Differential effects of altered precipitation regimes on soil carbon cycles in arid versus humid terrestrial ecosystems

Altered precipitation affects soil enzyme activity related to nitrogen and phosphorous but not carbon cycling: A meta-analysis

Altered precipitation significantly influences soil function in terrestrial ecosystems. As a bioindicator of soil function, soil extracellular enzyme activity (EEA) plays a crucial role in mediating ecosystem responses to altered precipitation. However, the global patterns and regulatory mechanisms of altered precipitation impacts on soil EEAs remain unclear. We conducted hierarchical mixed-effects meta-analyses to explore the responses and regulators of carbon, nitrogen, phosphorus hydrolytic EEAs, and carbon oxidative EEA to changes in precipitation, using the largest dataset to date, comprising 1185 observations of 14 soil EEAs from 73 publications. The results indicated that soil nitrogen hydrolytic EEA increased by 14.3% under increased precipitation, while phosphorus hydrolytic EEA decreased by 8.8% under decreased precipitation, showing higher sensitivity to altered precipitation compared to carbon-degrading EEAs. These responses varied across ecosystem types and depended on the magnitude of precipitation manipulation (MPM). Specially, decreased precipitation significantly reduced phosphorus hydrolytic EEA in forests, while increased precipitation enhanced nitrogen hydrolytic EEA in grasslands. Furthermore, these effects were linearly correlated with MPM, deviating from the expected nonlinear double asymmetric model. The response of soil hydrolytic EEAs was predominately regulated by soil water content, organic carbon, and microbial biomass. These findings underscore the higher sensitivity of nitrogen and phosphorus cycling EEAs to altered precipitation compared to carbon cycling EEAs and extend the application of the double asymmetric model for understanding soil EEAs' responses to precipitation changes. This synthesis provides essential insights for predicting biogeochemical cycling and improving ecosystem models to evaluate ecosystem functions under altered precipitation.

Nonlinear responses of ecosystem carbon fluxes to precipitation change in a semiarid grassland

Carbon (C) fluxes in semiarid grasslands subject to precipitation variability play a critical role in the terrestrial C cycle. However, how ecosystem C fluxes respond to variability in precipitation (both decreases and increases precipitation along a gradient) remains unclear. In this study, we conducted a three-year field experiment in a semiarid grassland, with six precipitation treatments (precipitation decreased by 70%, 50%, and 30% [P-70%, P-50%, and P-30%], natural precipitation [P+0%], and precipitation increased by 30% and 50% [P+30% and P+50%]) to examine how variations in precipitation influence ecosystem C fluxes, specifically focusing on gross ecosystem productivity (GEP), ecosystem respiration (ER), and net ecosystem CO2 exchange (NEE). We found that both decreased and increased precipitation significantly altered the GEP (from -26% to 14%), but only decreased precipitation significantly reduced the ER and NEE (from 1% to 31%), relative to their values during natural precipitation. This suggests that ecosystem C fluxes are more sensitive to decreased precipitation, and respond nonlinearly to the precipitation gradient. Furthermore, structural equation modeling indicated that the soil water content was the primary controlling factor driving changes in ecosystem C fluxes. Our research underscores the nonlinear response of ecosystem C fluxes to changes in precipitation within semiarid ecosystems, particularly their sensitivity to extreme drought. Considering this nonlinear response, it is crucial to improve dynamic models of the C cycle and predict ecosystem responses to precipitation variability.

Whole-Profile Soil Carbon Responses to Concurrent Warming and Precipitation Changes Across Global Biomes

The joint effects of simultaneous warming and precipitation shifts on soil organic carbon (SOC)-the largest terrestrial carbon pool-remain poorly understood across large spatial extents. By evaluating a global dataset of SOC measurements in the top meter of soil through a space-for-change substitution approach, we show that, averaging across the globe, increased precipitation compensates for warming-induced SOC reductions regardless of soil depth and vice versa. Although additive effects between these two factors are predominant, negative interactive effects, which exacerbate SOC losses, are also common, particularly in tropical and subtropical grasslands/savannas and Mediterranean/montane shrublands. SOC responses vary widely across the globe, primarily correlated to baseline SOC content and local climatic conditions. Notably, SOC responses in tundra systems are opposite the responses in other ecosystems, showing positive and negative responses to warming and precipitation increases, respectively. Under a scenario of 2°C air warming with projected precipitation changes, global SOC stocks in the 0-1 m depth are projected to decrease by 13.1% ± 6.6% (mean ± 95% confidence interval, or 351 ± 100 Pg C). These results demonstrate that accurately predicting SOC dynamics under climate change necessitates explicit consideration of local climatic conditions and existing SOC content in relation to concurrent precipitation shifts and warming.

