*Article* **Transpiration Sensitivity to Drought in** *Quercus wutaishansea* **Mary Forests on Shady and Sunny Slopes in the Liupan Mountains, Northwestern China**

**Bingbing Liu 1, Pengtao Yu 1,\*, Xue Zhang 2, Jiamei Li 1, Yipeng Yu 1, Yanfang Wan 1, Yanhui Wang 1, Xiao Wang 1, Zebin Liu 1, Lei Pan <sup>3</sup> and Lihong Xu <sup>1</sup>**


**Abstract:** Forests in water source areas are important factors for water supply security, soil, and water conservation, and their water consumption from transpiration is strongly affected by site conditions, including the slope aspect. However, the lack of research on how the slope aspect interferes with the response of stand transpiration to drought has hindered researchers from developing climate-resilient forest–water coordinated, sustainable development plans for different stand conditions. This study was conducted on *Quercus wutaishansea* forests in the southern part of Liupan Mountain in northwest China, and two sample plots were built on sunny and shady slopes. The responses of stand transpiration to various soil moisture and meteorological conditions on different slope orientations were analyzed. The results showed better-growing stands on shady slopes transpired more and consumed more soil moisture than those on sunny slopes. The soil moisture on shady slopes decreased rapidly below the threshold level during the drought, leading to a limitation of stand transpiration; however, its transpiration recovered rapidly after the drought. In contrast, stand transpiration on sunny slopes was not affected by this drought and maintained its pre-drought rate. Our results suggested that stands with higher water demand on shady slopes were more susceptible to drought when it occurred. This indicated that in the case of frequent droughts, the vegetation should be managed according to the vegetation-carrying capacities resulting from different site conditions.

**Keywords:** *Quercus wutaishansea*; transpiration; slope aspect; drought; relative extractable water

#### **1. Introduction**

Plant transpiration accounts for over 60% of evapotranspiration (ET), and it is an essential component of hydrological cycles [1–4]. It has been found to be strongly affected by increasing frequency of drought in the whole world [5–10]. The security level of the water supply in water source areas was also seriously threatened, especially for waterlimited areas such as northwest China [11,12]. Therefore, it is essential to study the effect of drought on water consumption by stand transpiration to develop a reasonable integrated forest water management plan and to cope with climate change, especially in forested water source areas.

The studies on slope aspect changing drought effects on stand transpiration are scarce, although research focusing on the role of drought in stand transpiration is increasing [11,13,14]. It has been found that shaded and downslope slopes with sufficient soil moisture could enhance the resistance of trees to drought [15,16], and this effect would be further changed by tree size, i.e., the large trees would transpire more when faced with to drought [17,18]. Slope aspect also changes the site conditions and the soil moisture content [19,20]. Thus, slope aspect would

**Citation:** Liu, B.; Yu, P.; Zhang, X.; Li, J.; Yu, Y.; Wan, Y.; Wang, Y.; Wang, X.; Liu, Z.; Pan, L.; et al. Transpiration Sensitivity to Drought in *Quercus wutaishansea* Mary Forests on Shady and Sunny Slopes in the Liupan Mountains, Northwestern China. *Forests* **2022**, *13*, 1999. https:// doi.org/10.3390/f13121999

Academic Editor: Romà Ogaya

Received: 28 September 2022 Accepted: 21 November 2022 Published: 25 November 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

indirectly affect the transpiration water consumption of forest stands [11,15,20,21]. However, most previous research has focused on shady slopes where trees are widely distributed and on stands with different slope positions. There is a lack of studies on the different responses of stand transpiration to drought between shady and sunny slopes.

Forest stands with different growth characteristics are severely threatened by drought [15, 20,22–24], such as *Quercus wutaishansea* forests in the Liupan Mountains of northwestern China. In recent decades, the transpiration of *Quercus wutaishansea* trees in northwest China has been severely limited by the increased frequency of droughts [7,25]. Dominant trees under the same stand conditions are more susceptible to soil moisture deficits [17,18,26]. As an essential factor for stand growth, available soil moisture is essential for stand transpiration amounts [15,27], and, as a result, the trees grow better on shady slopes due to more available soil moisture [20]. However, increasing drought would lead to a decrease in the available soil moisture. This would lead to severe impacts on stands with high water requirements. Therefore, we hypothesized that the better-growing stands on shady slopes are more sensitive to drought than those on sunny slopes.

