Long-term forest studies in Alaska

Alaska is a big state, with millions of acres of forested land, but has only seven long-term forest research projects (compare this with the 700 or so in Washington and Oregon). “Long-term” when it comes to the lives of trees and a forest study means 25 to 40 years or more, explained John Yarie, professor of silviculture, during a lecture for the graduate seminar course NRM 692 on Jan. 30.

An upland site in a 24-year-long forest drought study.

“Research Directions in Forest Dynamics: Limiting Factors,” gave a glimpse of those studies and their importance. When looking at a forest to create a long-term management policy, managers look at several factors: regeneration rate, stand density (what year should one cut?), type of tree and the location, thinning intensity (how much should one cut?), rotation length. The idea is that managers would create a policy that outlives them—but this may not be flexible enough to deal with such things as climate change, economic uncertainties, or wildfire. The original resource plan with 100-year goals might not be the best one for a changing system, and only prove appropriate for, say, 10 or 20 years. Perhaps the management needs change from timber harvest to non-timber forest products, or to recreational or wilderness use, or ecosystem services such as watershed filtration. Adaptive forest management allows for change in future situations.

Adaptive forest management is defined as an approach in which the effects of policy and treatments (cutting, fertilizing, burning, watering) are continually monitored and used. Research results are incorporated to modify the way the forest is managed so that objectives and policy dynamically interact with each other. “Adaptive forest management,” explained Yarie, “is a formal commitment to improve management” over time.

Yarie described the limits to a forest’s growth: nutrient availability, climate (temperature, precipitation, moisture dynamics, water stress), belowground climate, topography, the parent material of the soil, time (How old is the forest? When was the last disturbance?), and light. In Alaska, nutrient availability, it turns out, does not appear to be that important—but water availability is, especially on non-permafrost systems. (The studies using fertilizer treatments were very inconsistent.)

Yarie showed the seminar attendees charts marking comparative growth relative to different treatments, looking at white spruce, birch, and aspen. Growth was measured by trunk diameter at 4.5 feet above the ground, a standard known as DBH, or diameter at breast height. Depending on the species, growth results could be quite different: in the birch studies, the control (unfertilized trees) grew faster than the fertilized ones. With young aspen (less than 75 years old), fertilizer had a large and sustained influence on growth even though it was only applied for five years: a pronounced positive result. However, with white spruce, results changed from year to year. An interesting result was that, looking at other possible influences on the results, the best predictor for growth was the climate of the previous August.

Looking at only one treatment doesn’t give a clear picture of what works well to produce the best result, however. Yarie showed how the fullest growth could be seen over time when a combination of thinning and fertilizing was used, rather than only thinning or only fertilizing.

Another long-term study that Yarie described had to do with the effect of drought. Climate change will likely lead to a drier climate, and that could become a big problem to forest managers for the Interior. Approximately 60 percent of the Interior’s precipitation falls between May and September, or about 195 mm out of a year-round current level of 325 mm. Snow makes up about 40 percent of precipitation, falling between October and April.

Summer rainfall exclusion platform. Note the spillover dams to guide water around each tree trunk.

Yarie created artificial drought conditions for his study by using summer rainfall exclusion platforms, studying the effects over the last 24 years. In recent years, he added spring runoff drought by eliminating winter snowmelt for both types of sites used: upland and lowland floodplain sites. The hypothesis was that upland sites would depend on rain and snow, and so be heavily affected by the artificial drought, but that the lowland sites wouldn’t. The results were just the opposite. Eliminating winter snowmelt did make a difference in the upland sites, but summer rainfall exclusion reduced growth for white spruce in floodplain sites.

Upland site with snowpack removed (see photo, top, before removal).

Upland site one week after snow removal.

This sparked a discussion among the seminar’s attendees about why this might be, and how greater winter snowpack affected moisture penetration into the soil: deeper snow, deeper penetration, more moisture in the soil in spring. Dr. Glenn Juday, an audience member and forestry professor, observed that satellite observations had shown a correlation between increased snowpack and greater photosynthesis, “which takes us up from the plot level to the landscape level,” he said.