Comparación en la estimación de material particulado PM10 usando imágenes satelitales LANDSAT 7, LANDSAT 8 Y MODIS en Quito
The deterioration of air quality in recent years is one of the main problems that most cities already cause lung diseases, lung cancer to the exposed population and Quito is no exception. Through the application of tools such as remote sensing it was possible to estimate one of the main pollut...
| Autor Principal: | Torres Saquinga, Nelly Stefanía |
|---|---|
| Otros Autores: | Vivanco Pérez, Valeria Lizbeth |
| Formato: | bachelorThesis |
| Idioma: | spa |
| Publicado: |
2018
|
| Materias: | |
| Acceso en línea: |
http://dspace.ups.edu.ec/handle/123456789/16071 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: |
The deterioration of air quality in recent years is one of the main problems that most
cities already cause lung diseases, lung cancer to the exposed population and Quito is no
exception. Through the application of tools such as remote sensing it was possible to
estimate one of the main pollutants emitted into the air called particulate material (PM10),
monitored by the Quito Metropolitan Atmospheric Monitoring Network (REMMAQ)
through the automatic and passive stations located along the city.
We used satellite images provided by remote sensors (Landsat 7, Landsat 8 and
MODIS) during the period 2003-2017, as well as generating a series of environmental
indicators from the multispectral bands using the ArcGis 10.5 software. In addition,
predictive models were created from miles, both by generalized linear regression (GLM)
and geographically weighted regression (GWR) from quarterly averages. The stations of
Carapungo and Guamaní exceeded the WHO maximum limit of 50 µg/m3 due to the
predominance of public transport, the construction industry and activities associated with
industrial sources. To predict the PM10 a multivariate matrix was used for each sensor and
it was determined that the images provided by the “Landsat 8” sensor and the format using
a GWR model with n = 218 observations tests the best criterion of Akaike AIC = -12.73,
R2 Aj = 0.8339, and its residues meet the validation criteria, thus providing the best fit of
PM10. |
|---|