Now I'm Thoroughly Confused

Ever since my father explained that trees, or at least older trees, don't do much of the CO2 scrubbing that we imagine they do (saplings are a lot more helpful than 200-year-old behemoths, for example, and the grass under a big old tree generally does more for the "environment" than the tree itself does), I've been alarmed by how little the average person truly knows about any of this environmental stuff. It's not obvious to me that many people understand even a small slice of the Global Warming puzzle, and who really knows if trees help at all. No, seriously:

Before compact fluorescent light bulbs and ethanol, the first line of defense against global warming was planting trees.

Forests, after all, cool the atmosphere by drinking in carbon dioxide from the air. A new study, however, published in Proceedings of the National Academy of Sciences reports that forests' other climatic effects can cancel out their carbon cleaning advantage in some parts of the world. Using a three-dimensional climate model, the research team mimicked full global deforestation and also studied the effects of lear-cutting in different regions of latitude, such as the tropics and boreal zones. Apparently, these natural carbon sinks only do their job effectively in tropical regions; in other areas, they have either no impact or actually contribute to warming the planet. In fact, according to this model, by the year 2100, if all the forests were cut and left to rot, the annual global mean temperature would decrease by more than 0.5 degree Fahrenheit. (Emphasis mine)

This is Scientific American, mind you, not exactly a supporter of the Bush administration or big business. And if you think that's confusing, read this:

Trees perform three major climate functions: They absorb carbon, which they pull from the atmosphere, creating a cooling effect; their dark green leaves absorb light from the sun, heating Earth's surface; and they draw water from the soil, which evaporates into the atmosphere, creating low clouds that reflect the sun's hot rays (a mechanism known as evotranspiration that also leads to cooling). These three factors—the second two being largely ignored in climate models up to this point, according to Caldeira—taken together created very different results in the primary latitudes studied: the equatorial tropic zone; the midlatitudes that include most of the U.S.; and the boreal areas, which are subarctic and include much of Canada, Russia and the northern extremities of the U.S.

In all three regions, forests dutifully perform their task of sucking carbon dioxide from the air, but light absorption and evotranspiration vary wildly. In tropical zones, forests have a significant, overall cooling effect. The soil is very wet and, so, via evotranspiration, the trees are covered by low-lying clouds that create a small albedo (power of light that is reflected by a surface). In nontropical areas, Caldeira explains, "the real significant factor is whether there's snow on the ground in the winter." If a forest covers a snowy expanse, "that has a strong warming influence," he notes, because of little cloud cover resulting from less efficiency in evaporating water. The poor cloud formation coupled with the intense absorption of light by the trees "far overwhelms the cooling influence of the carbon storage," he says.