Sphagnum response to nitrogen deposition and nitrogen saturation – a meta-analysis

Background: Sphagnum plants are sensitive to nitrogen deposition changes. However, previous literature is inconclusive on the response of Sphagnum species to nitrogen deposition, little is known about the nitrogen saturation of Sphagnum growth, and the effects of climatic, spatial and fertilization variables on the response of Sphagnum plants are still unclear. Methods: We conducted a meta-analysis of 66 experiments on nitrogen application to Sphagnum plants. The combined effect size of a random-effects model was used to analyze the effect of nitrogen deposition on Sphagnum growth. A mixed-effects model was established to determine the signicance of each factor for the response of Sphagnum plants to nitrogen deposition. The nitrogen saturation value for Sphagnum species was obtained by tting the regression function. Results: Sphagnum growth is promoted by moderate nitrogen deposition, and when nitrogen deposition exceeds the saturation value, Sphagnum growth is inhibited. The current nitrogen deposition saturation value for Sphagnum species is 19.34 kg ha -2 yr -1 globally and varies in different latitudinal zones. When Sphagnum plants are affected by excessive nitrogen deposition, low-water-table microhabitats and excessive temperature and precipitation can all exacerbate Sphagnum growth inhibition. Conclusions: Excessive nitrogen deposition will inhibit Sphagnum growth, and the nitrogen saturation values we obtained for Sphagnum species can be used to adjust fertilizer applications to protect Sphagnum plants from excessive chemical fertilizer application.


Introduction
Sphagnum is a dominant genus of peatlands and the decomposition of Sphagnum litter is extremely slow. The rate of Sphagnum productivity far exceeds the decomposition rate, thus resulting in the xation of large amounts of carbon in peatlands (Rydin, 2006). Peatlands are important global carbon sinks, with more than half of the carbon sequestration worldwide since the Holocene being performed by Sphagnum species (Gorham, 2012;Freeman, 2012). Sphagnum plants lack a root system and vascular tissue, and atmospheric inputs are their primary source of nutrients, making them sensitive to changes in atmospheric nitrogen deposition (Bobbink, 2003). Since the industrial revolution, the burning of large amounts of fossil fuels by factories, the high emissions of nitrogen oxides as a result of urban construction, and the application of large amounts of nitrogen fertilizers to agricultural elds has led to increased atmospheric nitrogen concentrations and deposition in peatlands via rainfall and dust (Galloway et al, 2004;Erisman et al,2011;Miriam, 2014). In recent decades, the area coverage of Sphagnum has greatly declined, and Sphagnum growth has been restricted; a rapid increase in nitrogen deposition is considered one of the possible causes (Gerdol, 2008;Sheppard, 2013;Juutinen, 2016;Shi XM, 2017;Wieder, 2019). It has been shown that increased nitrogen deposition may lead to a shift in the competitive balance in favor of vascular plants and decrease Sphagnum growth, thus reducing the rate of carbon sequestration in peatland ecosystems (Limpens, 2006). Breeuwer (2009) and others have also shown that a high nitrogen application has a negative effect on Sphagnum production, particularly at lower latitudes. Some studies have also shown that increasing nitrogen deposition promotes Sphagnum growth (Bonnett, 2010;Vitt, 2003), and Berg A (2013) et al. found that increased nitrogen levels explained 76% of the increase in Sphagnum biomass. However, an experiment conducted by Gaudig G (2020) et al. showed no effect of continued nitrogen fertilizer application on Sphagnum growth under high-nitrogendeposition conditions (38 kg.ha − 1 .year − 1 ). Overall, there have been no uniform conclusions about the reaction of Sphagnum to changes in nitrogen deposition among existing studies.
The conclusions about the response of Sphagnum to nitrogen deposition have varied among studies, which may have been due to the different factors considered. Previous literature has shown that precipitation, temperature, nitrogen deposition, and the considered microhabitat in the Sphagnum environment may affect the response of Sphagnum to nitrogen deposition (Heijmans, 2008;Breeuwer, 2009;Utstol-Klein, 2015;Krebs, 2016;van den Elzen, 2017). In addition, a number of studies have found that Sphagnum has a threshold for nitrogen deposition saturation. Nitrogen saturation values for Sphagnum in Europe have been determined based on greenhouse studies, nitrogen deposition gradient experiments, eld experiments on net primary productivity, and investigations of other biogeochemical effects, and the critical nitrogen load for bogs and barren bogs in Europe was considered to be 5-10 kg. ha − 1 . year − 1 and 10-15 kg. ha − 1 . year − 1 , respectively (Bobbink and Hettelingh 2011;Tipping et al. 2013;Wieder R K, 2020). When nitrogen deposition exceeds the critical load required by Sphagnum, excess nitrogen will in ltrate the peatland and stimulate vascular plant growth (Wieder, 2019), which may affect carbon accumulation in Sphagnum and gradually transform the ecosystem from a carbon sink to a carbon source.
The nitrogen cycle encompasses the migration, transformation and turnover of nitrogen in the Earth's atmosphere, biosphere, soil and hydrosphere, with atmospheric nitrogen deposition being an important part. Climate conditions such as temperature and precipitation vary greatly among latitudes, and the levels of economic, and agricultural development and environmental protection policies differ among countries; therefore, atmospheric nitrogen deposition may vary at different latitudes. Since Sphagnum is highly sensitive to nitrogen deposition, we conjecture that the response of Sphagnum to nitrogen deposition varies at different latitudes. Previous studies have focused on the Sphagnum community in a single region or a single latitudinal zone, and there is a lack of assessments of the spatial heterogeneity of the response of Sphagnum to nitrogen deposition on a larger spatial scale. It is still unclear how environmental factors at different latitudes affect the response of Sphagnum to nitrogen deposition, so quantitative assessments of such factors are needed. The nitrogen deposition saturation values for Sphagnum are lacking, and research on the nitrogen saturation values in different latitudinal zones is even rarer.
Meta-analysis is a method in which the results of a large number of independent studies with a common research purpose are quantitatively combined to analyze the characteristics of the differences among the studies and comprehensively evaluate the results (Ellenberg, 1988). Meta-analysis gives scienti c weight to each study, excluding as much bias as possible, and the results are more scienti cally valid than other statistical analysis methods. In this study, we extracted the results of 66 published and unpublished nitrogen deposition studies to analyze the response of Sphagnum growth to nitrogen deposition at different latitudes and to quantify the nitrogen saturation values. We divided the data into three zones by latitude: a lower latitudinal zone below 50°, a mid-latitudinal zone at 50-60°, and a higher latitudinal zone above 60°. A mixed-effects model was developed for different latitudes to analyze how environmental factors such as temperature, precipitation, microhabitat, and local nitrogen deposition affect the response of Sphagnum to nitrogen deposition. The nitrogen deposition saturation values for Sphagnum at different latitudes were determined by linear regression to analyze the differences among latitudinal zones. We explored the response of Sphagnum to nitrogen deposition and demonstrated that excessive nitrogen deposition inhibited Sphagnum growth. The determined nitrogen deposition saturation values for Sphagnum will provide scienti c indicators for arti cial nitrogen fertilization and for maintaining the carbon storage function of Sphagnum plants.

