalls are likely to cause flood and ecosystem N loss from topsoil via leaching and runoff (Greaver et al., 2016, Kasper et al., 2019), thereby potentially limiting production of N₂O. On the other hand, large precipitation events possibly induce abrupt N₂O emission pulse (Petrakis et al., 2017), in particular after a dry period (Groffman et al., 2009). In some cases, consistent high rainfalls can suppress N₂O emission (Rowlings et al., 2015). Hence, speculating heavy rainfall effects on soil N₂O emissions, based on past relatively long-term evenly increased precipitation manipulative experiments, may not be straightforward.

Previous limited studies have suggested that high intensity precipitation occurred in different periods of a season had discrepant impacts on multiple ecological processes (Craine et al., 2012). For example, experimental heavy rainfalls in late-growing season consistently reduced below-ground and total biomass while heavy rainfalls in mid-growing season had little effects in a semiarid grassland (Li et al., 2019). In another semiarid grassland, deluge at mid-season caused the greatest increase in soil respiration, aboveground net primary production and canopy greenness than those in early- and late-season (Post and Knapp, 2020). Jian et al. (2020) found that yield variability of early rice in China was negatively affected by the frequency of extreme precipitation from the end of flowering to doughty stages but not in other stages. Likewise, seasonal timing may also largely regulate heavy rainfall effects on soil N₂O emissions and soil N content. However, we still lack an understanding of whether responses of soil N₂O flux and N pool to heavy rainfalls depend on the seasonal timing.

Here, we conducted a 3-year field manipulative experiment in which heavy rainfall was respectively imposed in middle and late plant-growing seasons in a semiarid grassland of Inner Mongolia, China. The seasonal dynamic of N₂O fluxes, soil inorganic and total N content, soil functional genes were measured. The specific questions addressed in this study were: (1) what are the effects of heavy rainfall on N₂O fluxes and soil N pool? (2) what are the biological mechanisms for N₂O fluxes following heavy rainfalls? (3) how does the seasonal timing of an event heavy rainfall regulate its effects on soil N₂O fluxes and soil N pools?