TY - JOUR
T1 - A steady-state approximation approach to simulate seasonal leaf dynamics of deciduous broadleaf forests via climate variables
AU - Xin, Qinchuan
AU - Dai, Yongjiu
AU - Li, Xia
AU - Liu, Xiaoping
AU - Gong, Peng
AU - Richardson, Andrew D.
N1 - Funding Information:
We thank the researchers and investigators that are involved in the data collection and analysis at the AmeriFlux sites. This research is supported by National Key R&D Program of China (grant no. 2017YFA0604302 and 2017YFA0604402 ) and Key Projects for Young Teachers at Sun Yat-sen University (grant no. 17lgzd02 ). Research at the Bartlett Experimental Forest is supported by the USDA Forest Service’s Northern Research Station , the National Science Foundation ( DEB-1114804 ), and the Northeastern States Research Cooperative . ADR acknowledges additional support from the National Science Foundation’s Macrosystems Biology (awards EF-1065029 and EF-1702697 ). We also thank anonymous reviewers for their constructive comments.
Publisher Copyright:
© 2017 Elsevier B.V.
PY - 2018/2/15
Y1 - 2018/2/15
N2 - As leaves are the basic elements of plants that conduct photosynthesis and transpiration, vegetation leaf dynamics controls canopy physical and biogeochemical processes and hence largely influences the interactive exchanges of energy and materials between the land surface and the atmosphere. Given that the processes of plant leaf allocation is highly sensitive to climatological and environmental conditions, developing robust models that simulate leaf dynamics via climate variables contributes a key component to land surface models and coupled land-atmosphere models. Here we propose a new method to simulate seasonal leaf dynamics based on the idea of applying vegetation productivity as a synthesized metric to track and assess the climate suitability to plant growth over time. The method first solves two closed simultaneous equations of leaf phenology and canopy photosynthesis as modeled using the Growing Production-Day model iteratively for deriving the time series of steady-state leaf area index (LAI), and then applies the method of simple moving average to account for the time lagging of leaf allocation behind steady-state LAI. The seasonal LAI simulated using the developed method agree with field measurements from a selection of AmeriFlux sites as indicated by high coefficient of determination (R2 = 0.801) and low root mean square error (RMSE = 0.924 m2/m2) and with satellite-derived data (R2 = 0.929 and RMSE = 0.650 m2/m2) for the studied flux tower sites. Moreover, the proposed method is able to simulate seasonal LAI of deciduous broadleaf forests that match with satellite-derived LAI time series across the entire eastern United States. Comparative modeling studies suggest that the proposed method produces more accurate results than the method based on Growing Season Index in terms of correlation coefficients and error metrics. The developed method provides a complete solution to modeling seasonal leaf dynamics as well as canopy productivity solely using climate variables.
AB - As leaves are the basic elements of plants that conduct photosynthesis and transpiration, vegetation leaf dynamics controls canopy physical and biogeochemical processes and hence largely influences the interactive exchanges of energy and materials between the land surface and the atmosphere. Given that the processes of plant leaf allocation is highly sensitive to climatological and environmental conditions, developing robust models that simulate leaf dynamics via climate variables contributes a key component to land surface models and coupled land-atmosphere models. Here we propose a new method to simulate seasonal leaf dynamics based on the idea of applying vegetation productivity as a synthesized metric to track and assess the climate suitability to plant growth over time. The method first solves two closed simultaneous equations of leaf phenology and canopy photosynthesis as modeled using the Growing Production-Day model iteratively for deriving the time series of steady-state leaf area index (LAI), and then applies the method of simple moving average to account for the time lagging of leaf allocation behind steady-state LAI. The seasonal LAI simulated using the developed method agree with field measurements from a selection of AmeriFlux sites as indicated by high coefficient of determination (R2 = 0.801) and low root mean square error (RMSE = 0.924 m2/m2) and with satellite-derived data (R2 = 0.929 and RMSE = 0.650 m2/m2) for the studied flux tower sites. Moreover, the proposed method is able to simulate seasonal LAI of deciduous broadleaf forests that match with satellite-derived LAI time series across the entire eastern United States. Comparative modeling studies suggest that the proposed method produces more accurate results than the method based on Growing Season Index in terms of correlation coefficients and error metrics. The developed method provides a complete solution to modeling seasonal leaf dynamics as well as canopy productivity solely using climate variables.
KW - Climate change
KW - Ecosystem
KW - Growing production-day
KW - Leaf area index
KW - Phenology
UR - http://www.scopus.com/inward/record.url?scp=85034867295&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85034867295&partnerID=8YFLogxK
U2 - 10.1016/j.agrformet.2017.11.025
DO - 10.1016/j.agrformet.2017.11.025
M3 - Article
AN - SCOPUS:85034867295
SN - 0168-1923
VL - 249
SP - 44
EP - 56
JO - Agricultural and Forest Meteorology
JF - Agricultural and Forest Meteorology
ER -