| Water is a key factor of limiting vegetation restoration and rehabilitation on the Loess Plateau, and precipitation variation plays an important role in shaping ecosystem structure and function of this region. In this paper, we selected Salix psammophila and Caragana korshinskii growing in water/wind erosion crisscross region of northern Shaanxi as the objects, and explored the effects of manipulated precipitation variation(control treatment: normal precipitation; water addition treatment: water was added by 50% or so, drought treatment: water was removed by 50% or so) on shrubland water balance, plant growth, leaf water-photosynthesis and sap flow transpiration characteristics. The major objective were to reveal the physio-ecological mechanisms of the two shrubs adapting to the dry sandy environment, and provide better understanding for the distribution and management of these two shrubs in the future and the evaluation of the sustainability of ecosystems. The main results were as the follows:(1) Soil moisture consumption, water supply depth and evapotranspiration were different for different treatments in two shrubs. For S. psammophila, the variation of soil water content during June to September was 37.79 mm for the control, 47.81 mm for the water addition treatment and 6.17 mm for the drought treatment, the corresponding soil water supply depth at the end of the growing season was 2.2m, 1.8m and 1.8m, respectively. For C. korshinskii, the variation of soil water content during June to September was-39.80 mm for the control, 38.66 mm for the water addition treatment and-60.05 mm for the drought treatment, the corresponding soil water supply depth at the end of the growing season was >1m, 2.2m and 0.5m, respectively. Evapotranspiration of S.psammophila and C.korshinskii shrublands under three different treatments followed the order: water-added treatment > control treatment > drought treatment. Evapotranspiration accounted for a higher proportion of water input under drought treatment, but the proportion was relatively lower for water addition treatment.(2) S. psammophila and C. korshinskii had different responses to manipulated precipitation variation in leaf water potential, photosynthesis and growth. Leaf water potential for different treatments of S. psammophila was almost the same, but it showed significant differences in C. korshinskii. Net photosynthetic rate, transpiration rate and water use efficiency of S. psammophila showed significant responses to precipitation variation, while net photosynthetic rate and transpiration rate of C.korshinskii responded obviously to precipitation variation. These photosynthetic parameters responded earlier in S. psammophila than in C. korshinskii. Net photosynthetic rate of S. psammophila increased with water addition, while transpiration rate and water use efficiency were greater under drought treatment. Net photosynthetic rate and transpiration rate of C. korshinskii were greater under drought treatment. The growth of the branches within 10~25 mm diameter classes for S.psammophila and within 5~10 mm for C.korshinskii were increased by water supply, while growth of branches within 5~10 mm diameter classes for S.psammophila and branches less than 20 mm diameter class for C.korshinskii were decreased by drought.(3) Sap flow characteristics of C. korshinskii were affected by manipulated precipitation variation. The daily change of sap flow rate was almost a bimodal curve at the end of July, but it had become an unimodal curve in early October. The main factors that affected the sap flow rate were solar radiation and vapor pressure deficit. The relative sap rate increased with solar radiation and the trends could be described by S-type function. The relative sap flow rate of the control and water addition treatment showed an S-type relationship with vapor pressure deficit or integrated variant(VT). The relative sap flow rate of drought treatment increased with the increase of vapor pressure deficit or integrated variable variant firstly, but then the relative sap flow rate decreased significantly when VPD> 2.04 k Pa or when VT> 60.38 k Pa·(w·m-2) 0.5. VPD or VT of drought treatment corresponding to peak relative sap flow rate was larger than the control and water addition treatment. Single branch water consumption under different treatments was: water addition treatment > control > drought treatment, but the difference did not reach significant level. |
