Quoted from: Milner, Kelsey S., Dean W. Coble, Andrew J. McMahan, and Eric L. Smith. "FVSBGC: a hybrid of the physiological model STAND-BGC and the forest vegetation simulator." Canadian Journal of Forest Research 33, no. 3 (2003): 466-479. https://doi.org/10.1139/x02-161 

      The linkage was accomplished by making the STANDBGC model an extension to the FVS system (Fig. 2). When the BGC extension is turned on (the shaded portion in Fig. 2), the two models run in parallel. They can share state variables at FVS cycle boundaries (every 10 years by default). In the current linkage, only simple exchanges of tree dimensions and increments are made.

      STAND-BGC is invoked from within an FVS simulation via keywords entered by a user into an FVS keyword file. Information is passed from FVS to STAND-BGC at the start of every FVS cycle. This information includes program control variables (e.g., the FVS cycle number and current thinning status), shrub cover information (if the FVS COVER extension is invoked), and individual tree data such as species, tree record number, diameter at breast height, total height, crown ratio, and trees per acre.

      Once this transfer is made, STAND-BGC grows the entities for as many years as are in a particular FVS cycle. Then program control is passed back to FVS, and the trees are grown using FVS. When both models have completed an FVS cycle, a user-controlled exchange takes place. The STAND-BGC increments for height, diameter, crown ratio, and mortality may be used by FVS in place of its own predicted increments. Each model then calls its own report writers and produces its customary output. STAND-BGC is then reinitialized with the updated tree list from FVS, and the two models continue the simulation. This simple transfer allows the user to essentially replace the empirical growth engine in FVS with the STAND-BGC process engine.

      If the STAND-BGC increments are used to update the tree list in FVS, then STAND-BGC is reinitialized with exactly the tree list with which it ended the cycle. The resulting simulation is then identical to what would be produced if STAND-BGC had no connection to FVS. The FVS output reflects the growth of trees as predicted from STAND-BGC alone. In this mode, FVS provides (i) a method to input entities into STAND-BGC (trees via FVS tree lists and the Establishment model, shrubs via the COVER model); (ii) a way to simulate silvicultural treatments (thinning and planting); (iii) FVS output files reflecting how FVS summarizes this growth in terms of the stand’s merchantable volume, stand density index (SDI), crown competition factor (CCF), canopy structure, etc.; and (iv) the opportunity to simulate insect and disease effects on a stand either before or after such a stand is simulated in STAND-BGC.

      If the user does not specify use of STAND-BGC increments, then the FVS output is unaffected by STAND-BGC growth. However, STAND-BGC is still reinitialized with the FVS-grown tree list at the start of each cycle. For FVS, the resulting simulation is identical to what would be produced if FVS had no connection to STAND-BGC. For STAND-BGC, the output gives a cycle-length picture of the physiological functioning of the stand as defined by FVS, given the climate provided for that cycle.

      This linkage of the two models thus allows users familiar with FVS output to see how a climate-driven process model projects stand growth.