Richard D. Bowden

Managers increasingly seek to increase forest soil carbon but long‐term controls on soil organic matter (SOM) sources and stability are weakly understood. We used a 30‐year detrital input/removal treatment experiment in a deciduous forest to evaluate the importance of root and leaf litter to SOM. Inputs were assessed by excluding roots and leaves (litter) or by doubling litter inputs. In mineral soil, %SOM differed only at 0–10 cm ( p = 0.11), with concentrations in the no litter and no roots treatments of 26% and 9%, respectively, lower than the controls. Cessation of litter inputs had a stronger effect on SOM than cessation of root inputs, but root litter inputs may contribute stable SOM. Doubled litter increased mineral SOM.

Multiflora rose (Rosa multiflora [MFR]) is an invasive, nonnative plant that has invaded many temperate forests across the eastern United States, often outcompeting native plants for sunlight and other resources. Herbicides can control MFR, but they can also reduce nontarget plant species and threaten aquatic ecosystems. In a black cherry-red maple forest in the Erie National Wildlife Refuge in Pennsylvania, the US Fish and Wildlife Service introduced prescribed goat-browsing as an exploratory control method. In four treatments, browsed, browsed/herbicide, cut/herbicide, and an unmanaged reference, we evaluated preliminary effects of these treatments on MFR and non-MFR herbaceous vegetation. For MFR, the browsed treatment had 56% lower leaf/stem mass ratios and 35% shorter stem lengths than the reference; the leaf/stem ratio in the cut/herbicide treatment was 55% lower than the reference. Stem density was not reduced because goats did not kill the MFR plants in this first year of treatment. The herbicide treatment had fewer non-MFR plants than the reference treatment. Light levels at ground level did not differ among the treatments. Overall, 33% of trees in the browsed treatment were affected by the goats, with 9% being completely girdled; red maple and ironwood were the most commonly browsed species. Preliminary results suggest that goats can be an effective control for MFR, however long-term success will be best evaluated after consecutive treatment seasons. Goats may increase tree mortality and shift tree species composition in stands dominated by trees with high browsing rates, but effects on diverse stands may be less pronounced.

Core Ideas: N additions alter soil microbial community composition and reduce forest soil microbial biomass and enzyme activity. Litter decomposition and soil organic matter degradation was slowed by N additions. Reduced decomposition increases soil C, but long‐term effects on forest productivity are unknown. Eastern North American forests receive anthropogenically elevated nitrogen (N) deposition that alters soil processes and forest productivity. We examined N deposition effects on soil carbon (C) and N in temperate, N‐rich forest plots fertilized annually (100 kg N ha ⁻¹ y ⁻¹ ) since 1993. After nearly two decades, soil C in O, A, and upper 50 cm of B horizons of N‐addition plots was 17% greater (14.2 ± 0.7 kg C m ⁻² ) than control plots. Aboveground tree biomass growth and litterfall were not affected by fertilization. Fine root mass (0–1 mm) was 34% greater in N‐addition plots, but did not explain soil C increases. Rather, reduced decomposition of litter and soil organic matter drove C increases in N‐addition plots. Decomposition rates of black cherry, sugar maple, and mixed leaf litter were 43, 67, and 36%, greater, respectively, in control than N‐addition plots. Light fraction organic matter was greater in N‐addition plots than in control plots, due to either enhanced root production or decreased decomposition of soil organic matter. Soil respiration was reduced, and microbial biomass in O, A, and upper‐B horizons was lower in N‐addition plots than controls. The soil microbial community composition was also altered dramatically with N additions. Recalcitrant organic matter enzyme activity (peroxidase) was reduced in the O‐horizon by N addition. Available Ca, Mg, and K were reduced in O and A horizons by N fertilization. These results suggest that chronic elevated atmospheric N inputs can increase forest soil C storage by decreasing decomposition, however the long‐term stability of this additional C sequestration is unknown.

Above‐ and belowground litter inputs in a temperate deciduous forest were altered for 20 yr to determine the importance of leaves and roots on soil C and soil organic matter (SOM) quantity and quality. Carbon and SOM quantity and quality were measured in the O horizon and mineral soil to 50 cm in five treatments (control, double litter [DL], no litter [NL], no roots [NR], no inputs [NI]). After two decades of doubled litter addition, soil C and SOM did not increase. However, leaf litter exclusions reduced soil C (O and mineral horizons combined) by 24% in NL and 33% in NI treatments. In the mineral soil, the largest declines occurred in the 0‐ to 10‐cm depth (0.93–2.01 kg C m ⁻² ), although losses were observed throughout the entire solum. The NR treatments showed no losses of C. Thermal characterization of SOM quality differed among treatments in the 0‐ to 10‐cm depth. Patterns of CO ₂ evolution during SOM combustion revealed differences in SOM quality between surface and deeper horizons. Our work shows that the sources of litter are important in controlling soil C. Leaf litter made important contributions to maintaining current stocks of soil C; increased leaf litter did not increase soil C, but decreases in litter inputs resulted in rapid soil C declines. Root litter may ultimately provide more stable sources of soil C. Management activities or environmental alterations that decrease litter inputs in mature forests can lower soil C content; however, increases in forest productivity and the resulting increased litter production seem unlikely to increase soil C sequestration.

At the Harvard Forest Long-term Ecological Research Site Chronic Nitrogen Amendment Study, a red pine and a mixed deciduous stand showed immediate changes in soil respiration following nitrogen additions (low N: 5 g N m⁻² per year; high N: 15 g N m⁻² per year) during the initial year (1988) of the study. In the hardwood stand, soil respiration rates increased after N additions in the first year (control: 482.0 g C m⁻² per year; high N: 596.5 g C m⁻² per year). This increase is attributed to increased productivity in the hardwood stand compared to the pine stand; N additions are hypothesized to have increased either root or microbial activity, or perhaps both. In the second year, however, respiration in the fertilized hardwood plots was not different from the control plot. In the pine stand, annual soil respiration was 21 and 25% lower, respectively, in low N and high N plots than the control (429.9 g C m⁻² per year), with further reductions in the second year.