Andrew Larson, Seattle Public Utilities

Andrew Larson works in the Watershed Management Division of Seattle Public Utilities as a forest ecology intern. He is also pursuing graduate studies in Forest Ecosystem Analysis at the University of Washington.

I'm shaking things up today. I should be working at the Cedar River Municipal Watershed (CRMW). Instead, I'm at school. I have two major assignments that must be finished today. Thankfully, the deadline for the work I planned to do today at the watershed is still a week out. I'll go into work tomorrow, and perhaps Saturday, so I stay caught up.

I'd really rather not miss work. The Forest Ecology staff meets every other Thursday to review the various projects we are working on, report our progress, set new work priorities, and generally hash out any problems with which we have been struggling. These meetings always result in great discussions; I usually leave our staff meetings feeling like I just finished a university-level ecology class, with an emphasis in problem solving! Not today, though. Instead, I'm sequestered in the computer lab, cranking through a data analysis exercise for my forest community ecology class and some other statistics homework.

I can't complain too much about the homework, though, especially the data analysis exercise. I'm learning ordination, one of the major analytical tools used by ecologists. Ordination belongs to the family of analytical methods known as multivariate statistics. This sounds horribly complex and intimidating (at least it did to me when I first heard it). Ordination calculations are so tedious that they can only be performed by a computer, unless you have way, way more free time than most people, but the output is elegant and reasonably easy to interpret. Basically, ordination allows you to take a huge dataset made up of more measurements than anyone could ever analyze, and reduce it down to something that can actually be interpreted. Ordination finds the important stuff in the dataset and pulls it out to the front where you can see it.

Ecologists use ordination to tell how similar plant communities are to each other. For example, I might go out and install sample plots in different forests (actually, I do this all the time). At each location I would record all the plant species present, how much of each species there is, the size and density of the trees, and loads of other measurements. Instead of taking all these tedious detailed measurements, I could look at the forest and say, "Well, yes. This plot is different from the last plot. See, the trees are a little bigger. And there are more of these little blue flowers." But if I want to actually say something about how different (or similar) the different types of forests are, and eventually move on to asking why, I need to sift through dozens and dozens of measurements and come up with some numbers. Ordination is a technique to speed up this sifting process.

The best part is that ordination results can be graphed. I can look at the graph and actually see how the forests I'm studying compare to each other. It's wild when you think about it. I can spend an entire day measuring the trees and plants in sample plot, and eventually these measurements are reduced to a single dot on a graph.

This winter (after I'm an ordination expert), I will analyze some of the data we have collected from plots in the remaining old-growth forests in the CRMW. Knowing what you have on the landscape is a big part of natural resource management. The wildlife biologists want to know what types of habitat we have in the old growth. The ecologists are curious about the range of old-growth communities and conditions, so we have a better idea about what to be shooting for with our forest restoration projects. I'll use ordination to characterize the variability in the old-growth left in the watershed, giving us a much clearer idea of what kinds of old growth, and how much of each type, we have left. For now, though, I better get cracking on my homework. I need to learn ordination before I can use it!