*Article* **Landscape Analysis of Runoff and Sedimentation Based on Land Use/Cover Change in Two Typical Watersheds on the Loess Plateau, China**

**Xiaojun Liu <sup>1</sup> and Yi Zhang 2,3,\***


**Abstract:** Understanding sedimentation and runoff variations caused by land use change have emerged as important research areas, due to the ecological functions of landscape patterns. The aims of this study were to determine the relationship between landscape metrics (LMs), runoff, and sedimentation and explore the crucial LMs in the watersheds on the Loess Plateau. From 1985 to 2010, grassland was the dominant landscape in the Tuweihe (TU) and Gushanchuan (GU) watersheds. Unused land and cropland experienced the greatest transformations. The landscape in the study area tended to become regular, connected, and aggregated, represented by increasing of the Shannon's diversity index and the largest patch index, and decreasing landscape division over time. The landscape stability of the TU watershed was higher than that of the GU watershed. Annual runoff and sedimentation gradually decreased and a significant relationship was found between water and soil loss. Due to larger cropland area and lower landscape stability in the GU watershed, the sedimentation of the two watersheds were similar, even though the runoff in the TU watershed was greater. There were stronger effects of LMs on runoff than that on sedimentation yield. The Shannon's evenness and the patch cohesion index was identified as the key factors of influencing water and soil loss, which had the greatest effects on runoff and sedimentation. Results indicated that regional water and soil loss is sensitive to landscape regulation, which could provide a scientific understanding for the prevention and treatment of soil erosion at landscape level.

**Keywords:** land use/cover change; landscape; runoff; sedimentation; Loess Plateau

#### **1. Introduction**

Water and soil loss is a serious problem across the globe and can influence both the biological and physical properties of soil, particularly those related to infiltration rates, nutrient storage, overland flow velocity, and overall soil productivity [1–3]. Environmental services and ecological equilibrium are threatened when soil loss is greater than soil production [4]. About 75 billion tons of soil is eroded every year from terrestrial ecosystems across the world [5], and approximately half of the land is affected by water and soil loss [6]. Water and soil loss is, therefore, an important research area.

Previous studies have shown that vegetation can control soil erosion and help retain runoff. Many studies [7–10] have shown that a high vegetation cover can control water erosion. When rainwater falls on soil, the canopy, roots, and litter components of the vegetation can retain water, weaken the impact of splash erosion, and slow down runoff velocity. These processes of runoff and sediment production are affected by the soil structure, land use type, and vegetation growth patterns [11]. The types and changes of vegetation are the critical factors affecting water and soil loss [12]. A decrease in vegetation cover may result

**Citation:** Liu, X.; Zhang, Y. Landscape Analysis of Runoff and Sedimentation Based on Land Use/Cover Change in Two Typical Watersheds on the Loess Plateau, China. *Life* **2022**, *12*, 1688. https:// doi.org/10.3390/life12111688

Academic Editor: Dmitry L. Musolin

Received: 30 August 2022 Accepted: 14 October 2022 Published: 24 October 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

in a growth in erosion problems [13]. For example, deforestation and land reclamation on slopes can accelerate runoff and sedimentation [14]. There is a lot of evidence linking forest clearance and continual cultivation resulting in serious soil erosion [15] because cultivation can change soil properties, such as soil aggregates, permeability, nutrient content, etc., which increases the likelihood of soil erosion [16]. The composition and types of land cover are closely related to runoff process characteristics and sediment yield [17,18]. Excessive land development may weaken the protective action of vegetation on water and soil retention, and encourage runoff and soil erosion. Examples of irrational land uses are planting olive orchards in the Alqueva reservoir region [19], leaving land unused, the inappropriate planting of vineyards [20], and land abandonment. Studies of land use/cover change could help us understand the characteristics of runoff and sedimentation variations, improve eco-environmental stability, and promote the sustainable utilization of water and soil resources.

Water and soil loss is strongly related with land use in landscapes [21]. The relationship between runoff, sedimentation, and vegetation have received attention in recent years [22–24]. The spatial configuration and composition of plant communities has become a vital and widely applied factor in studies of the geomorphological processes related to erosion [25]. Patch level landscape analyses have indicated that forests, shrubland, and grassland patches lead to better soil properties and have consequently reduced runoff and sediment yield [26,27]. Changes in landscape pattern could have a large impact on erosion [28]. The current landscape distributions or variations can be characterized by landscape metrics (LMs), which were classified into three levels that are patch, class, and landscape level. Natural conditions and human disturbances can be remarkably reflected by landscape, including configuration, composition, and topography [29]. Assessing water and soil loss via key environmental parameters and quantifying the respective influence of LMs can facilitate the development of water and soil quality management strategies [30]. For example, Silva [31] found that LMs are sensitive to changes in the soil surface when erosion occurs. In the upper Du River watershed, LMs were found to account for almost 65% and 74% of the variation in soil erosion and sedimentation yield, respectively. In a previous study, four main contributing LMs were highlighted that were closely related to the variations in the erosion modulus [32]. Shi et al. [33] identified several LMs that were the main indices that influenced watershed soil erosion and sediment yield using partial least-squares regression. A recent study identified the largest patch index of farmland and the landscape index of forest as the key LMs for preferred landscape planning to protect the water quality [34]. Therefore, LMs can be used for both geomorphic evaluations and quantifications of water and soil when they are subject to runoff and sedimentation inputs [4]. Although quantitative research has analyzed the impact of LMs on soil and water loss, it is still not clear whether the impact is more significant for LMs on soil loss or water loss. In addition, the reason that caused the influence differences of the LMs on water and soil loss between different regions is subject of debate.

Severe soil erosion and water loss in the Loess Plateau in China has attracted widespread attention, since it restrained local socio-economic development and seriously threatened environmental security [35]. It is particularly challenging to establish the relationship between LMs, and runoff and sedimentation on the Loess Plateau. The semi-arid landscapes of the Loess Plateau are water-limited due to the high evaporation and relatively low rainfall. Therefore, this area is particularly sensitive to a deterioration in environmental quality. Investigating the quantitative relationships between LMs and water and soil loss is crucial if soil erosion is to be prevented in these seasonally affected environments [36]. This is of particular importance when attempting to predict runoff and sedimentation.

#### **2. Materials and Methods**

#### *2.1. Research Methodology*

By field investigation, data collection, and processing, this paper firstly explored the land use and LM changes of time series in two typical comparative watersheds; secondly, a

relation between runoff and sedimentation and LMs at the landscape level was derived by combining ecologically significant LMs; and, finally, the dominant LMs that influence water and soil loss were verified, and the difference exhibited from the two watersheds was discussed. by combining ecologically significant LMs; and, finally, the dominant LMs that influence water and soil loss were verified, and the difference exhibited from the two watersheds was discussed. *2.2. Study Area* 

By field investigation, data collection, and processing, this paper firstly explored the land use and LM changes of time series in two typical comparative watersheds; secondly, a relation between runoff and sedimentation and LMs at the landscape level was derived

*Life* **2022**, *12*, 1688 3 of 14

#### *2.2. Study Area* The Tuweihe and Gushanchuan rivers are tributaries of the Yellow River and are

**2. Materials and Methods**  *2.1. Research Methodology* 

The Tuweihe and Gushanchuan rivers are tributaries of the Yellow River and are located on the right bank of the middle stream. The two watersheds are located between 109◦260–110◦05<sup>0</sup> E and 38◦180–39◦26<sup>0</sup> N, and have areas of 4503.40 and 1263.11 km<sup>2</sup> , respectively (Figure 1). Their elevations range from 743 to 1517 m, with a high terrain in the northwest and a low topography in the southeast. They are affected by the northern temperate continental monsoon climate, and the two watersheds are arid and semi-arid regions, with annual mean temperatures of 8.5 and 7.3 ◦C, respectively. Their annual rainfall amounts are 417.4 and 430 mm, respectively. Concentrated rainfall occurs in the summer, and high evaporation and high intensity storms are the main reasons for the runoff and sedimentation losses in the watersheds. Quaternary loss is widespread in the hilly and gully regions where there is serious wind–water erosion. Two deep fully developed gullies have been cut and their erosion moduli are 2244 and 3299 t km–2 a −1 , respectively [37,38]. located on the right bank of the middle stream. The two watersheds are located between 109°26′–110°05′ E and 38°18′–39°26′ N, and have areas of 4503.40 and 1263.11 km2, respectively (Figure 1). Their elevations range from 743 to 1517 m, with a high terrain in the northwest and a low topography in the southeast. They are affected by the northern temperate continental monsoon climate, and the two watersheds are arid and semi-arid regions, with annual mean temperatures of 8.5 and 7.3 °C, respectively. Their annual rainfall amounts are 417.4 and 430 mm, respectively. Concentrated rainfall occurs in the summer, and high evaporation and high intensity storms are the main reasons for the runoff and sedimentation losses in the watersheds. Quaternary loss is widespread in the hilly and gully regions where there is serious wind–water erosion. Two deep fully developed gullies have been cut and their erosion moduli are 2244 and 3299 t km–2 a**−**1, respectively [37,38].

**Figure 1.** The location of Tuweihe and Gushanchuan watershed, China. **Figure 1.** The location of Tuweihe and Gushanchuan watershed, China.

#### *2.3. Research Methods 2.3. Research Methods*

#### 2.3.1. Data Sources 2.3.1. Data Sources

A digital elevation model (DEM) dataset was provided by the Geospatial Data Cloud, the Computer Network Information Center, Chinese Academy of Sciences A digital elevation model (DEM) dataset was provided by the Geospatial Data Cloud, the Computer Network Information Center, Chinese Academy of Sciences (http://www. gscloud.cn, accessed on 28 February 2020). It was processed by the Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model (ASTER GDEM) Version 1 and the spatial resolution was 30 m when the Universal Transverse Mercator (UTM, 49N) projection was used. The DEM data were subjected to a mosaic and clipping process, which allowed the study area to be generated. Then, the data were subjected to a depression detention, which meant that the flow generation and drainage

network extraction values could be derived and the control watersheds could be created. The hydrological sites at the outlets provided the annual runoff and sedimentation data for 1985–2010.

