The properties and processes of deep soil horizons remain an important gap in knowledge due to the long history of shallow soil sampling. The majority of soil carbon and nitrogen can be found beneath the A horizon in most soils, particularly those deeper than one meter to bedrock. Such soils are common in many parts of the world, especially the Pacific Northwest where the combination of age (hundreds of thousands of years in many places) and high precipitation lead to rapid development of subsoil pedogenic features. My dissertation seeks to explore deep soils to better understand the relationships between nutrient cycles and the impact of land-use change and forest harvesting on soil carbon. In a series of 36 soil profiles sampled to 3 meters depth across the Pacific Northwest, pedogenesis frequently extended deeper than the upper 2 meters that is arbitrarily defined as the maximum soil depth for soil taxonomy. The combination of landslides, volcanic activity, and flooding have buried soils in many forests across the region, and these horizons can be important repositories of plant nutrients. In several cases, B horizon development extended deeper than could be excavated with a backhoe (3+ meters). The diversity of parent materials, climate gradients (with both latitude and orography), and soil carbon and nitrogen cycles directly control exchangeable cation cycling across the Pacific Northwest. Soils that experience more precipitation and contain higher levels of carbon and nitrogen hold less exchangeable calcium and magnesium in the whole soil profile, and also have more deeply distributed stocks of exchangeable cations within the profile. Consequently, human disturbances that alter soil carbon can have repercussions for plant nutrition. Millions of acres of forest in the US are actively managed for timber production, but the type and intensity of soil disturbance varies considerably. In a meta-analysis examining the response of soil carbon to forest management from 112 publications, I found that harvesting reduces soil carbon by 11% overall. This loss is predominately driven by O horizon losses (-30%), but there were also losses in surface mineral soil (0-15 cm; -3%). Loss of soil carbon extends deep into the soil with increasing average losses at each depth interval examined; however, very few studies examined soils deeper than 30 cm, leading to extremely wide confidence intervals in deeper soil. Land-use change, even converting one forest type for another, can substantially alter soil carbon cycling, as well. In the Brazilian Cerrado, over half of the natural vegetation has been lost to agriculture, silviculture or urban development, with a substantial portion of the landscape planted with Eucalyptus trees. The shift in the aboveground plant community increases aliphatic functional groups in water-soluble organic matter (WSOM), which may lead to reduced microbial biomass in Eucalyptus plantations that lack native understory trees. The difference in radiocarbon age between WSOM and bulk soil carbon is smaller under Eucalyptus relative to Cerrado, suggesting either mineralization or leaching of aged organic matter under this land use. The consequences of land-use change extend deep into the soil profile, particularly in the Oxisol soils of Brazil which are especially reliant upon soil organic matter for critical ecosystem services like nutrient recycling and water holding capacity.