Premium
Measuring and modeling the impact of a severe drought on terrestrial ecosystem CO 2 and water fluxes in a subtropical forest
Author(s) -
Xie Zhenghui,
Wang Linying,
Jia Binghao,
Yuan Xing
Publication year - 2016
Publication title -
journal of geophysical research: biogeosciences
Language(s) - English
Resource type - Journals
eISSN - 2169-8961
pISSN - 2169-8953
DOI - 10.1002/2016jg003437
Subject(s) - environmental science , ecosystem respiration , primary production , evapotranspiration , ecosystem , eddy covariance , water content , terrestrial ecosystem , atmospheric sciences , precipitation , carbon cycle , water cycle , subtropics , carbon sequestration , hydrology (agriculture) , ecology , carbon dioxide , geography , biology , geology , geotechnical engineering , meteorology , engineering
A severe drought occurred in central and southern China in the summer of 2013. The precipitation dropped to less than 25% of the long‐term average, and temperatures were abnormally high for more than 2 months with return periods of 125 years and 301 years, respectively, which induced significant changes in the terrestrial ecohydrological cycle. In this study, the impact of the severe drought on a subtropical forest ecosystem was investigated using measurements from a newly established flux tower and simulations performed by the Community Land Model version 4.5 (CLM4.5). Based on in situ observations, we found that both gross primary production (GPP) and evapotranspiration experienced strong reductions of 76% and 40%, respectively, during the prolonged dry and hot spell. There was an exponential relationship between ecosystem respiration ( R eco ) and temperatures when soil moisture was not too dry, but R eco decreased along with GPP when the temperature exceeded 32°C and soil moisture was below 0.14 m 3 m −3 . The ecosystem even switched to a net source of carbon in late August. The model captured the variations in water vapor fluxes well, with correlation coefficients r > 0.88, but overestimated net ecosystem CO 2 exchange because it did not adequately represent carbon fluxes responses to water stress and failed to capture the nonlinear relationship between GPP and R eco during the drought period. The long‐term simulation suggested that water availability severely limited carbon sequestration, and both the underlying water use efficiency and inherent water use efficiency reached their maximum values.