The Characteristics of Habitat, Functional Traits and Medicinal Components of Eucommia Ulmoides From Guizhou

To nd out the genuine characteristics of Eucommia ulmoides produced in Guizhou.The habitat, functional characters and the content of medicinal components of Eucommia ulmoides in Guizhou were studied by using the method of sample survey combined with typical survey, related laboratory experiments and quantitative analysis. The results showed that the yield of Eucommia ulmoides plantation in Guizhou was divided into low altitude, low middle and high temperature rain slope latitude mixing, short sunshine type(A type), medium altitude, low longitude and latitude, high temperature rain, positive oblique steep slope, medium sunshine type (B type), Middle altitude, low longitude and latitude, moderate high temperature rain, Shady side and Sunny side have gentle deection steep slope, medium sunshine type (C type), High altitude, low longitude and latitude, low temperature moderate rain, positive gentle slope, long sunshine type (D type); Different types of Eucommia ulmoides plantation, Different habitat quality, B type is intensity karst rocky desertication habitat, A type is potential karst rocky desertication habitat,The C and D types are light and moderate rocky desertication areas respectively, The species diversity of shrub layer in Eucommia ulmoides plantation was higher in D type and B type, A type and C type followed; There was no signicant difference in root carbon content and leaf nitrogen content in 4 types of Eucommia ulmoides plantation, Among the four types of A (cid:0) B (cid:0) C (cid:0) D, there were signicant or extremely signicant differences in other indexes of plant functional traits; Both genipinic acid and aucubin had the highest content of root bark, followed by trunk bark and lowest leaves, Chlorogenic acid is the opposite, The content of geniposide was higher in trunk bark and lower in root bark and leaves; Genipinic acid is higher in D type, Aucubin is higher in A and D type, Chlorogenic acid has higher leaves content in B type, Geniposide was the highest in trunk bark of D type; The element enrichment coecient K and Mn leaves are the largest, the largest in trunk bark is Ca and Zn, Fe root bark is the largest; Effects of soil potassium, phosphorus, pH value and bulk weight on the functional traits of Eucommia ulmoides were signicant. The contents of medicinal components in root bark, trunk bark and leaves was inuenced by species diversity of shrub layer, The contents of geniposide in root bark, aucubin in root bark and trunk bark, genipinic acid in bark and chlorogenic acid in leaves were particularly affected by soil physical and chemical indexes and metal element contents, The functional traits of Eucommia ulmoides can affect the content of medicinal components in root bark, trunk bark and leaves, Especially on the root bark, trunk bark and leaves in the content of aucubin content; The content of medicinal components of Eucommia ulmoides was high and stable. The above research results have important theoretical reference signicance for the cultivation of Eucommia ulmoides and the cultivation of target medicinal components and the comprehensive exploitation and utilization of resources. ulmoides in Guizhou, cultivation of target medicinal components and comprehensive development and utilization of resources. total nitrogen, alkali-hydrolyzable nitrogen, rapidly available phosphorus, and soil moisture; between leaf phosphorus content and total potassium, organic matter, total nitrogen, and alkali-hydrolyzable nitrogen; between root nitrogen content and rapidly available potassium, rapidly available phosphorus, soil bulk density; between root carbon nitrogen ratio and soil bulk density, total potassium, total phosphorus, rapidly available phosphorus, and pH; between branch and leaf index and soil moisture, total phosphorus, rapidly available phosphorus, soil bulk density. The above suggests that soil potassium, phosphorus, pH, and bulk density have a big impact on the functional traits of Eucommia ulmoides, especially in soil pH value and bulk density. This reects the commonality that soil habitat pH value and bulk density affect plant growth in karst areas. and show good phosphorus, rapidly available phosphorus, soil bulk density, soil moisture, soil Mn, total nitrogen, alkali-hydrolyzable nitrogen, pH value, and soil Zn, followed by the correlation with soil Fe, soil Ca, and organic matter. There is poor correlation with total potassium and rapidly available potassium. Where, signicant correlation is exhibited between the content of leaf chlorogenic acid and organic matter, total phosphorus, rapidly available phosphorus, total nitrogen, alkali-hydrolyzable nitrogen, soil moisture, soil Mn content, soil Zn content, soil bulk density, soil total potassium; between trunk bark aucubin content and soil bulk density, total phosphorus, rapidly available phosphorus, total nitrogen, alkali-hydrolyzable nitrogen, pH value, soil moisture, soil Mn content, soil Zn content; between root bark geniposide content and soil bulk density, total phosphorus, rapidly available phosphorus, soil Mn content, total nitrogen, alkali-hydrolyzable nitrogen, soil moisture , soil Zn content; between root bark aucubin content and soil bulk density, total phosphorus, rapidly available phosphorus, total nitrogen, pH value, soil moisture, soil Mn content; between trunk bark genipinic acid content and soil Fe, total nitrogen, alkali-hydrolyzable nitrogen, soil Mn content, soil Zn content; between leaf aucubin content and soil bulk density, total phosphorus, rapidly available phosphorus, pH value, etc. The above suggests that soil physical and chemical indexes in Eucommia ulmoides habitat and the content of metal elements exert varying degrees of inuence on the content of medicinal components in root bark, trunk bark and leaves of Eucommia ulmoides. In particular, leaf chlorogenic acid, trunk bark aucubin, root bark geniposide, root bark aucubin, trunk bark genipinic acid are greatly affected by the above-mentioned soil physical and chemical indexes and content of metal elements. This study shows that genipinic acid content of Eucommia ulmoides in type B (intensity karst rocky desertication area) (18.89mg/g), type D (moderate karst rocky desertication area) (19.15mg/g) is higher compared to normal landforms of