In this study, we analyzed how the slope aspect affected the transpiration of *Quercus wutaishansea* forest under drought conditions. The stand transpiration during the same drought period were compared among slopes and among different drought periods on the slope. The effects of slope, available soil moisture, and meteorological conditions on stand transpiration are furtherly partitioned. These results would help us understand the response mechanisms of the transpiration of *Quercus wutaishansea* forest stands with different slope aspects to drought under climate change and would provide a theoretical basis for integrated forest water management in water-limited areas.

#### **2. Materials and Methods**

#### *2.1. Study Area*

This study was conducted in the Qiuqianjia Forest farm (106◦18 ~106◦29 E, 35◦31 ~35◦37 N), located in the Natural Reserve of the Liupan Mountains in the midwestern part of the Loess Plateau in Northwest China (Figure 1). The forest farm area is 118.42 km2, and the farm has an elevation range from 1612 to 2317 m a.s.l., of which 25% is covered by *Quercus wutaishansea* natural forest. This region has a semihumid continental climate with a mean annual temperature of 5.8 ◦C and mean annual precipitation of 618 mm, 87.6% of which is concentrated in the growing season (May–October) at the elevation of 1960 m a.s.l. The mean annual evaporation (1981–2010) is 1372 mm measured by an evaporation dish with a diameter of 20 cm, which is 2.22 times higher than the mean annual precipitation.

The soil in this region is dominated by the grey cinnamon soil type with a sandy loam texture, and the soil thickness is approximately 80 cm. The dominant tree species are *Quercus wutaishansea* pure forest and *Quercus wutaishansea* and *Populus davidiana* Dode mixed forest, among which *Quercus wutaishansea* pure forest is the most widely distributed, accounting for approximately 30.9% of the natural forest area. The shrub community consists mainly of *Crataegus pinnatifida* Bge., *Ostryopsis davidiana* Decne., and *Cotoneaster multiflora* Bunge. The herbaceous vegetation consists mainly of *Carex pediformis* C. A. Mey. and *Epimedium brevicornu* Maxim.

In this study, a northeast slope and a southwest slope were chosen as shady sample slope facing north and sunny sample slope facing south, respectively (Figure 1). Ten consecutive plots with an area of 30 m × 30 m were established on the sample slopes, i.e., the shady and sunny slopes as shown in Figure 1, and a detailed vegetation survey was made in May 2021. Parameters such as the height, diameter at breast height for each tree in the plots, and forest crown density were measured by the classical method in the vegetation survey. Then, two of the ten samples (Table 1) were selected as intensive plots, on which the sap flow and soil moisture were continuously observed in the growing season of 2021. These two intensive plots were similar in tree age and elevation (Table 1), and the soil thickness in them was greater than 80 cm, which did not limit the tree growth.

**Figure 1.** Geographical location of study plots in the Qiuqianjia Farm Forest: (**a**) relative positions of the study plots on the two slopes; (**b**) red dots indicate consecutive plots with the area of 30 m × 30 m on slopes; blue dots represent the intensive plots, in which the sap flow and soil moisture were continuously and simultaneously measured. (**c**,**d**) The pictures of *Quercus wutaishansea* forest on sunny and shady slopes are shown in the top right corner.


**Table 1.** The characteristics of sample plots on sunny and shady sample slopes. Plots P3 and P8 were selected as intensive plots. *DBH* indicates diameter at breast height; *LAI* indicates leaf area index.

#### *2.2. Sap Flow Measurement and Transpiration Calculation*

All the tress in the sample plots were divided into different classes according to their *DBH*. For example, the trees in the shady slope plot were divided into six classes, i.e., with the *DBH* < 10 cm, 10–15 cm, 15–20 cm, 20–24 cm, 24–27 cm, and >27 cm; and those in the sunny slope plot into five classes, with the *DBH* < 9 cm, 9–14 cm, 14–18 cm, 18–22 cm, and >22 cm classes, respectively. Then, one tree, which had the *DBH* and *H* values close to the mean values of the corresponding *DBH* class, was selected from each *DBH* class as a sample tree, and eleven sample trees in total (Table 2) were observed and their stem sap flux was measured.