The land use dataset was provided by the Cold and Arid Regions Science Data Center at Lanzhou, China (http://westdc.westgis.ac.cn, accessed on 8 May 2020). It was funded by major grants from the Chinese Academy of Sciences under the 'National Resources and Environment Survey and Dynamic Monitoring Using Remote Sensing' program (96-B02-01). Researchers experienced in interpreting the spectra, texture, and tone of such images created visual interpretations, which were based on Landsat Multispectral Scanner (MSS), Thematic Mapper (TM), and Enhanced Thematic Mapper (ETM) information. Their results were evaluated by field studies and the precision was as high as 95%.

Due to the large study area, the land use type subcategories were merged into the main categories, which provided a scientific rationality and flexibility for landscape change analysis. Six landscape types were established in the geographic information system (GIS) database: cropland, forest land, grassland, water area, urban and rural land, and unused land. The spatial analyst module in the ArcGIS system and conversion tools were used to transform the vector data for land use into raster data for the following analysis.

#### 2.3.2. Research Methods

Describing the characteristics and variations of the landscape using LMs and identifying relationships between landscape patterns and processes are the most common quantitative methods applied in landscape ecology research [39,40]. The Fragstats 3.3 landscape analysis software was used to determine the relevant LMs according to the Fragstats 3.3 operation manual. The LMs were then used to study the pattern properties of the watershed. Fragstats 3.3 can calculate more than 50 LMs. These metrics were divided into three levels representing three different scales. (1) Patch level: this reflected the structural characteristics of single patches in the landscape and provided the computational basis for the other levels. (2) Class level: this reflected the structural characteristics of multiple patches in the landscape. (3) Landscape level: this reflected all the structural characteristics of the landscape. This study used the landscape-level metrics, number of patches, patch density, the largest patch index, the landscape shape index, the perimeter area fractal dimension, the contagion index, the patch cohesion index, the landscape division index, Shannon's diversity index, and Shannon's evenness index. These indexes were used because they reflect area, density, proximity, diversity, and divergence [30]. The computing method and ecological significance of each metric are listed in Table 1.


**Table 1.** Description and ecological significance of landscape metrics in this study.


**Table 1.** *Cont*.

The significance and correlation analyses were undertaken by one-way ANOVA and multiple linear regression using IBM SPSS Statistics Version 2.0 software.

#### **3. Results**

#### *3.1. Land Use Changes in the Watersheds Life* **2022**, *12*, 1688 6 of 14

Table 2 describes the change characteristics of the six land use types over the 25-year period. The area of the Tuweihe River watershed is 4503.40 km<sup>2</sup> and grassland represented the greatest proportion of the land cover (between 38.13 and 53.49% over the 25-year period), followed by unused land (between 23.08 and 37.90% over the 25-year period). An analysis of the land-use transfer matrix showed that the unused land variance was largest over the study period at 33.82%. Between 1985 and 2010, 35.12% of unused land was turned into grassland, with 67.28% of the conversion occurring between 1985 and 1996 (Figure 2). Furthermore, 92.93 km<sup>2</sup> of cropland was returned to forest and grassland, with the largest changes occurring between 2000 and 2010 (76.52 km<sup>2</sup> ). The proportion of land converted from forest to other land uses was lowest at 5.46%. **Urban and Rural land** 8.70 8.62 9.03 18.65 6.12 4.51 6.32 8.53 **Unused land** 1374.55 801.00 943.91 909.61 1.05 0.56 0.56 0.55 The GU watershed has an area of 1263.11 km2. The largest proportion of the wetland was grassland (between 61.00 and 62.61% over the 25-year period), followed by cropland (between 30.37 and 32.50% over the 25-year period). From 1985 to 2010, the cropland and unused land areas gradually decreased, and forest land, grassland, and urban and rural land areas increased. The water area was stable but, in the TU watershed, the water area slowly decreased. The cropland area changed the most (52.85 km2). Between 2000 and 2010, 49.76 km2 of cropland was converted into forest and grassland, while unused land had the highest transfer ratio (51.95%) of all the landscape types.

**Figure 2.** Land use distribution and variations with time in Tu (**up**) and Gu (**down**) watershed*.* **Figure 2.** Land use distribution and variations with time in Tu (**up**) and Gu (**down**) watershed.

Table 3 lists the LMs for the TU and GU watersheds at four time periods. As time progressed, the TU watershed's number of patches, contagion, and patch cohesion values gradually decreased, whereas the largest patch index, the landscape shape index, perimeter area fractal dimension, landscape division, and Shannon*'*s diversity values tended to

to the variance. The patch density, the largest patch index, perimeter area fractal dimension, contagion, and patch cohesion were all lower in the TU watershed than in the GU

*3.2. Land Metrics and Landscape Stability (LS)* 

watershed.


**Table 2.** The change variations of land use with time in the study area (km<sup>2</sup> ).

The GU watershed has an area of 1263.11 km<sup>2</sup> . The largest proportion of the wetland was grassland (between 61.00 and 62.61% over the 25-year period), followed by cropland (between 30.37 and 32.50% over the 25-year period). From 1985 to 2010, the cropland and unused land areas gradually decreased, and forest land, grassland, and urban and rural land areas increased. The water area was stable but, in the TU watershed, the water area slowly decreased. The cropland area changed the most (52.85 km<sup>2</sup> ). Between 2000 and 2010, 49.76 km<sup>2</sup> of cropland was converted into forest and grassland, while unused land had the highest transfer ratio (51.95%) of all the landscape types.

#### *3.2. Land Metrics and Landscape Stability (LS)*

Table 3 lists the LMs for the TU and GU watersheds at four time periods. As time progressed, the TU watershed's number of patches, contagion, and patch cohesion values gradually decreased, whereas the largest patch index, the landscape shape index, perimeter area fractal dimension, landscape division, and Shannon's diversity values tended to increase. The patch density values remained almost the same over the 25 years. The LMs in the GU watershed changed over the 25-year period but there was no obvious pattern to the variance. The patch density, the largest patch index, perimeter area fractal dimension, contagion, and patch cohesion were all lower in the TU watershed than in the GU watershed.


**Table 3.** Annual variations in landscape indices (units: see Table 1).

The cropland, forest, and WAR land use types had the highest LS values. The grassland had the lowest average LS value, particularly between 2000 and 2010. During this period, the character stability (CS) and density stability (DS) of urban and rural land in the TU watershed was 0.409 and 0.881, respectively, followed by a DS of 0.591 for unused land between 1985 and 1996. In the GU watershed, the lowest LS value occurred for forest land, followed by unused land and urban and rural land, which had the lowest values between 1985 and 1996. In general, the LS values for the TU watershed were higher than those for the GU watershed. The LS values for grassland and unused land increased over time, whereas the cropland and urban and rural land declined.

**Watershed Time Number** 

**Tuweihe** 

**Gushanchua n** 

**of Patches** 

**Patch Density**  **Largest Patch** 

**Landscape Shape** 

#### *3.3. Variation in and the Relationship between Annual Runoff and Sedimentation 3.3. Variation in and the Relationship between Annual Runoff and Sedimentation*

The Mann–Kendall trend results showed that runoff and sedimentation tended to decrease over time (*p* < 0.01). The peaks of annual runoff and sedimentation were correlated with each other (Figure 3). Up to 2010, the runoff rate in the TU and GU watersheds was 52.52% and 80.95% of the annual runoff in 1956, respectively. The annual reduction in runoff volume was 3.75 and 1.35 million m<sup>3</sup> , respectively. Sedimentation in 2010 decreased by 97.26% and 99.77%, respectively, relative to the value in 1985, and the average sedimentation reduction per year was 0.21 and 0.46 million tons in the TU and GU watersheds, respectively. The runoff in the TU watershed, which has a larger area than the GU watershed, was higher than in the GU watershed, but annual sedimentation was much the same (1.40 million tons). The rank-sum test showed that there was a break point for annual runoff and the sedimentation process in the two watersheds. In the TU watershed, the break points for runoff and sedimentation occurred in 1981 and 2001, respectively, while it was 1999 for both runoff and sedimentation in the GU watershed. The Mann–Kendall trend results showed that runoff and sedimentation tended to decrease over time (*p* < 0.01). The peaks of annual runoff and sedimentation were correlated with each other (Figure 3). Up to 2010, the runoff rate in the TU and GU watersheds was 52.52% and 80.95% of the annual runoff in 1956, respectively. The annual reduction in runoff volume was 3.75 and 1.35 million m3, respectively. Sedimentation in 2010 decreased by 97.26% and 99.77%, respectively, relative to the value in 1985, and the average sedimentation reduction per year was 0.21 and 0.46 million tons in the TU and GU watersheds, respectively. The runoff in the TU watershed, which has a larger area than the GU watershed, was higher than in the GU watershed, but annual sedimentation was much the same (1.40 million tons). The rank-sum test showed that there was a break point for annual runoff and the sedimentation process in the two watersheds. In the TU watershed, the break points for runoff and sedimentation occurred in 1981 and 2001, respectively, while it was 1999 for both runoff and sedimentation in the GU watershed.

time, whereas the cropland and urban and rural land declined.

*Life* **2022**, *12*, 1688 7 of 14

**Table 3.** Annual variations in landscape indices (units: see Table 1).

**Perimeter Area Fractal Dimension** 

1985 1393 0.31 20.30 36.48 1.60 36.47 97.79 0.91 1.32 0.734 1996 1332 0.30 41.04 35.32 1.58 39.82 98.72 0.80 1.24 0.690 2000 1343 0.30 37.39 36.16 1.58 38.46 98.60 0.83 1.27 0.706 2010 1340 0.30 34.03 35.94 1.57 38.36 98.44 0.86 1.27 0.707

1985 938 0.74 61.00 37.14 1.68 53.34 99.18 0.62 0.89 0.495 1996 909 0.72 62.50 36.34 1.69 55.22 99.21 0.61 0.85 0.476 2000 959 0.76 60.81 37.27 1.69 53.07 99.17 0.63 0.89 0.498 2010 928 0.74 61.79 35.73 1.68 52.59 99.14 0.62 0.91 0.506

**Contagion Patch** 

The cropland, forest, and WAR land use types had the highest LS values. The grassland had the lowest average LS value, particularly between 2000 and 2010. During this period, the character stability (CS) and density stability (DS) of urban and rural land in the TU watershed was 0.409 and 0.881, respectively, followed by a DS of 0.591 for unused land between 1985 and 1996. In the GU watershed, the lowest LS value occurred for forest land, followed by unused land and urban and rural land, which had the lowest values between 1985 and 1996. In general, the LS values for the TU watershed were higher than those for the GU watershed. The LS values for grassland and unused land increased over

**Cohesion Division Shannon's** 

**Diversity** 

**Shannon's Eveness** 

**Figure 3.** Interannual variation in annual runoff and sedimentation from 1956 to 2010. **Figure 3.** Interannual variation in annual runoff and sedimentation from 1956 to 2010.

The Pearson's correlation analysis revealed that there was a strong positive relationship between runoff and sedimentation in the two watersheds (*p* < 0.01, Figure 4). The The Pearson's correlation analysis revealed that there was a strong positive relationship between runoff and sedimentation in the two watersheds (*p* < 0.01, Figure 4). The coefficient of determination for the TU watershed, which has a higher landscape diversity, was 0.48, whereas the coefficient of determination for the GU watershed was 0.85. The sedimentcarrying capacity of the runoff (i.e., the slope of the regression line) in the GU watershed was greater than in the TU watershed carrying capacity. This showed that the sedimentation yields of the GU watershed were similar to those of the TU watershed, even though there was substantially less runoff (19.27% of that in the TU watershed). *Life* **<sup>2022</sup>**, *12*, 1688 8 of 14 coefficient of determination for the TU watershed, which has a higher landscape diversity, was 0.48, whereas the coefficient of determination for the GU watershed was 0.85. The sediment-carrying capacity of the runoff (i.e., the slope of the regression line) in the GU watershed was greater than in the TU watershed carrying capacity. This showed that the sedimentation yields of the GU watershed were similar to those of the TU watershed, even though there was substantially less runoff (19.27% of that in the TU watershed).

A Pearson's analysis was conducted to determine the effects of landscape on runoff

between the factors. More LMs were significantly (*p < 0.05*) or highly significantly (*p < 0.01*) correlated with annual runoff. When patch density, contagion, and patch cohesion rose, the annual runoff declined. In contrast, the LMs related to landscape diversity, such as Shannon's diversity and Shannon's evenness, were positively associated with annual runoff and sedimentation (*p < 0.01*). These relationships implied that the increase in patch density and area led to a decrease in runoff. Furthermore, contagion and patch cohesion had direct impacts (*p < 0.05*) on erosion (coefficients of determination of 0.773 and 0.738,

Patch density −4.457PD + 5.010 0.916 0.003 \*\* Contagion −0.113contagion + 8.191 0.738 0.028 \* Patch cohesion −0.717cohesion + 71.936 0.773 0.021 \* Shannon's diversity 3.312SHDI−3.361 0.930 0.002 \*\* Shannon's evenness 12.280SHEI−4.937 0.934 0.002 \*\*

Patch cohesion −0.043cohesion + 4.294 0.760 0.024 \*

The Chinese government initiated the Grain for Green Program (GGP) in 1999 and this nationwide project has gradually changed the national land use structure [41]. The Loess Plateau was particularly affected by the program because it was considered as a priority region [42]. Large areas (Table 2) have been converted to various alternative

**Figure 4.** Linear relationship between annual runoff and sedimentation. **Figure 4.** Linear relationship between annual runoff and sedimentation.

*3.4. Response Relationships between Runoff, Sedimentation, and LMs* 

**Table 4.** Regression relationships between LMs, runoff, and sedimentation.  **LMs Regression Equation R2 Sig.** 

**Sedimentation** Contagion −0.006contagion + 0.474 0.693 0.04 \*

\* Significant at *p* < 0.05; \*\* significant at *p* < 0.01.

respectively).

**4. Discussion** 

**Runoff** 

#### *3.4. Response Relationships between Runoff, Sedimentation, and LMs*

A Pearson's analysis was conducted to determine the effects of landscape on runoff and sedimentation (Table 4). The results showed that there were significant correlations between the factors. More LMs were significantly (*p < 0.05*) or highly significantly (*p < 0.01*) correlated with annual runoff. When patch density, contagion, and patch cohesion rose, the annual runoff declined. In contrast, the LMs related to landscape diversity, such as Shannon's diversity and Shannon's evenness, were positively associated with annual runoff and sedimentation (*p < 0.01*). These relationships implied that the increase in patch density and area led to a decrease in runoff. Furthermore, contagion and patch cohesion had direct impacts (*p < 0.05*) on erosion (coefficients of determination of 0.773 and 0.738, respectively).


**Table 4.** Regression relationships between LMs, runoff, and sedimentation.

\* Significant at *p* < 0.05; \*\* significant at *p* < 0.01.

#### **4. Discussion**

The Chinese government initiated the Grain for Green Program (GGP) in 1999 and this nationwide project has gradually changed the national land use structure [41]. The Loess Plateau was particularly affected by the program because it was considered as a priority region [42]. Large areas (Table 2) have been converted to various alternative landscapes in the study watersheds. More check dams were constructed in Tu watershed than in Gu watershed, which play a vital role in intercepting sediment. It was confirmed by the lower coefficient determination in the relationship between runoff and sedimentation in TU watershed (Figure 3). Both watersheds have been subjected to continuous deforestation and conversion of cropland to forest. In the process, patch connectivity developed, which led to species migration and other ecological processes. This was confirmed by the increase in the largest patch index, patch cohesion, and contagion values. The landscape shape index decrease in the TU watershed showed that many patches were affected by anthropogenic activities, which led to a regular and simple patch pattern. This was confirmed by the decrease in the perimeter area fractal dimension values.

In the TU watershed, number of patches decreased with time, but patch cohesion and contagion increased, which indicated that good connectivity was formed by merging a landscape type with species migration and other ecological processes [43]. The variable decreases in landscape shape index and perimeter area fractal dimension illustrated that many patches were being affected by human activities, which also showed that the landscape consisted of regular and simple patches. Large stretches of grassland were recreated in 1996, which led to the lowest value for Shannon's diversity. The lower patch density and area parameters resulted in a complex landscape system in the TU watershed. Therefore, the Shannon's diversity value for the TU watershed was higher than that of the GU watershed. The LS for the urban and rural land in the TU watershed decreased over time due to increased anthropogenic activities (Table 5). More than five programs, including the conversion of cropland to forests program, have been initiated since 1978 in an attempt to control desertification and soil loss. Furthermore, afforestation has also increased in China over the last decade [44]. Therefore, the LS values for grassland and unused land increased after 2000 and interference due to human activities declined across the two landscape types.


**Table 5.** Landscape stability variation characteristics.

The cropland landscape was the key factor affecting soil conservation [45] in the study area and there were more check dams in the TU watershed according to the field investigation, which caused the annual sedimentation yield in the TU watershed to be similar to the yield for the GU watershed, even though the annual runoff in the TU watershed was significantly higher (*p* < 0.01, Figure 3). Fragmented natural landscape indicated intensive agricultural activities, which caused more serious erosion and soil nutrient loss [46]. Higher landscape stability of TU watershed further confirmed its controlling function on sedimentation with higher runoff. In addition, the variation coefficient for annual sedimentation was higher than that for annual runoff, which indicated that the sedimentation was more susceptible to environmental effects than runoff.

Runoff and the sediment deposited in water is contained by the spatial pattern of the landscape [47]. The LMs synthesize the retardation capacity and spatial position, and reflect the potential risk of water and soil loss. For example, the Shannon's diversity value is not a biodiversity metric but, rather, focuses on the unbalanced distribution of various patch types in the landscape. In the study area, the diversity of land uses and low degree of landscape fragmentation exerted significant positive influences on runoff (*p* < 0.01, Table 4). The contagion and patch cohesion values had significant negative correlations with annual runoff and sedimentation (*p* < 0.05), which indicated that water and soil loss decreased when external and internal patch connectivity improved. Most LMs were significantly

related to annual runoff, which showed that the landscape had a greater effect on runoff than sedimentation. This means that it can be used as an ecological indicator to predict runoff, relative to sedimentation. The LS changes showed that there was an abrupt runoff change in 1981, which the land use data did not show. The sedimentation-to-runoff ratio was lowest in 2001, and the LS values for forest land and grassland were also at their lowest compared to 1985–1996 and 2000–2010 (Table 5). In the GU watershed, the lowest LS for forest land occurred in 1999 (1996–2000), which indicated that there had been a sharp increase in forest land. This could have caused the sharp decrease in runoff and sediment deposition in 1999. It, therefore, appears that a breakpoint usually occurs when the LS for forest land and grassland is small, which suggests that there has been a major expansion in these types of land use.

A Pearson's analysis was conducted to determine the factors that most strongly influence the annual decrease in runoff (DR) and sedimentation (DSe) (Table 6). The results showed that the correlations between DSe and LS, and the different land use types were not significant (*p* > 0.05). However, the DR was positively correlated to the DS for grassland (P = 0.740, *p* < 0.05), which meant that annual runoff in the watershed could be significantly reduced if the grassland had a high DSe. When the grass patches were highly connected, runoff could be effectively intercepted [48]. Therefore, the LMs and LS effects on runoff became more significant.

**Table 6.** Pearson's analysis between the DR, DSe, and LS values for the different land use types.


\* Correlation is significant at the 0.05 level (two-tailed). DR and DSe mean decrease in annual runoff and sedimentation, CS and DS mean stability of character and density. URL means urban and rural land.

