Forestry in the state of Idaho has increased dramatically over the last decade. This extensive increase in "Forest Management", as it is called, can be attributed to two causes. First consumer demand has increased. Second, controversy over forestry practices in the pacific north west has driven cutting inland. A Geographic Information System presents itself as a powerfully efficient tool for the analysis of forest lands, on both the large and small scales.
The forests of Idaho are a massive natural resource. Much of the forest in this state is managed by the United States Department of Agricultures’(USDA) Forest Service. "Approximately two-thirds of the forest lands in Idaho are federal lands, almost all in the National Forest System", said Jay O'Laughlin of the University of Idaho. The National Forest Service (NFS) works along with private corporations to make use of the natural resource that is entrusted to them. My project hopes to incorporate some of the most obvious variables related to the analysis of a forest in a GIS.
Idaho is a candidate for a relatively trouble free GIS forest analysis. The NFS and Bureau of Public Lands (BPL), and private timber companies hold a large portion of Idaho’s land mass, therefore, much of the state is potentially available for forest management. Population centers in the state (Boise and Pocatello; Coeur d’ Alene being the exception) are generally located in the southern part of the state where forests are less dense. These factors all contribute to the notion, or assumption, that a large portion of Idaho is potentially "manageable".
There are many different ways to "harvest" a forest. The two most widely known are clear cutting and selective cutting. Selective cutting, in general, involves removing specific trees from a plot of land while leaving a number of trees behind. Clear cutting removes all or nearly all of the trees from a plot of land. This method, in many cases, is much more economically feasible and hence the primary method of choice.
State and national laws place restrictions on forest practices. These limitations attempt to protect both the human and ecological systems around the cut forest.
As a hypothetical example let’s say for a minute that no restrictions were placed on forest practices. In the shorthanded point of view it would be most economical for a timber company to remove all of the trees from a large continuous area. This certain area would in most cases be adjoining former cutting grounds. The simple rule of economies of scale could lead one to this conclusion. Other reasons include: Forest equipment would only have to be mobilized once. Only one or two substantial roads would need to be constructed and when the area was exhausted the roads would not need to be maintained.
Well, this might not sound like a bad idea, but the other side must be looked at. After our hypothetical region has been exhausted of its timber very few trees will be left. Basic hydrology and soil conservation information reveals that tree removal predisposes the soil to significant erosion. In addition, rainfall formerly held and transpired by the trees, now has no choice but to flow into streams and then rivers. In fact if the area is large enough the extra water that enters rivers can lead to sever flooding, like northern Idaho has been experiencing over the past few years. Potentially the largest problem revolves around the now barren forest. Where will new trees come from? All the mature, seed bearing trees are gone. The forest will either need to be replanted or regeneration will occur naturally very slowly. Other problems include stream/river siltation, or increased turbidity, increasing water temperatures, and loss of animal habitat.
Now that the hypothetical example has been examined, one can see that certain restrictions are needed to protect both the human and forest ecosystems. Restrictions involving roads and road placement, clear-cut size, forestable slopes, forestry around streams, around rivers and roads and many others have all been implemented.
A summary of Idaho’s Best Management Practices (BMP’s), directly pertaining to this forestry analysis follow:
1. Slopes exceeding 45% grade can only be forested by aerial methods - e.g., Helicopter
2. Class I streams (fish bearing streams) are required to have a buffer zone of 75’.
3. Class II streams (small headwater streams) are required to have a buffer zone of 30’.
Although these are actual laws in Idaho other laws or rules can preempt them. In some cases 75’ of buffer zone is not enough, in others it is more than enough. Correspondence with Ms. Sharon Gray of the University of Idaho clarified some of the current laws. She mentioned that trees can actually be cut from the 75’ buffer zones but a shade requirement for the stream needs to be met. The "requirements [are] for 75% of current shade" over the stream and "a minimum number of...trees" need to be left dependent on stream width and tree diameters found within the 75’ buffer zone (Idaho BMP’s, 1996 and Idaho Forest Practices Act, 1996).
Some National Forests have different BMP’s. Ms. Gray went on to say that "buffer strips for these lands [NFS lands] are currently regulated by interim agency guidelines called PACFISH and INFISH, calling for 300-foot no-cut buffer zones on each side of fish-bearing streams and 150-foot buffers on other streams. Site-specific buffers and timber harvesting may be implemented after land managers perform [a] "watershed analysis", a process in its infancy."
In addition to the BMP’s, forest management zones commonly incorporate a buffer zone along major roads to preserve aesthetic beauty (Maine BMP’s, 1992).