**Table 2.** Characteristics of the sample trees selected for sap flow measurement in intensive plots.

\* "+" indicates that the calculated value is higher than the measured value; "−" indicates that the calculated value is lower than the measured value. Total sapwood area represents the sum of the sapwood areas corresponding to the *DBH* classes of the sample plots.

The sap flow density of the sample trees was measured at breast height (1.3 m above ground) using a thermal diffusion probe (SF-L, Ecomatik, Munich, Germany). These probes consisted of two sensors 20 mm long and 2 mm in diameter (S0, a heated sensor powered by a constant current of 12 volts; S1, an unheated sensor). These sensors were inserted 20 mm outside the xylem at breast height on the north-facing side of the trunk (rather than the side exposed to sunlight). Before insertion, the outer bark was peeled off. Each probe was coated with silicone gel to ensure good thermal contact between the probe element and the sapwood. After insertion, the exposed bark was covered with silicone gel to reduce evaporation from the wood surface and then covered with aluminum foil to avoid physical damage and the thermal effects of solar radiation. Sap flow data were recorded every 5 min by a logger (CR1000x, Campbell Scientific Inc., Salt Lake City, UT, USA).

The sap flow density (*JS*, mL·cm−2·min−1) (flow per unit of sapwood area) of an individual sample tree was calculated using Equation (1) [13]:

$$J\_S = 0.714 \times \left(\frac{\Delta T\_{\text{max}} - \Delta T}{\Delta T}\right)^{1.231} \tag{1}$$

where *JS* is the sap flux density (ml·cm−2·min<sup>−</sup>1); <sup>Δ</sup>*<sup>T</sup>* is the temperature difference between the two needles; and Δ*Tmax* is the maximum value of Δ*T* every night.

The transpiration of each *DBH* class was summed to obtain the transpiration of the stand per unit area. The daily stand transpiration (*T*, mm) was extrapolated using Equation (2):

$$T = \frac{\sum\_{i=1}^{n} A\_i \times J\_S}{S} \times 60 \times 24 \div 1000 \tag{2}$$

where *S* is the projected area of the plot (m2); *n* is the number of *DBH* class in the sample plots; *Ai* is the sum of the sapwood area of all trees in each *DBH* class (cm2), which was calculated from the *DBH* of all trees in the sample plots by the relationship between sapwood area and *DBH* for *Quercus wutaishansea* forest using Equation (3):

$$A\_S = 0.7122 \times DBH^{1.7189} \quad \left(R^2 = 0.95, \ n = 12\right) \tag{3}$$

This equation was established by us based on twelve core samples from our sample plots, where *AS* (cm2) is the sapwood area of the tree.

#### *2.3. Weather and Soil Moisture Measurements*

The weather conditions, including the precipitation (TE525MM, *P* (mm)), air temperature (HMP115, *T* ( ◦C)), solar radiation (PQS1, *RS* (W·m−2)), relative humidity (HMP155, *Rh* (%)), and wind speed (034E, *<sup>U</sup>* (m·s−1) were monitored for the period from June to October, 2021, and recorded every 5 min with an automatic weather station (CR1000X, Campbell Scientific Inc., Salt Lake City, UT, USA) in an open area away 100 m from the sample plots.

The daily potential evapotranspiration (*PET*; mm·day−1) was estimated using Equation (4) [28] based on the measured weather data:

$$PET = \frac{0.408\Delta (R\_n - G) + \gamma \frac{900}{T + 273} lI(e\_s - e\_a)}{\Delta + \gamma (1 + 0.34 lI)} \tag{4}$$

where Δ (kPa· ◦C−1) is the slope of the relationship between vapor pressure and air temperature; *Rn* (MJ·m−2·day−1) is the net radiation; *<sup>G</sup>* (MJ·m−2·day−1) is the soil heat density; *γ* (0.053 kPa· ◦C−1) is the psychrometric constant; *T* ( ◦C) is the mean air temperature at a 2 m height; *<sup>U</sup>* (m·s−1) is the wind speed ata2m height; *es* (kPa) is the saturation vapor pressure; and *ea* (kPa) is the actual vapor pressure.

