Soil Carbon and Nitrogen Stocks: Underestimation with Common Sampling Methods, and Effects of Thinning and Fertilization Treatments in a Coastal Pacific Northwest Forest
Soil contains substantially more carbon (C) than the atmosphere and vegetation combined. Consequently, the fate of soil organic C (SOC) in response to changes in climate, land use, and management is of great concern. The ability to accurately quantify and compare SOC stocks over time is an important part of understanding carbon-climate feedbacks and constraining climate models, particularly because soils with larger standing SOC stocks are most at risk to release considerable C to the atmosphere. The objectives of this study are to 1) review the global importance of SOC and deep soil biogeochemical processes; 2) compare soil sampling methods and approaches to quantify and compare SOC stocks over time; and 3) examine SOC and nitrogen (N) response to thinning and fertilization treatments in a coastal Pacific Northwest forest. Three soil sampling methods for estimating SOC stocks were compared: clod, core, and excavation. As it is considered the least biased method, the excavation method was used as the standard by which other methods were compared. Soil was sampled at an intensively managed Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) plantation in northwestern Oregon, USA. Management regimes – no treatment (control), thinning treatments, and fertilization treatments – were randomly assigned to nine 0.2-ha plots established in 1989 in a juvenile stand. Prior to harvest in 2015, soil bulk density and chemical analysis samples were collected in the middle of soil depth layers 0-10, 10-20, 20-50, 50-100, and 100-150 cm. Forest floor samples were also collected. Soil was generally non-rocky (<2% fine to medium gravel content by weight) with a sandy clay texture. The core method significantly (Tukey’s HSD, α = 0.1) underestimated soil bulk density at all depths below 10 cm. Soil organic carbon concentrations tended to be lowest for the clod method and highest for the excavation method. As all SOC stock calculations rely on the relationship between SOC mass and soil mass, the underestimation of these parameters by clod and core methods, respectively, similarly affects the fixed depth, genetic horizon, and mass based approaches to quantify SOC stocks. The clod and core methods underestimated the SOC stock (Mg ha-1) to a depth of 150 cm by 22% and 36%, respectively. Most of this difference occurred below 20 cm, where the majority of SOC stocks were contained across all soil sampling methods. During a single rotation of ~40 years, thinning treatments significantly reduced SOC (Mg ha-1) and N (kg ha-1) by 25% and 27%, respectively, compared to no treatment. Most of this loss occurred in deeper soil layers (below ~20 cm). Across all management regimes, deeper soil layers comprised the majority of SOC and N stocks. This study shows that 1) commonly used soil sampling methods for measuring soil properties such as SOC should not be assumed to be interchangeable; 2) accurately quantifying and comparing SOC and N stocks requires sampling deep soil; 3) regional and global SOC and N stocks may be largely underestimated due to shallow sampling and the frequent use of core methods; and 4) forest management can significantly impact both surface and deep SOC and N on decadal timescales.