Dépôt numérique
RECHERCHER

The use of a land use regression model to predict NO2 air pollution in two small areas of Auckland

Kingham, Simon; Pattinson, W.; Pearson, Amber; Longley, I.; Campbell, Malcolm; Apparicio, Philippe (2013). The use of a land use regression model to predict NO2 air pollution in two small areas of Auckland Air quality and climate change , vol. 47 , nº 3. p. 40-44.

Ce document n'est pas hébergé sur EspaceINRS.

Résumé

Land-use regression (LUR) modelling is widely used in many parts of the world, especially in Europe and North America. Only one published study has previously been carried out nationally in New Zealand for PM10; none for NO2. One previously identified limitation of LUR has been its limited effectiveness in small areas at fine spatial scales. In this study, LUR was used in two small areas of Auckland: Otahuhu and Mangere. Adjusted R2 figures of 0.72 (Mangere) and 0.44 (Otahuhu) were comparable with other studies. The low R2 found for Otahuhu may relate to the size of the monitoring area (3.5 km2), whereas the Mangere area was ten times larger (35 km2), while the extent of the modelled area was larger again: 10 km2 at Otahuhu and 70 km2 at Mangere. From these findings, we conclude there may be a lower limit on the size of study area used in LUR modelling, and that this area is somewhere in the range of 4-30 km2. It is hypothesised that in smaller areas there is not enough variability in the local geographic features used to as accurately predict the monitored NO2 values.

Type de document:
Mots-clés libres:
Centre: Centre Urbanisation Culture Société
Date de dépôt: 22 oct. 2013 20:09
Dernière modification: 11 janv. 2017 18:51
URI: http://espace.inrs.ca/id/eprint/1649

Actions (Identification requise)

Modifier la notice Modifier la notice