Just after Hurricane Hugo roared over the Atlantic coastal plain in 1989, U.S. Forest Service research soil scientist Bill McKee (now retired) visited Francis Marion National Forest in South Carolina. Some of its wet pine flats were so badly damaged that they looked like they had been clearcut.
McKee was joined by Michael Aust and Jim Burger, professors of forestry at Virginia Tech, along with partners from the University of Georgia, Weyerhaeuser, and Westvaco (now WestRock) to consider issues associated with salvage logging the damaged stands.
“Most of the sites had only three to five trees left in the overstory. They were really hammered by Hugo, and the salvage logging that took place afterward had left knee-deep — or deeper — ruts in some places,” says Aust.
It was unusual for logging operations to take place during wet conditions on national forest lands, because harvest equipment is known to create skid trails that compact soil and jeopardize future forestry operations.
But the hurricane had taken down so many trees that forest managers were concerned about increased fuel loads and wildfire risk.
The logging was permitted as a way to reduce this risk, and afterward McKee, Trettin, Aust, and their colleagues designed a long-term study to learn how the wet weather logging might affect the next generation of forest trees.
The scientists planned the study around four different site preparations: bedding, disking, bedding with disking, and none.
Bedding moves soil into linear raised beds, and disking is a tillage practice that can break up large clumps of compacted soil. In the combined treatment, areas were disked and then bedded.
Loblolly pine seedlings were then planted across the four different treatments and left to grow.
“This is a great illustration of the value of our experimental and national forests, as the integrity of the original plots remained intact,” notes Trettin.
The team was led by Charles Neaves, at the time a master’s student working with Aust and Chad Bolding. The scientists took tree measurements in the loblolly pine stands and collected soil samples across all four treatments. Would they find differences between the site prep and control areas?
“We could still see a few skids,” explains Aust. “But it looked like the site had pretty much recovered.”
In areas with bedding site preparation, tree seedling survival was greater, which led to larger total stand volume than in the areas with only disking or no site preparation.
That response is attributed to the depth of aerated soil available to tree roots, which was increased. Macroporosity, which allows water to enter and exit the soil quickly, was also improved in the bedded areas.
These soil conditions supported root development and nutrient availability for the regenerating loblolly pine seedlings.
Rooting activity from the pines helped to incorporate organic matter into the soils. Wet and dry cycling created space between soil particles for water and air to move freely. Soil organisms, observed during data collection as abundant and active, assisted with both.
Adding disking to bedding did not add significant benefits for either pine productivity or soil conditions. “Bedding is an efficient treatment for improving pine establishment and productivity on poorly drained sites,” states Trettin.
Across all treatments, the sizes of individual trees did not differ significantly. However, trees within the footprint of equipment traffic were smaller than those located outside of the skid trails. These findings support the current best management practice of limiting the extent of soil disturbance associated with wet weather logging.
“The differences between overall stand biomass and individual tree biomass could be important, depending on the landowner’s goals,” adds Aust. “If more valuable wood products are your primary goal, then you could manage for that through spacing.”
“We saw the soils recover regardless of treatment, which is a real positive for sustainable forestry operations,” concludes Trettin.
For more information, email Carl Trettin at email@example.com.