Two widely distributed plant species, couch grass Elymus repens L. and nettle Urtica dioica L., were collected in May 2021. The study area was located in the south of St. Petersburg city (59°53${}^{\prime }$ N, 30°38${}^{\prime }$ E). The plants were collected on 3 May, 10 May, and 25 May from a small (~1 m${}^2$) site. The climatic conditions during sampling are shown in Table 1. On each date of sampling, the plant samples were taken three times: at 9:00, 14:00, and 19:00. A completely randomized design was used. To ensure the reliability of the experimental data, at least three plants of each species were collected each time. The soil was taken from the plant roots. To take samples of the rhizosphere soil, the soil was carefully shaken off the roots. Then soil residues adhering to the roots were collected. Soil at the site was loamy sand (sand 75%, silt/clay 24%, silt 1%). Just after sampling, the plants were washed carefully by tap water to remove from the plant surface dust and soil particles. Roots were separated from leaves, the plant and soil samples were air-dried up to a constant weight. The soil samples were sieved through a 2 mm mesh to remove non-soil materials including plant fragments. The pH (1:2.5 H${}_2$O) of the soil samples was determined.

For digestion of plant and soil samples, each sample was weighed into the microwave digestion vessel (SK-15 high-pressure PTFE-TFM-Teflon vessels), then 9 mL of concentrated HNO${}_3$ were added, the vessels were closed, and the samples were heated in the Milestone Ethos up (Milestonem Srl, Sorisole, Italy) microwave oven. The programme was the following: heating for 10 min to 180 °C and holding at 180 °C for 9.30 min, then the vessels were cooled down to room temperature. The samples were diluted to 50 mL with ultrapure water. The ICP-OES determinations of Na, Mg, P, K, Ca, Mn, Fe, Zn, and Ba were carried out with an Agilent 5110 VDV ICP-OES, equipped with a SPS 4 autosampler, an U-series concentric glass nebulizer, a cyclonic glass double pass spray chamber and an 1.8 mm semi-demountable torch. Scandium (wavelength 335.372 nm) and yttrium (wavelength 377.433 nm) were used as internal standards to correct for sensitivity drift and matrix effects in the measurements. The ICP-MS determinations of Co, Ni, Cu, and Zn were carried out using an Agilent 8900 Triple Quad ICP-MS equipped with a SPS 4 autosampler, a MicroMist glass concentric nebulizer, a quartz Scott double-pass spray chamber and a 2.5 mm quartz torch. The parameters for ion lenses and torch position were tuned automatically before each ICP-MS measurement. Helium cell gas was applied to reduce interferences. For quality control purposes, blanks and microwave digested certified reference material Tomato Leaves NIST 1573a (National Institute of Standards and Technology, USA) were also measured as standard. The concentrations of elements determined in the NIST 1573a were in a good agreement with the values given in the certificate.

To obtain Raman spectra of light from the soil samples, an OPTEC-785TRS-2700 Raman spectrometer (spectral range –200 … + 2700 cm${}^{-1}$, covers the terahertz range, spectral resolution 3.5 cm${}^{-1}$) and an OPTEC-785SLRam Raman spectrometer (spectral range 240–3500 cm${}^{-1}$, spectral resolution 8 cm${}^{-1}$) were applied. The OPTEC-785SLRam spectrometer uses the architecture of the “cloud” version of the system. It includes the selection of measurement characteristics, processing of spectra (smoothing and spectral contour and compensation of the baseline through a polynomial processing procedure) and identification of target substances using the Hit Quality Index HQI and PeaksMatchingAndSubstract PMAS algorithms with spectral databases located on a cloud server. To obtain the Raman spectra, short exposure times and the minimum power of the laser source were applied. This compensated the luminescence diffusion bands. The compensation was carried out with polynomial processing of the spectral contour. The number of replicates when taking spectra was set at 100.

The statistical analysis of the experimental results was performed with help of Statistica for Windows 8.0 Software packages (StatSoft, Tulsa, OK, USA). The normality of the distribution of the analytical data was verified using the Shapiro-Wilk test. Outliers were excluded from further analysis. Mean concentrations of elements were calculated and analysis of variances was carried out to assess statistically significant (p 0.05) differences between groups of samples. Pearson correlation analysis, cluster analysis and principal component analysis (PCA) were applied in order to identify main factors affecting the uptake of different elements as well as interactions between the elements in plants and in the rhizosphere. To avoid misclassification caused by different orders of magnitude of variables, the data for cluster analysis and the PCA were normalized.

The concentrations of different elements in plants varied significantly over daytime on all dates of sampling. Fig. 1 illustrates daily dynamics of essential plant nutrients Mg, K, and Mn in roots and leaves of the couch grass and nettle collected on different dates. In most cases the differences between concentrations of the elements in the plants collected during the days at different times were statistically significant, and often the level of the p values was below 0.01. The statistically insignificant differences are listed in the Figure captions.

Supplementary Table 1 illustrates differences between concentrations of other elements in the plants collected on different dates during daytime. As the data of the Supplementary Table 1 suggests, the concentrations of many elements in roots and leaves of both plant species collected at different times of the days are statistically significantly (p 0.05) different.

Although couch grass and nettle were collected simultaneously from the same site, there were statistically significant (p 0.05) differences between concentrations of many elements in the plants. In roots of couch grass, the concentrations of Ca and Ba were lower, and the concentrations of Na, Co, Ni, and Zn were higher than concentrations of the elements in roots of nettle. On the other hand, the concentration of K was higher and the concentrations of Ca and Fe were lower in leaves of couch grass compared to concentrations of the elements in leaves of nettle.

Fig. 2 shows the changes in the concentrations of Mg and P in roots and leaves of couch grass and nettle taken at different dates from the same site. During May, the variations in the concentrations of the elements in the two plant species were different. In couch grass, the concentrations of Mg and P were the highest in the plants sampled on 10 May and were lower on two other dates. On the other hand, in nettle, the concentrations of the elements were higher in the plants collected on 3 and 25 May and lower in the plants harvested on 10 May.

The concentrations of Mg in roots and leaves of the couch grass and nettle collected on 10 May differed statistically significantly (p 0.05) compared to the concentrations of the element in the plants collected on 3 and 25 May. However, statistically significant differences between concentrations of P in the plants collected on 10 May and on two other dates were found only for roots of the plants. The differences between concentrations of P in leaves of the couch grass and nettle sampled on 10 May and on 3 and 25 May were statistically insignificant.

The concentrations of other elements were also not the same on different sampling dates. For example, the concentrations of Na in roots of couch grass were lower, and in leaves were higher in the plants collected on 10 May compared with those in the plants harvested on other dates. In nettle, the Na concentrations were more or less similar regardless of the date of sampling. In leaves of couch grass and in roots of nettle collected on 10 May, the concentration of K was minimal, while in the nettle leaves it was higher than the K content in leaves of the plants collected on 3 and 25 May. In roots of couch grass, the K concentrations slightly increased from beginning to the end of May. In roots of the couch grass and nettle collected on 10 May, the concentration of Fe was higher than in roots of the plants harvested on 3 and 25 May. In leaves of the couch grass, the Fe content slightly decreased during May, and in leaves of nettle, the concentration of Fe was lower in the plants collected on 10 May. The concentration of Ba in roots of the couch grass collected on 10 May was higher compared to the Ba content in roots of the couch grass taken on other dates. In leaves of couch grass, the Ba concentration increased during May. In roots of nettle, the Ba content increased, and in leaves it slightly decreased during May.

The temporal variations in the element concentrations in plants can be caused by various factors. One of the important factors might be climatic conditions at the site on different dates of sampling. It is known that light and temperature are among the key stimuli that can affect the plant biological rhythms in plants [22]. For example, it is commonly supposed that soil temperature can influence the uptake of elements by plants. At low soil temperature nutrient uptake can be reduced as a result of an increase of soil water viscosity and low activity of root nutrient transport [23]. On the other hand, it was reported that under lower temperature the uptake of K, Ca, Mg, Na, and Mn in sweet pepper increased [24].

In Table 1 are shown the changes in the soil/air temperature and air humidity. On 3 May, the temperature of the air and soil constantly decreased during the day, and the air humidity increased. There was also a rain from middle of the day till night. On 10 May, the air and soil temperature increased from morning to evening, while air humidity decreased. On 25 May, the temperature was higher in the middle of the day. At the same time, the concentrations of elements in plants growing in the soil did not always follow the changes in the soil temperature and often varied differently in the daytime. This indicates that the short-term variations in the uptake of elements by couch grass and nettle were not directly related to the daily changes in the soil/air temperature and air humidity. This may be due to the fact that plants have evolved light and temperature sensors capable of maintaining homeostatic balance [25].

It is also well-known that soil pH can affect the solubility of elements in the soil [26]. The relationships between pH and concentrations of elements in plants can be both positive and negative [27]. It was shown that the main factor influencing the changes in the pH in the rhizosphere soil is the net release of H${}^{+}$, HCO${}_3$${}^{-}$ or OH${}^{-}$ in response to the imbalance between cation and anion uptake by roots [28]. Since the concentrations of elements in plants vary during daytime, it would be interesting to compare the temporal variations in the bioaccumulation of elements and possible short-term changes in the pH of the rhizosphere soil.

Fig. 3 illustrates the daily changes in the pH of the soil taken from roots of couch grass and nettle. On all dates of sampling, the maximum values of the pH of the rhizosphere soil of both plant species were observed in the middle of day. But as was shown in Fig. 1 and in Supplementary Table 1, the daily variations in the plant element concentrations do not always follow the changes in the soil pH. Besides, the temporal changes in the concentrations of elements in couch grass and nettle growing at the same place and harvested simultaneously were often different.

On the whole, it is difficult to identify any one crucial factor that would be responsible for dynamic process of element uptake. Considering that various environmental conditions during plant growth can affect the structure and anatomy of the plants, thereby affecting the uptake of elements, the observed fluctuations are result of multiple effects of various factors. It is hardly possible to explain the changes in the plant element concentrations by any one of the external factors; each of them contributes to the uptake process. Since in many cases the daily fluctuations of elements in couch grass and nettle were different, it can be assumed that these effects, at least in part, are species-specific.