Data acquisition
We summarized data from 66 nitrogen experiments with Sphagnum in 48 experimental peatland areas in Europe, North America, Asia, and South America (Fig. 1). The published data were mainly obtained from the Web of Science database, with "Sphagnum", "nitrogen", and "fertilization" as keywords, and the screening conditions were as follows: (i) The study had to evaluate two or more levels of nitrogen applied to Sphagnum and ensure that the temperature, solar radiation, moisture conditions and other nutrients were consistent with those of the control. (ii) Fertilization studies had to have been carried out in the eld or indoor culture experiments in which materials were collected outdoors. (iii) The variable re ecting the Sphagnum growth must have been production. In addition, this selection of literature was supplemented with a search of Google Scholar, and data from a total of 45 peer-reviewed studies were collected (S3). To avoid publication bias to the greatest extent possible, in addition to the abovementioned published data, a portion of the data was obtained from relevant unpublished experiments shared by Sphagnum researchers. All data were used after receiving the author's permission.
The nitrogen deposition response of Sphagnum has been studied the most in North America and Europe, and Sphagnum tends to grow in mid-and higher latitudes where the environment is cold and humid. Therefore, when latitude is analyzed as a continuous variable, the signi cance of areas with more experimental data and widespread distribution of Sphagnum is weakened, while the signi cance of areas with less experimental data and sparse distribution is ampli ed. This will lead to the nal results showing an insigni cance of latitude, but we believe that the possibility of regularity with latitude cannot be excluded. We divided the experimental data into three latitudinal zones according to the number of experimental data points and the growth habit of Sphagnum to ensure that the amount of data in each latitudinal zone was su cient for meta-analysis, and the results were convincing.

