Forest Type Derivation

Average Forest Type Maps

Forest Type Derivation:

We have constructed a forest type map for each scenario, based on some rules (as explained below). We can use these forest type maps to get an idea as to how the forests will be potentially distributed in a 2xCO₂ climate.
Rules to classify 80 eastern species into forest types.: Importance values were summed for each species in the list, with the county being classed according to the highest ranking class. Species were assigned based on associations according to Hansen et al. (1992).

class 1   White/Red/JackPine  {eastern_white_pine, eastern hemlock}

class 2   Spruce/Fir          {balsam fir,  northern white cedar}

class 3   Longleaf/Slash_Pine {slash_pine, longleaf_pine}

class 4   Loblolly/Shortleaf  {shortleaf_pine, loblolly_pine,

                               Virginia_pine}

class 5   Oak/Pine            {eastern_white_pine, shortleaf_pine,

                               Virginia_pine, northern_red_oak,southern_red_oak,

                               loblolly_pine, water_oak, willow_oak,

                               post_oak, scarlet_oak}

class 6   Oak/Hickory         {hickory, bitternut_hickory, pignut_hickory,

                               shagbark_hickory, mockernut_hickory,

                               white_oak, scarlet_oak, chestnut_oak,

                               northern_red_oak, post_oak, black_oak,

                               sweetgum, tulip_tree}

class 7  Oak/Gum/Cypress      {swamp_red_oak, willow_oak, sweetgum,

                               American_elm, baldcypress, pond_cypress,

                               red_maple, water_tupelo, swamp_tupelo}

class 8  Elm/Ash/Cottonwood   {red_maple, American_elm, black_ash,

                               white_ash, sycamore, eastern_cottonwood,

                               willow, black_willow}

class 9  Maple/Beech/Birch   {red_maple, sugar_maple, American_beech,

                              yellow_birch, black_cherry, black_walnut}

class 10  Aspen/Birch        {quaking_aspen, bigtooth_aspen, paper_birch}

An additional rule set was needed for the oak/pine forest type, because it was a sum of many major oaks and pines, yet the class was intended to reach those with mixtures of at least 50% oak and 25-50% pine species (Merz, 1978). For this, if the above algorithm determined the class to be oak/pine, the following statements were applied:

* if loblolly/shortleaf (class 4) was greater than oak/hickory (class 6), then the county would be reclassed from oak/pine to loblolly/shortleaf, because pine component is >50%;
* if oak/hickory (class 6) was greater than twice that of loblolly/shortleaf (class 4), then the county would be reclassed from oak/pine to oak/hickory, because there was likely less than 25% pine component;
* if neither of the above apply, the county remains classed as oak/pine (class 5).

Average Forest Type Maps:

This is the average forest type map by averaging the IVs for each species for all 5 GCM scenarios and applying the rules above to get the forest type map.

Another map of interest would be the Forest type agreement - which show how many of the 5 GCMs agree the forest type to be. There are areas of major confluence as well as disparity among the models.

Literature Cited:
Hansen, H. M, T. Frieswyk, J.F. Glover and J.F. Kelly., 1992. The Eastwide Forest Inventory Data Base: Users Manual. USDA Forest Service, GTR NC-151, North Central Forest Experiment Station.

Merz, R.W., compiler. 1978. Forest Atlas of the Midwest. North Central Forest
Experiment Station, USDA Forest Service. St. Paul, MN. 48 pp.