Mapping of Urban Air Quality

Mapping of Urban Air Quality

CUSTOMER NEEDS

Urban planners, decision-makers and city leaders, as well as responsible officials of the environment or health care, need a comprehensive overview of the urban climate in order to understand the results of their urban policy, to assess the effects of air pollution, as well as to foresee the effects of new regulations. Networks of air quality measuring instruments have been established in major cities comprising a few measuring stations. They are means for alert and are also valuable to researchers and decision-makers for the analysis of air quality. Air quality is highly variable within a city. It is namely but not uniquely a function of the intensities of the air advection and turbulent diffusion around the pollution sources. These sources may have a large extension like for exhaust gas of vehicle motors. Loirestu'Air gather together different partners such as governmental organisation, local community, industrials, protection of environment association. It is responsible of the management and implementation of the survey network and is committed to survey air quality and to provide quantitative measurements of pollutants for Nantes and for the river estuary of the Loire. Loirestu'Air is therefore interested in a detailed mapping of the pollutants over the city of Nantes, which is not possible with its present means and resources. It is also willing to improve the forecast of the ventilation of the city for the dissemination of the pollutants.

METHODOLOGY

Among other tools such as in situ measurements and numerical modelling, remote sensing from space and from helicopter plays a major role. It offers an unique opportunity to collect various information in different wavelengths and above all at the different scales required for a comprehensive description of the urban system.

The investigation presented here particularly aims at the assessment of the potentialities of satellite imagery for the urban micro-climate and air quality. ERS SAR imagery may help in identifying urban morphological features and their typologies with relation to the air flow drag. SPOT and LANDSAT images are more related to the changes in visibility induced by air pollution. The selected site is the city of Nantes, along the Loire River and close to its mouth at the French Atlantic coast. The study of the city of Nantes has been undertaken for few years by means of ground measurements and surveys, aerial and satellite images as well as the creation of a data base using a geographical information system (GIS). Hence a clear picture of this urban system already exists which has guided the exploitation of the satellite data.

RESULTS

ERS SAR Imagery

The temporal variability of the SAR signal over the city has been assessed by analysing five images, which have been acquired on descending satellite passes. Beside the speckle effect, the quality of the signal is highly variable from one image to the other. Meteorological effects make contrasts between objects and their surroundings to be more or less pronounced. Urban features are mostly present in each image, but the structures are not always well perceived within a single image. It was concluded that it is necessary to have several images, whereby their redundancy allows a better exploitation of the urban features and decreases the level of speckle. Screening of the average SAR image (depicted in Figure 2) clearly indicates that the perception of the roads is highly dependent on the look direction of the SAR system.

The main factors for the perception of the morphological features are the height of the buildings, its orientation relative to the spacecraft orbit, its horizontal surface, its materials. The structure function or the multiresolution analysis, by means of wavelet transform,  provides a good discrimination between unbuilt areas, residential areas, industrial areas, and large groups of buildings. Several studies have demonstrated that urban morphological features and their typologies are well perceived in SAR imagery once properly processed. Thus, aerodynamic roughness parameters can be deduced. These parameters affect the air flow drag and accordingly the air ventilation of the streets, so the diffusion of pollutants. Maps of aerodynamic roughness derived from SAR data may serve as an input to numerical models for simulating and forecasting the diffusion of pollutants.
 

Optical imagery

Several satellites are carrying optical sensors. The SPOT series has a spatial resolution of 20 m and a revisit time of 26 days, the LANDSAT series has spatial resolution of 30 m for Band 1-5 and 7 and of 120 m Band 6 and a revisit time of 16 days, and different Russian missions of opportunity provide resolutions as good as 2 m with different revisit times. These sensors measure the energy reflected or emitted by objects on the ground in selected spectral bands, thus the optical properties can be deduced allowing the detection and recognition of specific objects. Since these sensors cover large areas they well suited for mapping purpose. However, the revisit period of about twenty days is too long for monitoring of a burst of pollution, but is still sufficient to build up a set of cases for designing and validating of models.

For LANDSAT data acquired on May 22, 1992, we found high correlation between black particle concentration and TM Band 6 data (thermal infrared) as depicted in Graph 1 and between TM Band 7 data (short-wave infrared) and nitrogen dioxide (NO2) concentration as depicted in Graph 2. The correlation coefficients are 0.96 and 0.98, respectively.

These preliminary results are very encouraging. The analysis of several cases will permit to conceive models explaining the perception of the pollution by satellite sensors. These models should take into account in a detailed manner the various optical properties of the urban landscape, which can be provided by specific processing of satellite or aircraft observations. Given a pollutant quantity, the models can compute the expected satellite signal while the "inverse" models will be used to retrieve pollutant quantities from space- or airborne imagery. Once applied to images periodically acquired over a city, these models will provide a detailed mapping of the air quality.

Presently, a simple model is used based on a linear regression between ground measurements and satellite observations. When applied to other measurements, this model provides estimates of the loading in black particulates (BP) with a relative RMS error of approximately 80%, which is still acceptable. Indeed, it is comparable to the RMS error obtained by interpolation techniques applied to the ground measurements. It offers the advantage of a better rendering of the local effects and of extrema. Figure 3 displays the map obtained by the linear regression method. It gives the BP concentration in µg/m³ with respect to the grey level (black: 0 µg/m³ up to white: 34 µg/m³).
 

Graph 2: Correlation between nitrogen dioxide concentration (NO₂ in µg/m³), as measured during the satellite overpass, and the satellite measurements in near infrared (about 2 µm, in arbitrary units) for May 22, 1992. The correlation coefficient is 0,98 and the confidence level is 0.95.
  Figure 3: Map of the loading of black particles over the city of Nantes. The grey level gives the concentration of the black particles contained in the air in µg/m³.

---

DELIVERABLES

Roughness map of towns as derived from SAR imagery (still under assessment) and the respective maps of pollutant concentrations are to be delivered from projects like these.

---

COST / BENEFIT

The price of a pollutant map is about 14 KECU and of a roughness map is about 13.5 KECU, both include the purchase of the respective satellite imagery.

---

CUSTOMER OPINION

The use of satellite data for the mapping of urban air quality is suitable and provides unique information. Indeed, the weak density of measuring stations does not allow an accurate mapping of pollutants, and urban roughness map is a tedious task when achieved by ground survey. Results are promising. Efforts are undergoing in order to increase the quality and accuracy of the maps of pollution elements.

---

REFERENCES

Customer References

Name:	Luc Lavrilleux
Company:	Loirestu'Air
Address:	4 rue Alfred Kastler  44070 Nantes cedex 3

Provider References
 

Name:	Ludovic Basly, François Cauneau, Thierry Ranchin, Lucien Wald
Company:	Ecole des Mines de Paris,  Groupe Télédétection & Modélisation
Address:	BP 207  06904 Sophia Antipolis cedex  France
Tel:	(33) 4 93 95 75 75
Fax:	(33) 4 93 95 75 35
E-mail:	wald@cenerg.cma.fr
URL:	http://www-cenerg.cma.fr/

Scientific References

Name:	Ludovic Basly, François Cauneau, Thierry Ranchin, Lucien Wald
Company:	Ecole des Mines de Paris,  Groupe Télédétection & Modélisation
Address:	BP 207  06904 Sophia Antipolis cedex  France
Tel:	(33) 4 93 95 75 75
Fax:	(33) 4 93 95 75 35
E-mail:	wald@cenerg.cma.fr
URL:	http://www-cenerg.cma.fr/

 

Modified on 06/27/2003 at 12:21 Creator: sophia webmaster