Green walls have a broad spectrum of applications, particular in urban areas. They have the unique capability of combining design and conservationist aspects, while being used as grey water treatment facility or acting as insulation and cooling system on a building and neighborhood scale.
Nonetheless, green walls have not been studied as extensively as other green infrastructures, like green roofs or bio retention cells - despite their versatility.
Consequently, there is an urgent for data collection and model development for green walls. The most important aspect in this context is the derivation of water demand models, as many aspects of green wall operation and functionality depend on optimized irrigation.
This is of particular importance in the face of climate change and urbanization, where green infrastructures provide a well-suited solution to mitigate the effects of these trends.
However, with the increasing realization of green infrastructures new challenges and conflicts arise. As drinking water resources become scarce, questions whether “ enough ” water is available. In this context, the use of rain and grey water as potentially suitable alternatives to irrigate green walls has yet to be analyzed.
To avoid potential conflicts and ensure optimized irrigation, models to quantify the water demand of green walls are required.
Thus, this project is aimed at developing a new modeling approach combining deterministic and stochastic models for plant behavior with irrigation and environmental data, monitored at an experimental green wall. To integrate the diverse data set, innovative machine learning techniques are applied.
By offering a way to quantify green wall water demand and the consequences of such structures on water resources, the resulting model is intended offer scientists and practitioners a tool to assess the water demand for green walls, given the specific on-site conditions, find suitable resources (e.g. grey water) and communicate requirements and advantages to all stakeholders.