Effects of Stumping and Meteorological Factors on Sap Flow Characteristics of Haloxylon ammodendron in Ulan Buh Desert, Northwestern China

Abstract

The shrub/dwarf tree Haloxylon ammodendron is a prevalent woody plant used to combat desertification in the arid and semi-arid regions of northwestern China. Despite its drought resistance, artificial stands of this species experience significant degradation approximately ten years post-afforestation. Stumping, which involves cutting a portion of the above-ground part of shrubs/trees, is a common practice aimed at reducing water consumption and enhancing the growth of these stands. However, the impact of stumping on the sap flow of H. ammodendron remain inadequately understood, posing challenges to the sustainable management of these artificial stands. In this study, we monitored the sap flow of H. ammodendron subjected to various stumping treatments in the Ulan Buh Desert using the PS-TDP8 tree sap flow monitoring system. Concurrently, we measured several meteorological factors with an automatic weather station. We examined the changes in sap flow velocity following stumping and its response to meteorological factors to elucidate water use during growth. Our findings indicate that both the change in sap flow velocity and characteristics were closely associated with the degree of stumping. The initiation time of sap flow for H. ammodendron under different stumping treatments was earlier than that of the control group. The daily mean value and daily accumulation of sap flow followed the order: 50% stumping > control (no stumping) > 75% stumping > 100% stumping. Sap flow velocity and daily sap flow accumulation increased at 50% stumping but decreased at 75% and 100% stumping. Stumping altered the relationships between sap flow velocity and meteorological factors, with the correlation coefficient between these variables decreasing as the degree of stumping increased. The sap flow following stumping was primarily influenced by both the degree of stumping and meteorological factors. These results may contribute to a better understanding of water transport during the growth of H. ammodendron following stumping.

1. Introduction

Haloxylon ammodendron, commonly referred to as saxaul, is planted on a large scale for sand-fixing afforestation in the arid and semi-arid sandy areas of northwestern China [1]. It stands as one of the most crucial shrubs used for desertification control in sandy deserts across Inner Mongolia, Gansu, Ningxia, and Xinjiang provinces. Over 20,000 hectares of artificially created saxaul forests exist on the northeastern fringe of the Ulan Buh Desert. Despite their significant role in mitigating sand movement, these forests undergo severe degradation approximately ten years post-afforestation due to excessive soil moisture consumption in sand dunes [2,3]. Such declines critically impair their protective function [4]. A common practice to manage degraded saxaul forests involves stumping, which entails cutting a portion of the above-ground part of shrubs/trees. While appropriate stumping can stimulate sprouting, enhance vigour, and potentially increase biomass, repeated stumping can stress the plant, markedly reduce its growth rate, and potentially lead to elevated mortality rates.

Plants absorb water through their roots and facilitate the flow of sap in the xylem via a process known as transpiration pulling. This process can be characterized by stem sap flow, which encapsulates the traits of water transport and usage within plants [5,6]. Sap flow refers to the water movement through a specific section of the stem diameter over a certain period. It is not only crucial for individual tree survival and growth, but also plays a significant role in maintaining the water balance of a forest ecosystem [7,8]. The velocity of sap flow provides a logical basis for analyzing the adaptation processes of desert plants [9]. Thermal diffusion sap flow meters enable automatic, continuous, real-time dynamic monitoring under natural plant growth conditions without disrupting their normal growth and physiological activities [10,11,12]. Consequently, these meters are convenient for monitoring the sap flow of trees.

