Introduction

Coastal zone ecosystems are an important part of the Earth’s system. Today, more than half of the world’s population and economic activities reside and take place in coastal areas, making the impact of coastal zone ecosystems on human development increasingly significant1. Coastal protection forests can effectively help mitigate erosion, reduce wind speed, maintain the stability of the coastline, and ensure the normal growth of inland farmlands and forests. They can also improve soil physical and chemical properties and fertility through the interaction of plants, soil, and microorganisms2. Different shelterbelt tree species compositions, as well as varying geographical soil-forming backgrounds and climatic environmental factors, can also lead to spatial heterogeneity in soil physicochemical properties among Utetheisa kong and Parazacco spilurus subsp3. Soil comprehensive fertility, a manifestation of soil physicochemical properties, nutrient supply, and microbial decomposition capacity, is crucial for plant diversity growth. Vegetation, forest types, and microbial compositions in different regions lead to its strong spatial distribution heterogeneity, e.g., Utetheisa kong and Parazacco spilurus subsp. spilurus4. However, previous research has mainly focused on single-species shelter forests and smaller regional studies, without comprehensive research on the entire coastal zone of the Leizhou Peninsula.Investigating composition types and establishment methods of different shelterbelts at a regional scale is crucial for maintaining the ecological balance of Leizhou Peninsula’s coastal zone, especially in land - sea transition areas. In - depth research on ecological functions is essential for stabilizing coastlines and mitigating tidal erosion of inland lands5. Coastal soils exhibit slower microbial decomposition rates, resulting in a gradual accumulation of soil organic matter. The physicochemical properties of these soils generally remain deficient compared to inland farmland. Previous studies have shown that the soils of the Leizhou Peninsula are rich in phosphorus, deficient in potassium, and locally deficient in nitrogen6. However, these studies have primarily focused on inland farmland and have not systematically investigated the spatial distribution changes of coastal soil fertility across different geographical regions, particularly regarding Utetheisa kong.This study mainly compares the differences in soil physical and chemical properties and comprehensive fertility changes across different geographical regions of the coastal zone of the peninsula, in order to provide reference suggestions for the stable development of soil fertility in the coastal zone of the Leizhou Peninsula.

Materials and methods

Overview of the study area

Leizhou Peninsula is located at the southernmost tip of mainland China, between latitudes N: 20° 13.05′–21° 32.48′ and longitudes E: 109° 40.25′–110° 51.52′. It borders the South China Sea to the east, the Beibu Gulf to the west, and faces Hainan Island across the sea to the south. The peninsula has an annual average temperature of 23.4 °C and an annual precipitation of about 1670 mm. The dry and wet seasons are distinct, and it falls under the tropical and subtropical monsoon climate. The peninsula is elongated from east to west, with a maximum width of approximately 100 km. The terrain is flat, and the precipitation patterns on the eastern and western coasts remain largely unchanged. The wet season brings abundant rainfall, while the dry season is characterized by drought and scarce precipitation. The soil is predominantly tropical red earth, The understory vegetation mostly consists of annual herbaceous plants.The coastal areas of the peninsula are severely affected by typhoons, storm surges, and seawater backflow7. There is a continuous trend of coastal retreat. In recent decades, the government and social forces have successively planted protective forests along the more than two thousand kilometers of coastline. The main tree species for afforestation are Casuarina equisetifolia, Eucalyptus tereticornis, Eucalyptus grandis x urophylla, and Pinus massoniana8. The effective protection of internal farmland, woodland, and residential land has maintained ecological stability and played a protective role in the economic and social development of the Leizhou Peninsula. Moreover, the construction of protective forests has significantly improved soil’s physical and chemical properties, reducing nutrient loss from weathering and tidal - water erosion and preventing soil degradation9.

