Soil C:N:P Stoichiometry in Two Contrasting Urban Forests in the Guangzhou Metropolis: Differences and Related Dominates

Abstract

Carbon (C) sequestration and nitrogen (N) and phosphorus (P) accumulation in urban forest green spaces are significant for global climate regulation and alleviating nutrient pollution. However, the effects of management and conservation practices across different urban forest vegetation types on soil C, N, and P contents and stoichiometric ratios remain largely unexplored. We selected forest soils from Guangzhou, a major Metropolis in China, as our study area. Soil samples were collected from two urban secondary forests that naturally regenerated after disturbance (108 samples) and six urban forest parks primarily composed of artificially planted woody plant communities (72 samples). We employed mixed linear models and variance partitioning to analyze and compare soil C, N, and P contents and their stoichiometry and its main driving factors beneath suburban forests and urban park vegetation. These results exhibited that soil pH and bulk density in urban parks were higher than those in suburban forests, whereas soil water content, maximum storage capacity, and capillary porosity were higher in urban forests than in urban parks. Soil C, N, and P contents and their stoichiometry (except for N:P ratio) were significantly higher in suburban forests than in urban parks. Multiple analyzes showed that soil pH had the most pronounced negative influence on soil C, N, C:N, C:P, and N:P, but the strongest positive influence on soil P in urban parks. Soil water content had the strongest positive effect on soil C, N, P, C:N, and C:P, while soil N:P was primarily influenced by the positive effect of soil non-capillary porosity in suburban forests. Overall, our study emphasizes that suburban forests outperform urban parks in terms of carbon and nutrient accumulation, and urban green space management should focus particularly on the impact of soil pH and moisture content on soil C, N, and P contents and their stoichiometry.

1. Introduction

Ecological stoichiometry is the study of quantitative relationships among multiple chemical elements in ecology systems and their influence on biological processes and ecological functions [1,2]. Carbon (C), nitrogen (N), and phosphorus (P), as fundamental elements for life on Earth [3], regulate plant development and ecological balance by driving biogeochemical cycles and energy flows, thus maintaining ecosystem functionality and stability [4,5,6,7]. Soil C, N, and P concentrations and stoichiometric ratios are often used to assess nutritional limitations, key ecological processes, and elemental interactions, influencing both plant nutrient absorption and soil nutrient availability [8,9]. Soil C:N:P stoichiometry serves as a valuable metric for analyzing and interpreting changes in plant–environment relationships within ecosystems, as well as for assessing nutrient dynamics [10,11,12]. Therefore, exploring the contents and stoichiometric ratios of soil C, N, and P could improve our comprehension of nutrient limitations and biogeochemical cycling in terrestrial ecosystems.

The impact of human activities on soil C, N, and P contents and stoichiometry is multifaceted, with land use practices and agricultural management measures being the most prominent influences [13,14,15]. Alterations in land utilization and related management strategies may influence soil properties, and plant and microbial communities, thereby altering elemental cycling processes and C:N:P ratios, establishing a new dynamic equilibrium within the soil system [16,17]. The effects of alterations in land utilization on soil physicochemical properties and the balance between soil C, N, and P inputs and outputs is largely due to alterations in vegetation types. Different plant compositions influence soil C, N, and P dynamics directly or indirectly via plant litter decomposition, root exudation, biological nitrogen fixation, and mineralization, as well as resident animals, insects, and microbes [18,19,20]. Furthermore, different species exhibit variations in their ability to utilize and absorb key soil elements such as soil C, N, and P [9,21,22]. Various analyses suggested that land use history, vegetation type, and management practices are strongly linked to soil C accumulation and nutrient dynamics [23,24].