Variable description
We used the annual production of Sphagnum as a response variable; nitrogen application, background nitrogen level, mean annual precipitation, mean annual temperature, mean July temperature (data for the Southern Hemisphere were converted to maximum average summer temperatures), and microhabitat as explanatory variables, of which microhabitat as a categorical variable; and the rest (nitrogen application, background nitrogen level, mean annual precipitation, mean annual temperature, mean July temperature) as continuous variables. The annual production of Sphagnum refers to the annual production or Sphagnum biomass (g.m − 2 .yr − 1 ) and was averaged over several years if the experiment was performed for a few years. Nitrogen application refers to the total amount of nitrogen (g.m − 2 .yr − 1 ) applied to the experimental area in one year or the last year of data collection if the experiment was conducted over several years. Background nitrogen is the amount of wet nitrogen deposition (g.m − 2 .yr − 1 ) that occurred in the experimental area. When data were not available in the literature, relevant nitrogen deposition data for the study area were obtained from local meteorological websites or from other studies. Annual mean precipitation is the annual mean precipitation (mm.yr − 1 ) within the peat bog in which the experiment was located, or if data were not available for that peat bog, data for the area surrounding the bog were obtained. Temperature variables included annual mean temperature and July mean temperature. The mean temperature in July has been shown to have the highest correlation with the summer mean temperature (Limpens, 2011); therefore, it was used as the summer temperature variable in the study area. There were mainly two types of microhabitats, hummock and hollow (include lawn), and the microhabitat variable was used as a measure of the marsh water table.

Methods
Meta-analysis is the quantitative, scienti c synthesis of research results. In a meta-analysis, one or more outcomes in the form of effect sizes are extracted from each study (Jessica, 2018). The effect size represents the degree of difference between the control group and the experimental group and can indicate the changes in Sphagnum growth following increased nitrogen deposition. In this study, log response ratios were used as effect sizes in the meta-analysis. The effect size and within-study variance for each study were calculated based on the mean, sample size, and standard deviation of the response variables using the following formula: Eq. (A.1) is used to calculate the effect size, y i , and Eq. (A.2) is used to calculate the within-study variance, V i , where X e is the mean value of the experimental group, X c is the mean value of the control group, S e is the standard deviation of the experimental group, S c is the standard deviation of the control group, N e is the sample size of the experimental group, and N c is the sample size of the control group. A random-effects model was established in this study, with Sphagnum production as the response variable, and the combined effect size was used as the basis upon which to analyze the effect of nitrogen deposition on Sphagnum growth. On the basis of the random-effects model, a mixed-effects model was established by introducing explanatory variables to analyze the signi cance of each factor to the reaction of Sphagnum to nitrogen deposition, with the mixed-effects model both taking into account the variation among studies and explaining the large heterogeneity in the random-effects model. Veroniki et al. (2019) stated that restricted maximum likelihood (REML) is the best method for calculating interstudy variance for both dichotomous and continuous data because it yields unbiased parameter estimates, so the REML method was used to calculate the interstudy variance here. To identify the overall reaction of Sphagnum to nitrogen deposition across all studies, the combined effect sizes were used, which were weighted as the reciprocal of the within-study variance and the sum of the between-study variances (Ellenberg, 1988). According to the mixed-effects model established to t the regression function (Y = ax + b) to the values of the response variable, the tted function Y represents the combined effect size of the explanatory variables, and x represents the value of the explanatory variable. When the combined effect size (Y) is 0, the response variables are not affected by the explanatory variables, which means that the critical value of the explanatory variables has been reached, from which we can obtain the nitrogen saturation value required for the Sphagnum growth. Since there may be redundancy in the results obtained by including all the signi cant factors in the model, the listed signi cant factors were ranked on the basis of the mixedeffects model, all the possible models were analyzed, and the AIC values were compared among models to identify the optimal model. All the above modeling and testing procedures were implemented in R studio version 3.6.1 using the 'metafor', 'ggplot2', 'glmulti' and 'readxl' packages (Viechtbauer, 2010). Meta-analysis model diagnosis is usually performed with a failsafe number or a funnel plot, both of which can be used to determine the reasonableness and stability of the model. The failsafe number was rst proposed by Rosenthal (1979) to assess the degree of publication bias by calculating the minimum number of unpublished studies needed to overturn the conclusions of the meta-analysis. Meta-analysis results are considered reasonable and highly stable when the calculated failsafe number is greater than 5k + 10 (where k is the number of studies included in the analysis). A funnel plot indicates the accuracy of the model by specifying the symmetry of Egger's regression test for quantitative calculations. When the pvalue is greater than 0.05, the model is considered to have high reliability, with a funnel with strong symmetry indicating high reliability of the conclusion (Egger M, 1998).

