A. Carvalho

Fire activity in Portugal and its relationship to weather and the Canadian Fire Weather Index System

The relationships among the weather, the Canadian Fire Weather Index (FWI) System components, the monthly area burned, and the number of fire occurrences from 1980 to 2004 were investigated in 11 Portuguese districts that represent respectively 66% and 61% of the total area burned and number of fires in Portugal. A statistical approach was used to estimate the monthly area burned and the monthly number of fires per district, using meteorological variables and FWI System components as predictors. The approach succeeded in explaining from 60.9 to 80.4% of the variance for area burned and between 47.9 and 77.0% of the variance for the number of fires; all regressions were highly significant (P < 0.0001). The monthly mean and the monthly maximum of daily maximum temperatures and the monthly mean and extremes (maximum and 90th percentile) of the daily FWI were selected for all districts, except for Braganca and Porto, in the forward stepwise regression for area burned. For all districts combined, the variance explained was 80.9 and 63.0% for area burned and number of fires, respectively. Our results point to highly significant relationships among forest fires in Portugal and the weather and the Canadian FWI System. The present analysis provides baseline information for predicting the area burned and number of fires under future climate scenarios and the subsequent impacts on air quality.

Procedures for estimation of modelling uncertainty in air quality assessment

The main objectives of this work focus, firstly, on a review of the current existent methodologies to estimate air quality modelling uncertainty, and, secondly, in the preparation of guidelines for modelling uncertainty estimation, which can be used by local and regional authorities responsible for air quality management. From the application exercise, it was concluded that it is possible to define a subset of statistical parameters able to reproduce the general uncertainties estimation. Concerning the quality indicators defined by EU directives, the results show that the legislated uncertainty estimation measures are ambiguous and inadequate in several aspects, mainly in what concerns the error measures for hourly and daily indicators based on the highest observed concentration. A relative error at the percentile correspondent to the allowed number of exceedances of the limit value was suggested and tested, showing that is a more robust and appropriate parameter for model performance evaluation.

Regional-scale weather patterns and wildland fires in central Portugal

The characteristic evolution of the synoptic- and meso-scale wind, temperature and humidity pattern during wildland fire events in Portugal was determined by lagged covariances for the period 1980 to 2001. The daily burnt area was chosen as the parameter to be correlated with atmospheric fields provided by the European Centre for Medium-Range Weather Forecasts Reanalysis (ERA40) data. The related time series consisted of wildland fires with a daily burnt area of more than 500 ha in central Portugal. Five days in advance of a fire event, a strong positive anomaly existed in the surface pressure and in the 500-hPa geopotential field, both appearing to the west of the Iberian Peninsula and moving towards Brittany until lag zero. In advance of the fire event, the flow above Portugal came from the north, turning to easterlies at lag time zero and finally coming from the south-east during the post-event phase. Surface wind statistics taken at Castelo Branco supported these results. Smoke plumes from wildland fires detected by satellites indicated a similar flow structure, which was quite different to the averaged summertime flow above Portugal. Cross-covariance regression between the Iberian thermal low and burnt area showed that the peak amount of burnt area occurred up to 3 days after the appearance of a thermal low. This suggested that in the pre-phase of a wildland fire, heated air is transported from the peninsula’s centre towards Portugal.

Impact of forest fires on particulate matter and ozone levels during the 2003, 2004 and 2005 fire seasons in Portugal.

The main purpose of this work is to estimate the impact of forest fires on air pollution applying the LOTOS-EUROS air quality modeling system in Portugal for three consecutive years, 2003-2005. Forest fire emissions have been included in the modeling system through the development of a numerical module, which takes into account the most suitable parameters for Portuguese forest fire characteristics and the burnt area by large forest fires. To better evaluate the influence of forest fires on air quality the LOTOS-EUROS system has been applied with and without forest fire emissions. Hourly concentration results have been compared to measure data at several monitoring locations with better modeling quality parameters when forest fire emissions were considered. Moreover, hourly estimates, with and without fire emissions, can reach differences in the order of 20%, showing the importance and the influence of this type of emissions on air quality.

Plans and Programmes to improve air quality over Portugal: a numerical modelling approach

PM10 limit values have been surpassed in the northern region of Portugal, and, according to European legislation, Plans and Programmes (PP) must be designed and implemented to reduce those PM10 levels. The Air Pollution Model (TAPM) was applied over the study region, to the year 2004, for the reference situation with the current PM10 emissions; and, for the reduction scenario with PM10 emissions re-estimated considering the implementation of all measures, in order to investigate the impact on air quality of all PP measures. The results point to a 19% reduction in the PM10 levels over the Porto urban area.

