Stormwater runoff from impervious surfaces in urban areas causes water pollution and flash flooding. Trees can help solve this problem by capturing and absorbing rain. However, there is often not enough space in cities for both trees and conventional stormwater control systems. Various bioretention systems that incorporate trees are used in cities, but they are expensive and complex, which makes their implementation difficult. We are investigating a system that may prove to be less complex, less costly, and better for growing trees. Our full-scale prototype of this system was constructed in 2020 adjacent to a parking lot on the Virginia Tech campus in Blacksburg, VA. It comprises a belowground gravel bed surrounding a soil bed where trees are planted. The gravel bed stores a large volume of parking lot runoff that is then absorbed by the tree roots and expelled to the air through transpiration. Our long-term study of this prototype aims to understand how well the system captures and stores runoff and how well the trees perform compared to control trees in a standard planting nearby. We are monitoring tree crown and root growth, photosynthesis, and leaf stomatal conductance. Growth and physiology of the trees in the system has been comparable to the control trees since being planted in May 2020.