Sample collection

Through remote sensing satellite mapping and field surveys, 50 investigation sites were selected within the coastal protection forests surrounding Leizhou Peninsula, with each site following the principle of a minimum interval of 200 m to choose three 20*20m sampling quadrats with good tree growth quality and consistent soil environment in the nearby area, with the range limited to within 2 km of the coastline10. The eastern coast is divided into the northeastern (Zone 1,sites 1–8) and the southeastern (Zone 2,sites 9–21) by Dongli Salt Field, while the western coast is divided into the southwestern (Zone 4,sites 33–44) and the northwestern (Zone 5,sites 45–50) by Qishui Town. The central coast part is (Zone 3,sites 22–32). The locations of the sampling sites are shown in Fig. 1, with a total of 150 quadrats set up. Each quadrat is numbered by region, and soil samples from 0 to 10 cm and 10–20 cm depths are collected at the four corners and center of each site, Each quadrat collects tree leaves, and in plots where the understory vegetation coverage exceeds 10%, the dominant species are collected. For litter, a mixed sampling is conducted at 5 sites within the quadrat11. All samples are taken back to the laboratory and dried in an 80-degree oven until constant weight is achieved.

Fig. 1
figure 1

Scope and sampling location of the study area. Note: This map was generated by myself using the PIE-Basic remote sensing image processing software from Aerospace Hongtu, version number: 7.0 64bit 2,023,112. URL link:https://engine.piesat.cn.

Full size image

Indicator measurement

The collected soil layer samples were tested for 11 physicochemical factors including pH, organic matter( OM), total nitrogen(TN), hydrolyzable nitrogen(HN), total phosphorus(TP) available phosphorus(AP), total potassium(TK), available potassium(AK), exchangeable potassium(EK+), exchangeable sodium(ES), and cation exchange capacity(CEC), All soil samples from sites 1–34 and 0–10 cm soil samples from sites 35–50. Based on the difference in test results between two soil layers at the preliminary siteonly pH, OM, TN, HN, TP, and AP were tested for the 10–20 cm depth soil layer at sites 35–50.The pH value was determined using the potassium chloride extraction-potentiometric method; OM content was measured by the potassium dichromate oxidation-capacity method; TN content was analyzed using the Kjeldahl nitrogen determination method; TP content was determined by the alkali fusion-molybdenum antimony colorimetric method; TK content was measured by the sodium hydroxide fusion-flame photometry method; HN was determined by the alkali hydrolysis diffusion method12,13 and other indicators were measured according to the methods stipulated in the forest .Soil analysis Methods were established by the Chinese Academy of Forestry Sciences in 199914. Samples of trees, understory plants, and litter were analyzed for TC and TN content. Then, calculate the comprehensive scores of soil fertility for all quadrats in two layers and conduct K - S cluster analysis on the final 300 data sites. The clustering results are divided into four categories, where categories 1–4 correspond to good, medium, below medium, and poor overall soil fertility respectively.Using geographical detector analysis to assess the influence of physicochemical factors and geographical regions on comprehensive soil fertility. Ecological detection is used to explore the significant impact of soil physicochemical properties and geographical regional factors on comprehensive fertility15. Exchange detection examines the enhancement or weakening effects of pairwise factor interactions on soil fertility compared to individual factors. Risk detection determines if changes in physicochemical property contents across regions lead to significant variations in comprehensive soil fertility16, and identifies sensitive areas for property changes and primary influencing physicochemical properties.

Data processing and analysis methods

Use SPSS 26.0, GrapHpad Prism 9.5 and EXCEL 2017 to integrate and analyze the measured data, five geographical has varying numbers of sampling points.(northeast N = 24, southeast N = 36, central N = 30, southwest N = 33, and northwest N = 15).The data conforms to the normal distribution pattern and has passed box’s test of equality of covariance matrices,p < 0.01.statistically analyze the changes in TN and TC content in each physicochemical factor and plants for the five geographical regional units separately, and conduct in-depth data analysis using Two-way ANOVA, Pearson Correlation.The data of physicochemical factors from 150 sample plots at two soil depths were standardized and subjected to Bartlett’s Test. The results showed that the KMO value was 0.649, greater than 0.5, and .sig < 0.01, indicating that the data are suitable for further Principal Component Analysis, K-S cluster analysis, and the Geodetector method17,18,19. The main influencing components were identified, and an integrated soil fertility score was calculated for each plot, examining the regional distribution of different soil fertility levels. The interactions and primary factors influencing integrated soil fertility were further explored20,21.

