TY - JOUR
T1 - Forest fluxes and mortality response to drought: model description (ORCHIDEE-CAN-NHA r7236) and evaluation at the Caxiuanã drought experiment
AU - Yao, Yitong
AU - Joetzjer, Emilie
AU - Ciais, Philippe
AU - Viovy, Nicolas
AU - Cresto Aleina, Fabio
AU - Chave, Jerome
AU - Sack, Lawren
AU - Bartlett, Megan
AU - Meir, Patrick
AU - Fisher, Rosie
AU - Luyssaert, Sebastiaan
N1 - This research has been supported by the Centre National de la Recherche Scientifique, Institut écologie et environnement (grant no. 16-CONV-0003 and “Make Our Planet Great Again Scholarship” grant).
PY - 2022/10/24
Y1 - 2022/10/24
N2 - Extreme drought events in Amazon forests are expected to become more frequent and more intense with climate change, threatening ecosystem function and carbon balance. Yet large uncertainties exist on the resilience of this ecosystem to drought. A better quantification of tree hydraulics and mortality processes is needed to anticipate future drought effects on Amazon forests. Most state-of-the-art dynamic global vegetation models are relatively poor in their mechanistic description of these complex processes. Here, we implement a mechanistic plant hydraulic module within the ORCHIDEE-CAN-NHA r7236 land surface model to simulate the percentage loss of conductance (PLC) and changes in water storage among organs via a representation of the water potentials and vertical water flows along the continuum from soil to roots, stems and leaves. The model was evaluated against observed seasonal variability in stand-scale sap flow, soil moisture and productivity under both control and drought setups at the Caxiuanã throughfall exclusion field experiment in eastern Amazonia between 2001 and 2008. A relationship between PLC and tree mortality is built in the model from two empirical parameters, the cumulated duration of drought exposure that triggers mortality, and the mortality fraction in each day exceeding the exposure. Our model captures the large biomass drop in the year 2005 observed 4 years after throughfall reduction, and produces comparable annual tree mortality rates with observation over the study period. Our hydraulic architecture module provides promising avenues for future research in assimilating experimental data to parameterize mortality due to drought-induced xylem dysfunction. We also highlight that species-based (isohydric or anisohydric) hydraulic traits should be further tested to generalize the model performance in predicting the drought risks.
AB - Extreme drought events in Amazon forests are expected to become more frequent and more intense with climate change, threatening ecosystem function and carbon balance. Yet large uncertainties exist on the resilience of this ecosystem to drought. A better quantification of tree hydraulics and mortality processes is needed to anticipate future drought effects on Amazon forests. Most state-of-the-art dynamic global vegetation models are relatively poor in their mechanistic description of these complex processes. Here, we implement a mechanistic plant hydraulic module within the ORCHIDEE-CAN-NHA r7236 land surface model to simulate the percentage loss of conductance (PLC) and changes in water storage among organs via a representation of the water potentials and vertical water flows along the continuum from soil to roots, stems and leaves. The model was evaluated against observed seasonal variability in stand-scale sap flow, soil moisture and productivity under both control and drought setups at the Caxiuanã throughfall exclusion field experiment in eastern Amazonia between 2001 and 2008. A relationship between PLC and tree mortality is built in the model from two empirical parameters, the cumulated duration of drought exposure that triggers mortality, and the mortality fraction in each day exceeding the exposure. Our model captures the large biomass drop in the year 2005 observed 4 years after throughfall reduction, and produces comparable annual tree mortality rates with observation over the study period. Our hydraulic architecture module provides promising avenues for future research in assimilating experimental data to parameterize mortality due to drought-induced xylem dysfunction. We also highlight that species-based (isohydric or anisohydric) hydraulic traits should be further tested to generalize the model performance in predicting the drought risks.
U2 - 10.5194/gmd-15-7809-2022
DO - 10.5194/gmd-15-7809-2022
M3 - Article
SN - 1991-959X
VL - 15
SP - 7809
EP - 7833
JO - Geoscientific Model Development
JF - Geoscientific Model Development
IS - 20
ER -