Stumping is a highly effective method for shrub management due to its ability to reduce stand density, mitigate vegetation decline, and significantly enhance vegetation productivity. Additionally, stumping contributes to the stability of shelter forest structure and its function in windbreak and sandstorm prevention [13,14]. An increase in stumping intensity within a certain range has been found to decrease the water consumption of old branches on the ground, while increasing the water use for germination of new branches per unit area [15]. This promotes the growth and development of new saxaul branches [16]. The vigor of plant life activities post-stumping can be assessed by measuring the sap flow velocity of plant trunks. Numerous studies have examined the sap flow of natural saxaul in various desert regions [17,18] and the relationship between sap flow and meteorological factors [19,20]. For instance, it has been shown that sap flow velocity is correlated with vapor pressure deficit and solar radiation, but not soil water conditions in tree species in the semiarid Loess Plateau region of China [21]. Sap flow velocity could first increase with growth age, but decrease as the saxaul grows older [22]. Another study shows that the key factor affecting the stem sap flow velocity may change with age: for saxaul forests that were younger than 15 years, it was temperature that played a key role in the stem sap flow velocity, while vapor pressure deficit was the key factor affecting the stem sap flow when the saxaul was 20 years old [23]. In contrast, there is limited understanding of the effects of stumping on saxaul’s sap flow. Consequently, it remains unclear how the sap flow of saxaul and its response to meteorological factors vary with different stumping treatments.

Artificial saxaul forests, characterized by varying afforestation years and diverse stand structures, offer valuable resources for investigating sap flow dynamics. In this study, we undertook a field experiment examining the sap flow of artificial saxaul forests [15]. We sampled saxaul plants subjected to different stumping treatments in the northeastern Ulan Buh Desert, monitored the sap flow of these plants alongside meteorological factors, and analyzed the relationships between sap flow velocity and meteorological factors under various stumping treatments. Subsequently, we dissected the impact of stumping degrees on sap flow characteristics. Our findings provide both technical support and a theoretical foundation for managing artificial saxaul forests to mitigate plantation degradation from a sap flow perspective.

2. Materials and Methods

2.1. Study Site

The research area is situated in the northeastern Ulan Buh Desert, within the Yellow River Basin (106°46.573′ E, 40°26.337′ N). This region falls within the mid-temperate zone and experiences a semi-arid continental climate. The average temperature stands at 6.8 °C, with an annual sunshine duration of 3229.9 h. The mean annual precipitation approximates 140.3 mm, while the potential annual evaporation is around 2372.1 mm. The prevailing wind direction is from the northwest, with sandstorms typically occurring between November to May of the subsequent year. The soil composition is primarily aeolian sandy soil, with fine sand being the dominant type of sand present. Crescent and conical dunes are the predominant forms of sand dunes in this area. The terrain lacks significant undulations, with dune heights not exceeding 10 m. Natural vegetation primarily consists of Artemisia ordosica and Nitraria tangutorum, while plantations mainly include Caragana korshinskii, Haloxylon ammodendron, Hedysarum scoparium, and Calligonum mongolicum.

2.2. Experimental Design and Sampling

The experimental design is shown in the Figure 1. The experiment was conducted in a 200 m² plot for the artificial sand-fixing saxaul forest under similar environmental conditions. The experimental plot is characterized by a relatively flat and non-undulating terrain. The forest was established over a period of 20 years, resulting in a relatively simple age structure. The density of the saxaul forest was measured at 3 m × 2 m, with vegetation coverage accounting for 15% of the plot. The average soil water content on a mass basis ranged from 0.35% to 2.24%. Despite some instances of plant mortality, the overall stand was relatively sparse. Healthy saxaul individuals were selected at random as the experimental sample plants. To facilitate comparison of growth index variations in saxaul before and after stumping, the sampled saxaul plants were labelled with 0%, 25%, 50%, and CK (control) in 2021. The growth index of the unstumped saxaul (from 2021) is presented in

Table 1. Growth index of Haloxylon ammodendron before and after stumping treatment.

Year	Treatments	Ground Diameter (cm)	Plant Height (m)	Crown Diameter (m)	New Branch Length (cm)
2021 no stumping	0%	11.75	2.91	2.65	7.13
	25%	7.25	2.39	2.36	5.37
	50%	12.25	2.69	2.38	6.14
	CK	10.83	3.10	3.35	8.24
2022 after stumping	0%	-	0.98	1.07	24.41
	25%	-	1.24	1.51	14.17
	50%	-	1.68	2.63	9.36
	CK	-	3.12	3.35	5.95

.