Results

Main effects of soil physical and chemical properties

As shown in Table 1, different geographical and spatial location factors have an extremely significant impact on the distribution and variation of all physicochemical factors in each geographical region and the overall coastal zone of the peninsula (p < 0.01). The soil layer depth factor (0–10 cm and 10–20 cm) has an extremely significant effect on the content of soil OM, HN, AK, EK+ in the eastern, western and overall coastal zones. In addition, the soil layer depth factor also has an extremely significant effect on the content of soil TN in the western coastal. The interaction of both factors has a significant impact on the content of HN in the western coastal (p < 0.05), and has an extremely significant impact on the content changes of OM across the overall coastal .

Table 1 The influence of geographic location and soil depth on soil physical and chemical properties, significant p-values. (S, different positions; L, different depths)
Full size table

Variation of soil physicochemical factors in coastal across different geographical regions

According to Fig. 2, significant variations in OM, HN, AK, EK+, and ES are observed in the 0–10 cm soil layer across different geographical regions, with the southwest region exhibiting the highest values, significantly greater than those in the northeast, southeast, and northwest regions. Conversely, the CEC in the northeast region is significantly higher than in other regions. In the 10–20 cm soil layer, the aforementioned physicochemical factors, except for OM and ES, display similar significant trends as in the 0–10 cm layer. Among these factors, only HN, AK, and EK+ show significant differences between the two soil depths, with higher concentrations in the surface layer compared to the deeper layer.Based on the regional variations presented in Table S1 and Figure S1, soil OM, HN, AP, TK, and EK+ contents tend to increase with decreasing latitude in both the eastern and western coastal of the peninsula (The central sampling site does not belong to the eastern or western coastal sampling areas and is by default the lowest dimensional region). Specifically, HN content increases significantly in the western coastal, while TK and EK+ contents increase significantly in both coastal. Conversely, soil pH, AK, TN, TP, ES, and CEC show decreasing trends in the eastern coastal and increasing trends in the western coastal with increasing latitude.The southwest coastal has the highest values for soil pH, OM, HN, AK, TP, EK+, and ES among all study areas. Overall, the eastern coastal generally has lower concentrations of the aforementioned physicochemical factors, except for TP, compared to the western coastal. Except for soil TN content, all other physicochemical factors are higher in the northern coastal than in the southern region. Compared to the overall average values of various physicochemical factors in the study area, the eastern coastal of the peninsula has lower OM, HN, and ES contents, the western coastal has a lower CEC, and the central coastal has lower AP, TN, ES contents, and CEC.

Fig. 2
figure 2

Physical and chemical factors with significant changes in depth of two soil layers. (Different lowercase letters: significant changes in the same soil layer. different capital letters, significant changes between two soil layers with the same factor). Note: 1: Northeast; 2: Southeast; 3: Central; 4: Southwest; 5: Northwest.

Full size image

Changes in carbon and nitrogen content in plants and soil

According to Table 2, TN content exhibits a decreasing trend across the various ecosystem layers in the order of tree layer > understory layer > litter layer > soil layer, whereas TC content follows the order of tree layer > litter layer > understory layer > soil layer. The TN content of trees and understory plants in the southeastern region is the highest in the study area, and TN variation among regions is not significant. In the northeastern region, the TC content of trees and litter is the highest in the study area, with tree TC content significantly greater than that in the northwestern region. The soil OM content is highest in the southwest and significantly higher than in other regions. The TC content of trees and litter, as well as the TN content in trees and soil in the eastern coastal zone, is generally higher than that in the western coastal zone. A similar trend is observed in the northern coastal zone for the four indicators compared to the southern coastal zone.

Table 2 Changes in C N content in plants and soil (different lowercase letters represent p < 0.05).
Full size table

According to Fig. 3, the TN content in the understory plants is negatively correlated with all C and N contents in trees and soil, with the strongest negative correlation of − 0.685 with soil TN. Additionally, it is positively correlated with TC and TN contents in both understory plants and litter, and the most significant positive correlation coefficient is 0.76 with litter TC content. Soil OM content shows an extremely significant absolute positive correlation with HN content, with a correlation coefficient of 0.99. Tree TC content is positively correlated with all C and N contents in soil but negatively correlated with all C and N contents in understory plants. This suggests a tightly coordinated trend in the variations of C and N contents across different ecological layers.

