There’s a lot of power in a rough estimate. If you’re trying to figure out whether something is worth doing, you could really go deep into the weeds trying to capture all the costs and benefits. But here’s the thing: Usually you don’t need an *exact* answer in order to make the right decision.

For example, say you’re having a big party, a hundred people, and you want to make special decorated cupcakes. You do some quick, cocktail-napkin calculations and you guesstimate it would take 10 hours. Well, in reality it might be 9 hours, or it might be 11 hours—it doesn’t matter! It’s way too long.

That’s what I did in a recent post. I had wondered, if everyone on earth planted a tree—all 7.5 billion of us—how much carbon dioxide would it pull out of the air? My answer, based on some very rough assumptions, was that it could cut atmospheric CO_{2} levels by around 6 percent.

Some people disagreed with my result, and that’s fine. I’m sure it’s wrong. (For one thing, I modeled the carbon content of trees without branches.) But as a first cut, it tells us what we want to know: Planting trees can make a difference, even if it’s not in itself a solution to climate change. Besides, trees are beautiful.

Of course 7.5 billion was a fanciful number. And planting that many trees has its own complications, such as where the land is and how it was used before. But what about 20 million? Yes, we can do that, and for carbon reduction, every contribution matters. That’s the aim of the #TeamTrees project—to plant 20 million new trees. And you can help.

What’s that? You live in Manhattan? Or the Mojave? No sweat: The Arbor Day Foundation will plant a tree for every dollar you pledge—which, if you do some cupcake-type math, is a lot less than it would cost you in time and money. So the goal is to raise $20 million for reforestation by January 1, 2020. Go #TeamTrees!

### How Much Land Would It Require?

Just for fun, how about a new estimation problem. How much land do you think it’ll take to plant 20 million new trees? What’s your intuition? Are we talking about a forest the size of Rhode Island? Even bigger?

I’m going to show how I’d come up with a ballpark estimate. And if you don’t agree with my answer, you can change the assumptions and make your own!

Step one is to estimate the area that one tree would require, then we’ll scale it up. Obviously the species matters, but I’m picturing a pine tree. You see these pretty close together in forests, with a separation of maybe just 3 meters. So from a bird’s-eye view, each tree is a circle with a radius of 1.5 meters.

I’ll assume the trees are in a square grid layout. Actually, there’s a whole field of math called “packing problems,” and if you worked it out, you’d find there is a better way to pack circles together to minimize the total space. But remember, this is an approximation. 80/20 rule. We’re looking for 80 percent of the truth with 20 percent of the effort.