Community Research and Development Information Service - CORDIS

Periodic Report Summary 1 - STRPM (Assessing Temporally and Spatially Resolved PM2.5 and AirTemperature Exposures For Epidemiological Studies Using SatelliteBased Methods)

Assessing Temporally and Spatially Resolved PM2.5 and Air Temperature Exposures For Epidemiological Studies Using Satellite Based Methods:

In this research project I aimed at a) developing better statistical exposure assessment methods to better handle currently exposure datasets, which are misaligned in both time and space. We developed and validated computationally efficient models that now allows us to accurately estimate PM2.5, PM10 and temperature at a very high spatial (1×1 km) and temporal (daily) resolutions across Italy and France. b) Make use of these generated estimations in a study assessing the effects of maternal exposure to PM2.5 and temperature on fetal growth. This 3rd aim is in progress.

The research has advanced substantially over the first tow years. We have recruited a MA student (Adar Rozenfeld), a PhD student (Maayan Yitshak-Sade) and a post-doc (Meytar Sorek Hamer) all working on the various project aims. We have started with aim 1 and aim 2 (The development and validation of estimation models between the years 2000-2013 across Italy and France) and have finished these aims in accordance of the proposed timetable from the work plan. We have finished the gathering of the spatial, temporal and satellite data. In order to get ready for model runs we needed to initially prepare the spatio-temporal predictor databases used in the models (Spatial predictors, meteorological data, NDVI, land use etc.). This complex process was carried out during the first 10 months of the project. In addition during the same time period we acquired and processed all satellite data (MODIS satellite- AQUA and TERRA platforms between the years 2000-2013) and validated all the datasets.
We followed this data base preparation stage with initial model calibration and fitting (again following the timetable from the work plan). We ran separate PM2.5/10 and air temperature models both for France and Italy. We introduced new statistical and geo-statistical methodologies in the calibration stage (such as new interpolation techniques and region specific model fitting). This model runs resulted in generation of estimated exposures for each day for each grid cell in each country. In addition, we have prepared the findings and methodology for publication in 3 separate papers (one for the PM model and two for the temperature model in France and Italy) which should be submitted in the following weeks (the France paper has been accepted for publication).
All models performances were excellent. For example in France, model performance was excellent for both days with available satellite temperature and days without satellite temperature observations (overall means out-of-sample R2=0.955 and R2=0.946 respectively). We demonstrate in France how satellite temperature can be used reliably to predict daily temperature at high resolution across France for use in multiple studies looking at the effects of fine resolution temperature exposure on various health outcomes. In Italy, the PM models displayed good fitting, with mean CV R2=0.64 and little bias (average slope of predicted VS observed PM10 ~ 0.95). We show how, for the first time, we are able to reliably estimate long-term and short-term effects of air pollution nationwide, even in areas lacking monitoring data.
We are now moving on towards working on aim 3 of conduct a nationwide study in France to estimate the association between exposure to PM2.5 and air temperature and fetal growth.

This study results is advancing European environmental exposure assessment by producing high resolution spatio-temporaly resolved exposure models. These models now allow us to estimate both short and long term exposure effects in both urban and rural areas, reducing exposure measurement error and provide a sound epidemiological base for the magnitude of risks in future studies.

Reported by



Life Sciences
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top