The stumping treatment was conducted in March 2022. Four levels of stumping treatments were applied: control (CK) without cutting; 50%, 25%, and 0% resulting in 50%, 25%, and 0% canopy retention, respectively, that is, 50%, 75% and 100% of the canopy was cut, respectively (Figure 2).

2.3. Stem Sap Flow Measurement

Sap flow was monitored using a Granier sap flow system (PS-TDP8, PlantSensors, Australia) installed in 2021. The system comprises 15-mm-long sensor probes and a data logger (CR1000X-ST, Campbell Scientific, Logan, UT, USA). The sample plant necessitates a straight trunk, and the installation area of the probe ensures no damage to the upper and lower 30 cm of the trunk. The sap flow meter and probe were installed following the manufacturer’s instructions for the PS-TDP8 sap flow system. Data recording occurred at 10 min intervals. To minimize interference from sunlight and rain, the probe was secured with plastic foam and sealed with tin foil paper.

The formula for calculating the sap flow velocity [10] is as follows:

$$V_S=119\times {10}^{-6}\times {\left({\displaystyle \frac{\Delta T_m-\Delta T}{\Delta T}}\right)}^{1.231}$$

(1)

where V_s is sap flow velocity (m·s⁻¹). The value was multiplied by 3.6 × 10⁵, so the unit of V_s became cm·h⁻¹. ΔT_m is the maximum probe temperature difference within 24 h. ΔT is the instantaneous temperature difference between the two probes, and ΔT is obtained by dividing the voltage difference between the two probes of TDP by the empirical constant 0.04. While this equation calculates uncalibrated sap flux density, it is widely used for comparison of transpiration characteristics before and after certain experimental treatments, especially when transpiration characteristics rather than absolute amounts of water use are discussed [21].

2.4. Meteorological Factors Measurements

The meteorological factors solar radiation (Rn), air relative humidity (RH), and air temperature (Ta) were measured by an automatic HOBO weather station (Onset, Bourne, MA, USA). The time interval was set to 10 min, which was consistent with the time interval of sap flow measurement.

The saturated vapor pressure deficit (VPD) is calculated by the following formula:

$$VPD=(1-RH/100)\times a\times e^{[b\times Ta/(Ta+c)]}$$

(2)

where Ta is air temperature; RH is air relative humidity; parameters a, b and c are 0.611 kPa, 17.502 kPa, and 237.3 °C, respectively.

2.5. Statistical Analysis

We chose to analyze the daily fluctuations of sap flow data over several consecutive days, all occurring on the same date, in both September 2021 and September 2022. The trends observed in meteorological factors were analogous, with minimal differences noted. To minimize error, we selected representative data from several days. We employed Microsoft^® Excel^® LTSC MSO (16.0.14332.20734) 64-bit software (Microsoft, Redmond, Washington, DC, USA) for the analysis of daily sap flow variations. Correlation analysis and stepwise regression analysis were conducted using IBM SPSS Statistics 29 software (IBM, New York, NY, USA). Finally, charts were generated using both Excel (Microsoft, Redmond, Washington, DC, USA) and Origin 8 software (OriginLab Corporation, Northampton, MA, USA).

3. Results

3.1. Variation in Sap Flow Velocity with Stumping Treatment

Sap flow in saxaul exhibited pronounced diurnal variations across all treatments (Figure 3). The velocity of sap flow was notably higher during daylight hours compared to nighttime, displaying distinct peaks and troughs (Figure 3a,b). Despite the significant diurnal fluctuations, the onset time of sap flow remained relatively consistent across different treatments. Under identical meteorological conditions (on a given day), there were discernible differences in the maximum, minimum, and average sap flow velocities among the treatments (Figure 3a,b). Following stumping in 2022, the diurnal patterns of sap flow velocity underwent noticeable alterations (Figure 3b), yet these changes deviated from those observed prior to stumping (Figure 3a,b). Post-stumping, the sap flow velocity under the 50% stumping treatment was markedly elevated compared to both the control and other treatments (Figure 3d). Furthermore, the onset time of sap flow differed from that of the control and pre-stumping conditions. For the control group, the sap flow velocity in 2022 was marginally lower than in 2021 (Figure 3c,d). Conversely, post-stumping, both the range of variation and the mean sap flow velocity were significantly reduced, particularly after the 25% and 0% stumping treatments (Figure 3c,d).

While groundwater level was significantly lower in 2022 than in 2021 (F_1,6 = 12.36, p = 0.013) (Figure S1, Supporting Information), there was no significant variation in soil water content between 2021 and 2022 (F_1,18 = 1.673, p = 0.212) (Figure S2, Supporting Information). There was a statistically significant variation in leaf traits(Figure S3, Supporting Information), but net photosynthetic rate did not vary significantly with stumping treatment (F_3,56 = 1.241, p = 0.303) (Figure S4, Supporting Information). Stumping treatment significantly reduced the transpiration (F_3,56 = 28.62, p < 0.001), while there was no significant difference among the three stumping treatments (Figure S5, Supporting Information).

3.2. Frequency Distribution of the Sap Flow Velocity

The frequency statistical analysis revealed that the sap flow velocity was predominantly distributed between 0–1 cm/h, with a frequency exceeding 40% both prior to and following the stumping treatments (Figure 4). The distribution of sap flow velocity underwent significant changes post-stumping. Specifically, after stumping, the frequency of sap flow velocity within the 0–1 cm/h range only decreased in the 50% stumping treatment (Figure 4b vs. Figure 4f), while it increased in all other treatments (Figure 4). Following the 50% stumping treatment, there was an increase in the proportion of high sap flow velocity (Figure 4b vs. Figure 4f). After the 75% (Figure 4g) and 100% (Figure 4h) stumping treatments, the sap flow velocity was primarily concentrated in the lower range. Notably, after the 100% stumping treatment, the maximum sap flow velocity decreased from above 12 cm/h before stumping to below 3 cm/h, while the proportion of sap flow velocity within the 0–1 cm/h range increased from less than 45% to over 80% (Figure 4d vs. Figure 4h).

3.3. Relationships Between Sap Flow Velocity and Meteorological Factors

In general, the variation in meteorological factors was minimal throughout the experimental period (Figure 5). While the amplitude of air temperature, relative humidity, and vapor pressure deficit was marginally greater in 2021 (Figure 5a,b,d), no significant alterations were observed in these meteorological parameters pre- and post-stumping treatments (Figure 5e−h). The wind speed was significantly different between 2021 and 2022 (F_1,4 = 2014, p < 0.001). However, the average wind speed was low in both years (2021: 1.36 m/s; 2022: 1.56 m/s), with the average wind speed being 15% higher than that of 2021 (Figure S6, Supporting Information).

The correlations between the sap flow velocity of saxaul and various meteorological factors were found to significantly vary with stumping treatments. Prior to stumping, a significant positive correlation was observed between the sap flow velocity and atmospheric temperature (Ta) (Figure 6a), solar radiation (Rn) (Figure 6c), and vapor pressure deficit (VPD) (Figure 6d). Conversely, a significant negative correlation was noted with air relative humidity (RH) (Figure 6b). Among these factors, solar radiation emerged as the most influential, with correlation coefficients of 0.96, 0.91, 0.71, and 0.87 for canopy retention at CK, 50%, 25%, and 0% respectively. Following stumping, the correlation between sap flow velocity and Rn decreased, particularly in cases of severe stumping (0% canopy retention). However, the correlations between sap flow velocity and Ta, RH, and VPD increased. As the degree of stumping increased, the correlation between sap flow velocity and Rn became less pronounced compared to that of Ta, RH, and VPD (Figure 6e–h).

