Large-Scale Effects of Aridity on Leaf Nitrogen and Phosphorus Concentrations of Terrestrial Plants

Abstract

The leaf nitrogen (N) and phosphorus (P) concentrations of terrestrial plants make large contributions to ecosystem function and dynamics. The relationship between aridity and leaf N and P has been established through experimental studies. However, few studies have focused on the large-scale effects of aridity on the leaf N and P of terrestrial plants. In this paper, we used linear regression models to test the effects of aridity on terrestrial plant leaf N and P and the N:P ratio based on global datasets. We found that aridity had significant effects on the leaf N and P and the N:P ratio of terrestrial plants. The strongest relationships were between fern leaf P, the fern N:P ratio, tree leaf P, the tree N:P ratio, vine leaf N, and the tree N:P ratio. Aridity could be used to predict the P and N:P ratio of terrestrial plants, particularly those of ferns and trees, on large scales in arid environments. Our study contributes to maintaining ecosystem functioning and services in arid environments under climate change.

1. Introduction

The leaf nitrogen (N) and phosphorus (P) concentrations of terrestrial plants play important roles in ecosystem function and dynamics [1,2,3,4,5]. Leaf N has significant effects on ecosystem processes, for example, photosynthetic rate, litter decomposition, and plant production [3,6,7,8]. Phosphorus is a vital element for the genetic material, energy storage, and cell structure of terrestrial plant species [9,10,11]. The leaf N:P ratio is an indicator of the nutrient condition limited to plant species growth and living [12,13]. Previous studies [3,10,13] have explored the drivers of the leaf N and P of terrestrial plants on a large scale. For example, Chen et al. found that the leaf N and P of woody plants vary, depending on different environmental conditions (i.e., climates and soils) and growth forms [10]. Therefore, leaf N and P are directly related to the physiological and ecological processes of plants.

Aridity is characterized by a severe lack of available water to the extent of hindering the physiology and ecology of plant life [3,14,15,16,17]. Experimental studies [18,19,20,21] have demonstrated that aridity can regulate changes in the leaf N and P of terrestrial plants. Aridity can affect growth rate and energy metabolism in leaves; on a small scale, aridity can affect productivity, such as net primary production and net ecosystem production [22,23]. Furthermore, increased aridity may lead to the loss of ecosystem multifunctionality [24]. Leaf N and P may reduce the negative effects of aridity on ecosystem functions and services as their key functional traits [23,24]. Large-scale studies [10,25,26,27,28] have shown that climatic factors are the drivers of the leaf nitrogen and phosphorus concentrations of terrestrial plants on a large scale. Reich and Oleksyn showed that the global patterns of plant leaf N and P are related to temperature and latitude [29]. However, few studies have indicated whether aridity could affect the leaf N and P of terrestrial plants on a large scale.

The main objective of our study was to explore the effects of aridity on the leaf N and P of terrestrial plants on a large scale. Here, we proposed two hypotheses: (1) large-scale aridity affects leaf N and P, and (2) life forms regulate the effects of large-scale aridity on the leaf N and P of terrestrial plants. Based on the testing of these two hypotheses, we aimed to provide references for ecosystem services and biodiversity management. We obtained global data on leaf N and P from Tian et al. [4] and global aridity data from Zomer et al. [30]. We used linear regression models to explore the effects of aridity on the leaf N and P of terrestrial plants across different life forms.

2. Materials and Methods

We obtained a global dataset on the leaf N and P of terrestrial plants from Tian et al. [4]. This dataset includes leaf N and P and the N:P ratio belonging to 11,354 individual records based on 1291 field sites; the dataset incorporates records from 201 families, 1265 genera, and 3227 species on a global scale [4]. The species belong to five life forms, namely, ferns, herbs, shrubs, trees, and vines [4]. We averaged the values of leaf N and P and the N:P ratio across different plant species at each investigation site based on ferns, herbs, shrubs, trees, and vines. Then, we used log10 transformation for leaf N and P and the N:P ratio for further analysis. The distribution of the studied sites is shown in Figure 1.