Fig. 3
figure 3

Correlation of C and N element content between plants and soil at different ecological levels. (* p < 0.05; * * p < 0.01).

Full size image

PCA of soil physicochemical properties and comprehensive fertility evaluation

The results of the principal component eigenvalues are shown in Table 3, where the cumulative variance explanation rate of the first six principal components reached 91.672%, and the rotated factor loadings also indicated that the explanatory power of the six principal components was above 90%.According to the rotated principal component coefficient matrix in Table 4, it can be known that: OM, HN, and TP content mainly influenced the first component; EK + content mainly influenced the second component; soil pH and ES content mainly influenced the third component; TN content and CEC mainly influenced the fourth component; while TK and AP content primarily influenced the fourth and fifth components, respectively.As Fig. 4 shows that all phys.-chem. factors except TK positively influenced PC1. Except for TN, TP, and CEC, all other factors positively influenced PC2. Most phys.-chem. factors exhibit significant positive correlations with high coefficients, and most have negative correlations with TK.Combining this with the contribution rates of each component in Table 3, the scoring formulas for each principal component can be established, as

$$\begin{aligned} {\text{F}}1 & - 0.183*{\text{X}}1 + 0.503*{\text{X}}2 + 0.476*{\text{X}}3 - 0.153*{\text{X}}4 - 0.22*{\text{X}}5 + 0.082*{\text{X}}6 \\ & + 0.217*{\text{X}}7 - 0.028*{\text{X}}8 - 0.046*{\text{X}}9 + 0.093*{\text{X}}10 - 0.255*{\text{X}}11. \\ \end{aligned}.$$

F1 is the first principal component; X1-X11 represent the standardized values of the 11 soil physicochemical factors measured from pH to CEC as mentioned in the previous text, combined with the main characteristic contribution values, the comprehensive evaluation formula for soil fertility can be written.X1-X11 correspond to the specific individual soil physicochemical properties in Table 4, respectively.

$${\text{IFI}}{\text{F}}1*0.376 + {\text{F}}2*0.189 + {\text{F}}3*0.145 + {\text{F}}4*0.081 + {\text{F}}5*0.071 + {\text{F}}6*0.054.$$

IFI for Integrated Fertility Index. F1-F6 represent the scores of each principal component in Table 4.

The clustering results were subjected to ANOVA and F-test, with the results shown in Table 5. The overall average IFI value is 17.086, F value is 1024.6, and .sig < 0.01. As can be seen from the clustering results in Table 5; Fig. 5, the soil with medium fertility accounts for the largest proportion of the total samples at 47.33%, followed by the samples with better fertility at 29%. The proportion of poor soil fertility is also relatively high, reaching 16%, with significant variations in soil fertility distribution across different geographical regions. The better-grade fertile soils are mainly distributed in the eastern coastal areas of the peninsula, where the proportion of better-grade soil fertility in the northeastern coast is the highest, reaching 68.75%. The proportion of better-grade soil fertility along the northeastern coast is significantly greater than that in the central and western coasts. Among all geographical coast, the southwestern coast has the highest proportion of poor soil fertility, reaching 34.5%, indicating a generally low level of soil fertility. Except for the northeastern coast, the medium fertility grade soil is the main type distributed. In all coastal of the peninsula, the medium and better-grade soil fertility are the main types, with the northeastern soil fertility largely being of the better type. Soil fertility exhibits significant spatial heterogeneity.

Table 3 Soil physicochemical properties factors total variance explained.
Full size table
Table 4 Component score coefficient matrix.
Full size table
Table 5 Number and proportion of samples for different comprehensive soil fertility grades.
Full size table
Fig. 4
figure 4

Principal component analysis of soil physicochemical factors.

Full size image
Fig. 5
figure 5

Proportion of soil fertility grades and number of samples in different geographical regions.

