Bioswales contribute to climate resilience, as they positively impact the urban water infrastructure by improving the water balance and water quality. However, current bioretention design ignores the facts that different designs in different soils and climatic locations produce different performance results.
The effect of polder conditions on the hydraulic performance, as well as the effect of different storm-types and initial conditions will be researched in this report. To this end, 5 bioswales located in Rotterdam, The Netherlands are monitored on their hydraulic behaviour. The discharge of the drain, groundwater levels in and at the edge of the bioswale and the water level in the bioswale are monitored for 4 of the bioswales. One bioswale only has groundwater and surface water measurements. The polder conditions resulted in a low volume reduction and faster and stronger reacting drain. However, the peak reduction and peak delay were still quite good due to the low permeability of the top-soil. For smaller storms, even the goal for volume reduction could be met. It was found that the structural porosity governed the infiltration, but large plants can increase the permeability too much. In addition, the textural porosity should start high enough to allow for vegetation development.

Phenomena like urbanization and climate change will increase flooding problems, making it clear that more research and investments are needed in sewer systems to improve urban flood prevention. This research was done to determine factors influencing the sediment load to gully pots. It was assumed that in the measurement period, the retaining efficiency of the gully pots remains unchanged. Sediment levels in gully pots for a period of 3 months after cleaning were measured in the fall/winter period, in the cities of Rotterdam and The Hague. A probabilistic approach was used, consisting of a deterministic and random part. For the deterministic part, a generalized linear model is constructed to determine the influence of chosen parameters from the surrounding environment and the gully pot itself. For the random part, a normal distribution is used. The factors which were deemed to be relevant (p>0.05) are the tree factor (p=2,39E-06), inlet structure (p=2,05E-05), sand trap volume (p=9,52E-03) and road surface (p=1,28E-02). Parameters deemed not to be relevant are connected surface area (p=4,39E-01) and traffic intensity (p=3,33E-01). In further research, rainfall intensities should be included as well.