Soil volumetric moisture in the root zone (soil layers 0–10, 10–20, 20–40, and 40–60 cm) was simultaneously monitored using soil moisture and temperature sensors (5-TE, Decagon, Pullman, WA, USA) in the two intensive plots. Data were collected every 5 min by a data logger (EM50, Decagon, Pullman, WA, USA). The relative extractable water (*REW*) was calculated as the ratio of the actual extractable water to the maximum extractable water [29]:

$$REW = \frac{VSM - VSM\_w}{VSM\_\varepsilon - VSM\_w} \tag{5}$$

where *VSM* is the volumetric soil moisture of the 0–60 cm soil layer (%); *VSMw* is the wilting *VSM* of the 0–60 cm soil layer (%) (i.e., soil water potential equal to −1.56 MPa) [30]; and *VSMc* is the field capacity *VSM* of the 0–60 cm soil layer (%).

#### *2.4. Data Analysis*

In this study, the upper boundary line method was applied to determine the response functions between *T* and *REW* data [13,31]. The corresponding thresholds of stand transpiration on *REW* are determined based on the inflection points of the outer boundary line [32].

#### **3. Results**

#### *3.1. Meteorological Characteristics and the Occurrence of Drought*

The summer of 2021 (June to August) was a dryer but hotter summer with low precipitation and high temperatures compared with that for the last 36 years. Although the precipitation from June to October (463.7 mm) was similar to the average precipitation levels of the past 36 years (482.9 ± 109.5 mm), the precipitation from June to August was only 221.7 mm, which was less than the average of the same period for the past 36 years (344.9 mm), with 36% less precipitation (Figure 2a). Unfortunately, this lower precipitation was accompanied by higher temperatures, with an average of 14.08 ◦C from June to October, which was higher than the 36-year average (8.83 ± 0.72 ◦C) (Figure 2b).

**Figure 2.** Multiyear average of the monthly cumulative precipitation from 1982 to 2018 and the monthly cumulative precipitation from 2021 (**a**) and multiyear average of the monthly mean temperature from 1982 to 2018 and the monthly mean temperature from 2021 (**b**). The error bars indicate one standard error.

From June to 17 August, the soil water content continued to decrease. In the period of from 4 July to 17 August, the *REW* reached its lowest level, and soil drought occurred (e.g., the shady slope *REW* varied from 0.37 to 0.47 with a mean value of 0.40). After 17 August, the *REW* on the shady slope gradually recovered above the threshold (0.43), which was determined by the shady slope *T* response *REW*. We defined the following four periods based on the threshold value (0.43) of the shady slope *T* response *REW*: for the pre-drought period (1 June–4 July), *REW* > 0.43; for the dry period (4 July–17 August), *REW* < 0.43; for the late dry period (17 August–19 September), *REW* > 0.43; and for the late growing season (19 September–31 October), *REW* > 0.43. The *REW* trend on the sunny slope was similar to that on the shady slope, but its value was lower than that on the shady slope for all periods. For example, in the dry period, the *REW* on the sunny slope varied in the range 0.31–0.45 with a mean value of 0.37, which was 7.5% lower than that on the shady slope.

PET was significantly higher on the sunny slope than on the shady slope (personal correspondence of Li Jiamei) (Figure 3). The *PET* on sunny slopes was 0.2–4.8 mm·day−1, with a mean value of 2.3 mm·day<sup>−</sup>1, which was 1.2 times higher than on shady slopes (the range of 0.03–4.1 mm·day−1, mean value of 1.9 mm·day−1). The *PET* difference between shady and sunny slopes was significant before and after the drought. For example, low precipitation and high air temperature before and during the drought resulted in higher *PET* with mean values of 3.26 mm·day−<sup>1</sup> and 3.27 mm·day−<sup>1</sup> on sunny slopes, respectively, and 2.73 mm·day−<sup>1</sup> and 2.74 mm·day−<sup>1</sup> on shady slopes. During the latter part of the drought, *PET* gradually declined, with mean values of 1.95 mm·day−<sup>1</sup> and 1.56 mm·day−<sup>1</sup> on sunny and shady slopes, respectively. Late in the growing season, *PET* declined to its lowest value, with mean values of 0.90 mm·day−<sup>1</sup> and 0.61 mm·day<sup>−</sup>1, respectively.

#### *3.2. Sap Flow Density*

The average daily sap flow density was higher on sunny slopes than on shady slopes for all drought periods, i.e., the pre-drought, drought, and late drought periods (Figure 4a). Moreover, the change of daily sap flow density according to drought from the pre-drought period to late drought were very different between the sunny slope and shady slope. On the sunny slope, the sap flow density maintained a stable trend and did not decrease with the development of drought, i.e., there was no big difference of daily sap flow density among three drought periods. Conversely, on the shady slope, it sharply declined significantly (*p* < 0.05) during the drought period.

**Figure 3.** Precipitation variation, *REW* variation, and *PET* variation for shady and sunny slope sample plots from June to October 2021 (the *PET* of both slopes was calculated from unpublished data). A value of 0.43 was determined based on the stand transpiration response *REW* threshold value in 3.3. A *REW* above 0.43 indicated adequate soil moisture, and a *REW* below 0.43 indicated inadequate soil moisture.

**Figure 4.** Daily variation of sap flow density (**a**), and stand transpiration (**b**) on shady and sunny slopes. The shaded area indicates the standard deviation of the sap flow density.

#### *3.3. Stand Transpiration Differences between Shady and Sunny Slope Forests*

In contrast to the sap flow density situation, stand transpiration on the sunny slope was lower than that on the shady slope for the whole growing season (Figure 4b). The stand transpiration on the sunny slope varied from 0.30 mm·day−<sup>1</sup> to 0.36 mm·day−<sup>1</sup> on average, and no dramatic drop of stand transpiration occurred in response to the onset of drought. Surprisingly, on the shady slope, stand transpiration not only decreased significantly during the drought period (*<sup>p</sup>* = 0.004, mean 0.42 mm·day−1), which was 19.1% and 17.9% lower compared to that in the pre-drought and late drought periods, respectively, but also quickly recovered to the pre-drought level after drought (Figure 4b). However, in any case, the stand transpiration on the sunny slope was 23.4%–40% lower than that on the shady slope.

#### *3.4. Responses of Stand Transpiration to REW and PET in Forest Stands with Different Slope Aspects*

For both shady and sunny slopes, *T* increases as a saturated exponential function of *REW* (Figure 5). The effect of the *REW* on *T* remains stable when *REW* reaches a certain level, i.e., a threshold value. However, *T* responded to *REW* at different thresholds among different slope aspects, i.e., the response threshold of stand transpiration to *REW* was 0.43 for the shady slope and 0.33 for the sunny slope.

**Figure 5.** Transpiration responses to *REW* in shady and sunny slope stands.

The response of T to *PET* always followed an increasing saturation exponential function, i.e., T increased rapidly with increasing *PET* and gradually stabilized after it reached a threshold value. However, the response of T to *PET* under sufficient soil moisture was significantly stronger than that under insufficient soil moisture. For the sunny slope, this difference of T among different soil moisture levels was not significant (Figure 6).

**Figure 6.** Effects of *PET* on stand transpiration under sufficient and insufficient *REW* conditions.

#### **4. Discussion**

#### *4.1. Stand Transpiration Is More Sensitive to Drought on Shady Slope*

It has been widely recognized that slope aspect affects stand structure [33] and the forests on shady slopes grow better, such as with larger *DBH*, height, and *LAI*. Additionally, there is less solar radiation and lower *PET* on shady slopes, which accounts for 73.7% and 81.3% of that on sunny slopes in the Liupan Mountains, Northwest China, respectively (personal correspondence of Li Jiamei), and lower soil evaporation [20]. Thus, the stands on shady slopes have more available soil water for tree growth compared with those on sunny slopes (Figure 3).