Contrasting carbon cycle responses of semiarid abandoned farmland to simulated warmer-drier and warmer-wetter climates

Rewilding abandoned farmlands provides a nature-based climate solution via carbon (C) offsetting; however, the C-cycle-climate feedback in such restored ecosystems is poorly understood. Therefore, we conducted a 2-year field experiment in Loess Plateau, China, to determine the impacts of warming (∼1.4 °C) and altered precipitation (±25 %, ±50 %, and ambient), alone or in concert on soil C pools and associated C fluxes. Experimental warming significantly enhanced soil respiration without affecting the ecosystem net C uptake and soil C storage; these variables tended to increase along the manipulated precipitation gradient. Their interactions increased ecosystem net C uptake (synergism) but decreased soil respiration and soil C accumulation (antagonism) compared with a single warming or altered precipitation. Additionally, most variables related to the C cycle tended to be more responsive to increased precipitation, but the ecosystem net C uptake responded intensely to warming and decreased precipitation. Overall, ecosystem net C uptake and soil C storage increased by 94.4 % and 8.2 %, respectively, under the warmer-wetter scenario; however, phosphorus deficiency restricted soil C accumulation under these climatic conditions. By contrast, ecosystem net C uptake and soil C storage decreased by 56.6 % and 13.6 %, respectively, when exposed to the warmer-drier climate, intensifying its tendency toward a C source. Therefore, the C sink function of semiarid abandoned farmland was unsustainable. Our findings emphasize the need for management of post-abandonment regeneration to sustain ecosystem C sequestration in the context of climate change, aiding policymakers in the development of C-neutral routes in abandoned regions.

Variation of root traits and its influences on soil organic carbon stability in response to altered precipitation in an alpine meadow

Soil organic carbon (SOC) dynamics are strongly controlled by plant roots. Yet, how variation of root traits under precipitation change influences SOC stability remains unclear. As part of a 5-year field experiment manipulating precipitation including 90 % (0.1P), 50 % (0.5P), 30 % (0.7P) decrease, and 50 % increase (1.5P), this study was designed to assess the effects of changing precipitation on root traits and production dynamics by minirhizotron and examine how such influences regulate SOC stability in an alpine meadow on the Qinghai-Tibetan Plateau. We found that root length density (RLD), specific root length (SRL), root branching intensity (RBI), and root residue carbon input (RC input) exhibited no significant response, whereas root turnover (RT), root carbon (C), nitrogen (N) concentrations and C/N ratio were altered by precipitation change with nonlinear trends. Absorptive root RT positively correlated to manipulated precipitation within the interannual precipitation range in topsoil, but it showed no significant change under extreme drought treatment. Alpine meadows can maintain the SOC content and density under varied precipitation. However, it showed significant variation in aggregate stability and organic carbon (OC) distribution in aggregates in topsoil, which were mainly due to the strong direct effects of soil moisture and partly related to RLD and RC input of transport roots. Although subsurface soil aggregate stability and OC associated with aggregates were not modified, our results indicated a risk of SOC stability variation in subsurface soil if absorptive root RT and SRL changed. These findings provide vital information to predict responses of SOC dynamics of alpine meadow to future climate change.

Nitrogen addition does not alter symmetric responses of soil respiration to changing precipitation in a semi-arid grassland

Concurrent changing precipitation regimes and atmospheric nitrogen (N) deposition can have profound influences on soil carbon (C) cycling. However, how N enrichment regulates the responses of soil C fluxes to increasing variability of precipitation remains elusive. As part of a field precipitation gradient experiment with nine levels of precipitation amounts (-60 %, -45 %, -30 %, -15 %, ambient precipitation, +15 %, +30 %, +45 %, and +60 %) and two levels of N addition (0 and 10 g N m-2 yr-1) in a semi-arid temperate steppe on the Mongolian Plateau, this work was conducted to investigate the responses of soil respiration to decreased and increased precipitation (DP and IP), N addition, and their possible interactions. Averaged over the three years from 2019 to 2021, DP suppressed soil respiration by 16.1 %, whereas IP stimulated it by 27.4 %. Nitrogen addition decreased soil respiration by 7.1 % primarily via reducing microbial biomass C. Soil respiration showed symmetric responses to DP and IP within all the four precipitation variabilities (i.e., 15 %, 30 %, 45 %, and 60 %) under ambient N. Nevertheless, N addition did not alter the symmetric responses of soil respiration to changing precipitation due to the comparable sensitivities of microbial biomass and root growth to DP and IP under the N addition treatment. These findings indicate that intensified precipitation variability does not change but N addition could alleviate soil C releases. The unchanged symmetric responses of soil respiration to precipitation variability under N addition imply that N deposition may not change the response pattern of soil C releases to predicted increases in precipitation variability in grasslands, facilitating the robust projections of ecosystem C cycling under future global change scenarios.