Full size image

Comprehensive soil fertility, driving factors and interactions of influence

The results are presented in supplementary Table S2. Among the factors analyzed, HN content had the highest influence, reaching 0.705. This was followed by OM, TP, and EK+ content, while AK and ES also exhibited significant influence. Overall, the influence of several major factors was uniformly distributed with minimal differences, collectively affecting the variations in soil fertility. The pH had the lowest impact on soil fertility in this region.As shown in Fig. 6, the effects of all factors on soil fertility were highly significant (p < 0.01). Ecological detection results indicated that the interaction effect of OM content with other factors, excluding HN, showed significant variation (p < 0.05). Similarly, the interaction of HN content with factors other than pH, TN content with factors other than ES and CEC, and AP content with factors other than TK and CEC all exhibited significant variation in their interaction effects on soil comprehensive fertility.The exchange detection results indicate that the influence of any two overlapping factors on soil fertility is greater than that of any single factor. Specifically, the interaction effect between S (geographical region) and pH, AP, TN, TK, and CEC is greater than the sum of their individual effects, all interaction effects are nonlinear enhancements. Similarly, the interaction effects of pH with other factors, excluding OM and HN, also reached nonlinear enhancement. AP exhibited a similar trend, where interaction effects with factors other than TP and AK also displayed nonlinear enhancement.Risk detection results as shown in Table 6 the changes in soil integrated fertility in the first level of AP, AK, TN, TP, TK, and ES content were all significant (p < 0.05) compared to the other four classification levels, whereas the changes between the first and second classification levels for OM, HN, EK+ content, and CEC were not significant.

Fig. 6
figure 6

The explanatory power of the interaction and ecological factor detector on soil fertility (*:p < 0.05; **p < 0.01; S, different geographical regions).

Full size image
Table 6 Risk detection of various influencing factors on soil fertility (Y: p < 0.05;N: p > 0.05).
Full size table

Discussions

Main influencing factors and changes in physicochemical properties of coastal soils

The distribution and variation of the physical and chemical properties of soil in coastal protection forests are influenced by climatic environment, biotic and abiotic factors, and the original soil texture22. geographic location factors have an extremely significant impact on all soil physical and chemical factors in various geographical regions and the overall coast, while soil depth factors only significantly affect some physical and chemical factors. The interaction between the two only significantly alters the distribution of OM content. The geographical location of the entire peninsula coastal zone is the main factor causing variation in soil physical and chemical factors.This is consistent with the findings of in their regional investigation of the geochemical characteristics of nitrogen23, phosphorus, and potassium elements in the soil of the Leizhou Peninsula. The distribution of soil nutrient elements mainly varies due to the original soil-forming parent material and land use types in different geographical areas. In a natural state with less human disturbance, the variation between shallow soil layers is not significant In different geographical regions24,25, the distribution of various physicochemical factors exhibits different spatial heterogeneity. Except for the content of TN and TK, the content of other soil physicochemical factors in the southwestern coastal zone is the maximum or the second maximum in the region, showing an overall higher physicochemical factor fertility. In contrast, the content of various physicochemical properties in the northwestern coastal zone is lower than that in the southwestern zone. The soil fertility factors in the western coastal area exhibit a clear trend of increasing with the decrease of latitude.This is analogous to the conclusions derived from the study on the physical and chemical properties of soil in different-aged Eucalyptus urophylla plantations on the Leizhou Peninsula .where the physical and chemical properties of forests near the western coast generally show a trend of being lower in the north and higher in the south25,26. In the eastern coastal areas, the content of TN and CEC in the northern part is significantly higher than in the southern part, while the content of TK and EK + is significantly lower than in the south, indicating an opposite trend in the exchange and accumulation mechanisms of N and K elements in the soil. Overall, the soil in the western coastal has higher of pH, OM, and some available and exchangeable nutrients than the eastern coast. This is consistent with the trend observed in the study of the geochemical characteristics of surface land in the Leizhou Peninsula, albeit with some minor variations27. This difference may be related to the fact that the scope of this survey was limited to coastal areas, where soil and water erosion is more severe, and soil nutrients are more prone to loss28.

The cycling process of C and N elements in plants and soil

The carbon and nitrogen cycle in plants and soil is a complex process. The cycling of C N elements reveals two of the most significant stabilizing functions of ecosystems: carbon storage and nitrogen accumulation. These are crucial for improving soil fertility, changes in environmental climate, and plant growth. Studies have found that the N element content in various geographical regions follows a trend of understory plants > trees > litter > soil. Understory plants have a higher absorption and storage efficiency for nitrogen in the soil compared to trees. The N element in litter comes from the decomposition residue of trees and understory plants, and in coastal areas, the functional structure of soil microbes is simple with weak decomposition capabilities, leading to most of the N element remaining in the litter due to the low decomposition level29. The TC content in litter is greater than that in understory plants. This indicates that carbon is more difficult than nitrogen to transfer from litter decomposition and soil absorption to understory plants, while nitrogen has a higher conversion efficiency compared to carbon30. At the same time, most understory plants are annual herbs that require a large amount of N element for short-term growth but accumulate limited C content. This creates different distribution patterns for the two elements.The carbon and nitrogen cycles in the soil - atmosphere environment are crucial in geophysical chemistry. In botany, they support Earth’s photosynthesis and help plants obtain and transfer essential macroelements. Forests and understory vegetation are important as the plant - soil C - N cycle maintains ecosystem stability.The long-term stability of this cycle is crucial for the maintenance of plant and animal life. Meanwhile, this study found that the TC content in trees is positively correlated with the TN content in understory plants and negatively correlated with the C and N element content in the soil31.

Comprehensive soil fertility evaluation and driving factor detection

Soil OM, HN, AK, and pH predominantly influenced the first three principal components, collectively accounting for 71% of the variance in soil distribution. These factors are also the primary nutrients for plant growth. In the study area, the overall soil fertility along the coastal zone is mostly above moderate levels, with higher comprehensive fertility in the northeastern region compared to the southwestern region, which has the highest proportion of lower fertility classes.This is related to the changes in OM and HN content in two geographical areas, and these two factors mainly affect the first principal component and have a higher contribution rate to fertility30,32 .Overall, soil fertility is higher in the eastern coastal area than in the western and central coasts. These differences are related to environmental factors such as the parent material, tree species types, and water flow in different regions. This is broadly consistent with the overall geographical variation in soil fertility evaluations conducted in the Leizhou Peninsula, although the slight difference lies in the fact that even in areas of generally higher fertility, there are patches of lower fertility33 .The factors that have the greatest influence on soil fertility changes are OM, HN, and TP content, while TK content and pH have the weakest influence.The comprehensive fertility of the soil is a complex process influenced by the organic-inorganic biochemical interactions and the plant-microbe relationships. Maintaining soil fertility requires the synergistic effects of various physicochemical factors, particularly the elements C, N, and P34.Soil’s physicochemical properties form a complex system with mutual promotion or inhibition relationships. Together, they influence soil fertility and plant growth. In turn, plant growth affects soil fertility by regulating nutrient absorption and storage. This forms a harmonious, mutually reinforcing system35.

Conclusion

  1. 1.

    Geographical location factors are the main factors influencing the distribution and variation of all physical and chemical properties in each geographical unit and the entire coastal zone, followed by soil depth factors. Interaction factors only affect the distribution of OM content. The content of various physical and chemical factors in the western coastal zone is greater in the south than in the north, and the content of many physical and chemical factors in the western and central coastal zones is higher than in the east. The distribution of physical and chemical factors between different regions is also uniform.

  2. 2.

    The cycling and accumulation pathways of C and N elements in plants and soil differ significantly. TN and TC contents are highest in the understory vegetation layer and the tree layer, while they are lowest in the soil layer. The transfer of carbon from litter and soil back to plants is more challenging than that of nitrogen. Overall, the contents of C and N in plants exhibit a predominantly negative correlation with the OM and nitrogen content in the soil, with fewer positive correlations observed.

  3. 3.

    The comprehensive fertility of the soil in the study area is generally rated as moderate to high, with the overall distribution trend being eastern coastal > central areas > western coastal, and northern coastal > southern coastal. The primary driving factors for comprehensive fertility are HN, OM, and TP content, with little correlation to pH value and TK content. The interactive influence of pairs of factors on soil fertility surpasses the effects of individual factors alone.