Reaction of Sphagnum to nitrogen deposition at different latitudes
The combined effect sizes for all three latitudinal zones were negative, indicating that nitrogen deposition would inhibit Sphagnum production regardless of the latitudinal zone. P-values were greater than 0.05 in all zones except the mid-latitudinal zone, indicating that the results at lower and higher latitudes were not signi cant, which is related to the number of experiments. Data for mid-latitudes were su cient to conclude that the effect of nitrogen was signi cant, but the conclusions for lower latitudes and higher latitudes need more experimental results to determine the signi cance of this effect. The absolute value of the combined effect size decreased with increasing latitude, with a combined effect size of -0.1323 in the lower latitudinal zone, -0.0691 in the mid-latitudinal zone and only − 0.0449 in the higher latitudinal zone (Table 1). The absolute value was low at all but the lower latitudes, suggesting that the inhibitory effect of increasing N application on Sphagnum growth became less pronounced at higher latitudes, probably due to the lower background N levels in these areas.

Analysis of signi cant factors
Due to the regular differences in the combined effect sizes of Sphagnum production at different latitudes, it was necessary to further explore the factors related to the latitudinal reaction of Sphagnum to nitrogen deposition. Therefore, we established optimal models for Sphagnum production at different latitudes, and the results showed that the related factors also differed among latitudes (Table 2). Nitrogen deposition was found to be a signi cant factor in the optimal model for all three latitudinal zones, but the roles of the two nitrogen deposition-related factors (nitrogen application, background nitrogen level) differed at the different latitudes. Production at lower and higher latitudes was positively correlated with the background nitrogen level and negatively correlated with the nitrogen application rate, suggesting that the current rate of nitrogen deposition at higher and lower latitudes can enhance Sphagnum production, while continued nitrogen application or sustained increases in nitrogen deposition may inhibit production. The combined effect sizes of production for background nitrogen were 0.0306 at lower latitudes and 0.0921 at higher latitudes ( Table 2), indicating that background nitrogen application at higher latitudes more strongly promotes Sphagnum production. The combined effect size of production for nitrogen application at lower latitudes was the lowest, indicating that sustained increases in nitrogen application have the greatest impact at lower latitudes. The combined effect sizes of production for both nitrogen deposition factors were negative at mid-latitudes, indicating that the current rate of nitrogen deposition inhibited Sphagnum growth in the region.
At mid-latitudes, in addition to nitrogen deposition-related factors, microhabitat type was also a signi cant factor, with combined effect sizes of -0.1211 for production in hummocks and − 0.0453 for production in hollows, indicating that Sphagnum plants in hummocks have a stronger inhibitory effect when production is suppressed by nitrogen deposition. Moreover, the signi cant factors at higher latitudes were precipitation and temperature, both of which had negative combined effect sizes, suggesting that persistently elevated temperatures and rainfall suppress Sphagnum production. The pvalues for most of the explanatory variables in the model satis ed the t-test and showed signi cance. The higher latitudes in the three optimal models showed the highest percentage of explanation, reaching 85.69%, while the other latitudes had lower explanatory power.

Nitrogen saturation value
In this study, total nitrogen is de ned as the sum of applied nitrogen and background nitrogen. We calculated two types of nitrogen saturation values. One is the saturation value for nitrogen application: when the amount of applied nitrogen exceeds the saturation value, it will inhibit Sphagnum growth. The other is the saturation value for total nitrogen deposition: when the sum of background nitrogen and applied nitrogen exceeds this value, it will inhibit Sphagnum growth. The saturation value for nitrogen application can be applied to the agricultural fertilization and industrial cultivation of Sphagnum, and the saturation value for total nitrogen deposition can provide some basis for research on the physiological nitrogen requirements of Sphagnum. As shown in Table 3 and Fig. 2, the Sphagnum nitrogen saturation values were lowest at mid-latitudes, followed by lower latitudes, and were highest at higher latitudes. The p-values corresponding to the 95% con dence intervals of the t-tests for nitrogen saturation indicated signi cance across all analyses. When the explanatory variable was nitrogen application, both the slope and the intercept were negative in the linear regression function for mid-latitudes, resulting in there being no nitrogen saturation value. The saturation values for nitrogen application were always lower than those for total nitrogen. The difference between them essentially re ects the background nitrogen value in the region, which was higher than the global nitrogen saturation values in all zones except the mid-latitude zone, indirectly indicating the higher background nitrogen level at mid-latitudes.

Model validation
In this study, the failsafe numbers for the lower-and mid-latitude models were 231 (k = 42) and 1391 (k = 137), respectively, and both models passed the test. The higher-latitude model also passed the test, with a funnel plot p-value greater than 0.05. Thus, the model validation results show that our conclusions are reliable, and the funnel plots for each model are detailed in Appendix 5.

Reaction of Sphagnum to nitrogen deposition
The total carbon content of peatlands averages 47 ± 6% (Loisel, 2014), and a large part of this value depends on Sphagnum plants. The acidic, nutrient-poor and water-saturated environment in which Sphagnum species grow makes it di cult for dead plant tissue to decompose (Hayward and Clymo, 1982). In addition, the polyphenol metabolites produced by Sphagnum plants have antimicrobial properties and inhibit microbial decomposition (Jassey, 2013). Therefore, changes in Sphagnum biomass will greatly affect carbon xation in peatlands. The results of this study indicate that excessive nitrogen deposition will inhibit Sphagnum growth (Table 2), whereas nitrogen deposition not exceeding the saturation values can promote Sphagnum growth (Table 3). A meta-analysis by Limpens et al. (2011) also showed that low nitrogen application in areas with low nitrogen deposition promotes Sphagnum production, while high nitrogen application inhibits Sphagnum production. The effects of changes in nitrogen deposition on Sphagnum growth may further affect carbon sequestration. We conclude that when nitrogen deposition promotes Sphagnum growth, this increase in production will slow the decomposition of litter in the environment, thus enhancing carbon sequestration. Conversely, when nitrogen deposition exceeds the saturation value, the effect on carbon sequestration will be reduced.
The inhibitory effect of increasing total nitrogen deposition on Sphagnum production became less pronounced at higher latitudes, which may be related to human activities. The data from lower latitudes included in this study were mainly obtained from eastern Asia, with rapid development occurring at latitudes below 50°, where industrial development over the last two to three decades has resulted in the production of large amounts of nitrogen oxides. Together with the long-term application of nitrogen fertilizers in many places, this has led to the occurrence of excessive nitrogen in adjacent peatlands, such as those in southwestern and northeastern China (Li TT, 2018;Du EZ, 2014). It has been shown that nitrogen deposition in East Asia has increased by 0.6 kg.ha − 1 .yr − 2 over the last 20 years; however, nitrogen deposition peaked in 2010-2012 and is expected to decrease in the future (Geddes, 2017). We suggest that Sphagnum at lower latitudes is currently affected most by nitrogen deposition and that when the environment is managed according to the saturation values identi ed in our study, this effect will diminish, and the carbon sequestration capacity of Sphagnum will eventually recover.
Compared with that in the lower latitudinal zone, the Sphagnum in the mid-latitudinal zone has a relatively lower impact on nitrogen deposition. A study by Fritz et al. (2014) showed that the rate of nitrogen uptake by Sphagnum decreases sharply under prolonged nitrogen stress, which means that high nitrogen uptake may evolve in Sphagnum over the long term. The mid-latitudinal regions in this study mainly occur in Europe and North America, which have experienced a long industrial revolution and where the rate of nitrogen deposition is generally high. Furthermore, nitrogen deposition in North America and Europe has shown an increasing and then decreasing trend over the last century, and Geddes et al. (2017) showed that nitrogen deposition decreased by 0.6 kg.ha − 1 .yr − 2 in North America and 0.1 kg.ha − 1 .yr − 2 in Europe over the last 20 years. It has been shown that Sphagnum detoxi cation of excess nitrogen is slow and requires both energy and products of photosynthesis production. Therefore, a mechanism for the downregulation of nitrogen uptake has evolved in Sphagnum species to prevent excessive uptake of nitrogen and energy loss (Heeschen, 1996;Chiwa, 2018). To adapt to a long-term existence in a highnitrogen-deposition environment, Sphagnum plants will reduce the rate of nitrogen uptake and thus suppress Sphagnum production (Fritz, 2014). Adaption to high nitrogen loads may occur over several years or decades of increased nitrogen deposition. Although nitrogen deposition may decrease due to legislation, this adaptation to high nitrogen will be maintained long after the decrease (Sally, 2006;Philippine, 2008). Nitrogen deposition at mid-latitudes peaked earlier and has been declining for more than 10 years, and the adaptation to high nitrogen gradually disappeared. However, nitrogen deposition at lower latitudes has also reached its peak, but a strong adaptation to the high-nitrogen-deposition environment may still be maintained. Thus, Sphagnum plants at lower latitudes are more signi cantly inhibited by excess N than are those at mid-latitudes. We propose that the adaptation of Sphagnum to high nitrogen at lower latitudes will gradually disappear with reasonable arti cial intervention to reduce nitrogen deposition. This will bene t Sphagnum growth and facilitate carbon sequestration by Sphagnum species.
High latitudes are less inhabited than lower latitudes and remain relatively pristine, with less nitrogen deposition. Furthermore, the productivity of the entire ecosystem at these latitudes is limited by nutrient de ciencies and climatic conditions such as low and freezing temperatures (Diakova, 2016), making such regions less affected by the effects of increasing nitrogen deposition on the carbon sequestration capacity of Sphagnum plants than lower-latitude regions.

