Determining the stand age for net carbon source to sink transition
Overview
The Gifford Pinchot National Forest is a largely fragmented landscape, made up of different aged stands created over centuries of silviculture. The practice of silviculture is common in the western Cascade Mountains because Douglas-fir is a valuable timber tree. I am measuring carbon fluxes across this heterogeneous landscape in the youngest age class — early seral stands. I will use the Biome-BGC model to gap-fill missing data periods (e.g., winter months).
Eddy covariance fluxes over young stands are relatively easy to measure because (1) the EC tower does not need to be very tall and is thus, inexpensive and (2) open-path IRGAs are advantageous over closed-path systems over short canopies (there is less high frequency loss with the open-path IRGA versus the closed-path) and can be run off of solar-power charged, 12 V batteries.
Another question I am addressing with this study is at which age do early seral stands return to net carbon sink status after being initial net carbon sources. To answer this I am using a new version of the BIOME-BGC model (provided by the University of Montana, Missoula) which allows for multiple disturbance regimes (clear-cut) and estimates of residual carbon left on site after each clear-cut harvest.
The Model Simulation
Biome-BGC is run using site-specific parameters whenever possible (e.g., canopy LAI, C:N ratio in the needles, soil particle size, soil water holding capacity, etc.), collected from the early seral stands. Biome-BGC is also run using local meteorological data.
Image citation: http://www.ntsg.umt.edu/models/bgc/
Numerical Terradynamic Simulation Group, Univ. of Montana
Actual early seral stand disturbance history at Wind River: initial forest (mature/old-growth?), catastrophic fire around 1500, forest develops to old-growth conditions (~ 400 years old), clear-cut harvest in 1920, forest grows to 80-years old, a second clear-cut disturbance in the early 1990's, early seral stands are currently about 10 years old.
How I will run the Biome-BGC disturbance regime for the early seral stands: (1) “spin-up” the model for 6000 years to reach steady-state conditions in the soil carbon pools, (2) run a fire disturbance, (3) simulate forest growth for 400 years, (4) run a clear-cut disturbance, (5) simulate forest growth for 80 years, (6) run a clear-cut disturbance and prescribe how much biomass or CWD remains on site, (7) simulate forest growth for 10 years.
To represent actual site conditions, early Seral South will have essentially no CWD left on site after step (6) in the model while I will keep a large proportion of the clear-cut CWD on site at Early Seral North.