3.4. Influence of Meteorological Factors on Sap Flow Velocity

A multiple linear stepwise regression was conducted, utilizing sap flow velocity as the dependent variable and solar radiation, air temperature, air relative humidity, and VPD as independent variables (

Table 2. Regressions between sap flow velocity and meteorological factors.

Treatments		Entry Factor	R2	Regressions Equation	p
2021 before stumping	CK	Rn	0.912	Y = 0.434 + 0.011Rn	<0.01
		Rn, RH, VPD, Ta	0.920	Y = 0.807 + 0.011Rn − 0.020RH − 0.662VPD + 0.081Ta	<0.01
	50%	Rn	0.833	Y = 0.426 + 0.008Rn	<0.01
		Rn, Ta	0.879	Y = −1.090 + 0.006 Rn + 0.092Ta	<0.01
	25%	Rn	0.500	Y = 1.034 + 0.006Rn	<0.01
		Rn, Ta, VPD	0.625	Y = −3.852 + 0.005Rn + 0.068Ta − 2.569VPD	<0.01
	0%	Rn	0.759	Y = 0.234 + 0.002Rn	<0.01
		Rn, Ta, VPD	0.808	Y = −0.079 + 0.01Rn + 0.171VPD + 0.203VPD	<0.01
2022 after stumping	CK	Rn	0.857	Y = 0.370 + 0.007Rn	<0.01
		Rn, VPD, Ta, RH	0.940	Y = −1.223 + 0.005Rn + 6.136VPD + 0.477Ta − 0.010RH	<0.01
	50%	Rn	0.763	Y = 0.959 + 0.011R	<0.01
		Rn, VPD, T, RH	0.912	Y = −3.854 + 0.006 Rn + 12.247VPD + 0.886Ta − 0.011RH	<0.01
	25%	Rn	0.420	Y = 0.717 + 0.004 Rn	<0.01
		Rn, RH, VPD, Ta	0.540	Y = 5.784 + 0.004Rn − 0.049RH − 7.933VPD + 0.652Ta	<0.01
	0%	VPD	0.425	Y = −0.465 + 0.515VPD	<0.01
		VPD, Ta	0.565	Y = −0.684 + 3.257VPD + 0.261T	<0.01

). The sequence of meteorological factors and their respective influences on sap flow velocity varied between 2021 and 2022. Solar radiation emerged as the primary meteorological factor influencing sap flow velocity for data from 2021, accounting for 91.2%, 83.3%, 50.0%, and 75.9% of the variation in sap flow velocity for CK, 50%, 25%, and 0% canopy retention, respectively. Conversely, in 2022, solar radiation exerted the most significant impact on the sap flow velocity of control and 50% and 25% canopy retention, albeit with a lower correlation than in 2021. The sap flow velocity of the 0% canopy retention treatment was most strongly influenced by VPD, while the sap flow velocity was less affected by meteorological factors for treatments 25% and 0% canopy retention, with R² values for models being 0.540 and 0.565, respectively. These findings suggest an increased influence of morphological factors on sap flow velocity following stumping. Correlations between sap flow velocity and meteorological factors show that stem sap flow velocity after the treatment 50% and 25% stumping as well as in the control treatment was most strongly correlated with solar radiation, while it was negatively correlated with relative humidity. After the treatment 0% stumping, the sap flow velocity was most strongly correlate with air temperature and vapor pressure deficit, also negatively correlated with relative humidity (Figure S7, Supporting Information).

3.5. Main Factors Affecting Sap Flow Velocity

The multi-factor analysis of variance revealed that the primary factors influencing the sap flow velocity of saxaul prior to stumping (2021) were morphological differences, meteorological factors, and their interaction, with F values of 22.24, 10.57, and 4.64, respectively. The degrees of freedom for morphological differences and meteorological factors were 3 and 4, respectively (

Table 3. Effects of morphological factors and meteorological factors on sap flow velocity.