Our results showed that stand transpiration on shady slopes decreased significantly when soil drought occurred, but recovered quickly after the drought. At the same time, there was no significant change of stand transpiration on sunny slopes regardless of whether the drought occurred. This verified our hypothesis that stand transpiration on shady slopes was more susceptible to drought than that on sunny slopes. Larger trees on shady slopes were more susceptible to drought that occurred frequently [17,18,34,35]. Past studies can indirectly support our results, i.e., that tall trees must lift water to greater heights against gravity and pathlength-associated resistance; therefore, tall trees face greater hydraulic challenges [36–38]. Therefore, the tree growth situation of the stands and the amounts of available soil moisture on different slope aspects are critical in influencing stand transpiration.

#### *4.2. Stand Growth Characteristics Affect Transpiration Response to Drought*

The slope aspect shapes the stand growth characteristics [39]. In general, shady slope stands consume more soil water because of their larger sapwood areas, tree heights, and stand transpiration amounts [15,20,26,40]. Thus, large stand transpiration would accelerate the *REW* decline to the threshold, and the *REW* is insufficient to resist gravitational and distance-based transportation to higher tree tops, resulting in a rapid reduction in stand transpiration (Figure 7) [41]. The results of the present study validate that the *T* response to *PET* is significantly higher in the case of adequate *REW* than in the case of insufficient *REW* (Figure 6). As in past studies, better-growing stands have higher *REW* requirements, and the transpiration begins to be limited when soil moisture is insufficient to meet stand transpiration needs [8,42,43].

**Figure 7.** Vegetation profile and response to soil moisture on shady and sunny slopes.

Stands with lower tree height and *LAI* would consume less soil moisture [17,34]. In our study, the sapwood areas and tree heights of the sunny slope stands are smaller [20], and the stand transpiration amounts are lower (only 42% of that of the shady slope) (Figure 4), thus consuming less soil moisture [40]. Under extreme drought, soil moisture may not drop below or near the threshold due to low water consumption by stand transpiration (Figure 5). In the present study, the *LAI* and tree height on the sunny slope are very low to accommodate higher *PET* and lower *REW* [20,33], thus consuming less soil water. Our results demonstrate that there is a small threshold of the *T* response to the *REW* (0.33) and no difference in the *PET* response under different *REW* levels in this drought of 2021 (Figures 6 and 7).

*Quercus wutaishansea* is the main natural vegetation in forest water areas and grows mainly on shady slopes. Our research shows that the stand transpiration on shady slopes has already been affected by drought. With increasing drought, forests in water source areas will face more serious threats. This reminds us to pay more attention to the forest dynamics related to drought.

#### **5. Conclusions**

We analyzed the transpiration characteristics of forest stands on shady and sunny slopes and their responses to drought. We concluded that on shady slopes with better stand conditions, the trees grew better, consumed more water, and were more susceptible to drought fluctuations than on sunny slopes. On sunny slopes with poorer stand conditions, the trees exhibited long-term drought adaptation strategies, such as low LAI and tree heights. These results highlight that in future forest management, more attention should be paid to the forests, especially on shady slopes where forests are predominantly located.

**Author Contributions:** Conceptualization, B.L., P.Y., and Y.W. (Yanhui Wang); instrument installation, B.L., X.Z., L.P., and J.L.; investigation, B.L., Z.L., and X.Z.; formal analysis, B.L.; writing—original draft preparation, B.L. and P.Y.; writing—review and editing, B.L., P.Y., Y.W. (Yanfang Wan), Y.Y., X.W., B.L., and L.X. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was financially supported by the Central Public-Interest Scientific Institution Basal Research Fund of Chinese Academy of Forestry (CAFYBB2022XD003, CAFYBB2021ZW002), the National Natural Science Foundation of China (U21A2005, U20A2085, 42161144008), and the National Key Research & Development Program of China (2022YFF0801804, 2022YFF0801803).

**Acknowledgments:** We thank Yongqiang Hu and Jun Zhang of Liupanshan Forestry Bureau for their support in field work.

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

#### **References**

