
==== Front
Heliyon
Heliyon
Heliyon
2405-8440
Elsevier

S2405-8440(24)12932-5
10.1016/j.heliyon.2024.e36901
e36901
Research Article
How did the diversity of woody plants in urban forests vary with the urban-rural gradient in Qingdao, China?
Zhu Ruirui ∗
Xiu Danping
Xue Ruixin
Teng Shuo
College of Architecture and Urban Planning, Qingdao University of Technology, Qingdao, China
∗ Corresponding author.
30 8 2024
30 9 2024
30 8 2024
10 18 e3690115 2 2024
27 7 2024
23 8 2024
© 2024 Published by Elsevier Ltd.
2024

https://creativecommons.org/licenses/by-nc-nd/4.0/ This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
The distribution and diversity of woody vegetation are crucial for understanding the structure and ecology of urban forests. As urbanization accelerates, the construction and composition of urban forests vary significantly along the urban-rural gradient. Qingdao's urban forests offer an opportunity to test the relationship between the diversity of woody plants and the urban-rural gradient. We classified the urban-rural gradient using imperviousness and construction time, then investigated the diversity of woody plants in Qingdao's urban forests under different urban-rural gradients and tested the reasonableness of their allocation. Correlation analysis found that the diversity index of woody plants in urban forests was highly connected to the urban-rural gradient (by imperviousness: rMargalef Index = −0.589, rShannon-Wiener Index = -0.373, rPielouIndex = −0.170, rSimpson Index = 0.272/by construction time: rMargalef Index = −0.530, rShannon-Wiener Index = −0.360, rPielouIndex = −0.148, rSimpson Index = 0.272/0.174). With a decrease in urbanization density, the Margalef (H), Shannon-Wiener (H′), and Simpson (D) indices all decreased while the Pielou (E) index increased. The four diversity indices showed a substantial correlation with one another, but not with the Margalef and Pielou indices. The analysis utilizing the 10/20/30 rule of empirical demonstrates a clear irrationality in allocating shrub species in Qingdao's urban forests, and the distribution of tree species is reasonable. Based on the study results, strategies for optimizing and enhancing urban forests in Qingdao are proposed for different urban-rural gradients, respectively. This study can provide a scientific framework for urban biodiversity conservation and management in Qingdao and serve as a guide for urban forests and greening with comparable climates.

Keywords

Urban-rural gradient
Urban forest
Woody plant diversity
Evaluation of plant configuration rationality
==== Body
pmc1 Introduction

Urban forests comprise all woodlands, tree groups, and infrastructure found in cities and peri-urban locations [1,2]. Urban forest construction is an important component of urban construction [3,4], as it helps to mitigate climate change and the heat island effect, improve environmental pollution (air, rainwater), increase biological carbon sequestration, conserve biodiversity, and maintain the sustained and healthy development of urban ecosystems in a small area [[5], [6], [7], [8]]. Urban woods are typically more species-rich than surrounding urban areas [[9], [10], [11]]. However, human activities and urbanization continue to reduce forest cover and biodiversity [12]. A study showed that the interaction between environmental gradients and ecosystems affects the distribution and behavior of organisms along the gradient within the ecosystem, and for the first time, the urban-rural gradient was introduced into the study of the structure and function of forest ecosystems [13]. Studies on urban forests in Qingdao mainly include: exploring the causes of vegetation diversity patterns in urban forests in the context of urbanization [14], subdividing urban forests according to the gradient of distance from the point of interest, and analyzing the diversity of woody plants and landscape patterns in urban forests [15], and so on. At present, there are relatively few studies related to the spatial layout, structure, and inter-functional relationship of Qingdao's urban forests under the urban-rural gradient.

Changes in the service functions of urban and rural ecosystems have received increased academic interest in recent years, as humans and the environment interact more closely [[16], [17], [18], [19], [20]]. The urban-rural gradient is an effective instrument for evaluating the effects of urbanization on ecosystems [21], and it has increasingly emerged as a source of concern in the field of ecological research. The spatial difference between urban and rural environmental gradients can substitute for the temporal difference of global change, allowing the study of biological responses to urbanization to predict future climatic change [22]. Qingdao has a peculiar pattern that differs from the "spreading" expansion trend in plain cities [23]. The research method based on urban-rural gradient can compare and analyze the differences in forest ecosystems in Qingdao more intuitively, and provide a scientific basis for the study of the impact of urbanization on urban forest ecosystems. However, most of the current studies use a single indicator to classify the urban-rural gradient, which does not apply to Qingdao City.

Woody plants are an important component of urban forests and their diversity determines the biodiversity of forest ecosystems. Maintaining and regularly assessing diverse ecosystems and the entire range of biodiversity within them is essential for the long-term survival of humankind [[24], [25], [26], [27]]. Woody plant diversity has been studied at multiple levels, including species richness ([28], genetic diversity [29], ecological niche differentiation [30], and ecosystem function [31], among others. European and American researchers have revealed the complex relationship between woody plant diversity and ecosystem services through long-term ecological monitoring and experimental studies [32]. Existing studies have found a clear pattern of urban-rural gradients in similarities and differences in species diversity [33]. However, further research is still needed on how woody plants can respond more effectively to the impacts of climate change, and how to conserve and increase biodiversity in the process of urbanization.

This study used imperviousness [[34], [35], [36], [37]] and construction time [33] to classify the urban-rural gradient in Qingdao City. The study aims to investigate the relationship between woody plant diversity and the urban-rural gradient in urban forests in Qingdao, as well as to evaluate the rationality of urban forest allocation based on the findings. The methodology and results are equally relevant to other places with a comparable climate and level of urban development as Qingdao.

The remainder of the paper is organized into four main sections. Section 2 presents the case study area, the division of the urban-rural gradient, and the analytical methods used in the article. Section 3 presents the results, including the species composition of urban forests, index values of woody plant diversity, and their correlation with urban-rural gradients. Section 4 discusses the factors affecting woody plant diversity on different urban-rural gradients and the rationality of urban forest allocation. Finally, Section 5 provides the conclusions of our study.