A stepwise regression analysis was used to determine the most influential variables that were not strongly correlated with one another [49]. Every independent variable was subjected to an F test and then deleted if the F-value showed that the variable was not significant. Furthermore, the previous variable was deleted if the F-value was not significant when a new independent variable was added to the set. This algorithm was repeated until no independent variable could be added or deleted. The optimal regression model was then established after applying this method (Table 7). The variance inflation factors (VIF) were 0.446 and 2.244 for the TU and GU watersheds, respectively, which meant that the collinearity hypothesis could be rejected. The significance values were all lower than 0.05. Therefore, the selected LMs (Shannon's evenness and patch cohesion) were the most significant factors affecting annual runoff and sedimentation. When the dominant landscapes had greater ecological benefits, the annual runoff decreased. Furthermore, measures that promote the value of patch cohesion should be taken if the interception function of water and soil loss is to be improved.

This study applied DEM dataset processed by ASTER GDEM for landscape analysis to assess water and soil loss. Although it has been proved to be an appropriate application in the field of erosion estimation [50] and provided scientific basis for soil erosion prevention and land use management, it is still a worthy study to investigate the result difference with finer or coarser resolution. In addition, the relationship between LMs and erosion we established and discussed was based on the dataset collected in the focalized regions, which are typical watershed on the Loess Plateau. Considering the extension of the scientific research, more analysis should be executed with larger scale and different regions. There is, of course, conventional existing research that concluded that LMs, e.g., Shannon's evenness and patch cohesion, were significantly correlated with soil erosion in the whole region of Loess Plateau [51]. In terms of driving factors, vegetation cover and landscape variables are not the only factors that influence the erosion process; soil properties, climatic conditions, etc., also play a vital role in water and soil loss [52]. Therefore, to increase the validity of analysis and deepen the understanding of soil erosion processes, more related variables should be considered in further related research.


**Table 7.** Optimal regression model for LMs, runoff, and sedimentation.

#### **5. Conclusions**

From 1985 to 2010, the landscape of the study area tended to become regular, connected, and aggregated, while the annual runoff and sedimentation values gradually decreased. The LS values for grassland and unused land gradually increased, but they decreased for cropland and urban and rural land due to human activities. Due to larger cropland area and lower landscape stability in the GU watershed than that in the TU watershed, the annual sedimentation for the two watersheds was similar, even though the annual runoff in the TU watershed was greater than that in the GU watershed. The annual runoff was significantly and positively correlated with sedimentation (*p* < 0.01), and the coefficient of determination for the TU watershed (0.48) was substantially lower than that for the GU watershed (0.85). The LMs had more significant influences on runoff than on sedimentation (*p* < 0.01), especially given that density stability for grassland could significantly decrease the runoff in the study area. The Shannon's evenness and patch cohesion were the crucial factors for affecting water and soil loss, and the measures involving landscape and land use could have a greater influence on runoff than on sedimentation.

**Author Contributions:** Y.Z. and X.L. conceived the main idea of the paper. X.L. designed and performed the experiment. X.L. wrote the manuscript, and all authors contributed in improving the paper. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by National Natural Science Foundation of China. Grant number 42107365 and 42107368.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **Foliar Fertilizer Application Alters the Effect of Girdling on the Nutrient Contents and Yield of** *Camellia oleifera*

**Shuangling Xie <sup>1</sup> , Dongmei Li <sup>2</sup> , Zhouying Liu <sup>1</sup> , Yuman Wang <sup>1</sup> , Zhihua Ren <sup>1</sup> , Cheng Li <sup>1</sup> , Qinhua Cheng <sup>1</sup> , Juan Liu <sup>1</sup> , Ling Zhang <sup>1</sup> , Linping Zhang <sup>1</sup> and Dongnan Hu 1,\***


**Abstract:** Improving the economic benefits of *Camellia oleifera* is a major problem for *C. oleifera* growers, and girdling and foliar fertilizer have significant effects on improving the economic benefits of plants. This study explains the effects of girdling, girdling + foliar fertilizer on nutrient distribution, and the economic benefits of *C. oleifera* at different times. It also explains the N, P, and K contents of roots, leaves, fruits, and flower buds (sampled in March, May, August, and October 2021) and their economic benefits. The results showed girdling promoted the accumulation of N and K in leaves in March 2021 (before spring shoot emergence) but inhibited the accumulation of P, which led to the accumulation of P in roots and that of N in fruits in August 2021 (fruit expansion period). Foliar fertilizer application after girdling replenished the P content of leaves in March 2021, and P continued to accumulate in large quantities at the subsequent sampling time points. The N and P contents of the root system decreased in March. In October (fruit ripening stage), girdled shrubs showed higher contents of N and K in fruits and flower buds, and consequently lower relative contents of N and K in roots and leaves but higher content of P in leaves. Foliar fertilizer application slowed down the effects of girdling on nutrient accumulation in fruits and flower buds. Spraying foliar fertilizer decreased the N:P ratio in the flower buds and fruits of girdled plants. Thus, foliar fertilizer spray weakened the effects of girdling on the nutrient content and economic benefits of *C. oleifera*. In conclusion, girdling changed the nutrient accumulation pattern in various organs of *C. oleifera* at different stages, increased leaf N:K ratio before shoot emergence, reduced root K content at the fruit expansion stage and the N:K ratio of mature fruit, and promoted economic benefits.

**Keywords:** *Camellia oleifera*; girdling; foliar fertilizer; nutrient content; yield

Academic Editor: Othmane Merah

Received: 19 January 2023 Revised: 9 February 2023 Accepted: 17 February 2023 Published: 20 February 2023

10.3390/life13020591

**Citation:** Xie, S.; Li, D.; Liu, Z.; Wang, Y.; Ren, Z.; Li, C.; Cheng, Q.; Liu, J.; Zhang, L.; Zhang, L.; et al. Foliar Fertilizer Application Alters the Effect of Girdling on the Nutrient Contents and Yield of *Camellia oleifera*. *Life* **2023**, *13*, 591. https://doi.org/

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

## **1. Introduction**

*Camellia oleifera* Abel (Theaceae) is an evergreen shrub widely planted in 18 provinces and regions of China, including Hunan, Jiangxi, and Guangxi (in order of planting area) [1,2]. Unlike other fruit trees, *C. oleifera* is characterized by the coexistence of fruit and flowers as well as vegetative growth and reproductive growth. This results in different periods of distribution patterns of *C. oleifera* being different from other fruit trees. Therefore, scientific management of the nutrient content of *C. oleifera* is an important measure for maintaining its yield at high levels [1,3,4]. In recent years, scholars have conducted extensive research on the fertilization of *C. oleifera* forests [5]; however, fertilization in forestland was found to be difficult and expensive. Tree girdling, as a means of nutrient content regulation, has been widely used in citrus [6], grape [7], apple [8], kiwi [9], and other fruit trees [10]. The objective of girdling is to sever the phloem and prevent the flow of carbohydrates to the underground plant parts, thus promoting reproductive organ growth, flowering, and fruit development and quality [11,12]. However, girdling should be carried out when the nutrient content of the tree is sufficient. Foliar fertilizer is applied to plant

stems and leaves so that plants can absorb various nutrients through stems and leaves and improve their nutritional status [12]. Meanwhile, with the application of unmanned aerial vehicles (UAVs) for plant protection [13,14], the application of foliar fertilizer on *C. oleifera* trees is in the initial stages. Foliar fertilizer will become another important way of nutrient management in *C. oleifera* [15].

A large number of experiments on fruit trees show that girdling effectively reduces the N(Nitrogen), P(Phosphorus), and K(Potassium) contents of the leaves above the girdle [11,16]. Therefore, if the nutrients are not replenished immediately after the girdling, the tree can become weak and eventually die. Therefore, reasonable foliar fertilizer application can effectively improve the nutrient content of plant organs and promote the growth and development of plants [15]. Urea has a high N content and is often used as a common fertilizer for plant nitrogen supplementation [17]. Potassium dihydrogen phosphate contains P and K, and the plant utilization rate is high, which can promote the absorption of N and P by plants, and it has good water solubility, which is the first choice for foliar fertilizer [18]. Girdling can reduce flower and fruit drop. Furthermore, plants are also sprayed with gibberellin [18], naphthaleneacetic acid [19], or boron [20] to reduce flower and fruit drop. *C. oleifera* has been cultivated for a long time [21]; however, its management is extensive, and there has been no way to report the application of girdling. In a preliminary experiment, we showed that girdling could effectively increase the fruit yield of *C. oleifera*. However, the effect of girdling on nutrient distribution in *C. oleifera*, the need to supplement nutrients after the application of girdling, and the relationship between the nutrient characteristics of *C. oleifera* and yield remain unclear. In this study, 10-year-old *C. oleifera* trees were treated with girdling and foliar fertilizer, and the root, leaf, flower, and fruit samples were collected to determine and analyze the key phenological period of *C. oleifera*. Additionally, ripe *C. oleifera* fruits were picked to determine fruit quality. To explore the effects of girdling and foliar fertilizer application on the nutrient content of *C. oleifera* trees, the distribution regulation and stoichiometric ratio characteristics of these trees were analyzed in different periods. Moreover, to develop recommendations and obtain theoretical support for nutrient management in *C. oleifera*, the relationship between nutrient characteristics and yield was analyzed via path analysis.

#### **2. Materials and Methods**

#### *2.1. Plant Material and Study Site*

Ten-year-old clones of *C. oleifera*, with an average plant height of 2.70 m, ground diameter of 76.83 mm, east-west crown width of 2.55 m, north-south crown width of 2.62 m, and plant row spacing of 2.0 m × 3.0 m, were used in this study. The experiment was conducted in the Jiu long shan Township of Yushui District, Xinyu City, Jiangxi Province, China (27◦400 N, 114◦490 E). The study site has a subtropical monsoon climate, with an average annual temperature of 15 ◦C and abundant annual rainfall of 1680 mm (Resources come from Xinyu Meteorological Bureau). The woodland was planted in strips on a gentle slope, and the soil contained 25.12 mg kg−<sup>1</sup> available nitrogen, 5.73 mg kg−<sup>1</sup> available phosphorus, and 20.25 mg kg−<sup>1</sup> available potassium. All *C. oleifera* clones were planted in the same period and grown using the same management practices.

