Premium
Simulating phenological characteristics of two dominant grass species in a semi‐arid steppe ecosystem
Author(s) -
Yuan Wenping,
Zhou Guangsheng,
Wang Yuhui,
Han Xi,
Wang Yingshun
Publication year - 2007
Publication title -
ecological research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.628
H-Index - 68
eISSN - 1440-1703
pISSN - 0912-3814
DOI - 10.1007/s11284-006-0318-z
Subject(s) - phenology , steppe , environmental science , grassland , growing season , arid , ecosystem , leymus , grazing , vegetation (pathology) , atmospheric sciences , agronomy , ecology , biology , medicine , pathology , geology
Abstract Vegetation phenology has a strong effect on terrestrial carbon cycles, local weather, and global radiation partitioning between sensible and latent heat fluxes. Based on phenological data that were collected from a typical steppe ecosystem at Xilingol Grazing and Meteorological Station from 1985 to 2003, we studied the phenological characteristics of Leymus chinensis and Stipa krylovii . We found that the dates for budburst of L. chinensis and S. krylovii were delayed with increasing temperature during winter and spring seasons; these results differed from existing research in which earlier spring events were attributed to the changes in increasing air temperature in winter and spring. The results also suggested that water availability was an important controlling factor for phenology in addition to temperature in grassland plants. The classical cumulative temperature model simulated the phenology well in wet years, but not in the beginning of growing season in all years from 1985 to 2003. The disparity between the simulation and the observation appeared to be related to soil water. Based on our research findings, a water‐heat‐based phenological model was developed for simulating the beginning of growing season for these two grass species. The simulated results of the new model showed a significant correlation with the observation of beginning date of the growing season, and both mean values of the absolute error were less than 6 days.