Wildfires In Portugal Research Articles

Space-time clustering analysis performance of an aggregated dataset: The case of wildfires in Portugal

This study focuses on the use of space–time permutation scan statistics (STPSS) to assess both the existence and the statistical significance of clusters on aggregated datasets. The investigated case study is represented from the Portuguese Rural Fire Database (PRFD) where the fire occurrences are georeferenced to an administrative unit level. The main goals are: (i) assessing the robustness of the STPSS to correctly detect clusters on aggregated datasets; (ii) testing the existence of space–time clustering in the PRFD; and (iii) characterizing the detected clusters. A synthetic database was designed to assess the potential bias introduced by aggregation of the data on the performance of the STPSS method. Results confirmed the ability of the STPSS to correctly identify clusters, regarding their number, location, and spatio-temporal dimensions and provided recommendations about the parameters setting of the scanning window. Finally, a discussion of the identified clusters on the PRFD is presented.

Impact of wildfire-induced land cover modification on local meteorology: A sensitivity study of the 2003 wildfires in Portugal

Wildfires alter land cover creating changes in dynamic, vegetative, radiative, thermal and hydrological properties of the surface. However, how so drastic changes induced by wildfires and how the age of the burnt scar affect the small and meso-scale atmospheric boundary layer dynamics are largely unknown. These questions are relevant for process analysis, meteorological and air quality forecast but also for regional climate analysis. Such questions are addressed numerically in this study on the case of the Portugal wildfires in 2003 as a testbed. In order to study the effects of burnt scars, an ensemble of numerical simulations using the Weather Research and Forecasting modeling system (WRF) have been performed with different surface properties mimicking the surface state immediately after the fire, few days after the fire and few months after the fire. In order to investigate such issue in a seamless approach, the same modelling framework has been used with various horizontal resolutions of the model grid and land use, ranging from 3.5km, which can be considered as the typical resolution of state-of-the art regional numerical weather prediction models to 14km which is now the typical target resolution of regional climate models.The study shows that the combination of high surface heat fluxes over the burnt area, large differential heating with respect to the preserved surroundings and lower surface roughness produces very intense frontogenesis with vertical velocity reaching few meters per second. This powerful meso-scale circulation can pump more humid air from the surroundings not impacted by the wildfire and produce more cloudiness over the burnt area. The influence of soil temperature immediately after the wildfire ceases is mainly seen at night as the boundary-layer remains unstably stratified and lasts only few days. So the intensity of the induced meso-scale circulation decreases with time, even though it remains until full recovery of the vegetation. Finally all these effects are simulated whatever the land cover and model resolution and there are thus robust processes in both regional climate simulations and process studies or short-time forecast. However, the impact of burnt scars on the precipitation signal remains very uncertain, especially because low precipitation is at stake.

Effects of regional climate change on rural fires in Portugal

CR Climate Research Contact the journal Facebook Twitter RSS Mailing List Subscribe to our mailing list via Mailchimp HomeLatest VolumeAbout the JournalEditorsSpecials CR 57:187-200 (2013) - DOI: https://doi.org/10.3354/cr01176 Effects of regional climate change on rural fires in Portugal Mário G. Pereira1,2,*, Teresa J. Calado2, Carlos C. DaCamara2, Tomás Calheiros2 1Centro de Investigação e de Tecnologias Agro-Ambientais e Biológicas (CITAB), Universidade de Trás-os-Montes e Alto Douro, Vila Real, Portugal 2Instituto Dom Luiz, IDL, Universidade de Lisboa, Lisboa, Portugal *Email: gpereira@utad.pt ABSTRACT: Wildfires are a major problem in Portugal. Since 1980 an area equivalent to 3/5 of the forested surface has burned. The aim of the study was to assess the potential impact of regional climate change on wildfires in Portugal using an appropriate Burnt Area Model (BAM). Based on multiple regression analysis, the model was able to estimate the decimal logarithm of the monthly burnt areas in July and August using the Daily Severe Rating as a predictor in the pre-fire season (May and June) and the fire season (July and August). The BAM, which was able to explain 63% of the total observed variance from the 1980-2011 period, was then fed with simulated data by a Global Climate Model (GCM) for the present climate and for two 30 yr periods (2051-2080 and 2071-2100) of future IPCC emission scenario B1. Comparison analysis between the logarithm of burnt area in July and August under present and future climate conditions shows an increase in the mean values of 7 and 11% for the first and second 30 yr periods, respectively, and a decrease of 32% in the SD for the first (2051-2080) period but no distinction between the observed and the simulated values for the last (2071-2100) period. Obtained estimates with the developed approach consistently point towards an increasing risk of fire under future climate conditions, and thus an increasing likelihood of much larger burnt areas. KEY WORDS: Climate change impacts · Climate scenario · Forest fire · Multiple regression · Global Climate Model · Model for Interdisciplinary Research on Climate · MIROC · Fire risk Full text in pdf format PreviousNextCite this article as: Pereira MG, Calado TJ, DaCamara CC, Calheiros T (2013) Effects of regional climate change on rural fires in Portugal. Clim Res 57:187-200. https://doi.org/10.3354/cr01176 Export citation RSS - Facebook - Tweet - linkedIn Cited by Published in CR Vol. 57, No. 3. Online publication date: October 08, 2013 Print ISSN: 0936-577X; Online ISSN: 1616-1572 Copyright © 2013 Inter-Research.