The global aridity data (Global-AI_ET0) were downloaded from the Global Aridity Index and the Potential Evapotranspiration (ET0) Climate Database v3 [30] (https://doi.org/10.6084/m9.figshare.7504448.v3; (accessed on 15 Octorber 2022)). Global-AI_ET0 is available annually as one grid layer over the period from 1970 to 2000 [30]. This dataset shows moisture availability for the potential growth of reference vegetation and excludes the impact of soil-mediating water runoff events [30]. The spatial resolution of the aridity index is 30 arc seconds (~1 km at the square). The aridity index ranges from 0 (arid) to 1 (humid) [30]. Furthermore, we selected the bioclimatic variables, including the annual mean temperature, temperature seasonality (standard deviation × 100), annual precipitation, and precipitation seasonality (Coefficient of Variation; https://www.worldclim.org/data/index.html; (accessed on 15 Octorber 2022)).

We used linear mixed-effects models to explore the effects of aridity on the leaf N and P of terrestrial plants [31]. The leaf N and P and the N:P ratio were the response variables, the aridity index was the exploratory variable, and the species group was the fixed factor [31]. The simple and mixed regression models defined the relationships between aridity and leaf N and P and the N:P ratio with log10 transformation across different life forms (i.e., ferns, herbs, shrubs, trees, and vines). We ran the mixed and linear regression models (unitary regression models or bivariate regression models) in JMP 11 (SAS Institute, Cary, NC, USA). We used a high R² to quantify the large-scale effects of aridity on the leaf nitrogen and phosphorus concentrations of terrestrial plants. We used unitary regression models to explore the effects of other environmental factors (i.e., bioclimatic variables) on leaf nitrogen and phosphorus concentrations.

3. Results

There were significant relationships between aridity and leaf N and P and the N:P ratio on a large scale based on the linear mixed-effects models (p < 0.1). Furthermore, these relationships were significant for ferns, herbs, subshrubs, shrubs, trees, and vines (p < 0.1;

Table 1. The R² of relationships between aridity and leaf N and P and N:P ratio based on bivariate regression models.

Life Form	Leaf N		Leaf P		N:P Ratio
	R2	p-Value	R2	p-Value	R2	p-Value
Fern	0.062	0.725	0.371	0.099	0.562	0.016
Herb	0.004	0.427	0.016	0.026	0.008	0.140
Shrub	0.010	0.032	0.071	0.000	0.082	0.000
Tree	0.030	0.065	0.110	0.000	0.117	0.000
Vine	0.173	0.071	0.031	0.648	0.218	0.032

). The R² values were higher based on the bivariate regression models than the unitary regression models (Table S1). Based on the bivariate regression models, we found that the significant relationships (p < 0.1; Table 1) between aridity and leaf N and P and the N:P ratio depend on different life forms (Figure 2 and Table 1).

Leaf N was positively related to aridity for shrubs, trees, and vines (p < 0.1; Figure 1 and Table 1). There were significant relationships between leaf P and aridity for shrubs, trees, and vines (i.e., increasing and then decreasing), but the relationships were the opposite for ferns (p < 0.1; Figure 2 and Table 1). Aridity had significant effects on the leaf N:P ratio of ferns, shrubs, trees, and vines (p < 0.05; Figure 2 and Table 1). Positive and then negative effects exist in ferns and vines, and the opposite effects exist for shrubs and trees (Figure 2). The largest relationships for aridity were between fern leaf P (R² = 0.154; p < 0.1; Table 1), the fern N:P ratio (R² = 0.562; p < 0.05; Table 1), tree leaf P (R² = 0.148; p < 0.05; Table 1), the tree N:P ratio (R² = 0.140; p < 0.05; Table 1), vine leaf N (R² = 0.173; p < 0.1; Table 1), and the tree N:P ratio (R² = 0.218; p < 0.05; Table 1). We found that bioclimatic variables had larger effects on leaf N, leaf P, and the N:P ratio for ferns, shrubs, trees, and vines (p < 0.1; Table S1). Annual mean temperature had the largest effects on vine N (R² = 0.312; p < 0.05; Table S1); temperature seasonality had the largest effects on tree P (R² = 0.234; p < 0.05; Table S1); and annual precipitation had the largest effects on fern P (R² = 0.280; p < 0.1; Table S1), followed by precipitation seasonality (R² = 0.239; p < 0.1; Table S1).

4. Discussion

The significant relationships between aridity and leaf N and P and the N:P ratio indicate that aridity has large-scale effects on nutrient acquisition. Furthermore, these relationships depend on different life forms. The effects of aridity on leaf P and the N:P ratio could widely exist across different life forms (e.g., ferns and trees), but such effects may be weak for leaf N on a large scale. Numerous experimental and field analysis studies [18,19,20,21,22,23] have shown that aridity may affect leaf N and P and the N:P ratio. Our study provides modeled analysis evidence for the relationships between aridity and leaf N and P for terrestrial plants on a large scale. Berdugo et al. indicated that aridity could drive global ecosystem thresholds [32]. Hence, our study could support the assessment of ecosystem functioning and services on a large scale under climate change.