Because growing trees can be quantified in a aerial manner the decision was made to use the Albers Equal Area projection for this analysis. While distorting shape, Albers is known to give accurate results for area based calculations. All analysis was done with data projected to the following parameters:
Projection ALBERS
Datum NAD83
Units METERS
Spheroid GRS1980
1st standard parallel 29 30 0.000
2nd standard parallel 45 30 0.000
central meridian -96 0 0.000
latitude of projection's origin 23 0 0.000
false easting (meters) 0.00000
false northing (meters) 0.00000
The only exception to the above parameters was a 1:250,000 Digital Elevation Map which was in the WGS84 datum. Correspondence with Mr. David Maidment of the University of Texas concluded that the two different datum’s were similar enough to ignore the difference.
This study will incorporate many different data sources. Beginning with a forest density map for Idaho, one can start the process of determining which lands are legally available for management. Roads and hydrology data are removed along with their corresponding buffer zones from this density map. From an elevation map one can generate slope data, and from this slope map, exclusion slopes are removed from the same forest density map. Other large land areas that are not available for management include National and State Park lands, wilderness areas, and private land holdings. These are also removed from the forest density map. When the final forest density map is compiled, it can then be converted to a forest value map, which is composed of a forest speies map superimposed with values of each of those corresponding forest species.
Obtaining the 1:2,000,000 hydrology and road data for Idaho was a start. From the graphics below one can see the extent of buffering that will be needed for the entire state. Although the road map has the primary roads it is missing a significant number of smaller yet important roads. The hydrology seems to portray the bulk of the actual data. Notice the area with no streams in the southeastern portion of the state. This area is arid and it’s geology is characteristically cavernous where streams and rivers actually disappear into the subsurface. Craters of the Moon National Park is located in the northern portion of this area.
Idaho’s 1:2,000,000 Digital Elevation Model (DEM), below, gives elevations in meters above mean sea level. From this elevation model the slope map, (below right) was generated (see arc commands). The steepest slopes approach 40% grade in this generated slope map. In general this slope map approximates Idaho’s slopes; inaccuracies however, do exist. The bulk of the incorrect slopes occur in the mountainous regions. This presents a problem. Forest harvesting happens where the forest is thickest, in the mountains. The erroneous slopes can be attributed to the coarseness of the DEM grid. The DEM takes an average value over a certain area (1:2,000,000 = 1 km^2). This averaging process in effect smoothes out the topography, and therefore decreases the accuracy of the slope calculation. The two maps below were obtained from the NFS (see data dictionary or summary of work). Both maps use grids of 1000m cells. The forest grid has one variable, percent coverage. Zero percent coverage means no vegetative cover exists in that area. One hundred percent coverage means trees screen the entire land surface of that cell. The forest type grid gives the predominant specie of that cell. There are ten different forest types in this grid. Notice that most vegetation grows in the northern portion of the state, and that in general as elevation increases so does vegetation. This forest type value map below represents the value in dollars of the predominant specie in that cell per square kilometer. For example if a square kilometer of land was totally forested with mature ponderosa pine and that certain volume of wood is worth 200 million dollars per square kilometer at market it would have a certain color assigned to it. Assigning this value to all the ponderosa pine cells yields a map that is in essence a "predominant" value map, meaning that wood from that area predominantly sells for that price. This type of map has error, if the particular cell is assigned as a ponderosa pine cell that dosent mean that it is 100% ponderosa pine only that it is predominatly ponderosa pine. This value assigning was then done for all forest types which yielded the map below. Source - National Forest Products Association (kinda but not really - I mostly put it together on what I thought was about right - I got the NFPA data two days after finishing). The map below portrays the relative value of a forests’ wood at market. The map was constructed by taking the forest density map (above) and multiplying it by a normalized version of the forest type value map (also above). The higher the value, the more the trees in that cell will be worth at market, the units are in dollars per square kilometer.
In the analysis above it would have been more useful to show the actual value of the timber per land area. Unfortunately, timber companies like Boise Cascade, and Weyerhauser hold these values as proprietary information and it is not available to the lay citizen (Craig Campbell, April 1997 - and by default, all the others who did not answer my e-mail). Bill Willams of the U.S. Bureau of Land Management (BLM) stated that values of standing trees are not normally given but when they are "the value..[of]...standing trees depends on the value at the mill minus all costs to get the wood to the mill. The process of determining the value "in the woods" is called an appraisal (just like any valuation process) and is a science and an art specialty within forestry and the wood products industry."
(c) Copyright Mike McAdams, 1997
if you want something e-mail me at: