are inserted into the upper 20 cm of
soil for decomposition experiment.
Figure 1. Preparing roots for wood decompostion experiment. Short sections of roots are sealed on the ends using a neoprene paint.
Roots are important structural and functional components of forested ecosystems. A large proportion of forest production is allocated to roots, resulting in a large flux of carbon and nutrients into the belowground detrital system. The fate of this material is not well understood. This is because most previous studies of litter dynamics in forest ecosystems have concentrated on aboveground litter. Very few root decomposition studies have been conducted. Two USDA grants (CSREES 94-37107-0534, CSREES 99-00859) have allowed us to systematically test the main biotic and abiotic controls of woody root decomposition in the Pacific Northwest. The overall goal of this project is to understand how substrate quality, soil temperature and moisture content and their interactions control root decomposition. Our major objectives were:
1. To measure small and coarse root decomposition rates of Douglas-fir, hemlock, and ponderosa pine using chrononsequences to parameterize a model of root decomposition.
2. To examine the responses of respiration rate of decomposing roots to a wide range of thermal-moisture regimes in the laboratory. These data were also used to parameterize the root decomposition model as well as to define the interaction between these two abiotic factors.
3. To establish a long-term root decomposition experiment to test the predictions of the decomposition simulation model on the effects of substrate quality, size, temperature, and moisture on root decomposition.
|
Figure 2. Roots wrapped in nylon mesh are inserted into the upper 20 cm of soil for decomposition experiment. |
Our field studies have been conducted in Cascade Head Experimental Forest, H. J. Andrews Experimental Forests, and Pringle Falls Experimental Forest in Oregon. These three sites form a climatic gradient from warm and wet at Cascade Head to cool and dry at Pringle Falls. Several approaches including chronosequences, time series, laboratory incubation, and simulation modeling have been used to examine woody root decomposition in this region. A root decomposition simulation model (ROOTDK) was developed to evaluate the effects of root substrate quality, temperature, and moisture on root decomposition. A long-term (10 year) root decomposition time series study was also established that will continue to help us to understand controls of woody root decomposition, corroborate our root decomposition simulation model, and furnish a basis for further studies of woody root decomposition. We are also examining lignin and cellulose degradation as well as the different decomposer functional groups (i.e., white-rot versus brown-rot). Cellulose and lignin degradation is being examined using several approaches including Cross Polarization Magic Angle Spinning 13C NMR NMR (nuclear magnetic resonance), py-MBMS (Pyrolysis Molecular Beam Mass Spectrometry), and wet chemistry analysis. Fungi involved in woody root decomposition are being identified using molecular methods.
The chronosequence study indicated that decomposing roots started to release nitrogen after 20-30% mass loss, a point when the average C: N ratio was as high as 140 (Chen et al. 2001). The time series study further confirmed this by showing that decomposing fine and small roots started to release nitrogen during the earliest stages of decomposition (Chen et al. in press). Decomposing roots, especially fine roots, could be an important nitrogen source after large-scale disturbances such as clear-cut or forest fire (Chen and Harmon in press). We estimated that 70 Kg/ha of nitrogen could be released from dead wood roots annually after catastrophic disturbances in Douglas-fir old-growth forests. Our data also suggest that decomposing fine roots could release at least 20 Kg/ha of N annually in mature Douglas-fir forests in the Pacific Northwest. The nitrogen release from dead woody roots appears to synchronize with the demand of new forest growth. Forest harvests, especially clear-cuts, create as much as 200 Mg/ha of dead roots in Pacific Northwest of USA. It may be important to keep tree stumps on site after forest cutting because decomposing fine and small roots started to release nitrogen from the earliest stages of decomposition.
Separating roots into structural components (bark, wood, and resin cores) provided a better estimation of long-term mass loss than initial substrate indices, especially for woody roots larger than 1 cm in diameter (Chen et al 2001). None of the14 initial substrate quality indices we examined was significantly correlated to the decomposition rate-constants (k) in the chronosequence study. Similarly, in the time series study, none of 17 initial substrate indices was significantly correlated with k of the woody roots larger than 1 cm in diameter. Differences in root substrate quality of woody roots therefore corresponded more to physical structures than to chemical indices. Root wood had the fastest k, root bark the second, and resin cores the slowest. Western hemlock and ponderosa pine, species without resin cores generally, had a higher k of 0.033 to 0.077/year. In contrast, Sitka spruce, Douglas-fir, and lodgepole pine, species with resin cores, had a much lower k, ranging from 0.011 to 0.03/year. For fine and small roots (< 1 cm in diameter), we did find that several initial substrate quality indices were significantly correlated with k. For all fine roots, lignin-cellulose index (LCI) and lignin - polyphenol: N accounted for 83% of the variation of decomposition rate-constants (Chen et al in press). For small roots, polyphenol: N ratio was the best predictor of k, accounting for 73% of the variation (Chen et al 2001). In addition, we found that decomposition rates in the first year were correlated with the fraction of water-soluble matter, whereas the subsequent decomposition rates were correlated to lignin content. NMR and wet chemistry indicate cellulose content decreased over time, whereas lignin increased.
Rot types in dead roots varied with species. Lodgepole pine and ponderosa pine were dominated by white-rot, whereas brown-rot occurred mainly in Douglas-fir and Sitka spruce. Both rot types appeared in roots of western hemlock. Decomposer species may play more important roles than we initially expected in woody root decomposition. In the case of root wood, we found that species prone to white-rot had higher decomposition than those prone to brown-rot. The high frequency of white-rot in the two pine species may suggest dead roots contribute less to the formation of "stable" soil carbon at Pringle Falls than at Cascade Head or H. J. Andrews site. The presence of white-rot in ponderosa pine may also explain why roots of this tree species decompose faster than western hemlock. Molecular methods revealed two hundred distinct RFLP pattern groups, suggesting the fungi community is very diverse (Vandergrift et al. 2001). We had anticipated that Basidomycetes would be the dominant fungus in decomposing woody roots. However, in the roots examined Zygomycetes were the dominant taxa.
