F o r e s t . H a r v e s t . P l a n , . A l a s k a
R . S c o t t . C h e r b a

The community of Galena, Alaska was considering a bio-fuel option for producing energy. Residents were concerned that harvesting their forest would result in clear-cutting. A forester was consulted to create a harvest plan to meet their needs. Visualization of the harvest plan and subsequent regrowth was used to show the visual impact of the plan.

Phase 1. View west-southwest down the Yukon River toward Galena, Alaska.

Phase 2. Harvest

Phase 3. Harvest + 10 years

Phase 4. Harvest + 30 years

Phase 5. Harvest + 60 years

Click for PDF version
Harvest areas and forest composition (TCC data)















11 20




Technical Information

We worked with several constraints in this project, notably low DEM resolution and limited land cover information. A realistic and representative scene was based on publicly available data and aerial photos taken by the illustrator.

The land surface was generated from the best DEM data publicly available (USGS 2-arc-sec NED). Alas, for Alaska this means working with a data grid 26 meters wide and 62 m tall. 10-m or better resolution is recommended for working with terrain as close as we are to this one.

The land cover data provided by TCC (see maps below) gave us a place to start. Unfortunately, the data was created with harvesting in mind and lacks the necessary detail and information typical of forestry datasets. The biggest problem was coming up with a natural and realistic density and species distribution within land cover type areas.

13 land cover types are represented within the view. Each land cover type has a species set with representative height ranges and densities. Within each land cover type area, density of deciduous and conifer species are controlled by a grayscale image derived from Landsat 7 data. The image above shows what band 4 looks like in our area of interest (Landsat from July 2000).

Band 4 is near infrared and covers the EM spectrum from 0.760 - 0.900 µm. In this range, deciduous leaves have higher values and coniferous leaves have lower values. The brightness and contrast of the original Landsat 7 image were adjusted to yield the image above with a white to black range. Where a white pixel appears on the terrain the full density value of hardwoods was rendered. Black pixels render no hardwood trees. Intermediate grayscale values render a percentage of the density value based on where the value falls in the 0 (black) to 255 (white) gray scale.

For conifers, an inverse of the deciduous density image was used to control the density of spruce. White image values render the full spruce density and black value pixels render no spruce.

This method, combined with the TCC land cover dataset below, yields a reasonable and realistic distribution of species that would not otherwise be possible without a great deal of additional work and expense.

060623a_GLC-HarvestArea.pdf (2 MB) is a prospective harvest area for visualization. Section 35 (see DRG page for topo map version) is a relatively flat area on higher ground. There's a variety of land cover types surrounding it to add visual interest.

Galena_060620-0930.pdf (6.4 MB) includes the master land cover map of Galena, a color key to the map, the original 'calls' key provided by TCC, and several detailed renders. Area A is centered on the dunes areas we explored. The additional areas were rendered for more options.

EcosystemSpecies.zip is an Excel 2000 spreadsheet with the land cover types (Ecosystems) and fields for Ecosystem density and species type, density, and height range.

Copyright 2006-2010 R Scott Cherba All Rights Reserved
Images and movies may not be copied, distributed, or transmitted without permission.
Please contact the author to purchase use.