Urban water

Introduction

Urbanization and the induced reduction of vegetation cover, sealing of surfaces and constructions of buildings and roads alter the infiltration, the interception and the depression storage, the surface runoff and the evapotranspiration of urban catchments (Figure 1). These changes decrease the amount of groundwater recharge and base-flow, and increase the frequency and magnitude of storm flows. An adequate regulation and management of urban water demands an integrated approach.

Figure 1: Urban environment alters water flow.

Urban surface infrastructure in combination with man-made drainage change pathways of precipitation, and aging infrastructure as well as flooding incidents and pollution of receiving water bodies deteriorate groundwater and surface water quality. Further, man-made infrastructure, such as roofs or urban pavements, has traditionally been designed to have no or low depression storage and even small storm events quickly generate surface flow. To reduce storm water runoff at the source, precipitation retention and loss through infiltration and evapotranspiration should be encouraged. Urban vegetation as well as some man-made surfaces, due to their interception and adsorption capacities, reduce and/or delay water flow and increase the evaporation and infiltration rate. This points out the importance of emphasizing more on sustainable planning strategies (source control) to avoid natural disasters and safeguard the living quality in our future cities.

To improve urban water management, it is important to increase the accuracy of hydrological modeling by a better quantification of the processes that influence the water balance of urban catchments. A way to achieve this is through the use of hyperspectral remote sensing (RS) data. The latter data are characterized by a high spectral and spatial resolution, and thus allow for a better characterization of the urban surface and the vegetation properties than the traditional RS data. This leads to an improved parameterization of our hydrological simulation tools and allows to create reliable ecosystem service indicator maps to support urban water regulation policies (Figure 2).

Figure 2: Urban water regulation

Objectives

Our general aim is to determine the total storage, retention and evaporative capacity of the urban ecosystem in a detailed and spatially distributed way. The hydrological indicator maps will then help to analyze the impact of different urban planning scenarios on the water retention capacity of the urban ecosystem.

More specifically, we intend to explore the potential of high resolution hyperspectral and LiDAR data  at a micro-catchment scale to:

  • Correlate vegetation indices from remote sensing data with in situ measurements.

  • Improve the parameterization of urban hydrological models.

 

In order to assess the hydrological response at a city-wide scale we use a spatial metric approach and  lower resolution multispectral data. The main challenge when working with low resolution data is to account for small-scale hydrological processes that influence the livability of a city. Therefore we relate spatial metrics to hydrological processes based on the results of the high-resolution hydrological simulations. Finally we:

  • Explore the impact of different urban planning scenarios on the water retention at a city-wide scale.  

 

 

Our approach

We characterize the land surface cover at a high spatial resolution (2m) with an airborne hyperspectral APEX (Airborne Prism Experiment) image, while the new daily Proba-V (Project for OnBoard Autonomy – Vegetation) products (100m) are used for a detailed characterization of the seasonal variation of urban green. The Proba-V satellite, which provides daily vegetation data with a 100m resolution as NDVI maps contributes to an improved monitoring of the seasonal variation of vegetation characteristics. The Normalized Difference Vegetation Index (NDVI) is a numerical indicator that uses the visible and near-infrared bands of the electromagnetic spectrum to identify the amount of living green vegetation. Based on the high spatial resolution of the airborne image and the high temporal resolution of Proba-V we can create distributed leaf area index (LAI) maps to simulate interception storage in urban areas. The process-based and spatially distributed WetSpa-Python model is used to simulate the water balance at a 2m resolution for the Watermaelbeek catchment in Brussels. The input maps as well as the hydrological processes are validated with ground-truthing experiments at specific locations (Figure 3).

Figure 3: Methodology for high resolution modelling

To respond to urban water management and urban planning strategies it is important to assess the hydrological response at a city wide scale (Figure 4). Therefore we consider different temporal scales:

1.            Peak events: to manage flooding in urban areas, high-runoff events have to be analysed. Therefore, we have to simulate at a high temporal resolution (minutes) and focus on a limited number of peak flow events to assess the interception storage and runoff capacity.

2.            Seasonal water balance: to assess potential effects of evapotranspiration (e.g. cooling) and analyse the liveability of a city considering the urban water fluxes a monthly or seasonal water balance simulation is needed (using hourly or daily time-steps and considering seasonal and annual water balances).

To upscale the hydrological response from the micro-catchment (2m resolution) to the city-wide scale (20 -30m resolution) we use spatial metrics and relate them to hydrological processes. At a sub-catchment and building block resolution, urban structure metrics can be added (such as degree of fragmentation - private/public domain, connectivity - open/closed building blocks, etc.) in order to assess the different types of urban planning and achieve more realistic upscaling results.

Figure 4: Upscaling from micro-catchment to city-wide scale

Publications

Wirion Ch., Nguyen Ho Khanh, Bauwens W.  & Verbeiren B. (2015). Using remote sensing to describe surface properties for improved hydrological modeling. Proceedings UDM, 20th-23rd September 2015, Quebec (Canada), 9pp.

Verbeiren B., Khanh N.H., Wirion Ch., & Batelaan O. (2016). An Earth observation based method to assess the influence of seasonal dynamics of canopy interception storage on the urban water balance. BELGEO online 2/2016.

Wirion Ch., Suliga J., Van Griensven A., Bauwens W., Verbeiren B. (2016). Monitoring of vegetation changes for hydrological modelling in urban and wetlands areas based on Proba-V derived LAI. Proceedings Proba-V symposium, 26-28 January 2016, Ghent (Belgium).

Wirion Ch., Salvadore E., Bauwens W., Verbeiren B., (2016). Simulating urban interception based on detailed in-situ monitoring and remote sensing data. Proceedings Novatech, 28/06-01/07/2016, Lyon (France).

Wirion Ch., Bauwens W., Verbeiren B., (2017). High resolution modeling of the urban hydrological response. IEEE Conference proceedings (abstract + oral presentation). JURSE 2017 Dubai.

Wirion, C., Bauwens, W., Verbeiren, B. (2017). Location and time specific hydrological simulations with multi-resolution remote sensing data in urban areas. Remote Sensing, 9(7), 645.

Wirion, C., Bauwens, W., Verbeiren, B. (2017). Water balance simulation of the urban surface supported by remote sensing. Proceedings YWP , 5-7/07/2017, Ghent (Belgium).

 

Wirion, C., Bauwens, W., Verbeiren, B. (2017). Simulating the urban surface water balance based on detailed hyperspectral and frequent multispectral remote sensing data. Proceedings Multitemp conference, 27-29/06/2017, Bruges (Belgium).

This project is funded by the Belgian Federal Science Policy (Belspo) within the RESEARCH PROGRAMME FOR EARTH OBSERVATION - “STEREO III”.

© 2015 UrbanEARS