Experimental area
The experimental area was in Dingbian County, Shaanxi Province, China, in the transitional zone between the Mu Us Sands and the Loess Plateau (107°15'~108°22'E, 36°49'~37°53'N, altitude of 1303 ~ 1907m). It is a mid-temperate semi-arid continental monsoon climate with low precipitation (annual average 316 mm), high evaporation (annual average 2490 mm), strong and frequent wind and sandstorms, an annual average temperature of 7.9 ℃, an annual maximum temperature of 37.7 ℃, and an annual minimum temperature of − 29.4 ℃. The soils are mainly aeolian sandy soil and yellow sandy soil, with a poor nutrient content, and poor water and fertilizer retention capacity. The zonal vegetation is arid steppe and desert steppe, with xerophytes, psammophytes, halophytes, and mesophytic meadow plants, among which herbaceous plants mainly include Agriophyllum squarrosum, Psammochloa villosa, Agropyron cristatum, etc.
The experiment commenced in 2016 in a plantation on a loess hill named Weiliang Hill at Baiwanzi Town, which was established in 2001 with a planting spacing 1.0 m × 1.5 m. The soil was loessial soil covered with a layer of sand. The stand contained the initial planted maternal plants of Chinese sea buckthorn and their root suckers, with a canopy density of 0.8. Before the coppicing, the trees were uniform in growth with some premature aging characteristics, including death of some branches, leaves, and even trunks, and reduction in growth rate and reproduction capacity12.
Materials
Chinese sea buckthorn is an endemic species (subspecies) of China. It is a small tree or shrub of the genus Hippophae L. in the family Elaeagnaceae Juss. It is widely distributed in the arid and semi-arid regions of northern China, occupying 85% of the total area of Hippophae L. in China34. Its root system can be divided into vertical and horizontal roots. Vertical roots grow vertically or inclined downward, while horizontal roots extend horizontally with the topography at a certain depth of soil from the surface7,8. After the mother plant (the original parent for clonal propagation) is settled (or planted), the population spreads by the long-distance extension of horizontal roots, multiple branching, and root suckers. Each mother plant and its ramets (root suckers) are connected by horizontal roots to form a clone. Consequently, the stand consisted of multiple clones (Fig. 1). Clones, as an ecological and physiological unit, can respond integrally to environmental resource levels or heterogeneity, and their plasticity is mainly expressed at the levels of clonal components and ramet components35. Among them, the clonal components can be divided into the mother plant, daughter ramet, and root system (including horizontal and vertical roots); the ramet (including mother plant and daughter ramet) components can be divided into the leaves, branches, and trunk. The vertical roots and horizontal roots are divided into primary, secondary, tertiary, and quaternary levels; the primary roots are from the mother ramet, the secondary roots are from the primary roots, and so on.
Experiment layout
The stubble height was set at 3 gradients of 0, 10.0, and 20.0 cm, according to the regression experimental design requirements and pre-experiment results, with no coppicing as the control, denoted by 1, 2, 3, and CK, respectively. The field layout was a combination of a randomized block and Latin square design, with 4 replications; however, at least 8 replications were statistically available In this case, each treatment (plot) consisted of 10 clones (each consisting of a planted mother ramet and its root system and root tillers). The mother ramet of each clone was coppiced, and its stubble was marked and protected by applying paint, keeping the root-tillers (daughter ramets) connected to the mother ramet. The experiment commenced in April 2016 and was followed up on the 25th of each month until October, and so on for three consecutive years. The final investigation was conducted in 2018.
Survey methods
Each plot was divided into 10 quadrats of 1.0 m×1.5 m (initial plant spacing) centered on the mother plant, regarded as one clone. The parameters regarding root sprouting ability (quantity and growth) and clonal dispersal ability (root extension and branching) were measured. The root sprouting ability was represented by the number, height, and ground diameter of root suckers. The clonal dispersal ability was represented by the diameter, length, and number of first-order horizontal roots and total number of horizontal roots at all levels. The horizontal root extension ability was represented by the diameter and length of the first-order horizontal root. The branching intensity was represented by the number of the first-order horizontal root and total number of horizontal roots. After the determination of the parameters of the root sprouting ability, the quadrat with the closest sprouting ability to the average value of each plot was selected as the average standard clone. Then, all the underground components of the average standard clone were dug out for the measurement of clonal dispersal capacity.
Measurement of phenotypic traits
Seven phenotypic traits that are genetically relatively stable and easy to obtain and measure were selected to characterize the clone, namely the number of root sprouts (NRS), height of root sprouts (HRS), ground diameter of root sprouts (GDRS), length of the first-order horizontal root (1ST-LHR), diameter of the first-order horizontal root (1ST-DHR), number of first-order horizontal roots (1ST-NHR) and total number of horizontal roots (TNHR). The number of root sprouts were counted one by one. The height of the root sprout and length of the first-order horizontal root were measured with a steel tapeline (accuracy of 0.01 m). The ground diameter of root sprouts and the diameter of the first-order horizontal root were measured with a vernier caliper (accuracy of 0.01 mm). The measurements were repeated 3 times and the mean was recorded. Since the first-order horizontal roots usually extend beyond the quadrat, the length of the horizontal roots within the quadrat was taken as the reference index of extension ability and the diameter as the main index.
Determination of nutrient elements
The above-ground components of the average standard clones were divided into leaves, branches, and trunks, and the below-ground components were divided into horizontal and vertical roots. After the fresh weight was determined, the biomass was sampled and brought to the laboratory for the determination of dry biomass and nutrient content. The dry biomass was determined by oven drying at 105 ℃ to constant weigh. The dried sample was ground and sieved for nutrient content determination. N content was determined using the Kjeldahl method after H2SO4-H2O2 digestion, P content using the molybdenum–antimony anti-colorimetric method, K content using the flame photometer method, and Ca and Mg contents using the atomic absorption photometer method36–38. Finally, the nutrient reserves were calculated according to the nutrient content and the dry biomass of corresponding components (leaves, branches, trunks, vertical roots, horizontal roots), and the distribution proportion of different nutrient elements in each component (as a percentage of the total amount) was calculated. Nutrient reserve was calculated as:
M = T×B (1)
where M is nutrient reserve, T is nutrient content in dry biomass (g·kg− 1), and B is Dry weight biomass(kg).
Data analysis
Duncan multiple comparison was used to state statistical difference of the reported data between treatments. The causality of root sprouting and clonal dispersal relating to nutrient content and reserve was assessed by correlation analysis. The contributions of different components, nutrient elements, and stubble height to the integrated nutrient accumulation in clone were obtained by principal component analysis (PCA), and their relative importance was ranked according to the corresponding principal component score39.
We established m principal component equations for each nutrient element, and the equation expression was as follows:
F i =\(\sum _{j=1}^{m}{e}_{ij}{a}_{i}\) (i = 1, 2, 3, …, m;j = 1, 2, 3, …, m) (2)
where Fi is the score of the i-th principal component equation; eij is the eigenvector; ai is the factor loading of the i-th index of the nutrient element. Based on the data and Eq. (2), m principal component scores were calculated for each nutrient element.
The scores of the m principal components were weighted and summed to obtain the score of the principal component, which was the integrated index F of nutrient element accumulation. The formula is as follows:
F = α1F1 + α2F2 + α3F3+…+αmFm (3)
where F is the integrated index of nutrient accumulation; α is the weighting factor, which is the percentage of the contribution rate of each principal component to the cumulative contribution rate of m principal components; F1–Fm is the score of m principal components for each nutrient element.
The Duncan multiple comparison and the principal component analysis were performed using SPSS17.0. The regression analysis, equation construction and graphing were carried out using Excel 2016. The correlation analysis and graphing were conducted using Origin 2021.
Ethical approval
We confirm that all the experimental research and field studies on plants (either cultivated or wild), including the collection of plant material, complied with relevant institutional, national, and international guidelines and legislation. All of the material is owned by the authors and/or no permissions are required.