In this study, we selected the urban forests of China’s metropolis (Guangzhou) as our study area. Specifically, we collected soil samples from six urban artificially planted forest parks (72 samples) and two urban secondary forests (108 samples). We aimed to address the following two questions: (1) What are the differences in soil physicochemical properties, as well as in soil C, N, and P contents and their stoichiometry, of urban forest parks and urban secondary forests? (2) What are the main factors driving the differences in soil C, N and P contents and stoichiometry between urban forest parks and urban secondary forests? By comparing and analyzing soil samples from these two different environments, we aim to investigate how different urban forest types affect soil C, N, and P contents and their stoichiometric ratios. We hope to provide more targeted recommendations for urban forest ecosystem management, especially in the areas of nutrient management and sustainable development of urban green spaces.

2. Materials and Methods

2.1. Study Site

We conducted investigations in forests originating from two contrasting land use types: one involving old growth forest land that was cleared and then planted with trees and other plants for urban parks, and the other consisting of old growth forest land that was clear-cut for timber and then abandoned for natural succession. Here, we define the former as urban plantations and the later as natural secondary suburban forests. We selected forest stands planted in six urban parks (Huadu, Yushu, Maanshan, Haibing, Xunfengshan, and Yuexiu) and two suburban semi-natural secondary forest stands (Baiyunshan and Dalingshan) in Guangzhou, China (22.755–23.391° N, 113.617–113.238° E; Figure 1). These urban parks and forests have not undergone reconstruction or human logging for many years, and their vegetation remains well preserved. The region experiences a typical monsoon humid climate, with an average annual air temperature of 12.7 °C and a mean annual precipitation of 1450 mm.

2.2. Soil Sampling and Measurement

In July 2023, we collected a total of 72 soil samples from 6 urban parks across three soil depths: 0–10 cm, 10–20 cm, and 20–30 cm. Among them, 18 samples were collected from Yushu, 18 from Xufengshan, and 9 from each of the remaining 4 forest parks. Additionally, 108 soil samples were collected from 2 urban forests—90 from Baiyunshan and 18 from Dalingshan. Non-continuous standard plots were randomly established in each urban park and urban forest, ensuring a minimum distance of 150–200 m between plots.

Soil total C was measured using an elemental analyzer by high-temperature combustion of soil samples. Total N was measured using the Kjeldahl acid digestion method. Total P was determined by the sulfuric acid hydrolysis procedure. Soil pH was determined in 1:2.5 (w/v) soil water suspensions using a glass electrode. We used the ring knife method to quantify soil moisture and physical properties, included soil bulk density, soil water content, soil maximum storage capacity, and soil capillary porosity.

2.3. Statistical Analyses

All metrics were analyzed and compared using their mean values and standard errors (SE). A one-way analysis of variance (ANOVA), followed by Tukey’s HSD test, was used to examine the impacts of two different urban forests on C, N, and P contents and stoichiometry, as well as on soil physicochemical properties. Additionally, a two-way ANOVA was performed to examine the individual and combined effects of study site; soil depth; and their interaction on C, N, and P contents and stoichiometry, as well as on soil physicochemical properties. Pearson’s correlation analysis was applied to examine the relationships between C, N, and P contents and their stoichiometric ratios and physicochemical properties. Statistical significance was defined at p < 0.05. All ANOVA and Pearson’s correlation analyses were carried out using SPSS version 19.0 (SPSS Inc., Chicago, IL, USA).

Prior to fitting subsequent models, z-score standardization was applied to all metrics to ensure model convergence. To eliminate multicollinearity, variables with a variance inflation factor (VIF) exceeding 3 were omitted from further analyses [25]. The VIF was calculated using the R package ‘car’ [26]. Linear mixed-effects modeling (LMM) was fitted using the lme() function from the R package ‘nlme’ [27]. The R script employed was ‘model = lme(C, N and P contents and stoichiometry~effect variables, random = ~1|soil depth)’, where soil depth was included as a random effect to account for autocorrelation. Restricted maximum likelihood was used to generate unbiased parameter estimates. We calculated the relative importance (RI) of each variable based on standardized effects (β) from the LMM. These statistical analyses were carried out using R version 3.5.3 [28].

3. Results

3.1. Changes in Soil Physicochemical Properties of Two Urban Forests

Soil bulk density and pH values were consistently higher in urban parks than in urban forests across all soil layers (Figure 2a,b). Soil water content and maximum storage capacity were higher in urban forests than in urban parks (Figure 2c,d). Soil capillary porosity was greater in urban forests than in urban parks (Figure 2e), whereas soil non-capillary porosity exhibited only minor differences between the two urban forest types (Figure 2f). Soil bulk density, pH value, maximum storage capacity and capillary porosity increased with soil depth, while soil water content and non-capillary porosity decreased in both urban parks and forests (Figure 2).

3.2. Changes in Soil C, N, and P Contents and Stoichiometry of Two Urban Forests

Soil C and N contents were substantially higher in urban forests than in urban parks (Figure 3a,b), whereas soil P content was greater in urban parks than in urban forests (Figure 3c). Soil C:N, C:P, and N:P ratios were higher in urban forests than in urban parks (Figure 3d–f). Additionally, soil C, N, and P contents and stoichiometric ratios gradually reduced with increasing soil depth in both urban parks and urban forests (Figure 3).

3.3. Relationships Between Soil Properties and C, N, and P Contents and Stoichiometry in Two Urban Forests

In these urban parks, a bivariate analysis showed that soil bulk density displayed a pronounced negative correlation with soil C, N, C:P, and N:P (all p < 0.01) and was significantly positively linked to soil P (p < 0.05). Soil pH exhibited significant negative associations with soil C, N, C:P, and N:P (all p < 0.001), as well as with soil C:N (p < 0.05), but was positively correlated with soil P (p < 0.001). Soil water content was significantly negatively associated with soil N (p < 0.05). Soil maximum storage capacity and capillary porosity were significantly negatively correlated with soil C, N, C:P, and N:P (all p < 0.05). Soil non-capillary porosity exhibited significant positive correlations with soil C, N, C:P, and N:P (all p < 0.001) and a negative correlation with soil P (p < 0.05; Figure 4a).

In these urban forests, soil bulk density showed significant negative correlations with soil C, N, and P contents and stoichiometric ratios (all p < 0.001). Soil pH exhibited a significant negative correlation with soil C, N, C:N, C:P, and N:P (all p < 0.01). Soil water content was strongly positively correlated with soil C, N, P, and C:N (all p < 0.001) and also positively related to C:P (p < 0.01) and N:P (p < 0.05). Soil maximum storage capacity and capillary porosity were significantly positively correlated with soil C, N, P, and C:N (all p < 0.05). Soil non-capillary porosity was significantly positively related to soil C, N, P, and C:N (all p < 0.001) but negatively correlated with soil C:P and N:P (p < 0.01; Figure 4b).

3.4. Key Determinants of Soil C, N, and P Contents and Stoichiometry in Two Urban Forests

After eliminating multicollinear variables, the LMM results for urban parks showed that soil pH had the strongest negative effect on soil C, N, C:N, C:P, and N:P (all β < −0.31, p < 0.001; RI > 44.16%) but exhibited the strongest positive effect on soil P (β = 0.56, p < 0.001; RI = 75.08%). Soil non-capillary porosity exerted a significant positive impact on soil C, N, C:P, and N:P (all β > 0.50, p < 0.05; RI > 25.97%) but showed no significant influence on soil P and C:N (p > 0.05). In contrast, soil water content showed a significant negative effect on soil C:N (β = −0.24, p < 0.05; RI = 42.40%) and exhibited no significant influence on other soil indices (p > 0.05) (Figure 5).

In contrast, within urban forests, soil water content had the strongest positive influence on soil C, N, P, and C:N (all β > 0.26, p < 0.05; RI > 48.75%) but was not significantly associated soil C:P and N:P (p > 0.05). Soil non-capillary porosity had a significant positive influence on soil C, N, and P (all β > 0.14, p < 0.05; RI > 22.15%) but showed no significant impact on soil C:N, C:P, and N:P (p > 0.05). Soil pH had a significant positive effect on soil P (β = 0.15, p < 0.01; RI = 13.64%) but showed no significant effect on other soil indices (p > 0.05) (Figure 5).

4. Discussion

4.1. Differences in Soil Properties and C, N, and P Contents and Stoichiometry in Two Urban Forests

Numerous studies have demonstrated that vegetation restoration, as a key approach to controlling soil and water erosion, is closely associated with soil development, physicochemical properties, element cycling, and microbial formation [8,29,30,31]. Our results showed that urban forests had higher soil water content, maximum storage capacity, soil bulk density, and capillary porosity, whereas urban parks had higher soil pH and non-capillary porosity (Figure 2), emphasizing that differences in vegetation cover and management practices can significantly influence soil physicochemical properties. Meanwhile, our findings indicated that soil C and N contents in urban forest systems were significantly greater than those in urban forest park systems across all soil layers (Figure 2a,b). The connection between aboveground inputs and belowground microbial metabolism during vegetation restoration is crucial for understanding soil dynamics in long-term restoration processes [32,33,34,35]. Plant diversity, litter biomass, root exudates, and microorganisms are direct or indirect sources of soil C and N [36,37,38,39,40]. Soil C and N derived from plants and microorganisms in urban forests that have been undisturbed for a long time are generally higher than those in urban parks. Our findings suggest that maintaining an undisturbed or minimally disturbed soil structure in suburban natural secondary forests can enhance the accumulation of soil C and N compared to urban plantation forest parks. However, soil P content in urban forest systems were markedly lower than urban forest park systems in all soil layers (Figure 3c). This finding suggests that soil P declines with soil development progress, aligning with the findings from previous studies [11,41]. Our results revealed that soil C:N:P stoichiometric ratios were significantly higher in urban forest systems compared to urban forest park systems across all soil layers.

Meanwhile, our results demonstrated that in both urban forests and urban parks, soil bulk density, pH, maximum storage capacity, and capillary porosity increased markedly with increasing soil depth, whereas soil water content and non-capillary porosity decreased significantly (Figure 2). As soil depth increases, organic matter content generally decreases, while pressure from overlying layers compacts the soil, resulting in higher bulk density. This compaction alters soil pore structure by reducing the proportion of large pores (non-capillary pores) and increasing that of small pores (capillary pores), thereby enhancing the soil’s maximum water storage capacity while decreasing its actual water content. Furthermore, the increase in soil pH with soil depth may result from reduced inputs in organic acids and lower biological activity in deeper layers. In addition, we found that soil C, N, and P contents and their stoichiometry decreased dramatically with increasing soil depth (Figure 3). In the organic-rich soil layer, our analysis focused on surface soil (typically 0−10cm). Of them, the C:P ratio declined much more rapidly than the C:N ratio with increasing soil depth. This is mainly because of the relatively stable soil P content throughout the soil profile, in contrast to the rapid decline in C and N content with soil depth. Low soil P content always led to high C:P and N:P ratios, suggesting that soil C:N:P stoichiometry is mainly controlled by the P supply [42].

4.2. Key Determinants of Soil C, N, and P Contents and Stoichiometry Under Two Urban Forests

Previous studies demonstrated that soil physicochemical properties primarily regulated C, N, and P contents and their stoichiometry [43,44,45]. Our findings revealed that soil pH negatively influenced soil C, N, C:P, and N:P in urban park systems (Figure 4a), while it negatively influenced all C, N, and P contents and their stoichiometry in urban forest systems (Figure 4b). In urban parks, soil pH had a negative effect on soil C, N, and P contents and their stoichiometry (except for soil N:P ratio) and emerged as the most influential factor in the mixed-effects model (Figure 5). Prior findings have shown that soil pH negatively affects soil C, N, and P contents and their stoichiometry by inhibiting microbial activity, reducing nutrient availability, increasing the rate of organic carbon decomposition, and decreasing nutrient storage. Together, urban forest park management should focus on regulating soil pH to promote microbial activity and nutrient availability, thereby improving soil nutrient status and ecological function.