Hunan Zhuzhou (7.25mg/g) and Zhejiang Hangzhou (9.9mg/g) at the same latitude (Qing et al., 2018), which is also higher compared to normal landforms of Hebei Anguo (7.95mg/g), Henan Luoyang (11.33mg/g), Henan Kaifeng (10.00mg/g), Beijing (10.03mg/g), Jiangsu Xiangshui (10.97mg/g), Sichuan Guangyuan (9.58mg/g), Shanxi Yuncheng (9.88mg/g), etc. at high latitude. However, the potential and mild rocky desertication areas of type A (6.20mg/g) and type C (3.77mg/g) have lower content than the above areas. This study shows that genipinic acid content in trunk bark of Eucommia ulmoides is higher in type A (17.09mg/g), type C (22.09mg/g), type D (25.88mg/g) than in Hunan Jianghua (14.23mg/g), Xupu (14.86mg/g), Yuanling (15.32 mg/g) of the same latitude (Cai,2014), which is comparable to that of Hunan Baojing (20.46mg/g) and Shimen (18.73mg/g), but much higher compared to high-latitude Shaanxi Lueyang (7.15mg/g) and Ankang (3.54mg/g). The above analysis indicates that Guizhou's natural geographical characteristics exert an important inuence on genipinic acid content of Eucommia ulmoides, which is generally higher compared to the normal landform. In this study, trunk bark chlorogenic acid content of Eucommia ulmoides in type A (2.21mg/g) and type C (2.38mg/g) was higher compared to Hunan Jianghua (1.35mg/g), Xupu (1.48mg/g), Baojing (1.72mg/g), Yuanling (1.38mg/g), Shimen (1.59mg/g) of the same latitude (Cai,2014), which was also higher compared to Shaanxi Lueyang (1.20mg/g) and Ankang (1.15mg/g) of high latitude. Type B and D had comparable content to the above-mentioned regions. The leaf chlorogenic acid contents of Eucommia ulmoides in the three habitat types B, C, and D were respectively 16.37mg/g, 11.21mg/g, 9.81mg/g, which was comparable to Hunan Jianghua (11.03mg/g), Xupu (10.21mg/g) , Baojing (12.44mg/g), Yuanling (11.80mg/g), Shimen (9.22mg/g), Cili (15.01mg/g) of the same latitude, while the content in type B, C, and D was comparable to that of high-latitude Lueyang(11.54mg/g), Ankang(12.85mg/g). It shows that Guizhou's natural geographical characteristics also have a major inuence on chlorogenic acid content, which is higher than or comparable to that of normal landforms in overall. In this study, aucubin content in trunk bark of Eucommia ulmoides in type A (14.66mg/g), type D (12.59mg/g) is comparable to that of Hunan Jianghua (14.84mg/g), Xupu (15.71mg/g), Baojing (10.91mg/g), Yuanling (14.32mg/g), Shimen (16.73mg/g) of the same latitude [21] , but much higher than that of high-latitude Shaanxi Lueyang (3.75mg/g), Ankang (2.11 mg/g). divided into namely, Type slope latitude mixing, short sunshine, Type B with medium altitude, low latitude and longitude, high temperature rain, positive oblique steep slope, medium sunshine, Type C with medium altitude, low latitude and longitude, moderate high temperature rain, gentle deection steep slope in shady side and sunny side, medium sunshine, Type D with high altitude, low longitude and latitude, low temperature and moderate rain, positive gentle slope, long sunshine. Habitat quality differs between the 4 types of Eucommia ulmoides plantations, whose indexes display good correlation with the contents of genipinic acid, chlorogenic acid, geniposide and aucubin in root bark, trunk bark, and leaves. Affected by the topography and landforms, the average forest density and different climate, the four types of Eucommia ulmoides plantation communities have slightly different shrub-grass species diversity indexes. A good correlation is shown between the content of genipinic acid, chlorogenic acid, geniposide, aucubin in root bark, trunk bark and leaves of Eucommia ulmoides and species diversity index of the shrub-grass layer of the community. Among the 4 types of Eucommia ulmoides plantations, most of the functional traits of Eucommia ulmoides exhibit signicant or extremely signicant differences, showing signicant or extremely signicant correlation with the content of genipinic acid, chlorogenic acid, geniposide and aucubin in root bark, trunk bark and leaves. Plants adapt to the environment through a combination of a series of functional traits, which reects the plant’s adaptation to the environment and resource utilization strategies. It is possible to adjust the functional traits of Eucommia ulmoides to let it better adapt to the habitat of the planting area, thereby further increasing the content of medicinal components in root bark, trunk bark and leaves of Eucommia ulmoides. For the 4 types of Eucommia ulmoides plantations, root bark has signicantly higher contents of genipinic acid and aucubin than trunk bark and leaves. This study provides a new idea for seeking the medicinal resource site of Eucommia ulmoides, which carries important theoretical reference signicance for cultivation of Eucommia ulmoides in Guizhou, cultivation of target medicinal components and comprehensive development and utilization of resources


Introduction
Eucommia ulmoides is a mono-species genus deciduous tree unique to China, which is a rare and precious medicinal material unique to China, also a new type of pharmaceutical and food resource with antihypertensive effect (Bai, (Fu et al., 2019). At present, the planting area of Eucommia ulmoides in the country is about 5 million mu, accounting for more than 99% of the total in the world. The planting areas are mainly concentrated in Guizhou, Hunan, Sichuan, Hubei, Shaanxi, Henan, Chongqing, Gansu, Anhui, Jiangxi, etc., with an annual output value of 1 trillion yuan (Zhu, 2019;Du, 2020). Eucommia ulmoides plantations of Guizhou are mainly distributed in Zunyi City, Liupanshui City, Anshun City, Bijie City, Qiannan Prefecture and Qianxinan Prefecture. As of the end of 2013, the total area of arti cial and wild Eucommia ulmoides in Guizhou Province reached 36,500 hm , accounting for 10.3% of the total area of Eucommia ulmoides in the country. The total output and total output value reached 50,000 tons and 27 million yuan respectively . Moreover, Eucommia ulmoides is listed as one precious medicinal material out of ten genuine medicines in Guizhou (He et al., 2005;Wu, 2015). Therefore, systematic research on the genuine medicinal material Eucommia ulmoides produced in Guizhou carries important theoretical and practical signi cance for the economic and social development of Guizhou and the cultivation of eucommia ulmoides in Guizhou area.
At present, research of Eucommia ulmoides mostly focuses on its pharmacology and effects (Wang et  Du et al., 2016) and so on. Its study areas are mainly concentrated in Henan, Hunan, Shaanxi, Guizhou and Sichuan. However, there are few reports on Eucommia ulmoides habitat, functional traits, medicinal component characteristics and relationships in between. In particular, Eucommia ulmoides, as an authentic medicinal material in Guizhou, has not been reported under the special karst ecological environment of Guizhou. In view of this, the genuine medicinal material Eucommia ulmoides produced in Guizhou is taken as the research object to study its habitat, functional traits and medicinal composition characteristics and explore the relationship in between in an attempt to explore the local characteristics of Eucommia ulmoides from Guizhou, and provide a theoretical basis for the cultivation of Eucommia ulmoides in Guizhou, cultivation of target medicinal components and comprehensive development and utilization of resources.
Overview Of The Study Area And Methods

Overview of the study area
Guizhou Province is located in the eastern part of the Yunnan-Guizhou Plateau, 103°36′ ~ 109°35′ E, 24°37′ ~ 29°13′ N. The territory terrain is high in the west and low in the east, which inclines from the middle to the north, east and south, with an average elevation of 1,100 m. Dominated by plateau mountains, hills and basins, it has 92.5% of mountains and hills. There are many mountains in the territory, with one mountain stretching after another and presenting high mountains and deep valleys. There is Dalou Mountain in the north, slanting across the northern border from west to northeast; Miaoling Mountain spans the middle and south, with Wuling Mountain in the northeast, and towering Wumeng Mountain in the west. The highest altitude is 2 900.6 m, and the lowest altitude is 147.8 m. Karst landforms have quite typical development, with exposed area of carbonate up to 109,084 km 2 , accounting for 61.9% of the total land area of the province (Wang,2003). On the typical karst landform development areas, the karst area has environmental degradation (Cai,1996), showing di cult soil formation, lack of lime-soil transition layer formed by carbonate rocks (Yang,1990), thin soil, discontinuous soil cover, and low ecological capacity (Li et al., 2016;). The climate is warm and humid, belonging to subtropical humid monsoon climate zone. The annual frost-free period is 250 ~ 300 d. Warm in winter and cool in summer, it has average temperature of 3℃ 6℃ in the coldest month of January, which is higher compared to other areas at the same latitude. With average temperature of 22℃ ~ 25℃ in the hottest month of July, it is a typical area cool in summer. The annual precipitation is 1 000 ~ 1 400 mm, with obvious rainy season. At the same time, under the in uence of atmospheric circulation and karst topography, Guizhou has formed a diversi ed microclimate characterized by "one mountain with four seasons, ten miles with different weathers", thus forming a characteristic and diverse habitat unique to Guizhou that produces genuine Chinese herbal medicines in special habitat. Where, woody genuine medicinal materials include fructus evodiae, gallnut and so on. Based on the data from the fourth forest resource inventory of Guizhou Province, the distribution area of Eucommia ulmoides plantation was determined, and locations with a continuous distribution area of more than 3 mu of Eucommia ulmoides were selected for investigation. The survey factors include altitude, longitude, latitude, annual average rainfall, annual average temperature, annual extreme maximum temperature, annual extreme minimum temperature, average temperature in July, average temperature in January, annual average sunshine hours, planting years, provenance, forest type, soil lithology, soil type, landform type, slope, aspect, slope position, etc.
De-trend correspondence analysis (DCA) and redundancy analysis (RDA) were performed using Canoco for windows 4.5 software. 12 indexes of Eucommia ulmoides plantations at 90 distribution points in Guizhou Province, including altitude, longitude, latitude, annual average rainfall, annual average temperature, extreme maximum temperature, average temperature in July, average temperature in January, average annual sunshine hours, lithology, slope, and aspect were subjected to de-trend correspondence analysis (DCA) to obtain the gradient length of each ordination axis. The gradient lengths (LGA) of the four axes are 0.414, 0.198, 0.087, 0.181 in this study, which are all smaller than 3. Therefore, RDA analysis (Zhang,2004;Jan and Petr,2003) was adopted. Through RDA analysis, Eucommia ulmoides plantation in Guizhou Province can be divided into 4 types (Tab. 1, Fig. 1). That is, type A has low altitude, low, medium and high temperature rain slope latitude mixing, and short sunshine. Type B has medium altitude, low latitude and longitude, high temperature rain, positive oblique steep slope, medium sunshine. Type C has medium altitude, low latitude and longitude, moderate high temperature rain, gentle de ection steep slope in shady side and sunny side, medium sunshine. Type D has high altitude, low longitude and latitude, low temperature and moderate rain, positive gentle slope, long sunshine. For the convenience of addressing, we use Type A, Type B, Type C, and Type D as substitute names (the same below). In this study, typical plots were selected from the above 4 types of Eucommia ulmoides plantations, with 3 replicates in each type, that is, 3 plots. Thus, there are a total of 12 plots for the 4 types, each with an area of 20m×20m (Zhang and Xu,2019 Note: The seedlings of Eucommia ulmoides in the four sampling sites were all from the same batch of seedlings in Jiangjin District of Chongqing (formerly Jiangjin County of Sichuan Province) which were planted in 1996.

Sample collection and processing
According to the average DBH, tree height, crown width, under crown height and overall growth of Eucommia ulmoides in each plot, 3 plants with no disease and insect pests were selected from each plot, and a total of 36 samples in 12 plots were selected. The leaf samples of the tree samples were collected from the 3 positions of upper, middle, and bottom parts in the four directions of east, south, west, and north. A total of 12 leaf samples were collected from each sample tree, and 1 root bark sample and 1 trunk bark sample were collected from each sample tree. There are a total of 432 leaf samples, 36 root bark and 36 trunk bark samples in 12 plots. The samples were collected from May 29 to June 30, 2017. The trunk bark was sampled by girdling method at 130cm from the ground. A second girding cut was made at 30cm upwards with this as a starting point. Longitudinally cut a slit between the two girding cuts to peel off the trunk bark. The thick root bark was all dug out, removed with the soil attached to the thick roots to peel off the root bark. When sampling the soil, dig out all the roots of the sample tree and perform sampling in 3 layers of 0~10cm, 10~20cm and >20cm of the sample tree. Then, mix the 3 layers of soil samples, involving a total of 36 soil samples in 12 plots. Put each sample in a ziplock bag and record the sample name. The sample was taken back to the laboratory for drying after washing the soil attached to the root bark.
The leaf, trunk bark and root bark samples were respectively placed in an oven at 80°C to be dried to constant weight and crushed, and screened through an 80-mesh standard nylon sieve. The sieved samples were put into a ziplock bag, with name recorded. The soil was naturally air-dried, ground in a mortar, and stored in a ziplock bag after screened through 80 mesh and 20 mesh nylon sieves for later use. 10 ne roots that are clean and intact with a diameter of ≤ 2 mm were selected and dried at 70℃ after measurement of root length and root volume. After dry weight was calculated, the roots were crushed and screened through an 80-mesh sieve for later use, mainly to measure the content of root carbon, nitrogen, and phosphorus.
Weigh 1.0000g of dried root bark, trunk bark and sample leaf of Eucommia ulmoides respectively, put them in a 250mL Erlenmeyer ask, add 50mL 70% methanol, perform ultrasonic (30℃, 300W, 40Hz) extraction for 30min, transfer the solution to a 100mL centrifuge tube, centrifuge (4000r/min) for 60min, lter and dilute to 50mL, shake well, pass through a 0.45μm microporous membrane, take the ltrate, transfer it to a 2.5mL sample bottle to obtain the test solution. The chromatographic column was a ZORBAX Eclipse Plus C18 column (4.6×250mm, 5μm; Agilent); the column temperature was 30°C; the mobile phase was acetonitrile (A)-0.1% phosphoric acid aqueous solution (B) (97:3); the injection volume was 5μL; volume ow rate was 1.0mL/min; detection wavelength of genipinic acid, chlorogenic acid, and geniposide λ=235nm, detection wavelength of aucubin λ=208. Qualitative analysis was made according to the peak retention time of the reference substance, and content of different medicinal components was calculated according to the peak area.

Soil physical and chemical properties and determination of soil and ne root, root bark, trunk bark and leaf elements of Eucommia ulmoides
The soil moisture content was determined by gravimetric method, the soil bulk density was determined by cutting ring method, the soil organic matter was determined by the potassium dichromate external heating method, the soil total nitrogen was determined by the semi-micro Kelvin method, the alkalihydrolyzable nitrogen was determined by alkaline hydrolysis diffusion method. Soil total phosphorus was determined by HClO 4 -H 2 SO 4 digestion-Mo-Sb colorimetric method, rapidly available phosphorus was determined by 0.5mol/L NaHC0 3 extraction-colorimetric method, soil total potassium was determined by NaOH melting-ame photometry, and rapidly available potassium was determined by NH 4 OAc extraction-ame photometry, soil pH value was determined by pH meter. The content of metal elements in soil, root bark, trunk bark and leaves was determined by nitric acid-perchloric acid digestion-ame photometry.
The total carbon content of ne roots and leaf total carbon content were determined by potassium dichromate external heating method. The total nitrogen content of ne roots and leaf total nitrogen content were determined by H 2 SO 4 -H 2 O 2 digestion-Nessler's colorimetric method. The total phosphorus content of ne roots and leaf total phosphorus content were determined by H 2 SO 4 -H 2 O 2 digestion-vanadium molybdate yellow colorimetric method. The total potassium content of ne roots and leaf total potassium content were determined by H 2 SO 4 -H 2 O 2 digestion-ame atomic absorption spectrometry (Bao,1999). The chlorophyll uorescence parameter Fv/Fm was determined by a chlorophyll uorometer (MINI-PAM-, Walz, Germany).

Data Statistics and Analysis
SPSS 22.0 and Excel software were used for data statistical analysis, and Canoco for windows4.5 software was used for redundancy analysis (RDA). The analysis of variance was made using LSD multiple comparison (α=0.05), and Pearson was used for correlation analysis (α=0.05).
Results And Analysis Fig. 1 shows: The 90 distribution points can be divided into 4 types. Based on 12 indexes of clustering (i.e. altitude, longitude, latitude, annual average rainfall, annual average temperature, extreme maximum temperature, average temperature of January, average temperature of July, annual average sunshine hours, lithology, slope, aspect), these four types can be named as follows: Type A has low altitude, low, medium and high temperature rain slope latitude mixing, and short sunshine. Type B has medium altitude, low latitude and longitude, high temperature rain, positive oblique steep slope, medium sunshine. Type C has medium altitude, low latitude and longitude, moderate high temperature rain, gentle de ection steep slope in shady side and sunny side, medium sunshine.

Types of Eucommia ulmoides plantations in Guizhou
Type D has high altitude, low longitude and latitude, low temperature and moderate rain, positive gentle slope, long sunshine. Where, the factors with a greater impact on the type of Eucommia ulmoides plantation include altitude, longitude, latitude, and annual average sunshine hours. The less in uential factors are average temperature of July, extreme maximum temperature, slope, aspect and lithology.For distribution area of the 4 types of Eucommia ulmoides forests ( Fig. 2), there are 40 distribution points for type A, 12 for type B, 31 for type C, and 7 for type D, mainly concentrated in northern Guizhou-central Guizhouwestern Guizhou and southwestern Guizhou. Type A is mainly distributed in northern and southern Guizhou, type C is mainly distributed in central and western Guizhou, while type B and D are mainly distributed in smaller areas in southwestern Guizhou. Fig. 3 shows: In general, organic matter, total P, rapidly available P, total N, alkali-hydrolyzable N, pH value, soil moisture, Mn, and Zn content are the highest in type B, and these habitat indexes are basically the lowest in type A; Ca content is the highest in Type A, higher in type B, lower in type C, and the lowest in type D. Fe content and rapidly available potassium in various types show opposite trend compared to Ca content. The total K and soil bulk density are the highest in type A, followed by type D and Type C, with the lowest content in type B. The above indicates that different types of Eucommia ulmoides plantations have different habitat quality, but in general, Type B is intensity karst rocky deserti cation habitat, and Type A is potential karst rocky deserti cation habitat. Type C, D is in the middle, and the difference is signi cant. the highest root carbon nitrogen ratio and chlorophyll uorescence parameters, followed by type D, and there is little difference between type B and type C. Root potassium content, leaf dry weight, leaf carbon content, bark DBH ratio are the highest in type D, while type A and C have the lowest bark DBH ratio, and type C has the lowest leaf dry weight and leaf carbon content. Root phosphorus content is the highest in type B, higher in type C, lower in type A and the lowest in type D. Type B and type C equally have the highest branch and leaf index, followed by type D, and type A ranks the lowest. Except the insigni cant differences in root carbon content and leaf nitrogen content in the four types of Eucommia ulmoides plantations, other indexes of plant functional traits have signi cant or extremely signi cant differences between the four types of A, B, C, and D. The above suggests that type A Eucommia ulmoides has the best growth, followed by type C, while type B and D grow poorly.

Species diversity of the shrub-grass layer of the community
2.5 Content characteristics of medicinal components in root bark, trunk bark and leaves Fig. 6 shows: the content of genipinic acid is the highest in root bark, followed by trunk bark, with the lowest content in leaves. Root bark and trunk bark have the highest content in type D, followed by type C and type A, with the lowest content in type B. The leaf content is higher in type B and D and lower in type A and C. The content of chlorogenic acid is the highest in leaf, followed by trunk bark, with the lowest content in root bark. Root bark and trunk bark have the highest content in type A and C, followed by type B, with the lowest content in type D. Leaf content is the highest in type B, followed by type C and type D, with the lowest content in type A. Geniposide has the highest content in trunk bark, lower content in root bark and leaf. Root bark content is higher in type A and D and lower in type A and C. Leaf content is higher in type B and C and lower in type A and D. Trunk bark content is the highest in type D, followed by type C, with lower content in type A and B. Aucubin has the highest content in root bark, followed by trunk bark, with the lowest content in leaves. Root bark, trunk bark, and leaves have higher content in type A and D and lower content in type B and C. The above suggests that the different parts and habitats of Eucommia ulmoides lead to different content of medicinal components.

Element enrichment coe cients of root bark, trunk bark and leaves in different types of Eucommia ulmoides
Tab. 2 shows: Type B has the highest enrichment coe cient of K element in root bark, trunk bark, and leaf, followed by type C, with the smallest coe cient in type A and D. Leaf has the highest enrichment coe cients among the four forest types in overall, followed by root bark, while trunk bark has the smallest coe cient. Ca element enrichment coe cient in root bark, trunk bark, and leaf is the highest in type D, followed by type C, with the smallest coe cient in types A and B. Type D has signi cantly higher coe cient compared to the other three types. Trunk bark has the highest enrichment coe cient among the 4 forest types on the whole. Type A has the highest enrichment coe cients of Fe and Mn elements in root bark, trunk bark and leaf, followed by other types, and the overall difference is insigni cant. In terms of parts of Eucommia ulmoides, enrichment coe cient of Fe element is the highest in root bark, with smaller coe cient in trunk bark and leaf. Enrichment coe cient of Mn element is the highest in leaves as a whole, and little difference is shown between root bark and trunk bark. Type A has the highest enrichment coe cient of Zn element in root bark, trunk bark and leaves, followed by type C and D, with the smallest coe cient in type B. Enrichment coe cient of Zn element is the highest in trunk bark, with insigni cant difference between root bark and leaf. Tab. 3 shows: as a whole, in terms of functional traits, petiole length, leaf phosphorus content, root nitrogen content, root carbon nitrogen ratio, branch and leaf index are in good correlation with habitat; followed by leaf dry weight, chlorophyll uorescence parameters, leaf carbon content, root phosphorus content, crown length ratio, while speci c leaf area, leaf nitrogen content, root carbon content, and bark DBH ratio have poor correlation, indicating differences in the impact of habitat on the functional traits of Eucommia ulmoides. Where, petiole length, leaf phosphorus content, root nitrogen content, root carbon nitrogen ratio, branch and leaf index exert the biggest in uence. Seen from habitat, rapidly available phosphorus, total potassium, pH value, soil bulk density and functional traits exhibit good correlation, with signi cant correlations shown between petiole length and total potassium, organic matter, total phosphorus, total nitrogen, alkali-hydrolyzable nitrogen, rapidly available phosphorus, and soil moisture; between leaf phosphorus content and total potassium, organic matter, total nitrogen, and alkali-hydrolyzable nitrogen; between root nitrogen content and rapidly available potassium, rapidly available phosphorus, soil bulk density; between root carbon nitrogen ratio and soil bulk density, total potassium, total phosphorus, rapidly available phosphorus, and pH; between branch and leaf index and soil moisture, total phosphorus, rapidly available phosphorus, soil bulk density. The above suggests that soil potassium, phosphorus, pH, and bulk density have a big impact on the functional traits of Eucommia ulmoides, especially in soil pH value and bulk density. This re ects the commonality that soil habitat pH value and bulk density affect plant growth in karst areas.

Tab.3 Correlation between functional traits and habitat of Eucommia ulmoides
Organic Tab. 5 shows: on the whole, the contents of medicinal components in Eucommia root bark, trunk bark and leaves show good correlation with soil total phosphorus, rapidly available phosphorus, soil bulk density, soil moisture, soil Mn, total nitrogen, alkali-hydrolyzable nitrogen, pH value, and soil Zn, followed by the correlation with soil Fe, soil Ca, and organic matter. There is poor correlation with total potassium and rapidly available potassium. Where, signi cant correlation is exhibited between the content of leaf chlorogenic acid and organic matter, total phosphorus, rapidly available phosphorus, total nitrogen, alkalihydrolyzable nitrogen, soil moisture, soil Mn content, soil Zn content, soil bulk density, soil total potassium; between trunk bark aucubin content and soil bulk density, total phosphorus, rapidly available phosphorus, total nitrogen, alkali-hydrolyzable nitrogen, pH value, soil moisture, soil Mn content, soil Zn content; between root bark geniposide content and soil bulk density, total phosphorus, rapidly available phosphorus, soil Mn content, total nitrogen, alkali-hydrolyzable nitrogen, soil moisture , soil Zn content; between root bark aucubin content and soil bulk density, total phosphorus, rapidly available phosphorus, total nitrogen, pH value, soil moisture, soil Mn content; between trunk bark genipinic acid content and soil Fe, total nitrogen, alkali-hydrolyzable nitrogen, soil Mn content, soil Zn content; between leaf aucubin content and soil bulk density, total phosphorus, rapidly available phosphorus, pH value, etc. The above suggests that soil physical and chemical indexes in Eucommia ulmoides habitat and the content of metal elements exert varying degrees of in uence on the content of medicinal components in root bark, trunk bark and leaves of Eucommia ulmoides. In particular, leaf chlorogenic acid, trunk bark aucubin, root bark geniposide, root bark aucubin, trunk bark genipinic acid are greatly affected by the above-mentioned soil physical and chemical indexes and content of metal elements.

Correlation between content of medicinal components and functional traits
Tab. 6 shows: in overall, the content of medicinal components in root bark, trunk bark, and leaves of Eucommia ulmoides has correlation with functional traits.
Where, aucubin content in root bark, trunk bark, and leaves has a good correlation with functional traits, especially with leaf dry weight, chlorophyll uorescence parameters, leaf carbon content, root nitrogen content, root carbon nitrogen ratio, crown length ratio, branch and leaf index, followed by the correlation between root bark genipinic acid, root bark geniposide, leaf chlorogenic acid, leaf geniposide and functional traits. Signi cant relationship is exhibited between root bark, trunk bark and leaf aucubin content and leaf dry weight, chlorophyll uorescence parameters, leaf carbon content, root carbon nitrogen ratio, crown length ratio, branch and leaf index; between root bark geniposide content and chlorophyll uorescence parameters, leaf carbon content, root carbon nitrogen ratio, branch and leaf index, crown length ratio; between leaf chlorogenic acid content and root nitrogen content, branch and leaf index, petiole length, chlorophyll uorescence parameters, root carbon nitrogen ratio; between root bark genipinic acid and leaf dry weight, petiole length, leaf carbon content, etc. The above correlation re ects that functional traits of Eucommia ulmoides can affect the content of medicinal components in root bark, trunk bark and leaves, especially the content of aucubin in root bark, trunk bark and leaves.
Tab.6 Correlation between the contents of medicinal components in root bark, trunk bark and leaves and plant functional traits
This study shows that genipinic acid content in trunk bark of Eucommia ulmoides is higher in type A (17.09mg/g), type C (22.09mg/g), type D (25.88mg/g) than in Hunan Jianghua (14.23mg/g), Xupu (14.86mg/g), Yuanling (15.32 mg/g) of the same latitude (Cai,2014), which is comparable to that of Hunan Baojing (20.46mg/g) and Shimen (18.73mg/g), but much higher compared to high-latitude Shaanxi Lueyang (7.15mg/g) and Ankang (3.54mg/g). The above analysis indicates that Guizhou's natural geographical characteristics exert an important in uence on genipinic acid content of Eucommia ulmoides, which is generally higher compared to the normal landform. In this study, trunk bark chlorogenic acid content of Eucommia ulmoides in type A (2.21mg/g) and type C (2.38mg/g) was higher compared to Hunan Jianghua (1.35mg/g), Xupu (1.48mg/g), Baojing (1.72mg/g), Yuanling (1.38mg/g), Shimen (1.59mg/g) of the same latitude (Cai,2014), which was also higher compared to Shaanxi Lueyang (1.20mg/g) and Ankang (1.15mg/g) of high latitude. Type B and D had comparable content to the above-mentioned regions. The leaf chlorogenic acid contents of Eucommia ulmoides in the three habitat types B, C, and D were respectively 16.37mg/g, 11.21mg/g, 9.81mg/g, which was comparable to Hunan Jianghua (11.03mg/g), Xupu (10.21mg/g) , Baojing (12.44mg/g), Yuanling (11.80mg/g), Shimen (9.22mg/g), Cili (15.01mg/g) of the same latitude, while the content in type B, C, and D was comparable to that of high-latitude Lueyang(11.54mg/g), Ankang(12.85mg/g). It shows that Guizhou's natural geographical characteristics also have a major in uence on chlorogenic acid content, which is higher than or comparable to that of normal landforms in overall. In this study, aucubin content in trunk bark of Eucommia ulmoides in type A (14.66mg/g), type D (12.59mg/g) is comparable to that of Hunan Jianghua (14.84mg/g), Xupu (15.71mg/g), Baojing (10.91mg/g), Yuanling (14.32mg/g), content in type A (12.36mg/g), type D (8.88mg/g) was higher compared to Hunan Jianghua (2.64mg/g), Xupu (2.68mg/g), Baojing (3.3 mg/g), Yuanling (2.59mg/g), Shimen (1.76mg/g), and Cili (4.65mg/g) of the same latitude, which was far higher compared to high-latitude Shaanxi Lueyang (1.51mg/g), Ankang (1.23mg/g), while that of type B (3.15mg/g) and type C (1.18mg/g) was comparable to the above regions. The above re ects that aucubin content in trunk bark and leaves of Eucommia ulmoides produced in Guizhou is generally higher compared to other regions. Studies have shown that in Zunyi and Bijie of Guizhou, genipinic acid content in trunk bark of Eucommia ulmoides is 24.29mg/g and 24.92mg/g respectively, aucubin content is 20.84mg/g and 28.42mg/g, respectively; leaf genipinic acid content is 6.91mg/g, 5.63mg/g, respectively, aucubin content is 3.84mg/g, 2.48mg/g, respectively (Cai,2014). In Zunyi, the leaf genipinic acid content of Eucommia ulmoides is 8.97mg/g, aucubin content is 29.21mg/g (Qing et al., 2018). The results of this study indicate that genipinic acid, aucubin contents in trunk bark and leaf of the four types of Eucommia ulmoides exhibit basically consistent law with the research results of the two researchers in different periods. This implies that content of medicinal components of Eucommia ulmoides produced in Guizhou has relatively stable spatial and temporal distribution.
The above analysis reveals that, as a whole, the content of medicinal components in Eucommia ulmoides produced from Guizhou is generally higher than that of other regions (including regions of the same latitude and high latitude), with only some regions slightly lower compared to other regions, and the temporal and spatial distribution is relatively stable. Meanwhile, a close relationship is exhibited with natural regional characteristics of Guizhou. Studies have shown that effective ingredient content and pharmacological activity of Eucommia ulmoides leaves differ between different origins, which may be related to the content of metal elements (Peng et al., 2011). Soil alkali-hydrolyzable nitrogen exerts a great effect on the accumulation of aucubin, genipinic acid and chlorogenic acid in the bark of Eucommia ulmoides, followed by rapidly available potassium and organic matter (Cai,2014). This study also suggests that different elements in different parts of Eucommia ulmoides have signi cantly different enrichment coe cients between different habitats. These studies indicate that natural regional characteristics will impact content of medicinal components in Eucommia ulmoides. At the same time, based on the correlation between the content of medicinal components in Eucommia ulmoides from Guizhou and the habitat, this study shows that soil physical and chemical indexes and metal element contents in habitats of Eucommia ulmoides exert different effects on the contents of medicinal components in root bark, trunk bark and leaves of Eucommia ulmoides. In particular, leaf chlorogenic acid, trunk bark aucubin, root bark geniposide, root bark aucubin, and trunk bark genipinic acid are particularly affected by soil physical and chemical indexes and the content of metal elements. Moreover, this study indicates that the functional traits of Eucommia ulmoides can affect the content of medicinal components in root bark, trunk bark and leaves, with particular impact on content of aucubin in root bark, trunk bark and leaves. The above suggests that Guizhou's natural regional characteristics signi cantly affect genuineness of Eucommia ulmoides.
Therefore, high and stable content of medicinal ingredients of Eucommia ulmoides from Guizhou is its primary characteristic of genuineness, and such genuineness is inseparable from the natural regional characteristics of Guizhou karst plateau mountain.

Implications for afforestation
This study shows that genipinic acid content is the highest in the root bark, which is much higher than that in trunk bark and leaves. The genipinic acid content in the root bark of the four habitat types is ranked in descending order as 73.62 mg/g in type D, 46.15mg/g in Type C, 39.85mg/g in type A, 32.79mg/g in type B, while type D has a higher content than the other 3 types (Fig. 6), showing better root functional traits (Fig. 5). Genipinic acid content in root bark is negatively correlated with soil pH value and soil Ca (Tab. 5). Type D habitat is a mild rocky deserti cation area, where genipinic acid content in the root bark is positively correlated with root functional traits (Tab. 6). There is no report on the content of medicinal components in root bark. Therefore, if genipinic acid is deemed as the target medicinal component, root bark is the best medicinal resource part of Eucommia ulmoides, and it is best to cultivate the medicinal root bark in type D habitat, followed by type C habitat and type A habitat, but it is best not to cultivate root bark in type B habitat for the purpose of obtaining genipinic acid. This study shows that the content of chlorogenic acid is the highest in the leaves. The leaf chlorogenic acid content of the four habitat types of Eucommia ulmoides is ranked in descending order as 16.37mg/g in type B, 11.21mg/g in type C, 9.81mg/g in type D, and 4.37 mg/g in type A, while type B has a higher content than the other 3 types (Fig. 6), which is consistent with the study results (Xu, 2015) that the content of chlorogenic acid in Eucommia ulmoides is higher in leaf than in bark. Type B has worse leaf functional traits than type A, type D, which is because type B habitat is an intensity rocky deserti cation area (Fig. 5). When faced with stress, plant secondary metabolites have greater defense bene ts, with more secondary metabolites produced [41] (BARTO et al., 2005). Chlorogenic acid content is extremely signi cantly positively correlated with soil organic matter, total phosphorus, rapidly available phosphorus, total nitrogen, alkali-hydrolyzable nitrogen, soil moisture, soil Mn, soil Zn. Soil physical and chemical indexes greatly impact the accumulation of leaf chlorogenic acid content in Eucommia ulmoides, and type B habitat has better quality in overall (Fig. 3). Leaf functional traits are mostly negatively correlated with leaf chlorogenic acid content (Tab. 6). Therefore, if chlorogenic acid is taken as the target medicinal component, it is best to cultivate Eucommia ulmoides leaves on type B habitat, followed by cultivation of leaves in type C and type D habitats. This study reveals that geniposide content is the highest in trunk bark. The geniposide content in trunk bark of the four types of Eucommia ulmoides is ranked in descending order as 7.81mg/g in type D, 3.43mg/g in type C, 1.32mg/g in type A, 0.88mg/g in type B (Fig.6). The geniposide content in trunk bark is mostly negatively correlated with soil physical and chemical indexes (Tab. 5), showing a good correlation with functional traits (Tab. 6). Hence, if geniposide is taken as the target medicinal component, it is best to cultivate trunk bark of Eucommia ulmoides in type D habitat, followed by type C, and then type A and type B habitats. This study suggests that aucubin content is the highest in the root bark, and aucubin content in the root bark of the four types of Eucommia ulmoides is ranked in descending order as 47.99mg/g in type D, 45.19mg/g in type A, 22.66mg/g in type C, 19.57mg/g in type B (Fig. 6). The aucubin content in root bark is positively correlated with total potassium and soil bulk density, showing a good correlation with functional traits. Therefore, if aucubin is taken as the target medicinal component, it is best to cultivate root bark of Eucommia ulmoides in type D habitat, followed by type A habitat, and then type C and type B habitats. The above analysis indicates important implications of this study for afforestation. That is, in the process of Eucommia ulmoides afforestation, comprehensive consideration should be given to habitat, functional traits, medicinal component content of Eucommia ulmoides and their correlation. To acquire different target medicinal components, it is possible to implement classi ed afforestation and classi ed management, "plant according to the environment and conduct eld survey of medicines", thus making the best use of the land and medicine to achieve the maximum economic value, social value and ecological value.
Eucommia ulmoides plantations in Guizhou Province can be divided into 4 types, namely, Type A with low altitude, low, medium and high temperature rain slope latitude mixing, short sunshine, Type B with medium altitude, low latitude and longitude, high temperature rain, positive oblique steep slope, medium sunshine, Type C with medium altitude, low latitude and longitude, moderate high temperature rain, gentle de ection steep slope in shady side and sunny side, medium sunshine, Type D with high altitude, low longitude and latitude, low temperature and moderate rain, positive gentle slope, long sunshine. Habitat quality differs between the 4 types of Eucommia ulmoides plantations, whose indexes display good correlation with the contents of genipinic acid, chlorogenic acid, geniposide and aucubin in root bark, trunk bark, and leaves. Affected by the topography and landforms, the average forest density and different climate, the four types of Eucommia ulmoides plantation communities have slightly different shrub-grass species diversity indexes. A good correlation is shown between the content of genipinic acid, chlorogenic acid, geniposide, aucubin in root bark, trunk bark and leaves of Eucommia ulmoides and species diversity index of the shrub-grass layer of the community. Among the 4 types of Eucommia ulmoides plantations, most of the functional traits of  Community diversity index of shrub layer in four types of Eucommia ulmoides plantations