[originally posted on www.foreststeward.com on Dec. 24, 2010]

In the spirit of the holiday season, I am veering drastically from my typical format. Instead of waxing pedantic on a recent scientific article, I am waxing pedantic on Christmas trees. My Christmas gift is to share some data I happen to have that will provide a bit of scientific reasoning behind selecting your Christmas tree. The data I have pertain to species you would find in the California Sierra Nevada, but the principles for selecting a tree could be applied anywhere. Also note that I am talking about trees that have grown out in the forest. That is, trees the way they were meant to be. They are not Franken-trees growing at frighteningly unnatural speeds with perfect symmetry on farmland in Oregon. I have no scientific advice for choosing those trees since they all look the same.

Many factors go into selecting the right tree (e.g. traditions, smell, some affinity for Charlie Brown). But as a scientist, I am of course only interested in those factors that can be measured. Below I compare how some species differ according several measures of crown morphology. I integrate these measurements to conclude what the best species is, but your conclusion might be different depending on which measure of crown morphology you value most.

Crown density

Do you like a tree packed with foliage, or one with lots of gaps to emphasize the ornaments? Then you are concerned with crown density. It just so happens that I have measured crown density for several Christmas tree-sized trees in an open grown environment. I used a light sensor placed at the bottom of tree crowns in order to measure how much light each crown was blocking and then converted this to the number of leaves or branches present along a linear distance within the crown (i.e. it’s a measure of how dense the crown is). The results are below (means and standard errors of means).

As you can see, Douglas-fir is the clear winner here. White fir and incense cedar (which I believe is the next big thing when it comes to Christmas tree species) also have relatively dense crowns. Sugar pine is nice in that it has a high crown density, yet it also has clear horizontal layers space within the crown for ornaments. I must admit that this graph misleading for giant sequoia. Giant sequoia crowns look very dense because all of its foliage tends to be packed on the tips of the branches. There aren’t many leaves near the stem of the tree, which brings down the density measurement. So it is a good choice if you like high leaf density, but it is also very prickly. I haven’t figured out a way to measure prickliness yet.

Crown symmetry

Want a tree with perfect symmetry? Does it have to form a perfect upside-down cone? The graph below is an index of asymmetry that I made from mapping crown profiles of different tree species. I then averaged the absolute differences between crown radii at several points along the length of the crown. Finally, I relativized it to average crown width. These differences are not “real” when a statistical test is done, although I can find a difference when the data are not made relative to crown size. As I expected, giant sequoia has the most symmetrical crown (sugar pine tend to be less symmetric because of branches that stick out sporadically throughout the crown). So giant sequoia is the winner when it comes to symmetry. Its crown also has a lot of taper that forms a nice inverted cone.

Branch density and size

Want lots of branches but don’t want them too big or too small? Then branch metrics is your thing. Below is a comparison of the number and size of branches for Douglas-fir, giant sequoia, and ponderosa pine (I originally collected this data for a pruning study). The graph contains “box plots,” where each box spans 50% of the data points, while the whiskers span 80% of the data. It shows you where most of the measurements were concentrated, so it is a way of displaying central tendency. It is apparent that giant sequoia tends to have lots of branches, but they are also relatively small. Ponderosa pine, on the other hand, tends to have very large branches but not very many of them. Douglas-fir have small branches, and also not very many of them.

Integrating the data – which species is best?

You can see why Douglas-fir is popular… it has high crown density yet it accomplishes this with few branches that are not very big. It’s crown tends to be more asymmetric, but not by a lot compared to other species. I therefore declare Douglas-fir to be the best Christmas tree (at least out of the three that I measured in detail). On the other end of the spectrum is poor ponderosa pine. Its crown is very thin and it’s got huge branches aligned in whorls around the tree, creating huge gaps in the crown. I’m pretty sure Charlie Brown’s tree was a ponderosa pine, unless he grew up on the east coast in which case it was an Eastern white pine.

But as with forest management, your choice should depend on your objectives. If you want to place huge ornaments in the tree (a German friend of mine likes to place candles in the tree… doesn’t sound very safe) then a pine species might actually be the one for you since it has large branches with big gaps in the crown. But if you want a nice shaped tree and you’re OK with the prickles, then consider giant sequoia. You can’t find a lot of giant sequoia available for Christmas trees, but along with incense-cedar, I think it is poised to become the next co-big thing.

There are other metrics that I could add- crown volume, lop-sidedness, and I even have color differences in terms of spectral reflectance. Maybe I’ll add that next year. Time to celebrate… Merry Christmas and Happy New Year from Foreststeward.com!