Emission factors and detailed chemical composition of smoke particles from the 2010 wildfire season

This paper complements the information previously published (Atmospheric Environment 45, 641–649) on gaseous and particulate emissions from wildfires in Portugal for summer 2009, in an attempt at obtaining more extensive, complete and representative databases on emission factors and detailed chemical characterisation of smoke particles. Here, emission factors for carbon oxides (CO2 and CO), total hydrocarbons (THC), fine (PM2.5) and coarse (PM2.5–10) particles obtained for fires occurring in Portugal in summer 2010 are presented. The carbonaceous content (OC and EC), water-soluble ions, elements and organic composition of smoke particles were, respectively, analysed by a thermal-optical transmission technique, ion chromatography, instrumental neutron activation analysis (INAA) and gas chromatography-mass spectrometry (GC–MS). The particle mass concentrations were in the ranges 0.69–25 mg m−3 for PM2.5 and 0.048–3.1 mg m−3 for PM2.5–10. PM2.5 particles represented 91 ± 5.7% of the PM10 mass. The OC/EC ratios in fine and coarse particles ranged from 2.5 to 205 and from 1.7 to 328, respectively. The water-soluble ions represented, on average, 3.9% and 2.8% of PM2.5 and PM2.5–10 mass, respectively. In general, the dominant ions in the water extracts were Na+, NH4+, Cl− and NO3− for the PM2.5 fraction, and K+, Mg2+, Ca2+ and SO42− for the PM2.5–10 fraction. The K+/EC and K+/OC ratios obtained in this study were, on average, 0.22 ± 0.23 and 0.011 ± 0.014 for PM2.5 and 0.83 ± 1.0 and 0.024 ± 0.023 for PM2.5–10 particles, respectively. The K+/levoglucosan ratio was, on average, 2.0 for PM2.5 and 3.1 for PM2.5−10 particles. Levoglucosan was detected at mass fractions of 1.6–8.7 mg g−1 OC in PM2.5 and 2.7–56 mg g−1 OC in PM2.5–10. The dominant elements detected in the smoke samples were Na, Br, Cr, Fe, K, Rb and Zn. The most representative organic constituents in the smoke samples were acids, alcohols, terpenoid-type compounds, sugars and phenols, in both size fractions.

Atmospheric emissions from vegetation fires in Portugal (1990–2008): estimates, uncertainty analysis, and sensitivity analysis

Abstract. Atmospheric emissions from wildfires in Portugal were estimated yearly over the period 1990–2008 using Landsat-based burnt area maps and land cover maps, national forest inventory data, biometric models, and literature review data. Emissions were calculated as the product of area burnt, biomass loading per unit area, combustion factor, and emission factor, using land cover specific values for all variables. Uncertainty associated with each input variable was quantified with a probability density function or a standard deviation value. Uncertainty and sensitivity analysis of estimates were performed with Monte Carlo and variance decomposition techniques. Area burnt varied almost 50-fold during the study period, from about 9000 ha in 2008 to 440 000 ha in 2003. Emissions reach maximum and minimum in the same years, with carbon dioxide equivalent (CO2eq.) values of 159 and 5655 Gg for 2008 and 2003, respectively. Emission factors, and the combustion factor for shrubs were identified as the variables with higher impact on model output variance. There is a very strong correlation between area burnt and emissions, allowing for good emissions estimates once area burnt is quantified. Pyrogenic emissions were compared against those from various economy sectors and found to represent 1% to 9% of the total.