Taken into account that all living organisms are under constant control of the Earth’s rotation [1, 29], it can be assumed that the circadian clock may be an important factor affecting the daily variations of the element concentrations in plants. In this case the question arises why the diurnal element variations in the couch grass and nettle growing in the same site and collected simultaneously were often different. This might be due to the fact that these plants belong to different clades, couch grass is monocotyledon and nettle is dicotyledon. It was reported that mono- and dicotyledons can differ in the ability to accumulate various macro- and trace elements [30, 31, 32]. One may speculate that the plants are also able to give different reactions on the circadian changes. As a consequence, the daily fluctuations of the element concentrations in plants can be different for monocotyledons and dicotyledons.

The ratios of nutrient concentrations can serve as an important parameter of the state of plants and the rhizosphere soil [33]. Fig. 4 illustrates the K/Na ratios in roots and leaves of couch grass and nettle and in the rhizosphere soil of the plants collected on different dates. The K/Na ratios were the lowest in the soil taken from roots of the plants, higher in roots, and the highest values of the K/Na ratios were observed in leaves. It was typical for both couch grass and nettle collected during May and was mainly due to significant differences in the K concentrations in the samples. In plants, the higher K/Na ratios resulted from both lower Na and higher (especially in leaves) K concentrations as compared to those in the rhizosphere soil. The higher K to Na ratio is desirable for the plant development [34].

Results of correlation analysis showed that there were no statistically significant relationships between concentrations of elements in the rhizosphere soil and in roots of both plant species. There was a statistically significant (p 0.05) positive correlation between Ba (r = 0.83) and Cu (r = 0.70) concentrations in roots and leaves of couch grass. In nettle, the correlation between concentrations of K in roots and leaves was statistically significant (p 0.05) and negative (r = –0.71).

The taxonomical status of a plant is among the most important factors responsible for uptake mechanisms. Monocotyledons and dicotyledons differ significantly at the morphological and physiological level. As a result, the plants may have different ability to accumulate various elements [35, 36, 37]. These differences may also be related to their distinct root architecture [38] leading to different strategies of uptake of elements from the rhizosphere soil and their distribution between underground and aboveground plant parts.

Cluster analysis revealed that both roots and leaves of the plants collected on different dates were well separated into different groups - couch grass and nettle. The PCA showed that main contribution to the separation was provided by K, Ca, and Mg (leaves) and Na, Ni, Ca, Mg, and Ba (roots). It is also interesting that leaves and roots of the plants collected at 9:00 were separated from leaves and roots of the couch grass and nettle harvested at 14:00 and 19:00. This is illustrated on the example of the plants collected on 3 May (Fig. 5). According to the PCA data, Co (leaves) and Ni (roots) were responsible for the separation. However, similar separation between plants collected at 9:00 and at the two other times was found only in leaves of the couch grass harvested on 10 May and in roots of the nettle harvested on 25 May. It should be mentioned that certain contribution to the separation of the plants collected on 3 May at 9:00 from the plants collected at two other times might also make a rain that took place from middle of the day till evening (precipitation levels were 0.8 mm/h at 14:00 and 0.9 mm/h at 19:00).

Results of the Raman spectroscopy analysis of the soil taken from the plant roots are demonstrated in Fig. 6. Table 2 shows the compounds that were found in the soil samples.

On 3 May, the concentrations of organic compounds of the benzoic acid group in the rhizosphere soil of nettle slightly increased from morning to evening. On the other hand, there were rather serious differences between samples of the rhizosphere soil of couch grass collected at different times. For both plant species, the concentrations of the organic compounds in the rhizosphere soil were higher in the samples taken at 19:00.

At 9:00 on 10 May, the concentrations of the organic compounds in the soil taken from roots of nettle were lower that those in the soil collected at two other times. To the contrary, the concentrations of the compounds in the rhizosphere soil of the couch grass collected at 9:00 were higher than in the soil taken from roots of couch grass at other times.

On May 25, the concentrations of organic compounds which can be attributed to the group of benzoic acids were slightly different for the rhizosphere soil of nettle and couch grass collected at 9:00, 14:00, and 19:00.

The spectra of the rhizosphere soil of the nettle collected on 3 and 25 May were relatively similar, but there were some differences in the area of the wave shift of 1318 cm${}^{-1}$, 1396 cm${}^{-1}$, and probably 1576–1584 cm${}^{-1}$ (salicylic acid, syringic acid, protocatechuic acid) in the soil taken on 10 May. The spectrum of the rhizosphere soil of the couch grass collected on 3 May differed from the spectra of the rhizosphere soil of the couch grass collected at two other dates.

Probably, the differences in the accumulation of the organic acids in the rhizosphere soil of the plants might be due to different ability of the root exudates produced by couch grass and nettle to stimulate homeostasis at the signal level. It can also be assumed that the differences between the spectra of the rhizosphere soils of the plants in certain degree might be associated with the differences between uptakes of mineral elements by these two plant species.