Signi cant factors and future trends
The signi cant factors varied by latitude, and the nitrogen deposition-related factors were all found to be extremely signi cant, indicating that nitrogen deposition has a signi cant effect on Sphagnum growth ( Fig. 3.a). In the mid-latitudinal zone, the microhabitat type was also found to be signi cantly related to Sphagnum growth, which was not evident at lower or higher latitudes ( Fig. 3.b). This may be related to the signi cant differences in the retention of nitrogen in different water tables among hummocks, hollows at mid-latitudes. Since Sphagnum production is highly correlated with water table uctuations in the microhabitats in which the plants occur, a relative decrease in water level will result in a signi cant decrease in production (Gaudig, 2020). Higher-latitude areas have the most signi cant factors. In addition to nitrogen deposition, Sphagnum production is closely related to precipitation and temperature in this region due to the low level of human activity and is extremely sensitive to environmental changes ( Fig. 3.c).
In recent decades, there has been an increasing global temperature trend, with a particularly pronounced increase in minimum temperatures (Rosmann, 2016). We found that higher temperatures at higher latitudes will inhibit Sphagnum growth (Fig. 4). Mineralization of peat would likely increase with warming, while greater evaporation at higher temperature leads to a lower water table, which will promote the growth of vascular plants in peatlands, thus increasing shading and eventually suppressing Sphagnum growth (Richard, 2019). A sustained increase in temperature may cause the respiration rate of Sphagnum plants to surpass their photosynthetic rate, leading to a reduction in the synthesis and storage of organic matter and then a weakening of the role of Sphagnum in sequestering carbon in peatlands. With global warming, the annual precipitation range has also increased at mid-to high latitudes (Fig. 4), and the rate of increase in maximum precipitation has exceeded the rate of increase in minimum precipitation (Chou chia, 2012). Our study shows that increased precipitation at higher latitudes will inhibit Sphagnum growth. The average residence time of water in the Sphagnum layer after low-intensity precipitation is 10-30 min, but excessive rainfall > 5 mm may remain for only 0.5-5 min in the uppermost Sphagnum layer (Martina, 2005;Holden J, 2009). Therefore, the higher frequency of extreme precipitation conditions at higher latitudes over the last half century may have resulted in obvious nitrogen leaching, leading to the weakened metabolic activity and growth of Sphagnum plants. Furthermore, the intensi cation of extreme precipitation and extreme cold conditions has been underestimated globally over the last fty to sixty years (Fischer EM, 2014). The combination of these extremes with continued warming and increased precipitation will result in rapid and deep in ltration of rainwater and solutes (Holden J, 2009;Jeremy A,2016;McCarter, 2020), which will signi cantly impact Sphagnum growth and further affect the ability of peatlands to sequester carbon.