The dominant environmental limiting factors appear to have varied with site. Low soil temperature at Pringle Falls and high root moisture at Cascade Head appear to have limited the microbial processes and in turn slowed down root decomposition. In contrast, the combination of soil temperature and moisture regimes at H. J. Andrews was more optimal, resulting in the fastest decomposition in the three sites. Our laboratory incubation study further indicated temperature and moisture effects on root decomposition (Chen et al. 2000). The rate of carbon release from dead roots increased with temperature and reached the maximum at 30- 40 0C, and then decreased rapidly. The Q10 of dead roots was significantly influenced (p < 0.0001) by temperature, but not by species, decay class, and the direction of change. For example, at 5 - 10 0C, Q10 averaged 3.99 and then decreased to 1.37 at 30 to 40 0C. Over a range of 5-60 0C, Q10 could be predicted by a single exponential model using temperature as the independent variable. The optimum root moisture ranged from 100 to 275% depending on the species. When roots were too dry (< 50%) or too wet (> 300%) root moisture became a limiting factor. The ROOTDK model was used to evaluate the effects of changes of temperature and moisture on root decomposition. For Cascade Head, warming climate may decrease excess soil moisture limitation effect on root decomposition by increasing evapotranspiration. This drying should also allow the positive impact of increased temperature on root decomposition. Therefore decomposition of dead roots will increase at Cascade Head. In contrast, low soil moisture limitation will increase at Pringle Falls, resulting in a decrease in root decomposition. For H. J. Andrews, if the climate warming does not drive the moisture too far from the optimum range, the decomposition of dead roots will increase due to the more favorable thermal condition. ROOTDK simulations indicated woody root decomposition at unfavorable sites such as Pringle Falls and Cascade Head is more sensitive to climate change than favorable site such as the H. J. Andrews.
This research also has important forest management implications in soil strength, belowground wildlife habitat, and nutrient conservation. The presence of woody roots in soils directly influences soil strength. The mechanical reinforcement provided by the living woody root system diminishes over time when roots decompose. The occurrence of decomposition-resistant resin cores in woody roots slows down the decomposition of roots, which in turn prolongs the functional period of woody roots to maintain soil strength. Our root decomposition study provides the basis to evaluate the roles of dead root in soil strength. Decomposing woody roots of forests also create many belowground channels, which in turn may provide habitat for soil animals and wildlife. Several species rely on or at least use decomposing roots as their habitats or movement channels in Douglas-fir forests of western Cascade of Pacific Northwest including a variety of salamanders (e.g., the clouded salamander, ensatina, and western redback salamanders), shrews (e.g.,the Trouwbridge's shrews), the shrew-mole, the coast mole, the western red-backed vole, and the Townsend's chipmunk. In general, older forest stands produces more large roots, therefore create more heterogeneous habitats. These diverse belowground habitats may play certain roles in the diversity of wildlife in old stands. The increasing shift to short rotation plantation management from old stand forests in Pacific Northwest will certainly change belowground root size and composition pattern, leaving less dead root habitat for wildlife.
We finished the field part of the long-term time series experiment on woody root decomposition in the Pacific Northwest. We plan to examine the mass loss data and analyze changes in nutrients if funding can be secured. We are completing a study examining N dynamics in fine roots by use of 15N. At this point mass loss patterns are very similar to what Chen et al (2002) observed, with a very rapid period of mass loss occurring in the first 6 months, followed by a slower, steady decline in mass.
Chen, Hua; Harmon, Mark E. 2001. Determination of ash and nitrogen concentration of decomposing woody roots by near infrared reflectance spectroscopy [Abstract]. In: The Ecological Society of America 86th annual meeting: Keeping all the parts: preserving, restoring and sustaining complex ecosystems; Madison, WI. Washington, DC: The Ecological Society of America: 69. (Pub No: 3137)
Chen, H., M. E. Harmon, and R. P. Griffiths. 2001. Decomposition and nitrogen release from woody roots in coniferous forests of the Pacific Northwest: a chronosequence approach. Canadian Journal of Forest Research 31: 246-260. (Pub No: 2792)
Chen, H., Harmon, M. E., Sexton, J. and Fasth, B. 2001. Fine root decomposition and N dynamics in coniferous forests of the Pacific Northwest of USA. Canadian Journal of Forest Research, in press. (Pub No: 2793)
Chen, H., M.E. Harmon, R.P. Griffiths, and W. Hicks. 2000. Effects of temperature and moisture on carbon release of decaying woody roots. Forest Ecology and Management 138:51-54. (Pub No: 2653)
Chen, Hua. 1999. Root decomposition in three coniferous forests: effects of substrate quality, temperature, and moisture. Corvallis, OR: Oregon State University. 218 p. Ph.D. dissertation. (Pub No: 2619)
Vandegrift, Eleanor V. H.; Chen, Hua; Harmon, Mark E. 2007. Fungal genetic diversity within decomposing woody conifer roots in Oregon, U.S.A. Northwest Science. 81(2): 125-137. (Pub No: 4235)
Vandegrift, V. Eleanor; Chen, Hua; Horton, Thomas R.; Harmon, Mark E. 2001. Identification of fungi decomposing woody conifer roots in Oregon using molecular methods [Abstract]. In: The Ecological Society of America 86th annual meeting: Keeping all the parts: preserving, restoring and sustaining complex ecosystems; Madison, WI. Washington, DC: The Ecological Society of America: 225. (Pub No: 3138)
Original document posted February 21, 2002