2 Study area and methods

2.1 Study area

Qingdao is located in the southern part of the Shandong Peninsula, and its vegetation belongs to the warm-temperate deciduous broadleaf forest belt's southern deciduous oak forest subzone, with deciduous oak species being the most zonally characterized tree species [38]. As of 2017, Qingdao has a total area of 11,282 km2, of which around 700 km2 are urban woods, with seven districts and three county-level cities under its administration (Fig. 1). Qingdao's total population reached 6,452,200 in 2020 (Qingdao Bureau of Statistics, 2020), which renders it a mega-city.Fig. 1 Study area location.

Fig. 1

2.2 Methods

2.2.1 Delineation of the urban-rural gradient

This study established a fishnet grid based on the extent of Qingdao's urban boundary, and Qingdao was divided into 4742 2km × 2 km grids. The ratio of the area occupied by impervious surfaces within a unit area is equal to the value of imperviousness, and the imperviousness within each grid is computed and allocated to the urban-rural gradient value of each grid [14]. Based on the imperviousness calculation results for each grid, the urban built-up area boundary data from three years (1992, 2006, and 2020), and the field survey, Qingdao's urban development density was classified into four classes based on the time of construction and imperviousness (Table 1), and the urban-rural gradient map of Qingdao was generated in the ArcGIS 10.6 software (Fig. 2).Table 1 Qingdao urban-rural gradient.

Table 1Gradient Level	Urban-Rural Gradient	Impermeability (abbreviation: I)	Construction Time	
L1	High-density urban areas (HU)	90 % < I ≤ 100 % (IL1)	The built-up area before 1992 (CL1)	
L2	Medium-density urban area (MU)	70 % < I ≤ 90 % (IL2)	The built-up areas between 1992 and 2006 (CL2)	
L3	Low-density urban areas (LU)	40 % < I ≤ 70 % (IL3)	The built-up areas between 2007 and 2020 (CL3)	
L4	Non-urban area (NU)	0 % < I ≤ 40 % (IL4)	Unbuilt Area (CL4)	

Fig. 2 Qingdao's urban-rural gradient (a) by imperviousness (b) by time of construction.

Fig. 2

2.2.2 Sample Setup

Based on the 2020 land use type data of Qingdao and the 2018 vegetation type data of Shandong Province, a total of 1649 urban forest grids of 2 km × 2 km were constructed using GIS software to identify, extract, and overlay to obtain the Qingdao urban forest distribution map (Fig. 3). Random stratified sampling was used to select sample sites, and a total of 184 urban forest grids and 787 sample squares of 400 m2 were selected (Fig. 3 and Table 2).Fig. 3 Sampling locations.

Fig. 3

Table 2 Sampling location details.

Table 2Classification	By Imperviousness	By Construction Time	Total	
Urban-Rural Gradient	IL1	IL2	IL3	IL4	CL1	CL2	CL3	CL4	
Number of sample sites in Huangdao	6	9	10	18	0	15	9	19	43	
Number of sample sites in Pingdu	3	2	2	16	0	6	1	16	23	
Number of sample sites in Laixi	1	2	6	12	0	1	2	18	21	
Number of sample sites in Jimo	4	4	8	20	0	7	1	28	36	
Number of sample sites in Jiaozhou	0	1	4	10	0	0	3	12	15	
Number of sample sites in Chengyang	1	5	7	6	0	7	4	8	19	
Number of sample sites in Laoshan	1	1	5	7	0	5	2	7	14	
Number of sample sites in Shinan	0	1	0	2	1	1	0	1	3	
Number of sample sites in Shibei	2	2	0	0	4	0	0	0	4	
Number of sample sites in Licang	2	2	1	1	0	5	0	1	6	
Total	20	29	43	92	5	48	21	110	184	

From December 2022 to April 2023, a total of 12 types of urban forests were investigated, primarily road greens, ancillary greens, dark greens, waterfront greens, mountains, scenic greens, plaza greens, eco-park greens, countryside greens, orchards, nurseries, and amusement park greens. In the urban forest field sampling, a total of 188,197 woody plants were analyzed, and 190 plant species were recorded, including 113 tree species and 77 shrub species. These species are classified into 109 genera and 50 families.

2.3 Data analysis

2.3.1 Species composition

The statistics and calculations from the field research data were used to compare and analyze the composition of woody plants in urban forests in Qingdao, the number of species in each family, and the composition of trees and shrubs in the community. Importance value is a comprehensive quantitative index that expresses the relative importance of a plant species in a community with a comprehensive value [39], which can explain the occupation of this species in its living environment to some extent. In this study, we calculated the relative frequency, relative abundance, relative significance, and cover of woody plants to assess the importance of plant species in the urban forest of Qingdao City.

2.3.2 Diversity index

Margalef index [40]:(1) H = (S-1)/ln N

Shannon-Wiener index [40]:(2) H’ = -Σ Pi (ln Pi)

Pielou index [41]:(3) E = H’/ln S

Simpson index [41]:(4) D=∑i=1S1−Pi2

Where S denotes the total number of species; N denotes the total number of plants, and Pi denotes the relative plurality of each species.

2.3.3 Correlation analysis

Pearson correlation coefficient is used in statistics to assess the correlation (linear correlation) between two variables X and Y, and its value ranges from −1 to 1. Typically, P ＞ 0.05 indicates that the difference is not significant, 0.01 ＜ P ＜ 0.05 indicates that the difference is significant, and P ＜ 0.01 indicates that the difference is highly significant. We used Excel to calculate the diversity indices of urban forests grid sample data under different urban-rural gradients, and SPSS and Origin to analyze the correlation between its diversity and urban-rural gradients.

