Forest Practices > Monitoring and Evaluation > EFP Monitoring

Monitoring Yield Predictions

Many important monitoring questions relate to checking stand growth and yield. A good overview of the topic of comparing observed stand growth to expected (or predicted) growth is provided in the report Silviculture information needs and growth and yield monitoring. However, in many cases, monitoring objectives are broader than simply comparing observed conditions to predicted conditions. A good overview with this broader perspective on monitoring is provided in Selecting appropriate statistical procedures and asking the right questions: a synthesis. This web page provides an overview of some of the systems that are used to monitor (in its broadest definition) stand growth and yield: designed experiments, special surveys, informal field reviews, existing silviculture surveys and information systems, and PSPs.

Designed experiments

Installing new experiments

Designed experiments (trials) are the best method of learning about the response to treatment. However, they rarely provide wide coverage over the range of conditions present on a management unit. The characteristics of a good experiment are well understood. The paper Design of Experiments provides good advice to foresters contemplating installing new experiments:

Learning from existing experiments

Apart from establishing new experiments (trials), much can be learned by finding out about existing experiments, obtaining the progress reports (or data summaries), and visiting the plot locations. Observe treatment impacts and attempt to confirm (or refute) your key expectations and assumptions about the outcomes of treatments. The results of these field visits should feed back into your Enhanced Forestry program planning cycle, leading to improved inputs to silviculture planning exercises, refined treatment regimes, and improvements to the silviculture strategy. It will take some investigation to identify the designed experiments in your area that are relevant to your needs. The location of many experiments are indicated on forest cover maps. Three other useful sources of information are:

• MoF region and branch research scientists

  http://www.for.gov.bc.ca/hre/

• MoF Nelson Forest Region directory of research sites

  http://www.for.gov.bc.ca/nelson/research/eplocation/Intro.htm

• Forestry Canada — Forest ecosystem research network of sites

  http://www.pfc.cfs.nrcan.gc.ca/ecology/ferns/index_e.html

Special surveys

A wide variety of special surveys can be conducted to obtain monitoring information.

Paired-plot retrospective studies

Paired plot retrospective studies are one common method to assess the outcome of past treatments. In it’s typical form, locations are found where treated and untreated areas are located side-by-side. For example, a spaced stand adjacent to an unspaced stand or a planted area adjacent to unplanted ground. Plot pairs are established: one in the untreated side and one in the treated area. Great care is taken to ensure that the areas differ in only one aspect — the treatment. The difference between plots is assumed to be the result of the treatment. Weaknesses of this approach include the danger 1) that the sample locations are not representative of the population to which you want to apply the results and 2) that some hidden factor (other than the treatment) may have caused the observed differences. Examples of this monitoring method include:

• The Babine EFMPP retrospective study of juvenile spacing.

• Miller, R.E., R.E. Bigley, and S. Webster. 1993. Early development of matched planted and naturally regenerated Douglas-fir stands after slash burning in the Cascade Range. Western Journal of Applied Forestry 8(1): 5–10.

• Miller, R.E. and H.W. Anderson. 1995. Stand characteristics of 65-year-old planted and naturally regenerated stands near Sequim, Washington, USDA, Forest Service, Pacific Northwest Research Station. PNW-RP-482.

Analytical surveys

Analytical surveys are sometimes used to provide a preliminary assessment of treatment impact. Analytical surveys involve a statistical sample of two populations (for example, fertilized and unfertilized polygons within a management unit). The means are compared (for example, mean basal area growth rates) and the difference is interpreted as the response to treatment. As with paired-plot retrospective studies, the great weakness of this approach is that that some other factor (other than the treatment) may have caused the differences that you are attributing to treatment. The acceptability of this approach is hotly debated within the research community. A good introduction to the topic is provided in the paper Studies of uncontrolled events.

Informal field reviews

Informal field observations are the most common method of monitoring used by silviculturists. This approach can involve visiting a range of sites to observe stand condition some time after treatment to make inferences about the effects of treatments and the factors that influence treatment outcomes. These observations feedback into the design of silviculture treatment programs. Though this approach is widely used, it has many weaknesses including the danger that subjectively selected sites provide a distorted representation of the population and that the lack of measurements provides low rigor. However, this method is cheap — and used consciously or unconsciously by all silviculturists. Tips for improving the quality of monitoring information provided by this method include:

• Ask yourself whether the sample of blocks you’re viewing represent the population that you want to apply your conclusions to.

• Carefully look for some factors other than the treatment that may have caused the outcome that you’re observing.

• Consider not only the current state — but also the probable future dynamics — when you assess treatment impact.

• Consider whether treatment standards have changed over time in a way that might make conclusions based on stands treated in the past less relevant to the treatments that you are conducting today.

Existing silviculture surveys and information systems

The existing system of silviculture surveys and databases (e.g., ISIS, Phoenix, etc) provide some monitoring capabilities — for example, the ability to monitor the type and amount of treatments each year in a management unit. With some modification, the capability of silviculture surveys and databases to answer growth and yield monitoring questions could be greatly enhanced. However, it will be necessary to commit to improving the software and keeping the data current. Desirable modifications include retaining previous forest cover information and enhancing survey procedures. This topic is discussed in more detail in the report Silviculture information needs and growth and yield monitoring.

PSP programs

A network of Permanent Sample Plots (PSPs) can address many monitoring questions.

Subjective sample PSP programs

With few exceptions, PSPs in B.COctober 20, 2005locations. All of the ministry’s PSP programs are subjective sample PSP programs. These plots provide valuable records of how some specific stands are growing — but they have been kept out of many stand types that are of interest and they can not be used to indicate the average stand condition, growth rate, or response to treatment.

Representative sample PSP programs

When the location of PSPs is determined by a grid (or other non-subjective method) the PSP program can be termed a representative sample PSP program. NIVMA runs the largest program of this sort in the province today. However, other groups are establishing representative sample PSP programs, including:

• Province wide network of PSPs established on a grid

• Weyerhaeuser's representative sample PSP program in TFL 35.