We all know trees are essential to our survival.

Forest doing a great job. (photo credit:

  • Trees produce oxygen.  The big oak in my backyard is producing enough oxygen for 10 people to breath.
  • Trees clean the soil.  Phytoremediation –That’s a mouthful of sylables we depend on.
  • Trees are carbon sinks, locking up carbon and keeping it from acting as a greenhouse gas.
  • Trees fight soil erosion and  slow storm water runoff.  One big spruce can intercept more than 1,000 gallons of water each year, helping to recharge underground aquifers.

Here’s the scary part:

Studies by the United Nations Food and Agriculture Organization show that each year the world loses 35,000 square miles of forest. That’s enough to cover the state of Indiana.

Slash and burn in the Amazon.

We need to track the state of our global forests more accurately, and now we can.  Mutlu Ozdogan, an assistant professor of forest and wildlife ecology at the University of Wisconsin-Madison is automatically generating maps showing where trees have been harvested in the form of clear-cut areas over five-year intervals.

Tracking what’s happening on the ground in terms of the location, shape, and size of the harvested plots has never been easy, but Mutlu is using NASA’s Landstat satellite images, which views Earth in several different wavelengths of light.

To power photosynthesis, tree leaves absorb most of the sun’s visible red light, so removing trees greatly increases the visible red light reflection.  Although areas with diseased, wind-thrown or burned trees can also increase reflectance, these phenomena can usually be distinguished from tree removal, Ozdogan says. “Most harvesting is bounded by straight lines, while fires, wind throws and disease outbreaks have irregular shapes, so most of the time, they are easily distinguished from harvests.”

Once the technique is perfected, Ozdogan plans to extend it to the entire North American forests. “We have the computer resources to do this and once the algorithm is stabilized, we will move in that direction, mapping harvested areas, wall-to-wall for the entire country every few years, using free and open source tools.”

Ozdogan also envisions building a web platform at UW-Madison “where scientists, non-governmental organizations, policy makers and anyone else interested in this kind of data would be able to come and do the analysis themselves. They will be able to interpret results and draw their own conclusions from data that are free of subjectivity.”

Mutlu Ozdogan, assistant professor of forest and wildlife ecology at the University of Wisconsin–Madison, processes images from the Landsat satellite to reveal changes in forest composition over time. Pictured in this composite is the dry pine-barren forest of Bayfield and Douglas counties in northwestern Wisconsin in 1989 (left) and 1995 (center). Vegetation appears green in these true-color images. A computer algorithm compared the 1989 and 1995 images and identified yellow areas in the right panel as having been harvested in the interim. (Photo credit: Mutlu Ozdogan)

Ozdogan adds that the computer-generated maps can be merged with other datasets related to water conservation, insect and disease outbreaks, fire, weather and climate.

This is the kind of data we really need to make good decisions.

1 reply

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s