**Habitat Contribution of Water Source (%) Soil Water Rain River Water Ground Water Condensation Water** mild salinity 11.375 ± 2.071 8.525 ± 1.280 27.025 ± 0.330 46.1 ± 4.594 7.05 ± 1.100 moderate salinity 6.775 ± 2.507 4.575 ± 1.661 18.5 ± 5.750 66.4 ± 11.253 3.775 ± 1.375 severe salinity 27.9 ± 9.484 11.925 ± 2.258 20.1 ± 3.955 28.975 ± 10.150 11.125 ± 2.478

#### *3.3. The Contribution of Condensation Water to Plant Growth*

Due to strong fractionation in the leaves, there were no results for leaf water source, so we analyzed the water source for branches, stems, and shallow roots. Water sources were topsoil, shallow soil, deep soil, rain water, river water, groundwater, and condensation water (Table 1).

**Table 1.** Sources of water in different salinity habitats (mean ± SE).

Results revealed that the main contributors to plant growth in mild salinity habitats were soil water (11.375%), rain (8.53%), river (27.03%), groundwater (46.1%), and condensation water (7.1%). In the moderate salinity habitats, the main contributors were 1.8% (Topsoil), 2.18% (Shallow soil), 2.8% (Deep soil), 4.58% (Rain), 18.5% (River water), 66.4% (Ground water), and 3.78% (Condensation water). In the severe salinity habitat, the main contributors were 27.91% (Soil water), 11.93% (Rain), 20.1% (River water), 28.98% (Ground water), and 11.13% (Condensation water).

To summarize, results revealed that groundwater is the main contributor to plant growth in *Halostachys caspica* with the maximum contribution in the moderate salinity community (66.4%) followed by the mild salinity community (28.89%) and the minimum contribution in the severe salinity community (28.98%). Additionally, there were multiple water absorption strategies exhibited in *Halostachys caspica* growth, which ensures that when there are water restrictions other strategies can instead be employed. Lastly, the contribution of condensation water to plant growth in severe salinity habitats was the largest, with an average value of 11.13% and a maximum of 82%. In the mild salinity habitat, the average value was 7.1% and the maximum value was 56%. In the moderate salinity habitat, the average value was 3.79% and the maximum value was 36%.

#### *3.4. Water Migration Path in the Plant Body*

Generally speaking, when water is transported in a mature plant, hydrogen and oxygen stable isotope fractionation will not occur. Only the leaves or high salinity plant will have fractionation, thus, one can use hydrogen and oxygen stable isotope technology to quantify these compositions in the vast majority of land plants. The δ <sup>18</sup>O isotope is present at the earliest water source, thus, this value will gradually increase over time and the moisture migration path in the body of *Halostachys caspica* in different salinity habitats can be determined (Figure 9a–c).

In all three habitats, rain and condensation water directly contributed to the river water and supplied the soil with water through soil infiltration. Groundwater also supplies the river with water and through soil evaporation and transpiration into the atmosphere, plant roots are able to absorb river water, groundwater, and soil water, which all contribute to plant growth.

All of the plant structures, except leaves, had δ <sup>18</sup>O values similar to soil water values. Yang [27] also observed this in *Haloxylon ammodendron* at the study area and speculated that in addition to rain, river water, groundwater, and soil water, *Haloxylon ammodendron* may take advantage of condensation water through strong evaporation through their leaves as a result of day and night temperature differences. Moreover, Yang [27] calculated that the contributions of condensation water to plant growth were 7.1% (mild salinity), 3.78% (moderate salinity), and 11.13% (severe salinity), which indicates that the contribution of

condensed water to plants in different saline habitats is influenced by the soil properties of saline habitats. water), and 11.13% (Condensation water). To summarize, results revealed that groundwater is the main contributor to plant

Results revealed that the main contributors to plant growth in mild salinity habitats were soil water (11.375%), rain (8.53%), river (27.03%), groundwater (46.1%), and condensation water (7.1%). In the moderate salinity habitats, the main contributors were 1.8% (Topsoil), 2.18% (Shallow soil), 2.8% (Deep soil), 4.58% (Rain), 18.5% (River water), 66.4% (Ground water), and 3.78% (Condensation water). In the severe salinity habitat, the main contributors were 27.91% (Soil water), 11.93% (Rain), 20.1% (River water), 28.98% (Ground