Differential responses of soil extracellular enzyme activity and stoichiometry to precipitation changes in a poplar plantation

Changes in precipitation patterns can significantly affect belowground processes. Although soil extracellular enzymes play a vital role in several biogeochemical processes, our knowledge of how precipitation changes affect soil extracellular enzyme activity (EEA) and stoichiometry remains insufficient. In this study, we investigated the activities of C-acquiring enzyme (β-1,4-glucosidase), N-acquiring enzymes (β-N-acetylglucosaminidase and leucine aminopeptidase), and P-acquiring enzyme (acid phosphatase) under different precipitation scenarios [ambient precipitation (CK), 30% decrease in precipitation (moderate DPT), 50% decrease in precipitation (extreme DPT), 30% increase in precipitation (moderate IPT), and 50% increase in precipitation (extreme IPT)] in a poplar plantation. We found soil EEA exhibited more pronounced increases to moderate IPT compared to moderate DPT (positive asymmetry), the opposite trend (negative asymmetry) was observed under extreme precipitation; whereas soil EEA C:N:P stoichiometry exhibited negative asymmetry at moderate precipitation changes, and exhibited positive asymmetry at extreme precipitation changes. Under moderate precipitation changes, the asymmetry of soil EEA was mainly regulated by asymmetries of respective microbial biomass and litter mass; the asymmetry of soil EEA stoichiometry was mainly regulated by asymmetries of respective soil stoichiometric ratios and litter mass. Furthermore, under extreme precipitation changes, the asymmetries of soil EEA and stoichiometry were best explained by the asymmetry of soil moisture. Our results provide the first evidence of double asymmetric responses of soil EEA and stoichiometry to precipitation changes and highlight the need to consider this asymmetry when modeling the dynamics of biogeochemical cycling in forest ecosystems.

Contrasting effects of intensified dry-season drought and extended dry-season length on soil greenhouse gas emissions in a subtropical forest

Over two-thirds of the Earth's land surface is subjected to seasonal precipitation changes along with climate warming, including the subtropical forests that represent one of the Earth's most important carbon sink and source. However, few experiments have been conducted to understand the response of soil greenhouse gas (GHGs) emissions from these forests to seasonal changes in precipitation. Herein, we conducted a field experiment in a subtropical forest of southern China including two precipitation seasonality treatments: an intensified dry-season (Oct-Mar) drought and wetter wet-season (Jun-Sep) treatment (ID) and an extended dry-season (Apr-May) length and wetter wet-season treatment (ED); for both ID and ED, the annual precipitation amount was kept the same as under ambient control (AC). Compared to AC, the decreased annual CO2 emissions for ID were mainly due to decreased WFPS in Oct-Mar of 2013-2014 and increased WFPS during Jun-Sep of 2013; the increased annual CH4 uptake for ID was predominantly attributed to decreased WFPS in Oct-Mar of 2013-2014; the decreased annual N2O emissions for ID were mainly due to decreased WFPS in Oct-Mar of 2013; the increased annual N2O emissions for ID in 2014 were mainly attributed to increased WFPS in Jun-Sep (p < 0.05). Relative to AC, the increased annual CO2 and N2O emissions from ED were predominantly attributed to decreased WFPS in Apr-May and increased WFPS in Jun-Sep during 2013-2014, respectively (p < 0.05). The average annual CO2-equivalent CH4 and N2O emissions increased under ED but decreased under ID compared to AC (p < 0.05). Although our two precipitation manipulation scenarios simulated seasonal drought impacts without changing annual precipitation amount, ED and ID had distinct impacts on soil GHGs emissions, which have important implications for modeling the subtropical forests GHG emissions and managing the forests to mitigate climate change.

Divergent patterns and drivers of leaf functional traits of Robinia pseudoacacia and Pinus tabulaeformis plantations along a precipitation gradient in the Loess plateau, China

Changes in precipitation patterns in arid and semi-arid regions can reshape plant functional traits and significantly affect ecosystem functions. However, the synchronous responses of leaf economical, anatomical, photosynthetic, and biochemical traits to precipitation changes and their driving factors have rarely been investigated, which hinders our understanding of plants' ecological adaptation strategies to drought tolerance in arid areas. Therefore, the leaf traits of two typical plantations (Robinia pseudoacacia, RP and Pinus tabulaeformis, PT) along the precipitation gradient in the Loess Plateau, including economical, anatomical, photosynthetic, and biochemical traits, were investigated in this study. The results show that the leaf photosynthetic traits of RP and PT increase along the precipitation gradient, whereas leaf biochemical traits decrease. The anatomical traits of PT decrease with increasing precipitation, whereas no significant variation was observed for RP. Random Forest analysis show that LNC, LDMC, Chl, and PRO are leaf traits that significantly vary with the precipitation gradient in both plantations. Correlation analysis reveals that the traits coordination of RP is better than that of PT. The LMG model was used to determine driving factors. The results suggest that MAP explains the variation of PT leaf traits better (30.38%-36.78%), whereas SCH and SPH contribute more to the variation of RP leaf traits (20.88%-41.76%). In addition, the piecewise Structural Equation Model shows that the climate and soil physical and chemical properties directly affect the selected leaf functional traits of RP, whereas only the soil chemical properties directly affect the selected leaf functional traits of PT. The results of this study contribute to the understanding of the ecological adaptation of plants to environmental gradients and highlight that correlations among leaf traits should be considered when predicting plant adaptation strategies under future global change scenarios.