2.3.4 Evaluation of urban forest configurations

The 10/20/30 rule of thumb was proposed and verified by Santamour and Kendal et al. to be reasonable and universal [42,43]. Therefore, this study utilized the 10/20/30 rule at the species, genus, and family levels to evaluate whether urban forest allocation is reasonable. The specific evaluation method is to see whether the relative abundance of the most common species (genus and family) in the region is less than 10 % (20 %, 30 %), if it is less than that, it indicates that the configuration of the species in the region is reasonable, and vice versa is not reasonable [42].

3 Result

3.1 Community composition of woody plants in urban forests along different urban-rural gradients

From the field research data, it can be seen that Rosaceae (23.99 %), Juniperus(16.31 %), Cedrus deodara (10.60 %), Populus tomentosa (7.23 %), and Buxus megistophylla (18.88 %) are the most abundant families, genera, evergreen trees, deciduous trees, and shrubs in Qingdao City Forest, respectively.

The most common evergreen, deciduous, and shrub species in the urban forest of Qingdao were C.deodara (68.31 %), Platanus orientalis (55.19 %), and B.megistophylla (67.76 %) (Table 3). Evergreen trees with high relative importance values in Qingdao's urban forests mainly include C.deodara (10.60 %), Pinus thunbergia (6.94 %), Juniperus chinensis 'Kaizuca' (4.02 %), Juniperus chinensis (2.91 %), Pinus bungeana (1.42 %), etc. Deciduous trees with high relative importance values in urban forests mainly include P.orientalis (14.44 %), P.tomentosa (9.98 %), Prunus cerasifera 'Atropurpurea' (4.62 %), Ginkgo biloba(3.57 %), Prunus persica(3.31 %), etc. And shrubs with high relative importance values in urban forests mainly include Photinia × fraseri (23.75 %), B.megistophylla(12.76 %), J.chinensis 'Kaizuca'G (11.86 %), Buxus sinica var. parvifolia (11.36 %), Euonymus japonicus 'Aurea-marginatus' (4.77 %) and so on.Table 3 Important plant distribution in the urban woods of Qingdao.

Table 3Important Plant Species	Number of Trees	Number of Distributed Grids	
Evergreen Tree Species	Cedrus deodara	4188	125	
Pinus thunbergii	3670	89	
Juniperus chinensis 'Kaizuca'	1806	74	
Juniperus chinensis	1425	40	
Deciduous Tree Species	Populus tomentosa	2857	81	
Platanus orientalis	2417	101	
Prunus cerasifera	2150	82	
Ginkgo biloba	1407	67	
Prunus persica	1397	52	
Prunus × yedoensis	1161	68	
Salix babylonica	928	51	
Sophorajaponica Linn.	905	65	
Styphnolobium japonicum 'Golden Stem'	736	37	
Shrub Species	Buxus megistophylla	28077	124	
Juniperus chinensis 'Kaizuca'	26474	90	
Photinia × fraseri	24870	117	
Buxus sinica var. parvifolia	18753	86	
Euonymus japonicus 'Aureo-marginatus'	9611	43	

The greatest number of species and genera were found in the area with the lowest density of urbanization (L4), indicating that the urban forest in this area is well-grown, high in size, and widely distributed. However, as the density of built-up areas in cities dropped, so did the number of shrub species. As urban density decreased, there was an upward tendency in the number of plant species found in the built-up urban region. The chronological gradient from the early to the recent periods indicated a growing tendency in the number of cypress species in the urban forests of Qingdao. It was found that urbanization increased the diversity of woody plants in urban forests in Qingdao to some extent, but not always.

There are 35 tree species in Qingdao's urban forest during 1992–2020 that have changed over time but have always been present. Among them, six species showed a trend of increasing all the time, namely Yulania denudata (△1992-2006 = 0.024, △2007-2020 = 0.155), Pinus densiflora (△1992-2006 = 0.216, △2007-2020 = 0.436), S.babylonica (△1992-2006 = 0.776, △2007-2020 = 0.334), P.orientalis (△1992-2006 = 1.168, △2007-2020 = 0.7935), Pinus tabuliformis (△1992-2006 = 0.198, △2007-2020 = 0.115), and P.cerasifera 'Atropurpurea' (△1992-2006 = 0.251, △2007-2020 = 2.260). Eleven tree species, including P.thunbergii (△1992-2006 = −2.990, △2007-2020 = −1.609), J.chinensis 'Kaizuca '(△1992-2006 = −1.327, △2007-2020 = −0.601), S.japonicum 'Golden Stem' (△1992-2006 = −0.070, △2007-2020 = −1.534), and so on, have been diminishing. Between 1992 and 2020, 18 shrub species continued to exist in urban forests, with the B.megistophylla (△1992-2006 = 19.166, △2007-2020 = 8.993) and the J.chinensis 'Kaizuca' (△1992-2006 = 7.569, △2007-2020 = 14.348) showing an upward trend. Seven shrub species were consistently present over time and showed a decreasing trend, such as Pittosporum tobira (△1992-2006 = −0.361, △2007-2020 = −0.330), Hibiscus syriacus (△1992-2006 = −0.771, △2007-2020 = −0.146), Osmanthus fragrans 'Semperflorens' (△1992-2006 = −1.180, △2007-2020 = −0.011), Photinia serratifolia (△1992-2006 = −0.397, △2007-2020 = −0.846), Berberis thunbergii 'Atropurpurea' (△1992-2006 = −0.030, △2007-2020 = −1.686), Jasminum nudiflorum (△1992-2006 = −3.843, △2007-2020 = −0.357), Forsythia suspensa (△1992-2006 = −3.483, △2007-2020 = −1.057) (Table 4).Table 4 Woody plants time-series changes details.