Characterization of wildfires in Portugal

Forest fires severity has increased in Portugal in the last decades. Climate change scenarios suggest the reinforcement of this severity. Forest ecosystem managers and policy-makers thus face the challenge of developing effective fire prevention policies. The characterization of forest fires is instrumental for meeting this challenge. An approach for characterizing fire occurrence in Portugal, combining the use of geographic information systems and statistical analysis techniques, is presented. Emphasis was on the relationships between ecological and socioeconomic features and fire occurrence. The number and sizes of wildfires in Portugal were assessed for three 5-year periods (1987–1991, 1990–1994, and 2000–2004). Features maps were overlaid with perimeters of forest fires, and the proportion of burned area for each period was modeled using weighted generalized linear models (WGLM). Descriptive statistics showed variations in the distribution of fire size over recent decades, with a significant increase in the number of very large fires. Modeling underlined the impact of the forest cover type on the proportion of area burned. The statistical analysis further showed that socioeconomic features such as the proximity to roads impact the probability of fires occurrence. Results suggest that this approach may provide insight needed to develop fire prevention policies.

Summer 2009 wildfires in Portugal: Emission of trace gases and aerosol composition

Abstract In summer 2009, emissions of trace gases and aerosols from several wildfires occurring in Portugal were sampled. A portable high-volume sampler was used to collect sequentially, on quartz fibre filters, coarse (PM2.5–10) and fine (PM2.5) smoke particles. Tedlar air sampling bags have been used for complementary chemical characterisation of the gaseous phase. The carbonaceous content (elemental and organic carbon, EC/OC) of particulate matter was analysed by a thermal-optical transmission technique. The levels of almost 50 elements were quantified by inductively coupled plasma-mass spectrometry. The water-soluble ions were obtained by ion chromatography. Emission factors of species that are favoured by the smouldering phase (e.g. CO) were above the values reported in the literature for biomass burning in other ecosystems. The CO emission factors were 231 ± 117 g kg−1 biomass (dry basis) burned. Emissions of compounds that are promoted in fresh plumes and during the flaming phase, such as CO2, were generally lower than those proposed for savannah and tropical forest fires. The CO2 emission factors ranged from about 1000 to 1700 g kg−1. Total hydrocarbons, PM2.5, PM10 and OC presented variable emissions, but in general substantially higher than values reported for wildfires in African and Amazonian biomes. The emission factors obtained in Portugal were as follows (in g kg−1 biomass, dry basis): 6–350 for total hydrocarbons, 0.5–42 for PM2.5, 1–60 for PM10, and 0.2–42 for OC (in PM10). The organic carbon-to-elemental carbon ratios measured in the present study largely exceeded those obtained by other researchers. The aerosol mass was dominated by organic matter (OC/PM2.5 = 50 ± 18%, OC/PM2.5–10 = 36 ± 18%). The metal elements represented, on average, 1.23 and 0.91%, while the measured water-soluble ions accounted for 2.6 and 2.1% of the PM2.5 and PM2.5–10 mass, respectively. Carbonates accounted for 0.15–3.1% (average = 0.83%) of PM2.5–10. The dominant elements were B, Ti, Mn, Cu, Zn, Zr and Ba.

