Model Based Design of Tree Pits After the Stockholm Model

Tree pits, as a form of Low Impact Development (LID), serve a multifaceted role in urban environments, addressing not only drainage needs but also contributing to urban climate moderation while retaining space for multifunctional use on the surface. While many designs for tree pits exist, the Stockholm Model and its adaptations offer a low-tech approach with few points of failure. For it to be applied efficiently and tailor-suited to local conditions, a design process that includes all (competing) objectives is needed. Currently, a simplified model facilitating such a design process is not available.

Designing effective tree pits deals with managing competing objectives. On the one hand, urban drainage requires tree pits to use their volume for retention as effectively as possible, while the tree, on the other hand, can withstand saturated conditions only for a limited time. Uncertainties in soil parameters and in rainfall and drought period distributions in a changing climate further complicate the design process. To navigate these complexities a model-based design approach for tree pits offers significant advantages.

This study aims to develop and apply a fast-running, simplified model tailored for tree pit design in urban landscapes. Essential to this model is its capability to accurately predict key aspects of hydrological tree pit performance while retaining low complexity and short runtime. The key variables modeled are the retention performance and the soil moisture content including duration and depth of saturated conditions. Additionally, the model assesses drain-flow, overflow, and seepage into the underground. Transpiration losses can also be estimated. This is achieved by utilizing different elements from USEPA’s SWMM LID toolbox and its twozone groundwater model. The widely established submodels are combined to develop an integrated model in the python programming language. This model is applied to a case study to optimize the design of a tree pit in the Stockholm Model for different scenarios and targets variables.