Particulate matter and health risk under a changing climate: assessment for Portugal

The potential impacts of climate-induced changes in air pollution levels and its impacts on population health were investigated. The IPCC scenario (SRES A2) was used to analyse the effects of climate on future PM10 concentrations over Portugal and their impact on short-term population exposure and mortality. The air quality modelling system has been applied with high spatial resolution looking on climate changes at regional scale. To quantify health impacts related to air pollution changes, the WHO methodology for health impact assessment was implemented. The results point to 8% increase of premature mortality attributed to future PM10 levels in Portugal. The pollution episodes with daily average PM10 concentration above the current legislated value (50 μg·m−3) would be responsible for 81% of attributable cases. The absolute number of deaths attributable to PM10 under future climate emphasizes the importance of indirect effects of climate change on human health.

Validation of two Photochemical Numerical Systems Under Complex Mesoscale Circulations

Portugal has a quite extensive coastline associated with significant terrain features and sea/land breeze circulation, which results in a complex wind field with strong implications for the production and transport patterns of photochemical pollutants with a typical time-scale larger than 24 hours. Since the coastal zone is a major component in global budgets and global resources availability and utilisation, effective sustainable management strategies in such areas require a special ability to think beyond divisions between different types of resources and human activities. Portuguese littoral is the region with the highest population density and economic development. The resulting pollutant emissions place a high pressure on the environment and natural resources of the Portuguese urban coastal areas.

Forest Fires Impact on Air Quality over Portugal

The main purpose of this work is to estimate the air pollution effects of 2003 forest fires through the application of two air quality modelling systems (CHIMERE and LOTOS-EUROS) over Portugal and its intercomparison. Forest fire emissions were estimated based on specific southern European emissions factors, on type of vegetation and area burned, and incorporated in the emission input data of both modelling systems. Results showed a significant performance improvement when forest fires are taken into account. PM10 and O3 values can reach differences in the order of 30%, showing the importance and the influence of this type of emissions from local to regional air quality. The different results of the two models may give an indication of the uncertainty associated by using different models to investigate the impact of forest fires. Historical datasets of area burned, number of fires and air quality data were evaluated from 1995 to 2005 aiming to investigate a potential relationship between forest fire activity and air pollutants concentrations. The obtained results point to statistically significant correlations between fire activity in Portugal and PM10 and O3 levels in the atmosphere.

Ensemble Techniques to Improve Air Quality Assessment: Focus on O3 and PM

Five air quality models were applied over Portugal for July 2006 with an ensemble purpose. These models were used, with their own meteorology, parameterizations, boundary conditions and chemical mechanisms, but with the same emission data. The validation of the individual models and its ensemble for ozone (O3) and particulate matter was performed using monitoring data from 22 background stations over Portugal. After removing the bias from each model, different ensemble techniques were applied and compared. Besides the median, several weighted ensemble approaches were tested and intercompared: static (SLR) and dynamic (DLR) multiple linear regressions (using less-square optimization method) and the Bayesian Model Averaging (BMA) methodology. The goal of the comparison is to estimate to what extent the ensemble analysis is an improvement with respect to the single model results. The obtained results revealed that no one of the 4 tested ensembles clearly outperforms the others on the basis of statistical parameters and probabilistic analysis (reliability and resolution properties). Nevertheless, statistical results have shown that the application of the weights slightly improves ensemble performance when compared to those obtained from the median ensemble. The same statistical analysis together with the probabilistic measures demonstrates that the SLR and BMA methods are the best performers amongst the assessed methodologies.

Forest Fires And Air Pollution:A Local And A Global Perspective

A numerical system developed to simulate the effects of forest fires in the air quality is presented. This system integrates several related features, like emissions and progression of forest fires, and atmospheric flow. The interaction between the fire and the atmospheric flow considered as crucial is also taken into account in the system. Aiming to estimate the contribution of forest fires to the air pollution, two study cases (a local and a global one) were analysed. The first one is related to urban/wildland forest fires. The Etoile fire, occurred during the summer of 97 near Marseille in France was studied attempting to estimate its effects on the air quality of the city. Results indicate a considerable impact on the quality of the air of the suburban area of Marseille. Concerning the global scale impact of forest fires, greenhouse gases (GHG) emissions from Brazilian fires in 1997 were calculated and compared to the annual global GHG anthropogenic emissions. The contribution of Brazilian fires to the greenhouse effect is significant. An analysis of forest fires GHG emissions, oriented toward the problem of global warming and the goals of Kyoto Protocol, is also presented.