Deep soil: Quantification, modeling, and significance of subsurface nitrogen

Abstract

Nitrogen (N) is one of the primary limiting nutrients in Pacific Northwest forests, as well as many other terrestrial ecosystems around the world. Efforts to quantify total soil N and to monitor N cycling have often sampled soils to a depth of 0.2m, occasionally to 1.0m depth, or the bottom of the B horizon. However, tree roots often extend many meters into the soil redistributing water to the surface during droughts and contributing to nutrient uptake. This study examined the systematic sampling depth for ecosystem N analyses in the Pacific Northwest, and compared best-fit models of N in deep soil layers with observed quantities. At 22 sites across the Pacific Northwest Douglas-fir zone, O horizon and mineral soil bulk density samples were collected at depths of 0.1m, 0.5m, 1.0m, 1.5m, 2.0m, and 2.5m. Mineral soil was screened to 4.75mm and analyzed for total N content. Systematic sampling shallower than 2.0m produced significantly smaller estimates of total N. On average, only 3% of total soil N was in the O horizon, and 31% was below 1.0m depth (almost 2700kgha−1 of N). Over 45% of soil N was below 1.0m at three sites. A nonlinear mixed effect model using the Langmuir equation predicted total N to 2.5m with −12.4% mean error given data to 1.0m, and −7.6% mean error with data to 1.5m. Shallow sampling of soil N in studies of biogeochemical cycling, forest management impacts, or ecosystem monitoring at best provides a biased estimate and at worst produces misleading conclusions. Research and monitoring efforts seeking to quantify soil N or measure fluxes should sample deep soil to create a more complete picture of soil pools and changes over time.