Table 4Species name	Changes per 400 m2 between 1992 and 2006	Changes per 400 m2 between 2007 and 2020	Species name	Changes per 400 m2 between 1992 and 2006	Changes per 400 m2 between 2007 and 2020	
Yulania denudata	0.024	0.155	Metasequoia glyptostroboides	−0.576	−0.195	
Pinus densiflora	0.216	0.436	Acer palmatum 'Atropurpureum'	−0.889	−0.354	
Salix babylonica	0.776	0.334	Magnolia grandiflora	−1.212	−0.059	
Platanus orientalis	1.168	0.793	Prunus mume	−1.366	−0.289	
Pinus tabuliformis	0.198	0.115	Buxus megistophylla	19.166	8.993	
Prunus cerasifera 'Atropurpurea'	0.251	2.260	Juniperus chinensis 'Kaizuca' (shrubby)	7.569	14.348	
Pinus thunbergii	−2.990	−1.609	Pittosporum tobira	−0.361	−0.330	
Juniperus chinensis 'Kaizuca'	−1.327	−0.601	Hibiscus syriacus	−0.771	−0.146	
Styphnolobium japonicum 'Golden Stem'	−0.070	−1.534	Osmanthus fragrans 'Semperflorens'	−1.180	−0.011	
Yulania liliiflora	−0.041	−0.046	Photinia serratifolia	−0.397	−0.846	
Camellia japonica	−0.260	−0.016	Berberis thunbergii 'Atropurpurea'	−0.030	−1.686	
Robinia pseudoacacia	−0.340	−0.074	Jasminum nudiflorum	−3.843	−0.357	
Acer palmatum	−0.384	−0.135	Forsythia suspensa	−3.483	−1.057	

3.2 Diversity of woody plants in urban forests under different urban-rural gradients

Analysis of the study data revealed some differences in the Margalef index, Shannon-Wiener index, Pielou index, and Simpson index of woody plants in urban forests on different urban-rural gradients in Qingdao (Table 5). The tree layer of urban forest in Qingdao has greater richness and diversity than the shrub layer; the most dominant tree species in urban forest are still dispersed in the tree layer, and the species evenness of the tree layer is also greater than that of the shrub layer.Table 5 Qingdao's urban woods' woody plant biodiversity index.

Table 5Plant Type	Margalef Index (H)	Shannon-Wiener Index (H′)	Pielou Index (E)	Simpson Index (D)	
Tree layer	12.604	3.662	0.745	0.958	
Shrub layer	7.159	2.523	0.566	0.872	
Overall	18.196	3.344	0.619	0.923	

In this study, we used Kriging interpolation to import the research results into GIS software to obtain the distribution map of woody plant diversity (Fig. 4). The considerable overlap between highly urbanized areas and high biodiversity areas makes this evident. It implies a close relationship between plant diversity and urbanization level. It was found that most (about 68.58 %) of the Shannon-Wiener index of urban forest woody plants in Qingdao were concentrated between 1.62 and 2.43, followed by 1.21–1.62 and 2.43–2.84, and only a small proportion (about 3.01 %) of urban forest woody plants with H' > 2.84 or H ' ≤ 1.21 (Fig. 5).Fig. 4 Woody plant diversity index distribution (a) Shannon-Wiener index (c) Margalef index (e) Simpson index (g) Pielou index, high/low diversity index values cluster (b) Shannon-Wiener index (d) Margalef index (f) Simpson index (h) Pielou index.

Fig. 4

Fig. 5 Qingdao's woody plant diversity's normal distribution.

Fig. 5

The highest value of the Shannon Wiener Index is 3.11, which is from the grid sample numbered 658 in the L1 area (Table 6). This sample site is a typical park green space in the urban forest, Zhushui Hill Children's Park, which has a beautiful environment, rich species, high flow of people and is located in the main urban area of Qingdao City easy to travel, where visitors mainly carry out activities such as parent-child recreation, photography and scenic views, and recreational activities. The highest plant diversity in sample site 658 is also a result of meeting people's needs.Table 6 Qingdao Zhushuishan Children's Park Information (highest Shannon Wiener Index = 3.11).

Table 6

The urban forest with the lowest Shannon Wiener Index (H' = 0.32) is from the grid plot number 906 in the area of L4 (by imperviousness) and L2 (by construction time) (Table 7). This sample site is also a parkland in the urban forest, Laohushan Park (northern part), but it is sparsely populated by species, less traveled by people, and not easily accessible on steep terrain. Compared with the southern part of the park, which is close to the entrance and has gentle terrain, sample site 906 in the same Laohushan Park has few visitors and a simple plant composition, while the southern part of the park, on the other hand, is rich in species and has a high flow of people. This shows that the needs and preferences of visitors, topography, and ease of access are important factors affecting plant diversity in urban forests.Table 7 Details of Laohushan Park's northern region in Qingdao (lowest Shannon Wiener Index = 0.028).

Table 7

3.3 Woody plant diversity of urban forests in Qingdao and its correlation with urban-rural gradient

Our study found that the four diversity indices were the highest for both NU areas based on imperviousness classification and UA areas based on construction time (Fig. 6, Fig. 7).Fig. 6 Bar graph of biodiversity index based on imperviousness classification under various urban-rural gradients (“ totle” means the sum of shrubs and trees).

Fig. 6

Fig. 7 Based on the classification of building duration, a bar graph of the biodiversity index under various urban-rural gradients (“ totle” means the sum of shrubs and trees).

Fig. 7

The degree of artificial intervention in urban built-up areas is positively correlated with the diversity and abundance of woody plants in urban forests, which are expected to contribute to ecosystem services and animal protection policies due to their higher species diversity. This was demonstrated by the gradual decline of the Margalef index and Shannon-Wiener index from high-density urban areas to low-density urban areas. However, there was little difference between the Pielou and Simpson indices of each urban-rural gradient, suggesting that urbanization had less of an impact on these two indices and that the percentage of dominant species in each stratum was roughly equal.

Except for the Pielou index, which is the highest, all three diversity indices are lowest in the pre-1992 built-up area when examining the urban-rural gradient based on the time of construction. The Margalef and Shannon-Wiener indices were the highest in the built-up area between 1992 and 2006. The Pielou and Simpson indices for the tree layer were similar for all three urban-rural gradients within the built-up area. The Simpson's index for the shrub layer was also relatively similar, but the Pielou's index for the shrub layer became lower as the urban-rural gradient became lower. This suggests that the richness and diversity of woody plants in Qingdao's urban forests increased with the time of establishment, but the proportion of dominant species in the urban forests remained almost the same.

SPSS analysis revealed that the four diversity indices of woody plants were significantly correlated with the urban-rural gradient (Table 8, Table 9). There was a highly significant positive correlation between the four diversity indices, but there was no significant correlation between the Margalef index and the Pielou index. This indicates that the richness and diversity of woody plants in urban forests showed a significant decrease with the gradual decrease of urbanization or the time of construction, but the evenness showed a significant increase. Simpson index was significantly and negatively correlated with the urban-rural gradient, and the index of ecological dominance gradually decreased as the degree of urbanization decreased or the time of construction went from far to near.Table 8 Correlation between woody plant diversity indices and urban-rural gradients (by imperviousness).

Table 8	Urban-Rural Gradient	Margalef Index (H)	Shannon-Wiener Index(H′)	Pielou Index (E)	Simpson Index(D)	
Urban-Rural Gradient	1					
Margalef Index (H)	−0.589a	1				
Shannon-Wiener Index(H′)	−0.373a	0.763a	1			
Pielou Index (E)	0.272a	−0.117	0.491a	1		
Simpson Index(D)	−0.170b	0.458a	0.845a	0.638a	1	
a The connection is significant at the double-tailed 0.01 level.

b The correlation is significant at the 0.05 level (double-tailed).

Table 9 Correlation between woody plant diversity indices and urban-rural gradients (by construction time).

Table 9	Urban-Rural Gradient	Margalef Index (H)	Shannon-Wiener Index(H′)	Pielou Index (E)	Simpson Index(D)	
Urban-Rural Gradient	1					
Margalef Index (H)	−0.530a	1				
Shannon-Wiener Index(H′)	−0.360a	0.754a	1			
Pielou Index (E)	0.174b	−0.110	0.507a	1		
Simpson Index(D)	−0.148b	0.450a	0.850a	0.655a	1	
a The connection is significant at the double-tailed 0.01 level.

b The correlation is significant at the 0.05 level (double-tailed).

4 Discussion

4.1 Characterization of woody plant diversity under different urban-rural gradients and its influencing factors

4.1.1 Characterization of woody plant diversity

The results of principal component analysis (PCA) showed that PC1 and PC2 in the shrub group accounted for 92.6 % and 3.8 % of the total variance, respectively (Fig. 8 a&b). Principal component analysis showed that, except for the NA and UA gradients, which had an almost equal influence on urban forest shrub composition, several other urban and rural gradients did not have the same influence on shrub composition. The HU gradient had the highest influence on urban forest shrub composition in Qingdao, followed by the NA and UA gradients. The B1992 gradient had less influence on plant species composition. Among them, the similarity of the composition of plant species under different urban and rural gradients was high, except for the plants of Photinia × fraseri, B. megistophylla, J. chinensis 'Kaizuca', B. sinica var. parvifolia, and E. japonicus 'Aurea-marginatus' (Fig. 8b).Fig. 8 Plant species principal component analysis.

Fig. 8

However, the results of principal component analysis (PCA) showed that PC1 and PC2 in the arborvitae group accounted for 82.1 % and 9.1 % of the total variance, respectively (Fig. 8 c&d). Principal component analysis (PCA) showed that two gradients, B1992–2006 and MU, had a higher influence on the tree composition of urban Sen in Qingdao. Among them, except for a dozen species of arborvitae, such as P. tomentosa, C. deodara, P. thunbergii, Bambusa oldhamii, P. orientalis, P. cerasifera 'Atropurpurea', and J. chinensis, other arborvitae species had higher compositional similarity under different urban-rural gradients (Fig. 8d).

We also found that each urban-rural gradient stratum contained more deciduous tree species and numbers than evergreen tree species and numbers, as well as more tree species than shrub species but more shrubs than trees in total numbers (Appendix Table A1). This shows that urbanization is an important factor influencing the distribution of plant diversity [44].

4.1.2 Influencing factors of woody plant diversity

The extent of anthropogenic pressure has been identified as one of the variables influencing the flora composition of urban woods [45], and this effect was supported by the results of this study through the collection and analysis of field research data. We discovered that when the urban-rural gradient class declined (from L1 to L4), the average number of woody plants per sample in the urban forest reduced as well. Other factors influencing the distribution of urban vegetation include property ownership types, natural and social environments, cultural differences, resident education level, and management [46], and the distribution of urban forest species is likely to be influenced by the aforementioned factors as well. The findings of this study also revealed that urbanization and anthropogenic causes had a significant impact on woody plants in Qingdao's urban forests, with a more pronounced divergence along the urban-rural gradient.

The rapid development of urbanization has brought great challenges to both human and natural environments [47,48]. Urbanization intensity is often used as a quantitative indicator to study the intensity of urban expansion [48] or urbanization effects [49]. We found that regions with higher urbanization intensity had higher woody plant diversity in urban forests, but their Simpson indices were lower, which is consistent with previous findings [15].

The higher the degree of urbanization and the higher the demand of people, the richer the composition of woody plants and the higher the proportion of ornamental plants in urban forests. The geographical distribution and economic development level influence the composition of woody plants in Qingdao's urban woods. The mix of woody plants in urban forests located in dense metropolitan centers with high economic development levels (e.g., the major urban area and the area around Jiaozhou Bay) is richer, and the proportion of plants with ornamental value is larger. On the contrary, the composition of woody plants is less diverse, with a lower proportion of species with ornamental value and a higher proportion of plants with economic worth.

Studies have shown that urban forest woody plant diversity indices are significantly correlated with the urban-rural gradient and that different land use types and habitats influence vegetation composition, a finding supported by many comprehensive analyses and comparative studies of vegetation diversity by land use type [[50], [51], [52], [53]]. Urbanization has a significant impact on the distribution of plant diversity [44]. Urbanization has resulted in urban forest landscapes in high-density urban areas being almost entirely dominated by anthropogenic plant communities, which has led to spatial variation in plant diversity as a combination of heterogeneity in habitat conditions and anthropogenic features [54]. Urban areas also reflect the diversity of urban forest management regimes. The urban-rural gradient study is a simplification of the complex distribution pattern of species diversity, and our results amply demonstrate that biodiversity is significantly correlated with the urban-rural gradient.

4.2 Evaluation of rationality of urban forest allocation and optimization suggestions

4.2.1 Evaluation of rationality of urban forest allocation

After analyzing the 10/20/30 rule of empirical, it was found that the plant allocation at the family and genus level in urban forests in each urban-rural gradient level divided in two ways was more reasonable, but the relative abundance of the most common genus in built-up areas between 2006 and 2020 (23.81 %) was higher than the 20 % allocation that was more unreasonable (Fig. 9).Fig. 9 Relative abundance of the most common families and genera in urban forests at different urban-rural gradients (a) by imperviousness (b) by construction time (dashed lines indicate the 10/20/30 rule).

Fig. 9

The shrub species configuration was inappropriate in all gradient levels based on imperviousness, but the relative abundance of shrub species in the non-urban area was closest to 10 % (12.1 %). This suggests that of the four levels based on imperviousness, the non-urban zone has the highest rationality in the allocation of woody plants in the urban forest compared to the other three tiers. Except for the pre-1992 construction region, where shrub species allocation was more logical, the rationality of shrub species allocation was lower in the three tiers depending on the time of construction. The above issue of shrubs in urban forests being arranged irrationally is counterproductive to the preservation and promotion of plant diversity in urban forests.

4.2.2 Optimization and Enhancement strategies for plant configurations in urban forests

The analysis revealed that the urban forest configuration in Qingdao City has a significant configuration irrationality problem at the shrub species level, with the shrub species of the urban forest currently relying primarily on plants from the Rosaceae, Cypressaceae, Celastraceae, Buxaceae Dumort. and so on. Given the aforementioned issues, the research summarizes the suggestions for tree species selection for urban forests in Qingdao (Appendix Table A2).

The selection of urban forest shrubs in each urban-rural gradient tier by imperviousness can be appropriately reduced by the use of B.megistophylla, J.chinensis 'Kaizuca'(shrub), Photinia × fraseri, B.sinica var. parvifolia, and so on. Currently, the urban forest is dominated by the evergreen tree C.deodara, with a more balanced distribution of deciduous trees. Rosaceae, Cupressaceae, and Celastraceae plants make up the majority of the shrub species in urban forests found in built-up areas from 1992 to 2006, built-up areas from 2007 to 2020, and non-built-up regions. The utilization of shrubs like B.megistophylla, J. chinensis, and Photinia × fraseri can be suitably reduced by choosing shrub species for urban forests under these three gradients. At all scales, there is a fairly uniform distribution of evergreen tree species, but this is not the case for deciduous tree species, particularly in areas that were constructed before 1992 and between 2007 and 2020.

This study demonstrates that the process of urbanization has some bearing on species richness and that species richness tends to decrease with increasing urban development density. The construction of gardens can improve the city's ecological performance and have a positive impact on natural environmental protection during the process of creating an urban ecological environment. If the requirements are met, the development of an urban forest should take the variety of plants and the configuration's logic into full consideration. Therefore, it is suggested that in the future, the selection of tree species in urban forests under these four levels may minimize the use of C.deodara, while increasing the use of other evergreen tree species such as P.tabuliformis, Chamaecyparis obtusa, Chamaecyparis pisifera, L.lucidum, M.grandiflora, and so on. So that the ratio of trees and shrubs, and the ratio of evergreen plants to deciduous plants can reach a relatively more reasonable range.

Different landscape effects will be presented in different seasons as a result of the various plant characteristics, such as flowering and fruiting times, tree shapes, leaf shapes, and leaf colors. Through the reasonable screening and pairing of plants, there are landscape iterations at various times, increasing the diversity of spatial distribution. Urban forests should be built and maintained in a way that fully utilizes local resources, adapts to local conditions, respects nature, and embodies the idea of "adapting to nature, transforming nature" as well as the mutual borrowing between people and the environment, as well as between the environment and buildings.

They should also fully exploit the role that plants play in regulating the local microclimate [55]. The development and design of native plants should be strengthened during the process of constructing an urban forest plant landscape, and a plant landscape with native features should be actively developed. Utilize all of the abundant local tree resources in Qingdao, including the Osmanthus fragrans, Liquidambar formosana, Ailanthus altissima, Acer palmatum, and other species of plants with a more beautiful shapes. Not only can form a unique landscape in the process of landscape design but also can be combined with other related plants to achieve a good color and form beauty [56]. It is also recommended to prioritize species that are resistant to sea breeze/direct coastal sunlight, such as Eucommia ulmoides, Catalpa ovata, Acer palmatum, etc., based on the regional characteristics of Qingdao's coastal area[[57], [58]].

We also suggest that the selection of deciduous trees for urban forests in built-up areas before 1992 may appropriately increase the number of plants in the Magnoliaceae, Sapphiraceae, and Salicaceae families, including M.denudata, Y.iliiflora, K.paniculata, P.tomentosa and so on. Salicaceae family plants (e.g., P.tomentosa, etc.) should be reduced in the built-up area between 2007 and 2020. In addition, the number of plants of the families Magnoliaceae, Sapphiraceae, Rosaceae, Celiaceae, and Meliaceae (such as M.denudata, M.grandiflora, Y.iliiflora, K.paniculata, Prunus × yedoensis, L.indica, M.azedarach, etc.) can be adequately increased in the built-up region between 2007 and 2020. Details of the design strategies for urban forest planting on different urban-rural gradients are as follows (Table 10).Table 10 Urban forest plant community design strategies for different urban-rural gradients.