#### *2.2. Study Design*

A single factor experimental design was adopted. Girdling was applied at the flowering stage, and foliar fertilizer was sprayed after girdling. The treatment subjected to neither girdling nor spray foliar fertilizer served as the check control (CK). Each treatment contained 15 plants (45 plants total). Additionally, the treatments were implemented on three adjacent strips in the middle of the hillside, with one treatment per strip.

#### *2.3. Experimental Method*

#### 2.3.1. Girdling Technique

The experiment began in November 2020 (at the first flowering stage). *C. oleifera* shrubs showing uniform growth were selected and girdled 270◦ with a girdling cutter, completely severing the phloem but without damaging the xylem. The girdle was 2 mm wide, and was located on first-order branches at 10–20 cm above the main stem [22]. Finally, the shrubs were labeled according to the treatment.

#### 2.3.2. Foliar Fertilizer Spray Technique

A sprayer was used to spray the surface of *C. oleifera* leaves with a foliar fertilizer composed of 0.2% urea, 0.2% potassium dihydrogen phosphate, 0.2% borax, 50 mg L−<sup>1</sup> gibberellin, and 20 mg L−<sup>1</sup> naphthalene acetic acid (The concentration selection is determined on the basis of comprehensive consideration of previous studies). A total of 7.5 L of foliar fertilizer was applied to 15 *C. oleifera*, with an average of 0.5 L per tree. The fertilizer was sprayed 1 week after girdling on a day with no rain [23].

#### 2.3.3. Sample Collection

Plant organs were sampled in 2021; roots and leaves were sampled on 10 March (before spring shoot emergence), 25 May (after spring shoot emergence), 5 August (fruit expansion stage), and 15 October (fruit maturity stage); fruits were sampled only on 25 May and 15 October, and flower buds were sampled only on 15 October, because flower buds and fruits were either too small or not present at the other time points. Sampling was performed in triplicate on each date, with each replicate containing samples collected from five plants. After collection, the samples were brought to the lab, washed, fixed, dried to a constant weight, pulverized, and stored for testing.

On 15 October 2021, the yield per plant of *C. oleifera* was determined (all fruits from a single *C. oleifera* plant were picked separately, weighed, and recorded). At the same time, before fruit picking, 24 fruits were randomly picked in the upper, middle, and lower layers in the four directions of southeast, southwest, and northwest, with different treatments (one for every five trees, a total of three parts). After collection, it was brought back to the laboratory to determine the fresh weight of single fruit and single fruit seed kernels, and then the seed kernels were put into the oven and baked to constant weight, crushed with a mortar, and stored to determine the oil content of the seed kernels.

#### 2.3.4. Determination of Nutrient Content and Fruit Economic Characteristics

Before the test treatment, three parts of 0–20 cm rhizosphere soil were collected according to the "S" sampling method, dried naturally, passed through a 2 mm sieve, and sealed and stored.

The available nitrogen content was determined using the alkaline hydrolysis method [24]. The available phosphorus content was determined using the molybdenum blue colorimetric method [21]. The available potassium content was measured using an atomic flame photometer [1].

Weigh 0.1 g samples (leaves, roots, fruits, buds) and put them in a boiling tube, add H2SO4-H2O2, place a small funnel at the mouth of the tube, put it on the cooking furnace at 420 ◦C to boil until transparent, cool and set the volume to a 100 mL volumetric flask, let stand for 5–7 min, transfer to a 15 mL centrifuge tube for storage, and determine the concentrations of N, P, and K respectively.

The N content of various plant organs was determined using the automatic discontinuous analyzer (Smartchem 200, AMS, Rome, Italy) [22]; P content was determined using the molybdenum blue colorimetric method (GENESYS 180, Shanghai, China) [21]; and K content was determined with a flame photometer (FP6400, Shanghai, China) [1].

Weigh 0.5 g of the sample (dried seed kernels), put it into a folded filter paper packet, extract it using petroleum ether Soxhlet for 8 h, stand at 75 ◦C for 0.5 h, weigh the weight, calculate the oil content of seed kernels [1], the oil content of fresh fruits, and the oil yield per plant. The calculation formula is as follows:


## *2.4. Data Analysis*

The data were analyzed using SPSS 19.0. One-way analysis of variance (ANOVA), followed by Duncan's multiple range test (DMRT), were used to identify significant differences (*p* < 0.05) in various parameters among the different treatments. Path analysis was used to determine the relationship between oil yield per plant and nutrient characteristics. Graphs were prepared in ORIGIN 2021.

#### **3. Results**

#### *3.1. Effects of Girdling and Foliar Fertilizer Application on the Nutrient Contents of Various C. oleifera Organs*

The N, P, and K contents of *C. oleifera* leaves in each treatment at four time points, 10 March (before spring shoot emergence), 25 May (after spring shoot emergence), 5 August (after fruit expansion), and 15 October (after fruit ripening), are shown in Figure 1. Without any treatment (CK), the N nutrient content of leaves was low before spring shoot emergence; however, the N nutrient content of new leaves gradually increased after spring shoot emergence, reaching a higher level at the later growth stage (Figure 1a). The girdling + foliar fertilizer treatments significantly increased the N content of leaves before spring shoot emergence. Compared with CK, the N content of girdling and girdling + foliar fertilizer treatments were 145.05% and 162.44% higher, respectively. During the period from spring shoot emergence to fruit expansion, no significant difference in leaf N content was detected between each treatment and CK. However, N did not accumulate in leaves at the fruit ripening stage; instead, it was transported to the fruit. At the fruit ripening stage, leaf N content in the girdling treatment was 12.66% lower than that in CK, which was significantly lower than that in the previous period. Spraying foliar fertilizer after girdling reduced the N output of mature leaves, and the gap in the N content between the girdling + foliar fertilizer treatment and CK was relatively smaller. The P content of the leaves in CK reached the highest after spring shoot emergence, and gradually decreased with fruit expansion, oil transformation, and flower bud differentiation at the later stage, indicating that P was transported from leaves to other organs (Figure 1b). Girdling decreased the P content by 8.82% compared with CK before spring shoot emergence. However, the P content of leaves increased during the period from the emergence of spring shoots to the expansion of fruits, and then decreased at the later stage. When the fruits matured, the P content of leaves was significantly higher than that in the CK by 55.54%. Spraying foliar fertilizer after girdling significantly increased the leaf P content before spring shoot emergence but inhibited the accumulation of P in leaves after spring shoot emergence. The leaf P content of the girdling + foliar fertilizer treatment was significantly lower than that of the girdling treatment at three stages after spring shoot emergence, but was not significantly different compared with CK.

The effect of the girdling and foliar fertilizer spray on K accumulation was different from that on the N and P accumulations. The K content of leaves in the CK treatment showed an increasing trend until the end of fruit expansion and decreased at fruit maturity (Figure 1c). Girdling promoted the accumulation of K in leaves before spring shoot emergence. However, during the period from spring shoot emergence to fruit expansion, the leaf K content in the girdling treatment was significantly lower than that in CK, and the K output also decreased at the later stage. At the fruit ripening stage, the K content of leaves in the girdling treatment was similar to that in the CK. Except during fruit expansion, foliar fertilizer spray after girdling reduced the difference in the leaf K content between the

girdling and CK treatments at all time points. In other words, foliar fertilizer application weakened the regulation of girdling on the K content of *C. oleifera* leaves. between the girdling and CK treatments at all time points. In other words, foliar fertilizer application weakened the regulation of girdling on the K content of *C. oleifera* leaves.

**Figure 1.** Nutrient content of *C. oleifera* leaves in different stages of each treatment. Data are means ± SE (n = 3), uppercase letters represent the difference between the same treatment and different periods (*p* < 0.05), lowercase letters represent the difference between different treatments in the same period (*p* < 0.05). **Figure 1.** Nutrient content of *C. oleifera* leaves in different stages of each treatment. Data are means ± SE (n = 3), uppercase letters represent the difference between the same treatment and different periods (*p* < 0.05), lowercase letters represent the difference between different treatments in the same period (*p* < 0.05).

With the change in root growth time, the N, P, and K contents of roots showed dif‐ ferent dynamic regulation (Figure 2). Overall, the root N content increased after spring shoot emergence, slightly decreased during fruit expansion, and then increased at fruit maturity; the P content of roots increased at the emergence stage of spring shoot, and then decreased; and the root K content gradually increased with time. However, the change regulation of the different treatments was not completely consistent. With the change in root growth time, the N, P, and K contents of roots showed different dynamic regulation (Figure 2). Overall, the root N content increased after spring shoot emergence, slightly decreased during fruit expansion, and then increased at fruit maturity; the P content of roots increased at the emergence stage of spring shoot, and then decreased; and the root K content gradually increased with time. However, the change regulation of the different treatments was not completely consistent.

Girdling had no obvious effect on the root N content before and after spring shoot emergence (Figure 2a); however, the root N content was 16.44% after fruit expansion and 15.07% lower after fruit ripening compared with the CK. Therefore, girdling promoted N accumulation in the root system during fruit expansion but decreased N accumulation at fruit maturity. Application of foliar fertilizer after girdling reduced the root N content in each period, indicating that foliar fertilizer supplementation was not conducive to accu‐ mulation of N in the roots of girdled *C. oleifera* plants. Girdling had no obvious effect on the root N content before and after spring shoot emergence (Figure 2a); however, the root N content was 16.44% after fruit expansion and 15.07% lower after fruit ripening compared with the CK. Therefore, girdling promoted N accumulation in the root system during fruit expansion but decreased N accumulation at fruit maturity. Application of foliar fertilizer after girdling reduced the root N content in each period, indicating that foliar fertilizer supplementation was not conducive to accumulation of N in the roots of girdled *C. oleifera* plants.

