Relationship between ecological stoichiometry and community diversity of plant ecosystems in the upper reaches of the Tarim River, China

Aim Although it is commonly proposed that nutrient cycling can impact plant community diversity, this relationship has not been fully examined in arid and semi-arid zones. Here, we expand on the framework for evaluating the relationship between biodiversity and ecological stoichiometry by scaling up from the level of the community. Location The upper reaches of the Tarim River (Northwest China, 80°10’-84°36’E, 0°25’-41°10’N). Methods We used multivariate analysis of variance to compare the stoichiometric characteristics and species diversity indices of sampled plant communities. We also measured carbon (C), nitrogen (N), and phosphorous (P) content of plants. We then assessed correlations between community stoichiometry and species diversity through structural equation models (SEM) and redundancy analysis (RDA). Results We found that the differences between stoichiometric characteristics and community diversity indices were highly significant. The Margalef index was strongly related to C and P content. The Simpson’s index and Shannon-Weaner index were most strongly correlated with C content. Pielou’s index was closely related to C and N contents, and the C:N and C:P ratios were important at driving ecological dominance. Main Conclusions Our study highlights the importance of ecological stoichiometry in driving community assembly and diversity within a desert ecosystems in northwestern China. The relationship between eclogical stoichiometry in the desert plant community had an effect on species diversity, and it was a good indicator of plant community diversity.


60
Ecological stoichiometry is an important tool for studying ecological processes diversity is a complex measure of community structure and function, and it is also the 83 material basis for maintaining ecosystem stability and sustainable production (Vilela D 84 S et al. 2016;Pelini S L et al. 2016;Zhang C H et al. 2017;Saitta A et al. 2018). 85 Therefore, the use of ecological stoichiometry knowledge to study plant communities 86 and their relationship to species diversity can help to reveal the ecological strategies 87 of plants in specific environments, and is of great significance to the circulation and 88 balance mechanism of community materials. 89 Understanding nutrient element utilization of plant has long been focus of plant 90 ecology (Frost P C et al. 2012;Yan W et al. 2016). Since the 1980s, ecological 91 stoichiometry has been applied to ecology for the first time (Reinhardt S B et al. 92 1986 ), it has become an important method for plant ecology research, and has been 93 widely praised by ecologists all around the world(Allen A P. 2009; Castellanos A E et 94 al. 2018;Yu H et al. 2017;Yan W et al. 2016 ). Most of the researches focuses on 95 different ecosystem such as grassland ecosystem, forest ecosystem, wetland 96 ecosystem and so on (Yang Y H et al. 2011;Cao Y et al. 2017;Graż yna P et al. 97 2018). And the rare researches that pay attention to desert ecosystems. Many The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/432278 doi: bioRxiv preprint 5 103 communities are not the same. 104 The C:N:P ratio is also considered to be one of the major drivers of biodiversity 105 (Sardans J et al. 2011). For over a decade, evidence has accumulated suggesting a 106 general trend towards the effect of community ecological stoichiometry on plant 107 species diversity (Kerkhoff A et al. 2005;Liu Z et al. 2018). Importantly, this work 108 suggests that community ecological stoichiometry can be important for plant species were to investigate whether desert plant community stoichiometry accounts for 130 characteristics of trees, shrubs, and herbs, and to determine whether these 131 relationships have any consequences for community diversity.   T1sample zone were T11,T12,T13,T14,T15 and T16. T2 sample zone and T3 sample   151 zone naming rules were the same as T1 sample zone. Plots were spaced at intervals of 152 at least 1 km within each sample zone, so as to avoid problems with 153 pseudoreplication. This design resulted in a total of eighteen sampling plots. Plots 154 were spaced at intervals of at least 1 km within each sample zone, so as to avoid 155 problems with pseudoreplication. This design resulted in a total of eighteen sampling 156 plots.

157
Different plot sizes were used to sample trees, shrubs and herbs. Trees were 158 sampled within 20 × 20 m 2 plots, while shrubs and herbs were sampled within 10 × 10   In the above equations, S is the total number of species in a plot, P i is the ratio of 185 the number of individuals in a species to the total number of individuals in a 186 community. N is the total nuber of individuals in a plot.

