Using a land surface model to simulate Net Primary Productivity in China comparing with the process model derived by remote sensing

Abstract

In this study, we analyze Net Primary Productivity from 51-year (1950-2000) offline simulations using an Australian land-surface model CABLE (CSIRO Atmosphere Biosphere Land Exchange). The model annual NPP climatology with the estimated 2001 annual NPP from what was derived from a process model driven by remote sensing in deriving China NPP at 1-km resolution. The CABLE results are highly comparable with those, with annual NPP decreasing from southeast to northwest. Southern China has high annual NPP of more than 1,200 gC m <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-2</sup> year <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-1</sup> , while the Tibetan Plateau, north and northwest regions show very low annual NPP. CABLE is also skillful in reproducing NPP interannual variability in northeast China, with similar year-to-year fluctuations for the period of 1982 to 2000. But the model shows a different trend compared with the satellite-based approach. Further analysis shows that using a prefixed CO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> concentration in the model experiment appears to be the chief reason for it failing to simulate an upward NPP trend.