Girdling promoted the accumulation and output of root P at spring shoot emergence and fruit maturity (Figure 2b), respectively. The increase in root P was the largest (139.48%) before and after spring shoot emergence, and the decrease in root P was the largest (30.48%) after fruit maturation. Spraying foliar fertilizer after girdling inhibited the accumulation of P in the root system after spring shoot emergence. The root P content was 46.09% lower in the girdling + foliar fertilizer treatment than in the girdling treatment, which was similar to the root P content in CK. However, during other time periods, foliar fertilizer application after girdling had no obvious effect on the root P content. Girdling promoted the accumulation and output of root P at spring shoot emergence and fruit maturity (Figure 2b), respectively. The increase in root P was the largest (139.48%) before and after spring shoot emergence, and the decrease in root P was the largest (30.48%) after fruit maturation. Spraying foliar fertilizer after girdling inhibited the accumulation of P in the root system after spring shoot emergence. The root P content was 46.09% lower in the girdling + foliar fertilizer treatment than in the girdling treatment, which was similar to the root P content in CK. However, during other time periods, foliar fertilizer application after girdling had no obvious effect on the root P content.

Girdling had no significant effect of on the root K content in each period (Figure 2c). However, the application of foliar fertilizer after girdling significantly increased the root K content of spring shoots by 20.72% before emergence. Moreover, after the girdling

**Treatment**

treatment and until fruit ripening, the root K content was reduced by 17.87% and 25.32% after fruit expansion and ripening, respectively, compared with that of CK. The results indicated that spraying foliar fertilizer after girdling was not conducive to the accumula‐

tion of K in *C. oleifera* roots after spring shoot emergence.

**Figure 2.** Nutrient content of *C. oleifera* roots in different stages of each treatment. Data are means ± SE (n = 3), uppercase letters represent the difference between the same treatment and different pe‐ riods (*p* < 0.05), lowercase letters represent the difference between different treatments in the same period (*p* < 0.05). **Figure 2.** Nutrient content of *C. oleifera* roots in different stages of each treatment. Data are means ± SE (n = 3), uppercase letters represent the difference between the same treatment and different periods (*p* < 0.05), lowercase letters represent the difference between different treatments in the same period (*p* < 0.05).

Girdling had no obvious effect on N and K accumulation in fruits at the fruit expan‐ sion stage (Figure 3). However, after fruit ripening, girdling reduced the N and K contents by 28.96% and 10.21%, respectively, compared with CK. The effect of girdling on the fruit P content was observed at the stage of fruit expansion, but little effect was noticed during fruit ripening. Application of foliar fertilizer after girdling reduced the N, P, and K con‐ tents of the fruit, and the N, P, and K contents of fruit at the swelling and ripening stages Girdling had no significant effect of on the root K content in each period (Figure 2c). However, the application of foliar fertilizer after girdling significantly increased the root K content of spring shoots by 20.72% before emergence. Moreover, after the girdling treatment and until fruit ripening, the root K content was reduced by 17.87% and 25.32% after fruit expansion and ripening, respectively, compared with that of CK. The results indicated that spraying foliar fertilizer after girdling was not conducive to the accumulation of K in *C. oleifera* roots after spring shoot emergence.

in the girdling + foliarfertilizer treatment showed no significant difference compared with the CK. Girdling had no obvious effect on the N, P, and K contents of flower buds (Table 1). The N content of flower buds was increased in the girdling + foliar fertilizer treatment, which was 18.36% and 25.21% higher than that in the girdling and CK treatments, respec‐ tively. The P and K contents of flower buds were also increased in the girdling + foliar fertilizer treatment, but this increase was not significant compared with the girdling and CK treatments. Girdling had no obvious effect on N and K accumulation in fruits at the fruit expansion stage (Figure 3). However, after fruit ripening, girdling reduced the N and K contents by 28.96% and 10.21%, respectively, compared with CK. The effect of girdling on the fruit P content was observed at the stage of fruit expansion, but little effect was noticed during fruit ripening. Application of foliar fertilizer after girdling reduced the N, P, and K contents of the fruit, and the N, P, and K contents of fruit at the swelling and ripening stages in the girdling + foliar fertilizer treatment showed no significant difference compared with the CK.

**Table 1.** Nutrient contents in flower buds of *C. oleifera* in each treatment. **Nutrient N Content (g kg−1) P Content (g kg−1) K Content (g kg−1)** Girdling+ foliar fertilizer 14.70 ± 0.73a 2.62 ± 0.14a 4.73 ± 0.15a Girdling 12.42 ± 0.39b 2.35 ± 0.03a 4.15 ± 0.2a Girdling had no obvious effect on the N, P, and K contents of flower buds (Table 1). The N content of flower buds was increased in the girdling + foliar fertilizer treatment, which was 18.36% and 25.21% higher than that in the girdling and CK treatments, respectively. The P and K contents of flower buds were also increased in the girdling + foliar fertilizer treatment, but this increase was not significant compared with the girdling and CK treatments.

CK 11.74 ± 0.21b 2.58 ± 0.17a 4.44 ± 0.12a

Note Lowercase letters represent the difference between different treatments in the same period (*p*

**Figure 3.** Nutrient contents of *C. oleifera* fruits in each treatment. Data are means ± SE (n = 3), low‐ ercase letters represent the difference between different treatments in the same period (*p* < 0.05). **Figure 3.** Nutrient contents of *C. oleifera* fruits in each treatment. Data are means ± SE (n = 3), lowercase letters represent the difference between different treatments in the same period (*p* < 0.05).


*p* value 0.013 \* 0.353 0.145

< 0.05); \*.\* indicates p < 0.05


relative N content of leaves in the girdling and girdling + foliar fertilizer treatments was more than 60%, and the relative N content of roots was less than 40%. However, the effect of girdling + foliar fertilizer was more obvious than that of the girdling treatment about Note Lowercase letters represent the difference between different treatments in the same period (*p* < 0.05); \* indicates *p* < 0.05.

#### the relative leaf N content. After the spring shoot emergence (25 May 2021), the fruit was small, and there were no flower buds. Girdling had little effect on N distribution in roots *3.2. Effects of Girdling and Foliar Fertilizer on Nutrient Distribution in Different C. oleifera Organs*

and leaves at this stage. After fruit expansion (5 August 2021), the root N content was the highest in the girdling treatment among all three treatments, and the relative N content of leaves, roots, and fruits in the girdling + foliar fertilizer treatment were consistent with those in CK. After fruit ripening (15 October 2021), the flower buds were swollen and about to bloom, and the relative N content of flower buds was lowest in CK and highest in the girdling + foliar fertilizer treatment. Additionally, on October 15, the relative N con‐ tent of leaves was the highest in CK, indicating that girdling and foliar fertilizer applica‐ tion promoted the transport of N from leaves to flowers. Thus, girdling and foliar fertilizer application promoted the transport of N from leaves to other organs. The P distribution patterns in *C. oleifera* organs in different treatments at different stages were shown in Figure 5. At the spring shoot emergence stage, girdling decreased the content of P in leaves and increased P accumulation in the root system. The distribu‐ tion of P in roots and leaves in the girdling + foliar fertilizer treatment was similar to that in CK. During the period after spring shoot emergence, the relative root P content was the highest in the girdling treatment, indicating that girdling was conducive to the accumu‐ lation of P in the root system. However, foliar fertilizer spray on girdled plants slightly decreased the relative root P content. After fruit expansion, the relative P content of fruits Girdling and foliar fertilizer application changed the distribution pattern of N in *C. oleifera* organs (Figure 4). Before spring shoot emergence (10 March 2021), *C. oleifera* fruit had not developed, and flower buds were absent. In the CK treatment, the N nutrient content of roots was significantly higher than that of leaves, and the relative root N content was approximately 60%. Girdling caused more N nutrient accumulation in leaves. The relative N content of leaves in the girdling and girdling + foliar fertilizer treatments was more than 60%, and the relative N content of roots was less than 40%. However, the effect of girdling + foliar fertilizer was more obvious than that of the girdling treatment about the relative leaf N content. After the spring shoot emergence (25 May 2021), the fruit was small, and there were no flower buds. Girdling had little effect on N distribution in roots and leaves at this stage. After fruit expansion (5 August 2021), the root N content was the highest in the girdling treatment among all three treatments, and the relative N content of leaves, roots, and fruits in the girdling + foliar fertilizer treatment were consistent with those in CK. After fruit ripening (15 October 2021), the flower buds were swollen and about to bloom, and the relative N content of flower buds was lowest in CK and highest in the girdling + foliar fertilizer treatment. Additionally, on October 15, the relative N content of leaves was the highest in CK, indicating that girdling and foliar fertilizer application promoted the transport of N from leaves to flowers. Thus, girdling and foliar fertilizer application promoted the transport of N from leaves to other organs.

after girdling.

was the lowest, while that of leaves was the highest in the girdling treatment; however, the relative P content of roots showed no significant difference among the three treat‐ ments. At the fruit maturity stage, the relative P content of leaves was still the highest in the girdling treatment, and the P content in flower buds was relatively less, indicating that the girdling induced the transport of P from roots mainly to leaves. No significant differ‐ ence was detected in P nutrient allocation between the girdling + foliar fertilizer and CK treatments. In general, the root system of *C. oleifera* accumulated more P in the first half of the year, and more P was transferred to the leaves in the second half of the year when fruits had expanded and ripened. This effect was alleviated by spraying foliar fertilizer

**Figure 4.** Nitrogen distribution characteristics of *C. oleifera* under different treatments in different **Figure 4.** Nitrogen distribution characteristics of *C. oleifera* under different treatments in different periods.

periods. The P distribution patterns in *C. oleifera* organs in different treatments at different stages were shown in Figure 5. At the spring shoot emergence stage, girdling decreased the content of P in leaves and increased P accumulation in the root system. The distribution of P in roots and leaves in the girdling + foliar fertilizer treatment was similar to that in CK. During the period after spring shoot emergence, the relative root P content was the highest in the girdling treatment, indicating that girdling was conducive to the accumulation of P in the root system. However, foliar fertilizer spray on girdled plants slightly decreased the relative root P content. After fruit expansion, the relative P content of fruits was the lowest, while that of leaves was the highest in the girdling treatment; however, the relative P content of roots showed no significant difference among the three treatments. At the fruit maturity stage, the relative P content of leaves was still the highest in the girdling treatment, and the P content in flower buds was relatively less, indicating that the girdling induced the transport of P from roots mainly to leaves. No significant difference was detected in P nutrient allocation between the girdling + foliar fertilizer and CK treatments. In general, the root system of *C. oleifera* accumulated more P in the first half of the year, and more P was transferred to the leaves in the second half of the year when fruits had expanded and ripened. This effect was alleviated by spraying foliar fertilizer after girdling. **Figure 4.** Nitrogen distribution characteristics of *C. oleifera* under different treatments in different periods.

