Transpiration is fundamental to the understanding of the ecophysiology of planted forests in arid ecosystems, and it is one of the most uncertain components in the ecosystem water balance. The objective of this study was to quantify differences in whole-tree transpiration estimates obtained with a heat ratio probe in a secondary Qinghai spruce (Picea crassifolia) forest. To do this, we analyzed the sap flux density values obtained with sensors installed in (1) holes drilled in the preceding growing season (treatment) and (2) holes drilled in the current year (control). The study was conducted in a catchment in the Qilian Mountains of western China. The results showed that an incomplete diameter at breast height (DBH) range contributed to 28.5% of the overestimation of the sapwood area when the DBH > 10 cm and 22.6% of the underestimation of the sapwood area when the DBH < 5 cm. At daily scales, there were significant differences in both the quantity and magnitude of the sap flux density between the treatment and control groups. Furthermore, a linear regression function (R² = 0.96, p < 0.001), which was almost parallel to the 1:1 reference line, was obtained for the sap flux density correction for the treatment group, and the daily sap flux density and whole-tree transpiration were underestimated by 36.8 and 37.5%, respectively, at the half-hour scale. This study illustrates uncertainties and a correction function for sap flow estimations in young Qinghai spruce trees when using heat ratio sensors with minimal damage over multiple growing seasons.

The spatial distribution of soil moisture is a critical determinant for the success of vegetation restoration initiatives in semi-arid and arid regions. The Qilian Mountains, situated within a semi-arid zone in China, have been subject to significant water-induced soil erosion, which has led to extensive restoration activities, predominantly utilizing the species P. crassifolia. However, the interconnections between soil moisture and various land cover types within this region remain unclear, presenting challenges to effective woodland rehabilitation. This study examines the surface soil moisture dynamics in afforested areas with varying ages of plantation to determine the influence of tree planting on the moisture content of the upper soil layer. It investigates the characteristics and temporal patterns of surface soil moisture as the age of the plantation increases. The findings indicate that: (1) soil moisture levels follow a descending sequence from natural forest, through shrubland and grassland, to planted forest and mixed forest, with statistically significant differences observed between natural and mixed forests (p < 0.05); (2) young afforested areas (less than 50 years old) have lower soil moisture levels compared to natural forests, shrublands, or grasslands, and the ecohydrological impacts of afforestation become apparent with a temporal delay; and (3) the analysis using Generalized Additive Mixed Models (GAMM) and the application of Kriging interpolation to determine the spatial distribution of soil moisture reveals that in semi-arid and arid regions, several factors have a pronounced a non-linear relationship with the moisture content of the surface soil. These factors include the duration of afforestation, the position on the lower slope, the presence of shade on the slope, and the scale at which the study is conducted. Therefore, a comprehensive understanding of the dynamics of soil water content is essential to prevent the potential failure of artificially established forests due to inadequate soil moisture in their later stages.

Mathematical modeling is extensively used for ecohydrological processes because it facilitates data acquisition. However, modeling of soil moisture and heat remains challenging in dry ecosystems. In this study, we examined the performance of four models in simulating hydrological processes in a semi-arid mountain grassland (SMG), and in shrubland forming a transitional zone between the desert and an oasis (desert-oasis ecotone; DOE) in northwestern China. We used precipitation, air temperature, humidity, atmospheric pressure, and other meteorological variables to estimate moisture and temperature at different soil depths. Four methods were used to test model performance, including partial least squares (PLS) regression, stepwise multiple linear regression (SMR), back-propagation artificial neural network (BPANN), and neural network time series. Our results showed that BPANN had the best prediction accuracy and supplied a robust modeling framework capable of capturing nonlinear environmental processes by improving the stability of the weight-learning process. Soil depth in SMG for which model performance was optimized was 20cm for PLS and SMR. Additionally, artificial neural networks (ANNs) have a remarkable applicability compared to other algorithms for increased accuracy in time-series predictions; however, they could not depict soil moisture or temperature dynamics at 160cm depth in SMG, and at 10cm depth in DOE. Using conventional meteorological data as primary predictors, and avoiding the complexity of distributed hydrological models can be helpful in developing a regional capacity for soil moisture and heat forecasting.

The spatial pattern of soil water content (SWC) determines the success of vegetation restoration in semi- and arid regions. The Qilian Mountains in China are in a semi-arid area where decades of environmental degradation prompted large-scale restoration efforts with a native constructive species Picea crassifolia. However, the relationships between SWC and landcover are not clear in this area, hindering woodland restoration. In this study, we determined spatial distribution and characteristics of SWC in the Qilian Mountains in four main types of landcover. Our results revealed that 1. SWC decreased in the order of natural forest, brushland, grassland, planted forest, and mixed forest, but it was significantly different only in natural forest (p<0.05); 2. planted forest exhibited a lower SWC at young ages (less than 50 a) than natural forest, brushland or grassland, and the ecohydrological effect of afforestation exhibited hysteresis at long time scales; 3. a variogram analysis and Kriging interception of the spatial pattern of SWC showed that shrubland and grassland exhibited superior adaptability to local SWC compared with afforestation. The differences in SWC among various landcover types indicated that strengthening the maintenance and protection of natural forests is more important for restoration efforts than afforestation; furthermore, in semi-arid regions, shrubs and grasses are more suitable for use in ecological restoration than forest plantations.