Long-term experiments to better understand soil-human interactions
Abstract
Interactions between soils and people may be transforming global conditions, but the interactions are poorly understood. Changes in soils have proven difficult to quantify, especially in complex ecosystems manifesting large spatiotemporal variability. Long-term ecosystem experiments that evaluate soil change and demonstrate alternative choices are important to understanding changes, discovering new controls and drivers, and influencing decisions. Inspired by agriculture studies, like Rothamsted, the US Forest Service established in 1990 a network of operational-scale experiments across the Pacific Northwest to evaluate long-term effects of different forest management and disturbance regimes. With a strong experimental design, these experiments are now helping to better understand the long-term effects of managing tree harvesting (clearcutting and thinning), woody debris, and tree and understory species composition, and-serendipitously-the effects of fire. Initial results from the Southern Oregon experimental site indicate surprisingly rapid soil changes in some regimes but not others. We've also learned that rapid change presents challenges to repeat sampling. We present our sample-archive and comparable-layer approaches that seek to accommodate changes in surface elevation, aggregation and disaggregation, and mineral-soil exports. Thinning mature forest stands (80-100 yrs old) did not significantly change soil C in 11-yrs. A small upper-layer C increase was observed after thinning, but it was similar to the control. Significant increases in upper-layer soil N were observed with most treatments, but all increases were similar to the control. Leaving woody debris had little effect. The most remarkable change occurred when mature stands were clearcut and Douglas-firs were planted and tended. Associated with rapid growth of Douglas-fir, an average of 8 Mg C ha-1 was lost from weathered soil 4-18 cm deep. This contrasts with clearcuts where early-seral hardwoods and knobcone pines were established, that trended positively with 2 Mg C ha-1. Soil changes resulting from wild and prescribed fire were substantial. About 50% of the soil C (3-21 Mg ha-1) and 36% of soil N (41-650 kg ha-1) were lost from the upper profile (0-6.2 cm) compared to pre-fire conditions. Intense wildfire that killed most forest trees had about double the losses of C and N than forests burned at lower temperature with fewer trees killed. Average wildfire C losses were more than twice prescribed-fire losses. A long-term perspective is needed to compare episodic influences on soils, like harvesting and wildfire, to day-in, day-out effects of different species mixtures. Especially important is the effect of shrubs, that can rapidly achieve full leaf area but that lack the woody stem structure to store captured C as well as conifers. In theory, therefore, extending shrub cover will increase soil C. The annual profile soil C loss in Douglas-fir (-0.8 Mg ha-1yr-1), if continued beyond 11 yrs, would be similar to the effects of a fire-return interval of less than a third of the historical interval of about 100 years. National and regional soil-C monitoring would benefit from being grounded in existing experimental studies to help integrate large-scale changes with an unfolding understanding of processes in ways useful to decisionmakers.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2011
- Bibcode:
- 2011AGUFM.B43G..01B
- Keywords:
-
- 0414 BIOGEOSCIENCES / Biogeochemical cycles;
- processes;
- and modeling;
- 0428 BIOGEOSCIENCES / Carbon cycling;
- 0439 BIOGEOSCIENCES / Ecosystems;
- structure and dynamics;
- 0486 BIOGEOSCIENCES / Soils/pedology