Differential effects of grazing intensity on carbon sequestration in arid versus humid grasslands across China

Livestock grazing, as a primary utilization practice for grasslands, plays a crucial role in carbon cycling process and its budget. Whether the impacts of different grazing intensities on carbon sequestration vary with precipitation over a broad geographic scales across China's grasslands remains unclear. In the context of striving for carbon neutrality, we carried out a meta-analysis based on 156 peer-reviewed journal articles to synthesize the general impacts of different grazing intensities on carbon sequestration with different precipitations. Our results showed that light, moderate, and heavy grazing dramatically reduced the soil organic carbon stocks by 3.43 %, 13.68 %, and 16.77 % in arid grasslands, respectively (P < 0.05), while light and moderate grazing did not alter soil organic carbon stocks in humid grasslands (P > 0.05). Moreover, the change rates of soil organic carbon stocks were all tightly positively associated with those of soil water content under different grazing intensities (P < 0.05). Further analysis revealed strong positive relationships between mean annual precipitation with the change rates of above- and belowground biomasses, soil microbial biomass carbon, and soil organic carbon stocks under moderate grazing intensity (P < 0.05). These findings imply that carbon sequestration is relatively less tolerant to grazing disturbance in arid grasslands than humid grasslands, which may be primary due to the grazing-intensified water limitation for plant growth and soil microbial activities under low precipitation. Our study is of implication to predict carbon budget of China's grasslands and help adopt sustainable management to strive for carbon neutrality.

Soil biogeochemical responses to multiple co-occurring forms of human-induced environmental change

Human activities cause a multitude of environmental issues, including increased temperatures and altered precipitation patterns associated with climate change, air pollution, and other impacts of urbanization. One area highly affected by these issues is the Sonoran Desert, specifically the Phoenix metropolitan area where urbanization is among the most rapid in the United States. Most studies investigate these multiple environmental change factors independently or sometimes in pairs, but rarely all together as co-occurring forms of change. We examined how the simultaneous manipulation of increasing temperatures, altered precipitation patterns, nitrogen deposition, and urbanization influenced soil respiration and mineral N pools in the Sonoran Desert. Soil was collected from urban and exurban sites, from both nitrogen-fertilized and control plots. To simulate projected climate change, the soils were incubated in microcosm at the annual average Phoenix temperature as well a 2 ℃ increase under a factorial precipitation treatment of decreased frequency and increased pulse size. Our results show that C and N dynamics were altered by all four forms of environmental change. However, the dominance of significant 3- and 4-way interactions among the four environmental factors for both respiration and mineral N pools demonstrates that the impact of any given form of environmental change will depend on the levels of the other environmental factors. In other words, the cumulative effect of altered precipitation, fertilization, temperature, and urbanization on soil biogeochemical processes is not necessarily predictable from their individual impact.

Contrasting effects of altered precipitation regimes on soil nitrogen cycling at the global scale

Changes in precipitation regimes can strongly affect soil nitrogen (N) cycling in terrestrial ecosystems. However, whether altered precipitation regimes may differentially affect soil N cycling between arid and humid biomes at the global scale is unclear. We conducted a meta-analysis using 1036 pairwise observations collected from 194 publications to assess the effects of increased and decreased precipitation on the input (N return from plants), storage (various forms of N in soil), and output (gaseous N emissions) of soil N in arid versus humid biomes at the global scale. We found that (1) increased precipitation significantly increased N input (+12.1%) and output (+34.9%) but decreased N storage (-13.7%), while decreased precipitation significantly decreased N input (-10.7%) and output (-34.8%) but increased N storage (+11.1%); (2) the sensitivity of soil N cycling to increased precipitation was higher in arid regions than in humid regions, while that to decreased precipitation was lower in arid regions than in humid regions; (3) the effect of altered precipitation regimes on soil N cycling was independent of precipitation type (i.e., rainfall vs. snowfall); and (4) the mean annual precipitation regulated soil N cycling in precipitation alteration experiments at the global scale. Overall, our results clearly show that the response of soil N cycling to increased versus decreased precipitation differs between arid and humid regions, indicating the uneven effect of climate change on soil N cycling between these two contrasting climate regions. This implies that ecosystem models need to consider the differential responses of N cycling to altered precipitation regimes in different climatic conditions under future global change scenarios.