SPATIAL DISTRIBUTION OF HEAVY METALS RELEASED FROM ASHES AFTER A WILDFIRE

Heavy Metals (HM) in great amounts in soil and water resources can cause coercive effects in the environment and human health. Ash contains high quantities of HM that depend on combusted species, type and part burned and soil characteristics. After a fire, the HM released from ash can lead to a soil solution and water resources contamination. This liberation of HM in solution can be highly variable across the affected area. This work pretends to study the spatial variation of the HM ‐ Aluminium (Al3+), Manganese (Mn2+), Iron (Fe2+) and Zinc (Zn2+) ‐ in a Quercus suber and Pinus pinaster stand affected by a wildfire in Portugal, applying some interpolation methods. The results showed that on average across the plot, Al3+ was the HM released in higher quantities and Zn2+ in lower. The higher variability was observed in Zn2+ and in Fe2+. The interpolation methods assessed showed that polynomial regression (PR) method was the more accurate to predict the distribution of the HM across the plot. Al3+ and Mn2+ showed a rise in their concentration from south towards north section of the plot, and Fe2+ and Zn2+ a decrease from northwest to southeast section of the plot. The liberation of Al3+and Mn2+ is related with species burning severity, and Fe2+ and Zn2+ with plot topography. The fire evolution across the plot and the consequent rising temperatures can have higher impacts than burned species in HM spatial variability. Over time, with the decreasing ash pH, HM will become more mobile and will be released in soil solution, with potential coercive effects in the environment. Santrauka Dideles sunkiuju metalu (SM) koncentracijos dirvožemyje ir vandenyje gali sukelti pavoju aplinkai ir žmogaus sveikatai. Pelenuose aptinkamos dideles SM koncentracijos, kurios priklauso nuo sudegusio medžio rūšies, tipo, sudegusios medžio dalies ir dirvožemio savybiu. Po gaisro iš pelenu pasišalinantys SM patenka i dirvožemio tirpala ir gali sukelti vandens ištekliu tarša. SM išsiskyrimas iš pelenu teritorijos plote gali kisti. Darbe taikant interpoliacinius metodus tiriama SM — aliuminio (Al3+), mangano (Mn2+), geležies (Fe2+) ir cinko (Zn2+) ‐ pasiskirstymas teritorijoje po Quercus suber ir Pinus pinaster miško gaisro Portugalijoje. Iš rezultatu matyti, kad nagrinejamoje teritorijoje vidutiniškai didžiausios koncentracijos pasiskirste Al3+, o mažiausios ‐ Zn2+. Didžiausi koncentraciju pokyčiai buvo būdingi Zn2+ ir Fe2+. Iš taikytu interpo‐liaciniu metodu, prognozuojant SM pasiskirstyma teritorijoje, tikslesnis buvo polinomines regresijos metodas. Al3+ ir Mn2+ koncentraciju padidejimas nustatytas iš pietines teritorijos dalies šiaurinCs link, o Fe2+ ir Zn2+ ‐ sumažejimas iš šiaures vakaru dalies pietrytines teritorijos dalies link. Al3+ ir Mn2+ išsiskyrimas iš pelenu priklause nuo gaisro intensyvumo, o Fe2+ ir Zn2+ ‐ nuo teritorijos topografijos. Gaisro intensyvumo pokyčiai teritorijoje ir su tuo susijes temperatūros kilimas gali tureti didesnes itakos SM pasiskirstymui teritorijoje negu sudegusiu medžiu rūšys. Laikui begant ir kintant pelenu pH, SM tampa judresni, tad gali patekti i dirvožemio tirpala, sukelti pavoju aplinkai. Резюме Большие концентрации тяжелых металлов (ТМ) в почве и воде могут быть oпасны для окружающей среды и здоровья людей. Большие концентрации ТМ, выделяемые после пожара из золы, зависят от вида, типа, части сгоревшего дерева и свойств почвы. ТМ попадают в почву и могут вызвать загрязнение подземных вод. Выделение ТМ из золы на площади территории может меняться. В статье рассматривалось распределение после пожара на территории в Португалии таких тяжелых металлов, как алюминий (Al3+), марганец (Mn2+), железо (Fe2+) ir цинк (Zn2+). С этой целью применялись методы интерполяции. Из результатов анализа видно, что на исследуемой территории в среднем наибольшими были концентрации Al3+, а наименьшими Zn2+. Наибольшим изменениям подвергались концентрации Zn2+ и Fe2+. Из применявшихся методов интерполяции метод полиноминальной регрессии был наиболее точным для прогноза распределения ТМ на территории. Для Al3+ и Mn2+ было характерно увеличение концентраций с южной территории в сторону северной части, а для Fe2+ ir Zn2+ – уменьшение концентраций с северо-западной части в сторону юго-восточной территории. Выделение Al3+ и Mn2+ из золы после пожара зависело от интенсивности пожара, а в случае с Fe2+ и Zn2+ – от топографии территории. Изменение интенсивности пожара на территории и в связи с этим изменение температуры может оказывать бόльшее влияние на распределение ТМ на территории, чем вид сгоревших деревьев. С течением времени и изменением pH золы ТМ становятся более подвижными, могут попадать в подземные воды и тем самым создавать угрозу для окружающей среды.