Table 10

5 Conclusion

Understanding the distribution pattern of woody plant variety in urban forests and the factors that influence it is critical for urban forest planning as well as the protection and improvement of the urban environment. Based on a large amount of field survey data, we discovered that the majority of the common plants found in Qingdao's urban forests belong to the Rosaceae, Cupressaceae, Buxaceae, Oleaceae, and Celastraceae families. Additionally, we discovered that the area and dispersion of urban forests are greater in non-urban locations.

The present investigation revealed a strong correlation between the diversity of woody plants and the urban-rural gradient. The Pielou index showed a negative correlation with the urban-rural gradient, but the Margalef, Shannon-Wiener, and Simpson indices showed a positive correlation. Future studies can further enhance the relationship between woody plants in urban forests and urban-rural gradients by enlarging the sample size of the data and thoroughly accounting for all the variables that contribute to the definition of the urban-rural gradient. Meanwhile, this study also found that the urban forest in Qingdao has a large allocation problem at the shrub species level, and there is a large gap between the plant diversity of urban forests in different urban-rural gradients. We propose a series of relevant recommendations and specific optimization strategies for Qingdao's urban woods. This holds not only for the study region but also for the establishment and optimization of urban forests in other places with comparable climate conditions and patterns of urbanization. This holds not only for the study region but also for the establishment and optimization of urban forests in other places with comparable climate conditions and patterns of urbanization.

CRediT authorship contribution statement

Ruirui Zhu: Writing – review & editing, Resources, Project administration, Methodology, Funding acquisition, Conceptualization. Danping Xiu: Writing – original draft, Validation, Software, Investigation, Formal analysis, Conceptualization. Ruixin Xue: Visualization, Software. Shuo Teng: Validation, Software.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A Supplementary data

The following is the Supplementary data to this article:Multimedia component 1

Multimedia component 1

Acknowledgments

We would like to thank all participants for their help in the literature sorting and statistical analysis.

Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.heliyon.2024.e36901.
==== Refs
References