In suburban natural secondary forests, soil water content was strongly positively associated with all soil C, N, and P contents and their stoichiometry (Figure 4b) and emerged as the most influential factor after accounting for other variables (except for soil N:P) in this mixed model (Figure 5). This evidence together suggests that in undisturbed urban forests, soil moisture exerts a stronger influence on soil C, N, and P contents and their stoichiometry. Adequate moisture directly affects microbial activity and metabolism, facilitating nutrient mineralization and transformation, thereby influencing soil C, N, and P contents and their stoichiometry. Diverse plant communities and root activities provide nutrient sources; promote microbial growth; and significantly increase the C, N, and P content in the soil under sufficient moisture conditions. Maintaining soil moisture by reducing human disturbances is crucial for the ecological health of urban natural secondary forests. Conversely, in urban plantation forest parks, bivariate analysis indicated that soil water content was significantly positively correlated with only N and generally had a weak variance explanation for soil C, N, and P contents and their stoichiometry (except for C:N) in the mixed-effects model (Figure 4a and Figure 5). The main reason is that soil in urban parks has been artificially modified, leading to stable nutrient input, and artificial irrigation maintains relatively stable soil moisture, which weakens the influence of soil water content on soil C, N, and P content and stoichiometry.

Soil porosity directly influences soil aeration, water permeability, and root growth and serves as a vital regulator in soil moisture, nutrient availability, temperature, and microbial activity [46,47,48]. Plant root systems, organic matter content, soil physical structure, precipitation, and gardening interventions (e.g., artificial watering, nutrient enrichment, and plant trimming) influence both soil total porosity and non-capillary porosity both directly and indirectly [49,50,51]. Our analysis demonstrated that soil capillary porosity had a significant negative link with soil C, N, C:P, and N:P in urban plantation forest parks (Figure 3a). Two reasons underlie our findings. Firstly, vegetation in urban parks is dominated by lawns and shallow-rooted plants, resulting in reduced organic matter input. Secondly, frequent human interventions (pruning, leaf removal, and trampling) limit organic matter accumulation, and artificial irrigation alters water dynamics and accelerates nutrient leaching. However, soil non-capillary porosity was significantly positively related to C, N, P, C:P, and N:P in urban parks (Figure 4a). The positive relationship is likely because soil non-capillary pores enhance soil aeration, promote microbial diversity and enzyme activity, and increase nutrient build-up and availability. In contrast, soil capillary porosity and non-capillary porosity were significantly and positively related to soil C, N, and P contents and stoichiometry in urban forests (Figure 5b). Compared to urban parks, urban forest ecosystems have higher species richness, well-developed root structures, stable soil structures, and greater litter quantity and quality. These factors alter soil pore structure and enhance soil moisture retention capacity and nutrient cycling, ultimately increasing soil C, N, and P contents and altering C:N:P stoichiometric ratios. Overall, urban park management should reduce human disturbances, enhance organic matter accumulation, and implement appropriate irrigation. However, urban forest management should focus on protecting biodiversity and soil structure to promote soil porosity and nutrient cycling.

5. Conclusions

This study evaluated the differences in soil properties and C–N–P element cycling between various urban park plantations and suburban natural secondary forests in Guangzhou, China. Our findings revealed significant differences in soil physicochemical properties, as well as in soil C, N, and P contents and stoichiometric ratios between urban parks and urban forests. Specifically, in urban parks, soil pH had the strongest negative effects on soil C, N, C:N, C:P, and N:P, whereas in suburban forests, soil water content had the strongest positive effects on soil C, N, P, C:N, and C:P. These key results suggest that future urban green space management should consider the differences between urban parks and urban forests and regulate soil pH and soil moisture to promote and stabilize soil carbon sequestration and nutrient cycling. Naturally, reducing human disturbances and implementing effective management measures are also very important.

Of course, this study also has certain limitations, such as the lack of monitoring data across different seasons and the absence of long-term dynamics observations for the two forest types. In addition, it did not systematically integrate plant community composition and microbial functions to elucidate the potential mechanisms underlying C–N–P element cycling. Therefore, future studies should combine vegetation, microbial communities, and multi-season long-term observations to conduct in-depth explorations, thereby providing more robust scientific evidence for urban ecosystem restoration and sustainable management.