Treatments	Source	Type III Sum of Squares	df	Mean Square	F	p
2021 before stumping	Corrected Model	1364.74	19	71.83	8.66	<0.01
	Intercept	21,953.50	1	21,953.50	2648.10	<0.01
	Morphological factor	553.06	3	184.35	22.24	<0.01
	Meteorological factor	350.56	4	87.64	10.57	<0.01
	Morphological factor × Meteorological factor	461.11	12	38.43	4.64	<0.01
2022 after stumping	Corrected Model	4114.04	19	216.53	33.98	<0.01
	Intercept	11,733.50	1	11,733.50	1841.32	<0.01
	Morphological factor	3572.73	3	1190.91	186.89	<0.01
	Meteorological factor	292.77	4	73.19	11.49	<0.01
	Morphological factor × Meteorological factor	248.54	12	20.71	3.25	<0.01

), suggesting that morphological differences had a more significant impact on the sap flow velocity before stumping. Conversely, the main factors affecting the sap flow velocity after stumping (2022) were also morphological differences, meteorological factors, and their interaction, but with F values of 186.89, 11.49, and 3.25, respectively. Given that the degree of freedom of the model terms remained constant, this suggests a significant increase in the influence of morphological differences on the sap flow velocity (Table 3). The results indicate that the substantial changes in morphological structure due to stumping treatments strongly influenced the variation in the sap flow velocity after stumping.

4. Discussion

4.1. Diurnal Variation in Sap Flow Velocity

The sap flow velocity of saxaul exhibited significant diurnal variation (Figure 3). During daylight hours, the sap flow velocity exceeded that at night, with a notable diurnal fluctuation. Elevated temperatures in the afternoon may have prompted stomatal closure in the assimilating branches, thereby inhibiting branch and leaf transpiration and reducing the sap flow velocity. This observation aligns with findings from prior research by other scholars [24,25]. A brief “lunch break” phenomenon was observed in the 50% stumping treatment, after which the sap flow velocity rebounded and remained high. A similar “lunch break” was noted in the sap flow of saxaul on the southern margin of the Junggar Basin, potentially attributed to ecological, physiological, and biochemical factors [26]. While the initiation time of sap flow varied across different stumping treatments, it consistently preceded that of the control group. Research indicates that the onset of sap flow can fluctuate seasonally [25]. In this investigation, sap flow was monitored over two consecutive years during identical periods, suggesting that both the onset timing and diurnal variations were primarily influenced by individual morphological changes in saxaul induced by stumping treatments.

4.2. Influence of Different Treatments on Sap Flow

In this study, the sap flow of saxaul was primarily monitored within a 200 square meter range under similar environmental conditions. This allowed for a reasonable comparison of sap flow variations in saxaul subjected to different stumping treatments. Human intervention was employed to modify the morphological characteristics of saxaul to mitigate excessive water consumption, a strategy proven feasible in our findings. This aligns with the understanding that specific control measures are necessary to maintain soil moisture balance following several years of saxaul plantation growth [27]. Some research has shown that actual water consumption results can be comparable when observing different individuals of tree species under identical environmental conditions simultaneously [28], despite the fact that individual water consumption varies under real-world conditions [29]. The sap flow of saxaul under varying stumping treatments can indicate the transpiration efficiency and water usage status of individual plants under normal conditions [30]. The peak and daily mean values of sap flow were highest in the 50% stumping treatment and lowest in the treatment resulting in 0% canopy retention. Variations in physiological and biochemical characteristics led to differences in water consumption ability among individuals [31]. The alteration in sap flow of saxaul under different stumping treatments served as a mechanism for the shrubs to meet their transpiration needs, reflecting the relationship between sap flow and tree transpiration water consumption [32]. The mean and daily accumulation of sap flow in saxaul were found to be lower in treatments resulting in 25% and 0% canopy retention, compared to the control. This suggests that high-intensity stumping significantly reduces water consumption, and that morphological differences after stumping affected the distribution of sap flow velocity. Conversely, the sap flow was higher in the treatment with 50% stumping than in the control, indicating that the characteristics of sap flow are still influenced by tree height [33] and crown width [26], even after mild stumping. A previous study demonstrated significant differences in sap flow when the variance between variants of Royal Walnut (Juglans regia L.) was substantial [34]. These results suggest that the structure of the canopy, as well as its area and shape, play a crucial role in determining sap flow velocity during the growing season [34,35].