Although we found that the effects of aridity on leaf N were weak on a large scale, aridity can play an important role in the regulation of the leaf N of terrestrial plants on a large scale. Berdugo et al. showed that drastic reductions in leaf N concentration may occur at levels of aridity lower than 0.65 [32]. Hence, we should pay attention to the changes in leaf N in the drying lands around the world. Our results show that there were significant relationships between aridity and leaf N for terrestrial plants on large scales. Hence, the changes in leaf N have a large potential to affect ecosystem functioning and services (e.g., climate regulation, nutrient cycling, and livestock production) in drying regions [24,32,33,34]. To maintain the stability of ecosystem functioning and services, we should monitor aridity levels, as per the suggestions of Berdugo et al. [32].

Our results demonstrate that there were significant relationships between aridity and leaf P and the N:P ratio across different life forms on a large scale. Leaf P is directly related to the physiology of terrestrial plant species and can affect the ability of terrestrial plant species to adapt to aridity [9,10,11,12,13]. Nitrogen and phosphorus are widely recognized as the most important limiting nutrients controlling plant growth, ecosystem functioning, and terrestrial ecosystems dynamics [1,35,36]. Both leaf P and the N:P ratio are effective indicators of nutrient limitations for terrestrial plant species along the aridity gradient [1,14,17,23]. Furthermore, these two leaf indicators could be used to monitor the stability of ecosystem functioning and services [24]. For example, Reich et al. showed that leaf P can play an important role in the relationships between photosynthesis and nitrogen based on a cross-biome analysis on numerous species [37]. The relationship between photosynthetic capacity and nutrients may be constrained by low P in drying environments (Figure 2). Hence, the detection of leaf P and the N:P ratio is important for modeling vegetation nutrient stocks and cycling in arid environments. Leaf nitrogen concentration (LNC) is a key parameter of vegetation photosynthetic efficiency and yields, and it can be applied for scientific guidance in nitrogen (N) fertilization management [37,38,39,40,41,42]. A fast and accurate estimation of crops’ LNC is vital to indirectly understand crops’ growth status. Similarly, nitrogen is also a component of the chlorophyll molecule, which enables a plant to capture sunlight energy by photosynthesis, driving plant growth and grain yield [40]. Nitrogen plays a critical role within a plant to ensure energy is available when and where the plant needs it to optimize yield [37,40,41,42]. In contrast, phosphorus promotes early root growth, winter hardiness, and seed formation; stimulates tillering; and increases water use efficiency [40].

Furthermore, we found that the largest relationships existed between aridity and fern leaf P, the fern N:P ratio, tree leaf P, the tree N:P ratio, vine leaf N, and the tree N:P ratio; this indicates that aridity can affect leaf P and the N:P ratio for ferns, trees, and vines. The photosynthetic capacity and nutrient content of ferns depend on climatic changes at temporal and spatial scales [43,44]. Based on our results, the effects of climatic factors (e.g., annual mean temperature, temperature seasonality, annual precipitation, and precipitation seasonality) on the photosynthetic capacity and nutrient content of ferns were due to the limitation of leaf P caused by aridity. Furthermore, leaf P and the N:P ratio could affect the photosynthetic capacity and nutrient limitation of trees in arid environments [43,44,45].

Aridity is an important indicator of climate change [30]. Our results provide important references for the impacts of climate change on species and ecosystems at a global scale. Although temperature and precipitation are the drivers of plant functional trait distribution under climate change, aridity could be applied as an assessment of the effects of climate change on plant functions, closely associated with ecosystem functions and services, and biodiversity management [26,27,30,43,44,45,46]. Therefore, aridity monitoring should be widely used to evaluate ecosystem functions and services benefiting human beings under climate change.

5. Conclusions

Aridity significantly affects the leaf N and P and the N:P ratio of terrestrial plants on large scales. However, the effects are weak for leaf N and strong for leaf P and the N:P ratio. Aridity could be used to model P and the N:P ratio of terrestrial plants, particularly for ferns and trees on large scales in arid environments. Our study provides new evidence for the relationships between aridity and leaf N and P and the N:P ratio for terrestrial plants on large scales. These results contribute to monitoring ecosystem functioning and services under climate change.