**Figure 5.** Phosphorus distribution characteristics of *C. oleifera* organs in different treatments in dif‐ ferent periods. **Figure 5.** Phosphorus distribution characteristics of *C. oleifera* organs in different treatments in different periods.

dling had little effect on the distribution nutrients in roots and leaves. After spring shoot emergence, girdling caused more K to accumulate in the root system,reducing the amount

Girdling and foliar fertilizer application also affected the distribution of K in various

10 Mar

25 May

5 Aug.

15 Oct.

Girdling and foliar fertilizer application also affected the distribution of K in various *C. oleifera* organs (Figure 6). Before spring shoot emergence, leaves showed a higher K content in the girdling treatment than in CK; however, spraying foliar fertilizer after girdling had little effect on the distribution nutrients in roots and leaves. After spring shoot emergence, girdling caused more K to accumulate in the root system, reducing the amount of K transported to the leaves. However, spraying foliar fertilizer after girdling further increased the K content of leaves. The relative K content of fruits in the girdling treatment increased with fruit growth and expansion. Foliar fertilizer application after girdling promoted the transport of K from roots and leaves to fruits at the fruit maturity stage, and the relative K content of fruit was highest in the girdling + foliar fertilizer treatment. *Life* **2023**, *13*, x FOR PEER REVIEW 9 of 15 of K transported to the leaves. However, spraying foliar fertilizer after girdling further increased the K content of leaves. The relative K content of fruits in the girdling treatment increased with fruit growth and expansion. Foliar fertilizer application after girdling pro‐ moted the transport of K from roots and leaves to fruits at the fruit maturity stage, and the relative K content of fruit was highest in the girdling + foliar fertilizer treatment.

**Figure 6.** Potassium distribution characteristics of *C. oleifera* organs in different treatments in differ‐ ent periods. **Figure 6.** Potassium distribution characteristics of *C. oleifera* organs in different treatments in different periods.

#### *3.3. Effects of Girdling and Foliar Fertilizer Application on N:P and N:K Ratios in Different C. oleifera Organs 3.3. Effects of Girdling and Foliar Fertilizer Application on N:P and N:K Ratios in Different C. oleifera Organs*

Girdling had no significant effect on the leaf N:P ratio from spring shoot emergence to fruit expansion, but significantly increased the leaf N:P ratio before spring shoot emer‐ gence and decreased the leaf N:P ratio after fruit ripening (Table 2). Girdling had no significant effect on the leaf N:P ratio from spring shoot emergence to fruit expansion, but significantly increased the leaf N:P ratio before spring shoot emergence and decreased the leaf N:P ratio after fruit ripening (Table 2).

**Table 2.** The ratio of N: P and N: K of different organs in each period. **Date Treatment N:P N:K Leaves Roots Fruits Buds Leaves Roots Fruits Buds** Girdling+ foliar fertilizer 5.12 ± 0.07b 2.19 ± 0.13b – – 6.11 ± 0.12a 2.19 ± 0.09b – – Girdling 5.98 ± 0.28a 2.69 ± 0.10a – − 5.21 ± 0.26b 3.24 ± 0.07a − − CK 2.22 ± 0.11c 2.96 ± 0.06a − − 2.63 ± 0.16c 3.20 ± 0.06a − − Girdling+ foliar fertilizer 4.19 ± 0.17a 1.52 ± 0.06b − − 4.89 ± 0.13a5.76 ± 0.07ab − − Girdling 3.71 ± 0.16a 1.28 ± 0.03c − − 5.49 ± 0.24a 5.14 ± 0.35b − − CK 3.41 ± 0.31a 1.79 ± 0.07a − − 5.11 ± 0.24a 5.85 ± 0.17a − − Girdling+ foliar fertilizer 5.11 ± 0.59a 5.75 ± 0.76a 3.19 ± 0.35a − 6.74 ± 0.30a5.84 ± 0.36ab1.50 ± 0.06a − Girdling 4.60 ± 0.91a 6.45 ± 0.96a 3.52 ± 0.36a − 6.14 ± 0.08a 7.29 ± 1.75a 1.35 ± 0.13a − CK 6.13 ± 0.27a 4.37 ± 0.43a 2.88 ± 0.18a − 4.59 ± 0.02b 3.74 ± 0.36b 1.47 ± 0.05a − Girdling+ foliar fertilizer11.06 ± 0.61a5.78 ± 0.01a3.50 ± 0.14ab5.63 ± 0.20a7.17 ± 0.27b 7.50 ± 0.38a 1.37 ± 0.04a 3.10 ± 0.09a Girdling 8.11 ± 2.30b 5.82 ± 0.40a 2.79 ± 0.14b 5.27 ± 0.13b6.93 ± 0.13b 6.54 ± 0.28a 1.38 ± 0.11a3.01 ± 0.13ab CK 12.79 ± 0.46a6.11 ± 0.26a 3.71 ± 0.18a 4.59 ± 0.25b8.18 ± 0.22a 7.36 ± 0.18a 1.75 ± 0.03a 2.65 ± 0.03b Note: Data are means ± SE (n = 3). Lowercase letters represent the difference between different treat‐ The leaf N:P ratio in the girdling + foliar fertilizer was significantly lower than that in the girdling treatment but significantly higher than that in CK. During other periods, the effect of girdling + foliar fertilizer on the leaf N:P ratio was not obvious. In addition, during the leaf growth period, the N:P ratio in CK gradually increased, reaching a peak at the fruit maturity stage; however, in the other two treatments, the leaf N:P ratio first decreased and then increased at the fruit maturity stage. The root N:P ratio differed among the three treatments only in the first half of the year; after fruit expansion in the second half of the year, girdling had little effect on the root N:P ratio, regardless of the application of foliar fertilizer. The root N:P ratio decreased after girdling; however, after foliar fertilizer spray, the root N:P ratio decreased further, reaching levels significantly lower than those observed in CK, both before and after spring shoot emergence. The effect of girdling on the N:P ratio was not obvious at the fruit expansion stage, but was significant in mature fruits and flowers. Thus, in the CK treatment, the N:P ratio was significantly higher in fruits and significantly lower in flower buds. Application of foliar fertilizer after girdling increased the N:P ratio in mature fruits to different degrees, bringing the N:P ratio in fruits in the girdling + foliar fertilizer treatment closer to that in the CK treatment, although the N:P ratio in flower buds was higher in the girdling + foliar fertilizer treatment than in CK.

> The leaf N:P ratio in the girdling + foliar fertilizer was significantly lower than that in the girdling treatment but significantly higher than that in CK. During other periods,

> at the fruit maturity stage; however, in the other two treatments, the leaf N:P ratio first

ments in the same period (*p* < 0.05).


**Table 2.** The ratio of N:P and N:K of different organs in each period.

Note: Data are means ± SE (n = 3). Lowercase letters represent the difference between different treatments in the same period (*p* < 0.05).

Compared with CK, the leaf N:K ratio was significantly higher in the girdling treatment before spring shoot emergence and fruit expansion, and lower after fruit ripening; however, the difference in the leaf N:K ratio was not a significant difference between these two treatments during late spring shoot emergence. Application of foliar fertilizer after girdling significantly increased the N:K ratio before spring shoot emergence but had little effect on the N:K ratio at the other stages. In *C. oleifera* roots, girdling had little effect on the N:K ratio before spring shoot emergence and after fruit ripening but significantly reduced the N:K ratio during spring shoot emergence and increased this ratio after fruit expansion. Application of foliar fertilizer after girdling significantly reduced the root N:K ratio before spring shoot emergence but had little effect on the N:K ratio at each stage after spring shoot emergence. Girdling had no obvious effect on the N:K ratio in fruits but increased the N:K ratio in flowers to varying degrees. Girdling + foliar fertilizer treatment showed a significantly higher N:K ratio in flower buds than CK.

#### *3.4. Influence of Girdling and Foliar Fertilizer Application on C. oleifera Yield*

Significant differences were observed in the fresh fruit yield, fruit oil content, and per-plant oil yield among the different treatments (Table 3). Girdling significantly increased the fruit yield, while spraying foliar fertilizer reduced the fruit yield increase; nonetheless, compared with CK, the girdling and girdling + foliar fertilizer treatments showed 63.56% and 33.24% higher fruit yield, respectively. However, girdling reduced the oil content of fresh fruit, and the application of foliar fertilizer after girdling further reduced the oil content; compared with CK, the girdling and girdling + foliar fertilizer treatments showed 13.07% and 20.03% lower oil yield, respectively. In addition, the per-plant oil yield in the girdling treatment was significantly increased by 41.91% and 33.65% compared with the girdling + foliar fertilizer and CK treatments, respectively.