4.3. The Relationship Between Sap Flow and Meteorological Factors

Sap flow is influenced not only by the biological characteristics of a plant but also by environmental factors [10,36]. In this study, we identified solar radiation, air temperature, and vapor pressure deficit as the primary meteorological factors affecting sap flow in saxaul [20,37,38,39]. We found a significant positive correlation between sap flow and Rn (net radiation), Ta (air temperature), and VPD (vapor pressure deficit), while a significant negative correlation was observed with RH (relative humidity). While earlier studies conducted manipulation experiments to examine plant primary processes’ responses to environmental changes, such as photosynthesis’s response to drought [40,41], these were not designed to elucidate plant physiological adaptation mechanisms. Therefore, it remains crucial to investigate plants’ responses to environmental changes before, during, and after human intervention [42,43]. The notable alteration in sap flow patterns prior to stumping was primarily attributed to variations in solar radiation intensity [44,45]. However, in the treatment where 0% of the canopy was retained, VPD emerged as the first entry factor, accounting for 42.5% of the variation in sap flow velocity. For other stumping treatments and the control group, Rn was the primary entry factor, explaining 76.3% (50% canopy retention), 42.0% (25% canopy retention), and 85.7% (CK) of the variation in sap flow velocity, respectively. This approach more accurately simulates the response characteristics of sap flow velocity in saxaul to meteorological factors [18]. Our findings diverge from those of previous stepwise regression studies on plant sap flow and environmental factors in arid regions [46,47], potentially due to differences in tree species and climate regions.

4.4. Caveats

While our study was conducted with a rigorous methodology, it was confined to a relatively small spatial scale of a few hundred square meters. Consequently, the calculated stem sap flow velocity can only serve as a reference and not an accurate basis for determining water consumption at the population or ecosystem level on a larger geographical scale. Additionally, measurement errors are unavoidable due to the relatively small sample size. We hypothesize that the relative importance of various meteorological factors may differ in varying environmental contexts. Nonetheless, our research provides empirical evidence regarding the impact of human intervention on the water consumption of desert shrubs. Our findings suggest that the water consumption of saxaul forests could be reduced through the application of appropriate stumping treatments. Further research should focus on the response of sap flow to stumping treatments across different shrub species and those with varying afforestation ages.

5. Conclusions

The diurnal variation in sap flow of saxaul, when stumped, was evident. The initiation time for all stumping treatments occurred earlier than that of the control group (without stumping). Peak and daily mean values of sap flow varied across different stumping treatments, with the highest recorded in the 50% stumping treatment and the lowest in the treatment resulting in 0% canopy retention (100% stumping). The relationship between the sap flow of saxaul and meteorological factors altered with different stumping treatments. Compared to the control group (no stumping), the response of the sap flow to meteorological factors decreased in treatments resulting in 25% (75% stumping) and 0% canopy retention (100% stumping). Solar radiation had the most significant impact on the sap flow in the control group and in treatments resulting in 50% and 25% canopy retention, while VPD had the most significant influence in the treatment resulting in 0% canopy retention. This study indicates that water consumption and transport processes in saxaul can be significantly influenced by various stumping treatments. In conclusion, this research demonstrates that both the sap flow velocity of saxaul and its response to meteorological factors are significantly affected by stumping treatments. These findings have important implications for ecological protection and restoration in arid and semi-arid sandy regions. Future research will continue to monitor the effects of human disturbance and climate change on the sap flow of saxaul, integrating other research methods and results to elucidate the mechanisms of human disturbance on plant water use in dry areas.