*Forests* **2022**, *13*, x FOR PEER REVIEW 11 of 18

In the plant body, water does not simply migrate from the roots to the leaves then return to the atmosphere through evaporation. Rather, in *Halostachys caspica* communities under mild salinity conditions, water migration flows as follows: shallow roots → stems → branches → leaves; there are also short water circuits from shallow roots to deep roots. In moderate salinity conditions, the stems acts as a bifurcation point and one path of water migrates from: stems → shallow and deep roots, while the other path flows from: stems → branches → leaves. In severe salinity conditions, water migrates from: stems → shallow and deep roots →branches → leaves; there is also a water loop from the stems to shallow roots. growth in *Halostachys caspica* with the maximum contribution in the moderate salinity community (66.4%) followed by the mild salinity community (28.89%) and the minimum contribution in the severe salinity community (28.98%). Additionally, there were multiple water absorption strategies exhibited in *Halostachys caspica* growth, which ensures that when there are water restrictions other strategies can instead be employed. Lastly, the contribution of condensation water to plant growth in severe salinity habitats was the largest, with an average value of 11.13% and a maximum of 82%. In the mild salinity habitat, the average value was 7.1% and the maximum value was 56%. In the moderate salinity habitat, the average value was 3.79% and the maximum value was 36%.

In terms of soil water, there are different δ <sup>18</sup>O compositions in topsoil and shallow soil water along the vertical section. Shallow soil water δ <sup>18</sup>O composition showed a partially negative trend from bottom to top under the topsoil, which indicates that there is an ascending phenomenon. In mild and moderate salinity conditions, topsoil water and shallow soil water δ <sup>18</sup>O composition under the topsoil showed a partially positive trend up to the surface. This may be because mild and moderate salinity habitats are near the river and, thus, the soil surface absorbs more precipitation [23]. *3.4. Water Migration Path in the Plant Body* Generally speaking, when water is transported in a mature plant, hydrogen and oxygen stable isotope fractionation will not occur. Only the leaves or high salinity plant will have fractionation, thus, one can use hydrogen and oxygen stable isotope technology to quantify these compositions in the vast majority of land plants. The δ18O isotope is present at the earliest water source, thus, this value will gradually increase over time and the moisture migration path in the body of *Halostachys caspica* in different salinity habitats can be determined (Figure 9a–c).

**Figure 9.** *Cont*.

*Forests* **2022**, *13*, x FOR PEER REVIEW 12 of 18

**Figure 9.** (**a**) Soil-vegetation-atmosphere water migration diagram in a mild salinity community. (**b**) Soil-vegetation-atmosphere water migration diagram in a moderate salinity community. (**c**) Soil-vegetation-atmosphere water migration diagram in a severe salinity community. Note: the unit of δ <sup>18</sup>O is presented as percentages. The arrow means the direction of water transport.

#### **4. Discussions**

#### *4.1. The Contribution of Condensation Water to the Water Needed for the Growth Halostachys Caspica*

In the wild, the water that plants utilize come from precipitation, soil water, runoff (including melting snow and ice), and ground water [28]. In arid regions, there is little precipitation, so groundwater is the only water source available for plant survival, especially perennial plants [29,30]. Using hydrogen and oxygen stable isotopes to study the water sources of plants, we quantitatively analyzed the origin and direction of water and the selective absorption and water use in three ecological environments that differed in salinity levels. Groundwater was determined to be the main water source utilized for plant growth by *Halostachys caspica* in different salinity habitats found within the study area. The utilization of ground water in the mild, moderate, and severe salinity habitats accounted for 46.1%, 66.4%, and 28.98% of the total water use, respectively. Thereunto, the severe salinity plants relied on groundwater the least, followed by the mild habitat, while the moderate habitat was the highest, such that groundwater accounted for more than half of all water sources. Therefore, it is clear that the change of groundwater level has had the greatest influence on plant growth in the moderate salinity area.

Although groundwater is the main water source, soil water is probably the most important source of water for these plants. Due to the nature of the soil (i.e., particle size and porosity), there are differences in the composition of hydrogen and oxygen stable isotopes. In general, surface soil water is able to contact surface moisture, thus, there is relatively large evaporation intensity to produce hydrogen and oxygen stable isotopic fractionation; this is based on established theories that state there is "light" molecular priority to evaporate ( <sup>1</sup>H1H16O) compared to deeper soil water levels. Surface soil water is more enriched with heavier hydrogen and oxygen stable isotopes, so there are significant differences in the levels of hydrogen and oxygen stable isotopes. Soil profiles taken at different depths show that soil water hydrogen and oxygen stable isotopic composition are indeed different, such that there is relatively stable isotopic content up to the deepest layer [26,31]. Moreover, non-moving water was a result of evaporation close to the surface area [32].

The results of this study also revealed that the closer to the soil surface, soil water δ <sup>18</sup>O levels are partially positive, which is similar to the findings of other studies [33]. This is because the light isotope of oxygen evaporates, which leads to <sup>18</sup>O relative enrichment. As for δD values, soil water in the three salinity habitats had ∆δD values between those of rain and the other three natural water bodies (river water, condensation water and groundwater). This indicates that soil water comes primarily from rain, river water, condensation water and groundwater. Furthermore, soil water δ <sup>18</sup>O values were all partially negative relative to the natural water bodies and were similar to condensation water (−5.23614 ± 0.72364%). This further illustrates that soil water in the study area comes from condensation water. These results are similar to the findings of Zhu and Jiang [34], who conducted a study in the northern Loess Plateau. Namely, they found that atmospheric condensation water contributed between 0–20 cm to soil water, and that it is difficult to contribute to the soil water at the 30 cm soil layer and below.

Some scholars employ the use of thermal pulse technology, isotope tracer techniques, fluorescent tracer methods, thermal ratio, botanical anatomy, pressure chamber, and physiological measurements [35] and found that plant leaves absorb water to ease water deficit in the body of the plant through the leaf stomata, bristles, fissure, drainage organs, and other structures on the surface of leaves [6,36–38]. In a study conducted by Burgess et al. [39], isotopic tracer techniques were used on the leaves of *Sequoia sempervirens* located in the United States and found that their main source of water comes from the fog formed overnight. Ellsworth and Williams [40] conducted a study on 16 species of drought and semi-arid shrubs, and one species of mesophytic herbage kept under control conditions. They found that due to transpiration processes, there was a δD and δ <sup>18</sup>O enrichment found in leaf water, which is likely the source of water for young stems. Zhuang and Zhao [38] also conducted a study on whether the leaves of desert coat plants, *Bassia dasyphylla*, and glabrous plants,

*Agriophyllum squarrosum* (Linn.) Moq., absorbed condensation water. Results revealed that leaves of desert plants can absorb condensation water and through the process of photosynthesis, water retention and growth all respond to the presence of condensation water. Meanwhile, glabrous plants could not absorb condensation water. Vitarelli et al. [41] also conducted research on croton plants and demonstrated that the trichoid structure is the key structure through which plants absorb atmospheric water. Yan et al. [42] used a molecular ecology approach to research the intrinsic mechanism of the leaves of *Tamarix ramosissima*, a desert woody plant that absorbs condensation water from the canopy. Yang et al. [43] studied short-life desert plants and found that the leaf and stem can absorb condensation water, and increasing amounts of condensation water can significantly affect population dynamics. Cen and Liu [44] conducted research on the effects of simulated condensation water on the physiological characteristics of the leaf surface structure of *Leymus chinensis* and *Agropyron cristatum* under drought conditions and found that the aboveground biomass and root biomass increased with the presence of condensation water, and that condensation water can both protect and repair damage on plant leaf surface structures following stress induced by drought conditions. In this study, the leaf of *Halostachys caspica* could absorb the condensation water, this is because that hydrophilic polysaccharide compounds are found in cell walls of leaf epidermal cells, eutrophic cells and vascular bundle cell for photosynthetic organs of plants in arid region, these polysaccharides connect an extracellular network within the photosynthetic organs, and accept the moisture absorbed by the cuticle and transport it to the xylem. The microstructure on the surface of photosynthetic organs and the hydrophilic compounds inside which are the material basis of the canopy could absorb the condensation water [18]. In addition, In photosynthetic organs, the aquaporins (AQPs) regulating the plasma membrane and vacuolar membrane plays an important role in the process of transporting the condensation water in the canopy among cells through the symplastic pathway in desert woody plants, which is the molecular mechanism that absorbs the condensation water in the canopy [18].

The contribution of condensation water on plant growth in *Halostachys caspica* should not be overlooked, and the degrees of condensation water use vary under different salinity conditions. The degrees of condensation water use in mild, moderate, and severe salinity habitats were 7.1%, 3.78%, and 11.13%, respectively. This is roughly equivalent to the contribution of rain water in these habitats, which was 8.53%, 2.18%, and 11.93%, respectively. The degrees of condensation water use was relatively high. This may be because the surface soil salinity level was also high, which allows for greater absorption of condensation water, which infiltrated the soil where plant roots were then able to absorb and use the water for growth. Furthermore, salt crusts can reduce soil moisture evaporation, allowing more soil condensation water to infiltrate it. It was also found in this study that the degree of isotopic fractionation was positively correlated with salt tolerance; namely, the salt tolerance of plants lead to the fractionation of hydrogen isotopes [40].

#### *4.2. The Water Migration Path in Halostachys Caspica*

Under normal conditions, there are two paths of water migration in plants. First, roots transport water through the root system so that the cells in the soil absorb moisture from the soil through osmosis and root hair cells absorb soil moisture from the soil to the root hair cell through osmosis, which then transfer moisture to the plant root catheter through the intercellular osmotic differential and this in turn transfers moisture to each plant structure on the ground. Second, plant cells above the ground absorb moisture when guard cells on the plant leaves open, which allow a small amount of water vapor from the atmosphere to be absorbed by the plant tissue.

