| Water resources shortage has become a key issue that constrains the development in economics and society in arid and semi-arid areas and an important factor that limits the forestdistribution and biodiversity in arid and semi-arid ecosystems. Beijing mountainous area is drought-prone and water-short, which are also the key ecological shelter for Beijing and an important part for the whole North China Plain. To reveal the way forests survive and thrive in Beijing mountainous area, the water use patterns in different seasons need to be investgated to enhance understanding of how seasons change has altered plant moisture use dynamics. In this study, P.orientalis, R.pseudoacacia, Q.variabilis and P.tabulaeformis were selected to build the plant-soil water movement system. By field investigation, different samples based on different temporal and spatial scales were collected for isotopic analysis. By measuring isotopic compositions of different samples, plant-soil water movement was analyzed, and based on direct inference method and multi-source mass balance method; plant water use patterns were established. The major results were as follows:(1) the LMWL was δ2H=6.42δ18O-5.91(R2=0.90, n=52), with smaller slope and intercept. The d-excess went higher in winter and became lower in summer.δ18O in rainfalls had negative relationship with temperature and precipitation and showed seasonal patterns. Values of δ2H and δ18O were decreasing as the rainfall continued. When the precipitation was12.4mm and9.0mm, the△were above and below0, respectively. In each forest ecosystem, LAI all had positive relationships with the δ18O in the throughfall;(2) soil water content (SWC) varied under different site conditions, SWC of Q.variabilis and P.tabulaeformis were generally higher than that of P.orientalis and R.pseudoacacia. In dry season, heavy rain had little impact on the deeper soil water potential but the surface soil. And the soil moisture characteristic curves of different soil depths in different forest ecosystems varied significantly;(3) the average values of δ18O and δ2H in soil water samples increased as soil depths grew and the upper soil layer could be easily affected by environmental factors. Average values of δ18O and δ2H in groundwater remained constantly;(4) different plants had distinct isotopic signatures. The average values of δ18O and δ2H in P.orienlalis, Q.variahilis and understory shrubs were-6.95‰,-6.54‰,-6.69and-67.99‰,-67.91‰,66.07‰, respectively, while in R.pseudoacucia and P.tabulaeformis were-4.97‰,-4.92‰and-71.25‰,-73.16‰, respectively. And the coefficients of variation in δ18O and δ2H of different plants were similar to each other;(5) δ2H in the soil profile was affected sigfinicantly by the rainfall in dry season, but was less affected in wet season. The anticipant precipitation should reach a threshold before it could be utilized by plants, which was at least4.6mm. When it was12.1mm, it could fully recharge the soil water and the plants;(6) approximately50%of the root of P.orientalis distributed in0-20cm,70%of the root of R.pseudoacacia distributed in0-50cm, root of<2mm of Q.variahilis mainly distributed in0-60cm and root biomass of P.tabulaeformis was generally smallest;(7) water use patterns of different forest ecosystems had significant seasonal variations, which was was more affected by the rainfall in dry season, but was less affected in wet season; and soil moisture contents of different forest ecosystems were affected by different root distribution patterns;(8) the water use of different soil depths was affected by pre-precipitation, LAI, soil water content and root biomass. |
