Questions from the Field: 

Permanent Variable Plots

A newsletter reader recently asked what he should do with a few Variable Plots that were set up some years ago, tagged, and were now due for remeasurement. In this case, his concern was that the growth of the trees, let alone the appearance of a new one, might change the number of trees in a way that might be too variable. He was mostly interested in stand structure, and did not want any “misleading” changes in tree numbers.

There is some question here as to why the number of trees should not change. Do you really believe that the number in the forest did not change just because you have the same number in your plot? Let’s leave aside that issue for the moment, and give him what he wants, which is more stability in numbers of trees. It’s easily done.

In this case, we can go back to the most basic and the oldest use of permanent Variable Plot data discussed by Lew Grosenbaugh many years ago. It’s really quite simple. We just maintain the trees we got on the first sample. Each of them “represented” a certain number of trees per acre at the beginning, and we will keep that situation as time passes.

If a tree “represented 20 other trees/acre” at the time of plot establishment with a 25 BAF prism, then it still represents the same number of trees 10 years later. Leave the prism home, just like you left the plot radius tape home when you went the second time. What does change is the basal area, the volume, the value, etc. on all those 20 trees per acre as time progresses. Another of the original trees might have represented 15 trees/acre – the same reasoning applies to each measured tree.

As we get “ingrowth”, meaning trees that now become large enough to be measured, we can add those, and how to deal with these is in the textbooks and biometrics literature. You have the same problem, of course, with a Fixed Plot. Actually, a fixed plot just for ingrowth trees is not a bad idea. The basic message is that we will look for new small trees (with either system), but the large tagged tree numbers are constant - until they die and suddenly disappear from the plot.

In the case of Variable Plots, the number of trees/acre will probably not be calculated more efficiently than with Fixed Plots, but the other measurements will outperform Fixed Plots pretty consistently. Several Variable Plots, each with a few measured trees, will give better results for the amount of effort than a larger Fixed Plot. The reason is simple. Variable Plots get better precision where the action is – in the larger trees. That is why it is better for timber cruising, and that is why it is better for permanent plots and stand growth measurement.

The only limitation is for building computer models (although Variable Plots would work as well for testing models), and that’s a legitimate use for permanent plots. Also, if you have no idea at all about why you are putting in plots (some people do not) then Fixed Plots are perhaps slightly safer. They are more comfortable, and easier to understand. If you want stand growth, however, Variable Plots are a clear winner.

Why, then, are Variable Plots not used more ??

  1. People don’t know what they want to measure.
  2. You have to understand something slightly new, and fixed plots are so simple. The familiar comment for Variable Plots, yet again.
  3. Habit.
  4. The rumor of “problems” of some sort with Variable Plot computations.

All of these are strong psychological factors. They were factors when Variable Plot Sampling emerged 60 years ago, and they took a long time to get over. Fixed Plots took a long time to die in timber cruising, and they keep crawling back into view every so often. “We have never done it that way” is a powerful force.

There is a good argument for installing at least a few Variable Plots in any organization (or coop). This is the way to eventually prove, with local data, what differences actually occur. Otherwise, “the verdict is never in”. We went through the same thing with Variable Plots for cruising. They had to be directly compared to each other. In addition, you get some personal experience with doing the work and any field details involved. If someone tells you that “the computations are too difficult that way”, you know you are dealing with the wrong person.

There are new and better ways to deal with the computations for permanent Variable Plots than were taught in school 30 years ago, and I would strongly recommend measuring the distance to all the “in” tree centers for Variable Plot trees so these newer methods can be used. This only needs to be done once for existing trees. As new trees show up they are measured when they become “in” with the prism. Azimuth is easy to get, and I would take it, but it is not necessary for computations. The problem of volume quickly increasing if new trees “suddenly” appear with a prism has been solved for a generation, and growth is particularly well estimated with Variable Plots (for the same reason volume is well estimated with Variable Plots).

Only mortality continues to be a problem, and computer models that separate growth and mortality issues can take advantage of the very predictable growth patterns that can be efficiently measured. Mortality, until someone has a really new idea, will mostly depend on making a lot of “alive vs. dead” determinations on existing plots. There is a lot to be said for measuring growth and mortality in different ways. If all you need to know is how much is removed by mortality, all you need to measure are the few new dead trees at every remeasurement. In general, we measure too many trees for growth, and far too few for determining mortality. The problems need to be separated.

You don’t need new plots to do this – just put a Variable Plot into the middle of some of your existing Fixed Plots (with the distances noted). That way, the measurements are already done for you. Just make sure to include any “in” trees beyond the Fixed Plot border. It’s cheap and easy to do this, and the results will be convincing over time. If you already have a stem mapped plot, you can actually do this from the office (although checking for any trees “in” with the prism outside the border of the fixed plot should be done when you visit next time.

As an alternative, you can just stick with Fixed Plots forever – like a number of wise old cruisers did years ago when those whackos with prisms started to talk about an easier method. Who is going to believe that kind of baloney??



Originally published November, 2012

Return to Home


Back to Contents