The water migration paths in the three saline habitats were different and all had specific water movement patterns. In the mild habitat, the water movement path in plants followed as: shallow root → stem → branches → leaves and shallow root → deep root. In moderate habitats, stems acted as the bifurcation point where the water movement path followed as: stem→ branches → leaves and stem → shallow root → deep root. In

severe habitats, the water movement path went from stem to shallow root. These water movement paths occur in the plants' xylem, not the epidermal cells, this may be because the photosynthetic organs of *Halostachys caspica* have the ability to absorb the condensation water in the canopy and transport the water to the xylem [18]. Another discovery was made in the study of the *Haloxylon ammodendron*. When the photosynthetic organs of the *Haloxylon ammodendron* absorb the water from the canopy, whose water potential keeps increasing, and when the water potential of photosynthetic organ increased to a certain degree, which may establish the reverse water potential gradient, namely Ψ Photosynthetic organs > Ψ Secondary branches, the photosynthetic organs can transfer excess water to the main stem via a reverse water potential gradient, and this is conducive to the continuous absorption and utilization of the condensation water in the canopy, it can be seen that the reverse water potential gradient is the energy structure of plant photosynthetic organs capable of absorbing the condensation water in the canopy [18].

It is clear that the water movement path in the mild habitat begins in the shallow root, while the path in the moderate and severe habitats begin in the stems. Thus, water migration paths start in different parts of the plant depending on the degree of salinity in the habitat. This may be related to the salt and water potential differences in plant structures. Previous studies have shown that hydrogen isotopic fractionation in droughtresistant, salt-tolerant, woody plants is likely to occur when water is being absorbed in the roots [45]. Thus, it seems this will also affect condensation water use by *Halostachys caspica*.

To summarize, condensation water absorption and utilization in the study plants took two forms: (1) Direct absorption where leaves directly absorbed condensation water, which can add moisture to the plant surface, reduce the surface temperature of leaves, and reduce the water loss of surface evaporation [46], or replenish the evaporation consumption, as well as go to the plant body for growth; and (2) Indirect absorption, where atmospheric condensation in the soil constitutes the soil water, while the atmospheric condensation water replenishes the river and groundwater that plant roots absorb in addition to soil water. These results reflect the findings of Goebel and Lascano [17], who measured the δD and δ <sup>18</sup>O composition of different water sources of cotton and quantitatively analyzed the water content, including condensation water. Chen et al. [47] also conducted a study on 50 plant species in arid and semi-arid regions in Central Ningxia Province and found that the plant leaves could absorb water and even had the ability to take advantage of small amounts of rainfall.

#### **5. Conclusions**

In conclusion, this study revealed that: (1) Scale-like leaves can actively absorb condensation water, the condensation water absorbed by leaves can supplement the water consumed by evaporation to a certain extent, and the contribution of condensation water to plant growth in severe salinity habitats is the greatest (11.13%); (2) The migration path of water movement in the three habitats followed two main paths: (a) rainwater and condensation water were recharged through soil to compensate for groundwater, while some groundwater compensated for river water, and these were in part returned to the atmosphere by soil evaporation and plant transpiration; and (b) rainwater and condensation water directly compensated for the river, such that plant roots not only absorbed river water but also groundwater and soil water to assist with plant growth; (3) In mild salinity habitats, water movement paths in plants followed as: shallow root → stem → branches → leaves; and shallow root → deep root; in moderate habitats, stems acted as the bifurcation point and the path of water followed as: stem → branches → leaves, as well as: stem → shallow root → deep root; in severe habitats, the water movement path followed as: deep root → shallow root → stem → branches → leaves and finally returning to the atmosphere; there is also a water circuit from stem to shallow root. Although there are clear condensation water movement paths, this remains to be studied further, specifically in the xylem of the plant.

**Author Contributions:** Conceptualization, L.Q., X.H. and G.L.; investigation: L.Q., X.H. and J.Y.; methodology, L.Q. and X.H.; software, L.Q. and X.H.; writing—original draft, L.Q., X.H. and J.Y.; writing—review and editing, L.Q. and X.H.; supervision, G.L.; funding acquisition, G.L., L.Q. and X.H. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was funded by the National Natural Science Foundation of China (Nos. 41571034, 32101360, 31760168, 31660120) and the Doctor Starts Project of Xinjiang University (202115120003).

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

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** We thank Jing Cao, Xiao Ying, Ting-Quan Wang in Key Laboratory of Oasis Ecology of Xinjiang University for their indispensable help in fieldwork.

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

#### **References**