1 Dj N. Historical vegetation change in Oakland and its implications for urban forest management J. Arboricul. 5 19 1993 313 319
2 SalbitanoF Guidelines on Urban and Peri-Urban Forestry 2016 Food and Agriculture Organization (FAO) of the United Nations
3 Threlfall C.G. Kendal D. The distinct ecological and social roles that wild spaces play in urban ecosystems Urban For. Urban Green. 29 2018 348 356
4 Steenberg J.W.N. Duinker P.N. Nitoslawski S.A. Ecosystem-based management revisited: updating the concepts for urban forests Landsc. Urban Plann. 186 2019 24 35
5 Nowak D.J. Greenfield E.J. Tree and impervious cover in the United States Landsc. Urban Plann. 107 1 2012 21 30
6 Roman L.A. Human and biophysical legacies shape contemporary urban forests: a literature synthesis Urban For. Urban Green. 31 2018 157 168
7 Ostoic S.K. Urban forest research in the Mediterranean: a systematic review Urban For. Urban Green. 31 2018 185 196
8 Dobbs C. Kendal D. Nitschke C.R. Multiple ecosystem services and disservices of the urban forest establishing their connections with landscape structure and sociodemographics Ecol. Indicat. 43 2014 44 55
9 Alvey A.A. Promoting and preserving biodiversity in the urban forest Urban For. Urban Green. 5 4 2006 195 201
10 Almas A.D. Conway T.M. The role of native species in urban forest planning and practice: a case study of Carolinian Canada Urban For. Urban Green. 17 2016 54 62
11 Stewart G.H. The re-emergence of indigenous forest in an urban environment, Christchurch, New Zealand Urban For. Urban Green. 2 3 2004 149 158
12 Abunie A.A. Dalle G. Woody species diversity, structure, and regeneration status of yemrehane kirstos church forest of lasta woreda, north wollo zone, amhara region, Ethiopia International Journal of Forestry Research 2018 1 8
13 McDonnell M. Pickett S. Ecosystem structure and function along urban-rural gradients: an unexploited opportunity for ecology Ecology 71 4 1990 1232 1237
14 Liang H. Causes of Vegetation Diversity Patterns and Their Maintenance in Qingdao's Urban Forests in the Context of Urbanization 2014 East China Normal University 133
15 Yang J. The woody plant diversity and landscape pattern of fine-resolution urban forest along a distance gradient from points of interest in Qingdao Ecol. Indicat. 122 2021 107326
16 Gaglio M. Changes in land use and ecosystem services in tropical forest areas: a case study in Andes mountains of Ecuador International journal of biodiversity science, ecosystems services & management 13 2017 264 279
17 Haase D. Synergies, trade-offs, and losses of ecosystem services in urban regions: an integrated multiscale framework applied to the leipzig halle region, Germany Ecol. Soc. 17 2012 22
18 Yu D. Zhou X. Qi H. Low-carbon city planning based on collaborative analysis of supply and demand scenarios City Built Enviro 1 2023 7
19 Zhu R.R Yuan L.Y Pan Y.T Effects of natural sound exposure on health recovery: a systematic review and meta-analysis Sci. Total Environ. 921 2024 171052 38373459
20 Zhu R.R Liu W.Y Xue R.X. Linking wilderness mapping and ecosystem services: identifying integrated wilderness and ecological indicators to quantify ecosystem services of wilderness Ecol. Indicat. 160 2024 111915
21 Nagy R.C. Lockaby B.G. Urbanization in the South eastern United States: socioeconomic forces and ecological responses along an urban-rural gradient Urban Ecosyst. 14 2011 71 86
22 Jillian W.G. Clive G.J. Todd E.D. Urbanization effects on tree growth in the vicinity of New York City Nature 424 2003 183 187 12853954
23 Yang C. Study on the Evolution of Qingdao’s Urban Spatial Form since 1990’s 2010 Shandong Jianzhu University 84
24 PushpangadanP, RaviK and SantoshV, Conservation and Economic Evaluation of Biodiversity 1997 Oxford and IBH publishing Co. Pvt. Ltd. New Delhi
25 Wagg C. Plant diversity maintains long-term ecosystem productivity under frequent drought by increasing short-term variation Ecology 98 11 2017 2952 2961 28869781
26 MalikZA Phytosociological Behavior, Anthropogenic Disturbances and Regeneration Status along an Altitudinal Gradient in Kedarnath Wildlife Sanctuary (KWLS) and its Adjoining Areas 2014 Garhwal University (Srinagar Uttarakhand)
27 MalikZA Species richness and diversity along the disturbance gradient in Kedarnath Wildlife Sanctuary and its adjoining areas in Garhwal Himalaya India 2014 10918 10926
28 Solomou A.D. Sfougaris A. Predicting woody plant diversity as key component of ecosystems: a case study in Central Greece Int. J. Agric. Environ. Inf. Syst. 10 1 2019 1 20
29 Hamrick J. Godt M. Effects of life history traits on genetic diversity in plant species Phil. Trans. Biol. Sci. 351 1345 1996 1291 1298
30 Umarani M.S. Wang D. O’Dwyer J.P. D’Andrea R. A spatial signal of niche differentiation in tropical forests Am Nat. 203 4 2024 445 457 38489774
31 Ding J. Multiple trade-offs regulate the effects of woody plant removal on biodiversity and ecosystem functions in global rangelands Global Change Biol. 26 2 2019 709 720
32 Ding J. Eldridge D.J. Ecosystem service trade-offs resulting from woody plant removal vary with biome, encroachment stage and removal method J. Appl. Ecol. 61 2 2023 236 248
33 Xiao L. Urban-rural and temporal differences of woody plants and bird species in Harbin city, northeastern China Urban For. Urban Green. 20 2016 20 31
34 Xian G.G. An analysis of urban thermal characteristics and associated land cover in Tampa Bay and Las Vegas using Landsat satellite data Rem. Sensing Environ 2 104 2006 147 156
35 Zhang Y. Odeh I.O.A. Han C. Bi-temporal characterization of land surface temperature in relation to impervious surface area, NDVI and NDBI, using a sub-pixel image analysis Int. J. Appl. Earth Obs. Geoinf. 11 4 2009 9
36 Imhoff M.L. Remote sensing of the urban heat island effect across biomes in the continental USA Remote Sensing of Environment 114 3 2010 504 513
37 Xianlei M. A Multi-Scale Study of the Relationship between Impervious Surfaces, Vegetation, Water Bodies and Urban Heat Islands 2010 73
38 Chen Z. Landscape with Qingdao plants J. Shandong Univ. (Nat. Sci.) 1 1958 257 272
39 Jin Y. Plant community species diversity and comparison of its indices in arid desert land Arid. Land Geogr. V28 1 2005 7
40 Beisel J. Moreteau J. A simple formula for calculating the lower limit of Shannon's diversity index Ecol. Model. 99 2 1997 289 292
41 Magurran A.E. Ecological Diversity and its Measurement 1988 Princeton University Press New Jersey
42 Santamour F. Trees for Urban Planting: Diversity, Uniformity, and Common Sense 1990 57 65
43 Kendal D. Dobbs C. Lohr V.I. Global patterns of diversity in the urban forest: is there evidence to support the 10/20/30 rule? Urban For. Urban Green. 13 3 2014 411 417
44 Peng Y. Liu X. Research progress in effects of urbanization on plant biodiversity Biodivers. Sci. 15 5 2007 558 562
45 Fornal-Pieniak B. Wysocki C. Flora structure in rural parks on example of Sandomierska Valley Acta Sci. Pol. Silvarum Colendarum Ratio Ind. 5 2006 31 45
46 Kendal D. Williams N.S.G. Williams K.J.H. Drivers of diversity and tree cover in gardens, parks and streetscapes in an Australian city Urban For. Urban Green. 11 3 2012 257 265
47 Sahana M. Hong H. Sajjad H. Analyzing urban spatial patterns and trend of urban growth using urban sprawl matrix: a study on Kolkata urban agglomeration, India Sci. Total Environ. 628–629 2018 1557 1566
48 Yang J. Li S. Lu H. Quantitative influence of land-use changes and urban expansion intensity on landscape pattern in Qingdao, China: implications for urban sustainability Sustainability 11 21 2019 6174
49 Zhang D. Effects of urbanization intensity on forest structural-taxonomic attributes, landscape patterns and their associations in Changchun, Northeast China: implications for urban green infrastructure planning Ecol. Indicat. 80 2017 286 296
50 Jim C.Y. Liu H.T. Species diversity of three major urban forest types in Guangzhou City, China For. Ecol. Manag. 146 1 2001 99 114
51 Bourne K.S. Conway T.M. The influence of land use type and municipal context on urban tree species diversity Urban Ecosyst. 17 1 2013 329 348
52 Zhang D. Effects of forest type and urbanization on species composition and diversity of urban forest in Changchun, Northeast China Urban Ecosyst. 19 2015
53 Meng X.S. Composition of plant species and their distribution patterns in Beijing urban ecosystem Acta Ecol. Sin. 24 2004 2200 2206
54 Collins J.P. A new urban ecology Am. Sci. 88 2000 416 425
55 Zheng X. Research on Evaluation and Optimization of Garden Landscape in Nanjing Residential Community 2008 Nanjing Forestry University
56 Sun X. Song X. A study on the importance of plant culture in residential landscape plant configuration Xiandai Horticulture 43 23 2020 66 68
57 Meng X. Analysis on rotation timing of dynamic Rotating Latent-Energy-Storage Envelope (RLESE) Develop. Built Environ. 19 2024 100498
58 Meng X. Thermal performance assessment of the Dynamic Rotating Latent-Energy-Storage Envelope (DRLESE) during Winter Case Stud. Therm. Eng. 60 2024 104671