Saturation value
Our study shows that there is no saturation value for nitrogen application at mid-latitudes, which indicates that the current nitrogen application rate has already exceeded the nitrogen saturation value required for Sphagnum growth. The total nitrogen saturation values at mid-latitudes are also far below those at lower and higher latitudes (Fig. 5). Thus, a continued increase in nitrogen deposition would inhibit Sphagnum growth. Fortunately, the current mid-latitude nitrogen deposition rate is gradually decreasing. However, the results are also caused by the fact that the distribution of the data in this study is mainly concentrated in Europe, and data from more regions should be incorporated into future analyses. Moreover, the nitrogen saturation values were higher at higher latitudes. In cold, nitrogenstressed ecosystems, Sphagnum plants may derive a large amount of their nitrogen requirement from free amino acids, with a different nitrogen requirement compared to that in other regions (Krab EJ, 2009).
It is possible that Sphagnum plants are in uenced by the microclimate at high latitudes (Krab EJ, 2013). Sphagnum plants also associate with N-xing microbes that could provide nitrogen to their moss hosts (Carella P, 2018;W.K. Sexton, 2020), which needs to be explored further in future studies.
Differences in nitrogen saturation values at different latitudes can be a good indicator of peatland ecological restoration, especially in mid-and lower-latitude areas, which have high nitrogen deposition rates and high levels of human activities. Attention should be paid to agricultural fertilization, in addition to controlling nitrogen oxide emissions. According to the local nitrogen sedimentation background, scienti cally calculating the nitrogen application rate can assist in the development of reasonable fertilization measures, not only providing a basis for ecological protection but also facilitating carbon sequestration in ecosystems by Sphagnum species.

Conclusion
Page 13/23 The results show that excessive nitrogen deposition inhibits Sphagnum growth and that the inhibitory effect of increasing nitrogen application on Sphagnum growth decreases with increasing latitude. The nitrogen saturation value for Sphagnum varies with latitude, with the lowest value found at mid-latitudes. This indicates that mid-latitude areas are the most heavily in uenced by anthropogenic activities and that Sphagnum-containing ecosystems are approaching nitrogen saturation. The reasonable use of nitrogen saturation values for different latitudes can provide a theoretical basis for precision fertilization in agriculture and peatland protection.
Signi cant factors affecting the Sphagnum response to nitrogen deposition differ with latitude, with the exception of nitrogen deposition; microhabitat type is a major factor affecting Sphagnum growth at lower latitudes, with microhabitats possessing a higher water table being favorable to Sphagnum growth. Precipitation and temperature are also important to Sphagnum growth at higher latitudes, and a continuous increase in temperature and precipitation is detrimental to Sphagnum growth. Based on the signi cant nitrogen deposition-related factors affecting Sphagnum growth, a reasonable protocol can be speci ed for Sphagnum culture and further applied in commercial production.
Overall, the interpretation of the model of the Sphagnum reaction to nitrogen deposition remains inadequate due to the presence of unconsidered factors such as sunlight, symbiotic plants and the frequency of extreme weather. The light factor will affect Sphagnum photosynthesis, which directly relates to its growth and production. Symbiotic plants will impact the shading of Sphagnum plants, which will further relate to its light condition, and the nutrient uptake by symbiotic plants will affect the nutrient absorption by Sphagnum plants. Extreme weather, such as drought and freezing, will change the temperature and moisture conditions, which will further relate to the Sphagnum nitrogen uptake and growth conditions. Furthermore, we did not discuss interactions between factors, such as the interaction between temperature and precipitation and that between microhabitat and precipitation. To improve the explanatory ability of the model, subsequent studies could discuss the interactions between climatic and geographical factors. The spatial correlations between signi cant factors should be further investigated by collecting data from additional areas, which will help to improve the understanding of spatial differences in Sphagnum characteristics and of Sphagnum-related mechanisms and facilitate applications of the nitrogen saturation values. The development of high-resolution remote sensing satellite imaging and unmanned aerial vehicle (UAV) technology in recent years has provided the possibility to perform this aspect of research.
Declarations researchers who worked on the nitrogen experiments included in this meta-analysis; without their work, this paper would not have been completed.    Higher-latitude Sphagnum effect sizes were associated with annual temperature and annual precipitation. The dashed red lines represent an effect size of 0, the dashed black lines represent the upper and lower boundaries of the con dence interval, and the solid black lines represent the relationship between the effect size and annual temperature or annual precipitation. Each gray dot represents a nitrogen application study, with larger dots indicating greater weight.

Figure 5
Nitrogen saturation values of Sphagnum at different latitudes