208
The characteristic of plant community stoichiometry in different riparian zones 209 We found C concentration was greatest in T13 community in T1 zone and lowest in . CC-BY 4.0 International license author/funder. It is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/432278 doi: bioRxiv preprint 9 210 T31 community in T3 zone. Nitrogen concentration was greatest in T23 and was 211 lowest in T14. Phosphorous concentration was greatest in T34 and was lowest in T22.

212
All of the stoichiometric ratios (i.e., C:N, C:P and N:P) were greatest in zone 1 and 213 were lowest in zone 3 (Table 2). Significant difference in the stoichiometric 214 concentrations and stoichiometric ratios were observed among same zones.

215
Significant differences in stoichiometry and stoichimetric ratios were also observed 216 among the different zones. In addition, N and P concentrations were not significantly 217 different among T15, T25 and T35 types ( Table 2).

218
An extremely significant negative correlation but high slope was found between 219 plant community C:N ratio and N concentration, and between C:N ratio and N:P ratio

231
The T23 and T11 communities had higher Pielou's index values than the other 232 communities (Fig.3). Overall, the same general trends observed in the T1 zone plant 233 community diversity incies generally greater than T2 and T3 plant community 234 indices.

235
The correlation between plant community stoichiometry and plant community 236 diversity . CC-BY 4.0 International license author/funder. It is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/432278 doi: bioRxiv preprint 10 237 We used RDA to examine community diversity in more detail. The first axis 238 explained 64% of the variation, while the second axis only explained 2%. As such,  Simpson's index was strongly positively correlated with P concentration, and was 245 weakly positively correlation with C and N concentrations (Fig. 4). In contrast, the 246 C:N and N:P ratios had a strong negative correlation with Simpson's index, and the 247 C:P ratio showed a weak negative correlation (Fig. 4). A clear positive correlation 248 emerged between the Shannon-Wiener index and C, N and P concentrations (Fig. 4).

249
Likewise, a negative correlation among these ratios was found, and the remaining 250 indices showed these simliar patterns as well (Fig. 4).

251
Based on the above studies, it can be seen that the effects of concentrations and 252 ratios on plant community diversity index were different, but it is not possible to 253 conclude that the concentration or ratio had the greatest impact on plant community 254 diversity index. Therefore, we performed a Monte-Carlo test on three concentrations 255 and three ratios, and obtained the order of importance of the stoichiometric variables.

256
The results are shown in Table 3. The important sequencing indicated that the most 257 important element was N concentration (Table 3). We found that the N:P ratio was 258 not as important for plant community diversity indices (Table 3). In order to further 259 explore how C and N concentrations and C:N and C:P ratios potentially affect plant 260 community diversity indices, we conducted a correlation analysis between single 261 concentrations or ratios and plant community diversity indices. N, P concentrations 262 and C:N, C:P ratios had great effect to four diversity indices. So to further explore the 263 effects of these four important factors on the single plant community diversity index, 264 We conducted a test of the impact of key concentration or ratio on the diversity   In general, N and P are assumed to be the two most limiting elements to plant . CC-BY 4.0 International license author/funder. It is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/432278 doi: bioRxiv preprint 13 322 growth (Elser et al. 2000). We found a significant difference in N:P ratios among 323 plant communities. Specifically, the N:P ratio were greater than 16 in some 324 communities, indicating that these communities were strongly P-limited, and that      Table 2 The characteristic of C,N,P concentrations and ratios in different      Where R 2 represents the degree of fit, the closer R 2 is to 1, the better the fit. P 691 represents the correlation, P < 0.05 represents a significant correlation, P < 0.01 indicates the correlation, the smaller the angle indicate the stronger correlation. When 702 the angle is between 0° and 90°, there is a positive correlation between the two 703 variables; when the angle is between 90° and 180°, there is a negative correlation 704 between the two variables; when the angle is 90°, it means There is no correlation . CC-BY 4.0 International license author/funder. It is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/432278 doi: bioRxiv preprint 31 705 between the two variable.  The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/432278 doi: bioRxiv preprint