Application of Downscaled Meteorological Data in Hydrological Modelling to Assess the Impact of Climate Change on the Performance of Green Roofs

An increase in the occurrence of climate extremes has necessitated the integration of Nature-based solutions (NBS) such as green roofs into urban landscapes to help maintain hydrological balance. Green roofs are known to benefit biodiversity by adding vegetative spaces in urban areas and reducing the urban heat island effects. Runoff retention by green roofs helps delay the peak of the hydrograph, thereby preventing the overwhelming of drainage networks that often cause urban pluvial floods. Therefore, the design and planning of green roofs should be preceded by hydrological modelling studies to ensure their effectiveness against climate extremes that are highly likely in the future. As a high percentage of imperviousness generates quick hydrological responses from urban areas, it is necessary to perform hydrological modelling using fine-resolution meteorological data. This study proposes the downscaling of precipitation and temperature data from the climate model CMIP6 (SSP2-4.5 and SSP5-8.5) using the framework of Universal Multifractal (UM) theory.

Through UM, the meteorological fields can be characterised using two parameters: α (multifractality index) and  (the mean intermittency codimension). The UM parameters for precipitation and temperature fields were estimated from the observed data using Double Trace Moment analysis. The climate data for the future scenarios from the CMIP6 model were then downscaled to a 6-minute resolution using the estimated UM parameters, employing a double cascade simulation process. This methodology helps conserve the heterogeneity and intermittencies of the field while generating extreme events that are imperative for studying the performance of urban systems. Further, temperature data were used to generate evapotranspiration data using an empirical parameterisation specific to the regions considered in the study. All meteorological data generated at a 6-minute resolution were used as input in a hydrological model to assess the performance of green roofs.

The hydrological modelling was performed for five regions in France: Paris, Lyon, Marseille, Nantes, and Strasbourg. Each region has specific regulations to ensure that the performance of green roofs complies with “Zero-Emission” criteria. Zero-emission rules define reference rainfall events to be contained within green roofs, such that the runoff retention/detention, and discharge rates are within limits that are favourable for the developmental conditions of the region. Thus, the Zero-Emission Metric (ZEM) used to estimate the performance of green roofs was defined as the ratio of the number of reference rainfall events that comply with the zero-emission rules to the total number of reference rainfall events in the region. The reference rainfall events specific to the regions were generated by renormalizing the downscaled precipitation data. The observations from the study indicated a decrease in the return period of reference rainfall events in the future scenarios, implying an increase in their frequency of occurrence. The performance of green roofs was found to decrease for the future scenarios: SSP2-4.5 and SSP5-8.5, due to the emergence of frequent climate extremes in future. The insights from the study highlight the requirement for effective hydrological modelling studies using region-specific meteorological data at fine resolution to design NBSs that are resilient to future climate extremes.