#### *3.5. Path Analysis of Nutrient Content, Nutrient Stoichiometric Ratio, and Per-Plant Oil Yield*

Only the leaf N:K ratio in March, root K content in August, fruit N:P ratio in August, and fruit N:K ratio in October had significant effects on the per-plant oil yield, and the root P content in March and leaf N content in May had significant effects on per-plant oil production (Table 4). Among these effects, the effect of the N:K ratio on per-plant oil production per plant in March was positive, while those of the root K content in

August, fruit N:P ratio in August, and fruit N:K ratio in October on the per-plant oil yield were negative.

**Table 3.** Yield of *C. oleifera* in each treatment, oil content of fresh fruit, and oil production per plant.


Note: Data are means ± SE (n = 3). Lowercase letters represent the difference between different treatments in the same period (*p* < 0.05).



The root K content in August, fruit N:P ratio in August, and fruit N:K ratio in October had direct negative effects on the per-plant oil yield (Table 5). However, the root K content in August, fruit N:P ratio in August, and fruit N:K ratio in October had indirect positive effects on the per-plant oil yield by interacting with each other and the leaf N:K ratio in March, which offsets the direct negative effects among factors. The leaf N:K ratio in March had an indirect negative effect on the per-plant oil yield by interacting with other factors. Among the effects of these factors, the direct positive effect of the leaf N:K ratio in March was greater, and the indirect positive effects of the fruit N:P ratio in August and fruit N:K ratio in October were greater. In addition, the residual path coefficient was 0.089, indicating that the factors in this equation fully explained the variation in oil production per *C. oleifera* plant.

**Table 5.** Path analysis of tree nutrient factors on oil production per plant of *C. oleifera*.


Note: X1 is N:K ratio of leaves in March, X2 is root of K content in August, X3 is N:P ratio of fruits in August, X4 is N:K ratio of fruits in October.

#### **4. Discussion**

#### *4.1. Effect of Girdling on the Nutrient Contents of C. oleifera Plants*

Girdling is widely used in fruit trees [6–9] because it causes physical damage to phloem in the trunk, impeding the transport of carbohydrates and inorganic nutrients to the tree roots [25]. Studies show that girdling reduces the N, P, and K contents of leaves, and this effect is related to the healing time of girdling wounds [26]. The growth and development

and the nutrient distribution pattern of *C. oleifera* are different from those of most fruit trees. The results of this study showed that girdling of the *C. oleifera* trunk in November significantly increased the N content of leaves before spring shoot emergence in early March of the next year, which was conducive to the accumulation of N in roots at the fruit expansion stage (August), and promoted the transport of N from leaves to fruits at the fruit maturity stage (October). During the period from the girdling of *C. oleifera* to the emergence of spring shoots, the fruits formed in the previous year (2020) had been harvested, and the new fruits (in 2021) had not yet formed; N had been transported to the leaves through the xylem; and competition from other organs was non-existent. Additionally, the girdle in the phloem had not yet begun to heal and could not transport N down to the roots, which significantly increased the N content of leaves during this time. After the girdle healed in mid-May, N was normally transported to above- and below-ground organs. After the girdling, the roots obtained more N and grew better, and the N output of the above-ground leaves increased at the later stage. In the first half of the year, P accumulated more in the roots, so that the P content of leaves before and after spring shoot emergence was significantly lower. However, after the fruit expansion and until the fruit maturity, P was transferred to the leaves in large quantities in the girdling treatment, resulting in a significantly higher leaf P content compared with CK. This result could be attributed to the P uptake characteristics of plants [21]. In *C. oleifera*, the first half of the year is the key period of P uptake and accumulation by roots. Before girdle healing, the normal transport of P is affected, resulting in a greater accumulation of P in the root system. After girdle healing, a large amount of P is transported to the leaves, greatly increasing the leaf P content.

In this study, girdling increased the availability of K to the leaves before spring shoot emergence, for the same reason as that which is responsible for N accumulation in leaves during this period. However, after spring shoot emergence, the K content of *C. oleifera* leaves, roots, fruits, and flowers in the girdling treatment was lower than that in CK, whereas the relative K content of flower buds and fruits was higher in the girdling treatment, so that more K was allocated to fruits and flower buds [27]. These results indicate that girdling promotes the distribution of K to fruits and flower buds, which is conducive to seed setting. However, it should be noted that girdling could consume a large amount of K while promoting seed setting, which may lead to insufficient K availability in the plant, thus requiring supplementation over time.

#### *4.2. Influence of Ring Cutting on the Stoichiometric Ratios of N, P, and K in C. oleifera*

N, P, and K are the limiting nutrients affecting plant growth, and the N:P and N:K ratios could be used as the determinants of plant health [28]. When the N:P ratio < 14, plant growth is limited by N; when the N:P ratio > 16, plant growth is limited by P; and when the N:P ratio = 14~16, plan growth is limited by both N and P [29]. The results of this study showed that the N:P of each organ of *C. oleifera* plants was less than 14 in different periods, indicating that the growth of *C. oleifera* was severely limited by N. In the trees girdled in November, the leaves had accumulated a large amount of N in March of the following year, whereas the level of P was low, improving the leaf N:P ratio, which to a certain extent, alleviated the effects of N limitation on *C. oleifera* growth. After the girdle wound healed, phloem returned to its normal transport capacity, gradually reducing the leaf N and P contents to levels consistent with those in the CK. The N:P ratio of the *C. oleifera* roots was relatively low before and after spring shoot emergence, which was severely restricted by N. At the later stage, the root N:P ratio gradually increased. At the fruit maturity stage, large amounts of N and P were accumulated in the vegetative organs before girdle wound healing, which were gradually transferred to the reproductive organs, flower buds, and fruits at the later stage. Therefore, the N:P ratio in *C. oleifera* fruits and flower buds was relatively high at the fruit maturity stage, which reduced the limitation of N in fruits and flower buds to a certain extent. When the N:K ratio is less than 2.1, plant growth is mainly limited by K [30]. The results of this experiment show that the N:K ratio in each organ of *C. oleifera* at different stages was less than 2.1, indicating that the growth of *C. oleifera* plants

was limited by K. However, changes in the N:K ratio in different organs at different stages after girdling was the same as that in the N:P ratio, for roughly the same reasons.

## *4.3. Foliar Fertilizer-Induced Modification of the Effect of Girdling on the Nutrient Content of C. oleifera Organs*

Application of foliar fertilizer after girdling affected nutrient accumulation in roots, increased the nutrient contents of fruits and flower buds, brought the N, P, and K contents of leaves, roots, and fruits closer to those in the CK, and weakened the regulatory effect of girdling on the contents of N, P, and K in *C. oleifera*. This result might be related to the vegetative growth of the above-ground parts of *C. oleifera* plants treated with foliar fertilizer [4]; the root system absorbed more nutrients. In addition, foliar fertilizer application after girdling also weakened the effects of girdling on the distribution of N, P, and K in various organs during the period from spring shoot emergence to fruit expansion and reduced the deviation of tree nutrient distribution patterns from the control in this period. Foliar fertilizer increased the relative contents of P and K in ripe fruits and flower buds of girdled *C. oleifera* trees, which may be related to the less fruits treated with foliar fertilizer [31].

#### *4.4. Key Nutrient Characteristics of Girdled C. oleifera Trees for Increasing Yield*

Girdling improved the fruit yield and per-plant oil yield of *C. oleifera*, but it did not increase the oil content of fresh fruit. It may be because girdling greatly increased the number of fruits per plant, resulting in reduced seed kernel oil content. However, the substantial increase in yield compensated for the reduction in seed oil content, leading to a substantial increase in oil production per plant.

Path analysis of *C. oleifera* nutrient contents and per-plant oil and fruit yield revealed that the root K content in August, fruit N:P ratio in August, and fruit N:K ratio in October were important factors affecting per-plant oil production. In March, the leaf N:K ratio had direct positive effects on oil production per plant, while the other three factors had indirect negative effects. This indicates that the increase in the leaf N:K ratio in March could directly and effectively promote the increase in oil yield per plant. Girdling promoted the accumulation of N in *C. oleifera* leaves in March, increasing the leaf N:K ratio. At the same time, girdling promoted a reduction in the K content of roots in August, and large amounts of N and P were accumulated in leaves, which were later transported to fruits, but the N and P contents of fruits were relatively low. In October, large amounts of N and K were accumulated in leaves and roots, which were later transported to fruits, and the amount of K transported was greater than that of N, resulting in low N and P contents of fruits.

#### **5. Conclusions**

Girdling and foliar fertilizer application affected the nutrient contents of *C. oleifera* organs at the flowering stage. The improvement in the economic benefits of girdling on *C. oleifera* is more obvious. In the girdling treatment, N was accumulated in leaves before girdle wound healing. However, the accumulation of N and K in the roots was limited. With the gradual healing of the girdle wound and the recovery of phloem transport capacity, the underground plant parts began to accumulate N. P accumulated in the root system before girdle wound healing. After wound healing, P began to accumulate in the leaves. At the fruit ripening stage, three nutrients (N, P, and K) previously accumulated in the vegetative organs were gradually transported to the reproductive organs (fruits and flower buds), and the amount of K transported was greater than that of N and P, which promoted flowering and fruiting. Compared with the girdling, the application of foliar fertilizer after girdling weakened the overall effect of girdling on nutrient regulation in *C. oleifera* trees but promoted the accumulation of a large amount of nutrients in roots, flower buds, and fruits. **Author Contributions:** Conceptualization, S.X. and D.H.; methodology, S.X. and D.H.; software, S.X. and Y.W.; investigation, Z.L., Z.R., C.L. and Q.C.; resources, D.L.; writing—original draft preparation, S.X.; writing—review and editing, S.X., D.H., L.Z. (Ling Zhang), J.L. and L.Z. (Linping Zhang); visualization, S.X.; project administration, D.H.; funding acquisition, D.H. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research were funded by National Key R&D Program of China, grant number "2018YFD1000603", and Science and Technology Innovation Project of Forestry Department of Jiangxi Province, grant number Innovation Project [2020] No.2.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data that support the findings of this study are openly available in PubMed or available in other sources.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
