The overall goal of the research activities linked to the BELAIR 2015 campaign is the spatiotemporal estimation and monitoring of water and energy fluxes in urban areas.

The main focus in the BELAIR 2015 campaign was the validation of physical and thermal characteristics of urban surfaces and vegetation, at high and medium spatial resolution. The main idea is that these will contribute to an improved distributed hydrological parameterisation (high and medium resolution) for modeling purposes. In addition it should serve a surface energy balance modeling for hydrological validation and estimation of surface urban heat island intensity and its relation to urban form.

Executive summary

While more than two thirds of the world population are expected to live in cities by 2050, the attention given to urban ecosystems is relatively modest, compared to other types of ecosystems, like e.g. wetlands and forests.

Urban ecosystems are characterized by a strong dominance of the built environment, fragmenting green and blue spaces, consisting of parks, gardens, urban forests, green along the road network, rivers and ponds. Understanding the impact of spatially complex urban areas on ecosystem function and sustainability requires spatially distributed modeling of water and energy fluxes, based on a detailed parameterization of the built and nonbuilt urban environment. Remote sensing (RS) is indispensable to reach this objective. Next to the added value of an improved spatial characterization of urban areas, RS also offers the possibility for temporal monitoring (change detection).

The RS based spatiotemporal parameterization of urban areas for distributed water and energy balance modeling aimed at is targeting both urban surfaces and vegetation.

Relevant parameters include:

1. Urban surfaces: impervious surface fraction, roughness & adsorption properties, thermal properties, etc.

2. Urban green: vegetation type, vegetation fraction, leaf area index (LAI), interception storage capacity, wetness, etc.

Information on urban surface types, moisture content and imperviousness can be directly obtained from imaging spectroscopy. Previous experience has shown that shortwave infrared data, especially in the region of 1.31.9 μm is particularly valuable in separating different ecotopes. Hyperspectral imagery providing spectral profiles from 0.4 to 2.5 μm would be ideal for accurate landcover classification with reference to moisture content and for identifying specific characteristics of urban surfaces.

In particular, it will be investigated how parameters derived from hyperspectral sensing can contribute to the physical (energy balance) modelling of evapotranspiration (ET), being the largest water balance component. Novel is that ET energy balance modelling has so far been seldom applied in urban areas and neither with input of hyperspectral data. RS derived ET will be used in combination with spatially distributed hydrological modeling for improving the estimation of the different water balance components (runoff, recharge, …).

Detailed information on the characterization of urban and nonurban surface types, obtained from highresolution hyperspectral remote sensing will be used, in combination with mediumresolution thermal data (Landsat, ASTER) and existing largescale databases on the structure of the environment (UrbIS, GRB), to analyze the relationship between land surface temperature at the level of urban neighborhoods and specific neighborhood characteristics, including the presence and spatial distribution of green and blue spaces, the type of building materials used, the spatial configuration and 3Dcharacteristics of the built environment. This should lead to recommendations for urban planning regarding low and high impact development.

The SONIA site covers an urban transect in the eastern part of the Brussels Capital Region, including the upstream portion of the Woluwe Basin and the Sonian Forest. Note that the flight lines were extended to the North in the 2015 campaign and that an extra flight line at the west of Brussels city centre (along canal zone) was acquired as well.

The BELAIR 2015 campaign serves for the validation of physical (LAI, moisture, reflectance spectra) and thermal characteristics of urban surfaces and vegetation, at field level (high spatial resolution). These are planned to be used for validation of RS derived parameter maps based on Landsat, ASTER and Proba-V imagery (medium spatial resolution). The data will also be used within the frame of the STEREO III UrbanEARS project (

Data acquisition

Activities in the BELAIR 2015 campaign for the SONIA site focused on the URBAN subsite only. No measurements on the FOREST subsite were performed.


Below follows a list of collected in-situ data for SONIA URBAN. Some of this in-situ data collection is a repetition of data collected within the frame of the previous BELAIR 2013 campaign.

  • Spectral reference measurements (ASD spectroradiometer), amongst other for atmospheric correction of airborne APEX imagery
  • Measurement of LAI inside a number of 100x100m validation blocks
  • Mapping of the leaf area index (LAI) of over 100 individual trees in the urban zone (SunScan instrument)
  • Sapflow measurements (transpiration estimation) of 3 individual trees
  • Characterisation of 8 grass reference plots: (1) Top-soil moisture content, (2) Spectra, (3) Biomass
  • Surface temperature of grass and urban surfaces (IR gun)

Meteorological measurements: precipitation (tip-bucket pluviometer), temperature and relative humidity (mobile HOBO logger)


Hyperspectral APEX images of the SONIA site were acquired on June 30th 2015. Quicklooks are available on

Remark: Within the frame of a EUFAR campaign in addition hyperspectral AHS images (including 10 thermal bands) over the SONIA site were acquired. This parallel campaign is part of the STEREO III UrbanEARS project ( and falls outside BELAIR activities, but both campaigns are obviously linked.  


Targeted “overlapping” satellite data are: Landsat, ASTER and Proba-V


Active team members in BELAIR 2015 SONIA campaign:

Boud Verbeiren, Charlotte Wirion, Christian Anibas and Gert Ghysels (Vrije Universiteit Brussel, HYDR), Frank Canters, Frederik Priem, Kasper Cockx, Jonas van Breedam, Sophie Mossoux and Sven Vanderhaegen (Vrije Universiteit Brussel, CGIS), Ben Somers, Jeroen Degerickx, Vincent Smets, Jan Vanstockem (KULeuven, FNL),Frieke Van Coillie and Marie-Leen Verdonck (Universiteit Gent, FORSIT).


Supporting team members:

Kathy Steppe and Wouter Maes (Universiteit Gent, Laboratory Plant Ecology), Kris Vandekerckhove (INBO).


 Photo gallery


Contact PI: