New prognostic phenology in the Yale Interactive Terrestrial Biosphere Model: implications for plant volatile emissions and surface air pollution in the temperate zone between 1980s and 2010s

Abstract:

We describe the implementation of an improved prognostic phenology scheme in the Yale Interactive Terrestrial Biosphere Model (YIBs). The spring phenology of deciduous broadleaf forest (DBF) is predicted using the cumulative thermal summation with chilling requirements, and the autumn phenology is jointly regulated by temperature and photoperiod. The phenology of shrubland and grassland is sensitive to temperature, except for sites with climatological soil temperature > 12 °C, where plant phenology is controlled by water availability. The model represents frost hardening in evergreen needleleaf forest. An off-line version of the model is evaluated against carbon flux measurements at 145 FLUXNET sites across a wide range of ecosystem types. The simulated seasonality of the gross primary productivity (GPP) has high correlations R > 0.95 with observations at 15 out 20 deciduous broadleaf sites. The updated vegetation model has been embedded on-line inside the NASA ModelE2 global chemistry-climate model. We apply this framework to quantify the phenology changes between the 1980s and 2010s and compare those changes to satellite observations for this period. We isolate the impacts of the phenology changes on plant volatile emissions (isoprene and monoterpenes) and examine the consequences for surface ozone and PM_2.5 pollution in the United States and Europe.