Recent research has confirmed the importance of secondary vegetation regrowth and land management in the terrestrial carbon cycle. Reforestation in temperate regions, especially in eastern North America, has contributed to reduced warming. However, there is uncertainty in model estimates of these secondary land fluxes and the role of land use management factors such as shifting cultivation, irrigation and fertilization. The goal of this work is to explore the role of agricultural conversion/abandonment, wood harvesting, nitrogen fertilization, irrigation, and tillage, within the context of rising CO2 levels, warming climate, nitrogen deposition and ozone on ecosystem carbon and water dynamics using the process-based prognostic Terrestrial Ecosystem Model (TEM) – Hydro2 with the historical land cover transition dataset of Hurtt et al. (2006). The new version of TEM-Hydro2 includes multiple vegetation structural pools, enhanced disturbance modules and calibration, and a reduced-form open nitrogen model. We consider three experiments to explore historical terrestrial carbon dynamics for 1900-2006 in the U.S.: 1) potential vegetation, 2) transient vegetation based on the Hurtt et al. (2006) dataset including realistic agricultural management, and 3) same as previous case without the agricultural management. A preliminary simulation was run for a single pixel (0.5-degree) in the vicinity of Harvard Forest in MA using Ramankutty and Foley’s (1999) data, which contains 6% cropland in 1900, abandoned by the end of the century. This result shows that the process of cropland establishment and abandonment reduces 5.2% of annual NPP and 41.8% of cumulative NEP, reducing the strength of today’s carbon sink. Annual average runoff increased 6.1% and soil moisture 2.2% as a result of including past disturbance. We will present results extrapolated across the entire U.S. with the Hurtt et al. (2006) dataset for the three scenarios we outlined above.