VGTU talpykla >
Aplinkos inžinerijos fakultetas / Faculty of Environmental Engineering >
Konferencijų straipsniai >
Please use this identifier to cite or link to this item:
http://dspace.vgtu.lt/handle/1/890
|
Title: | Spatial Variation of Ash Total Carbon and Total Nitrogen after a Wildfire |
Authors: | Pereira, Paulo Ubeda, Xavier Martin, Deborah |
Keywords: | Total Carbon Total Nitrogen Wildfire Fire Severity Interpolation Methods Spatial Distribution |
Issue Date: | 2011 |
Publisher: | Vilniaus Gedimino technikos universitetas |
Citation: | Pereira, P.; Ubeda, X.; Martin, D. 2011. Spatial Variation of Ash Total Carbon and Total Nitrogen after a Wildfire, in The 8th International Conference “Environmental Engineering”: Selected papers. Ed. by D. Čygas, K. D. Froehner, May 19–20, 2011 Vilnius, Lithuania. Vilnius: Technika, 255–262. |
Abstract: | The present work aims to study the spatial variability of ash Total Carbon (TC%) and Total Nitrogen (TN%) after a wildfire occurred in Portugal. Ash TC% a mean of 26.72 % (±8.76 %) and ash TN% of 1.20 % (±0.32 %). We observed that the area recover very fast to wildfire and two years the great part of soil was cover by vegetation. The fire severity (assessed with ash color and CaCO3 content) was more related with ash TC%, than ash TN% due the different vulnerability and response of these two elements to fire temperature and as consequence severity. Previous to model data we tested their normality that only was achieved in the case of TC% after a neperian logarithmic transformation (ln). TN% respected gaussian distribution, however showed a high skewness, thus this variable was also ln transformed. Ash TC%(ln) presented an excellent spatial structure and the experimental model fits better with the rational quadratic model and ash TN%(ln) with the lineal model. Among all interpolation methods tested, the most accurate to interpolate ash TC%(ln) was IDW2 and the less suitable LP3. In relation to ash TN%(ln) the best interpolation technique was SPT and the less precise LP3. This allow us to observe differences that the spatial distribution of both variables is different, that is due the small scale variability observed in ash TC%(ln). In general the models were well performed because the residuals mean were closed to 0 and no differences between average of predicted and estimated values were identified. Ash TC%(ln) spatial distribution was in higher amounts in western, northern part and southeast and northeast corner of the plot and lower in the centre and south of the interested area and ash TN%(ln) was observed in higher amounts in the southwest and southeast part of the plot and lesser in the central part. |
URI: | http://dspace1.vgtu.lt/handle/1/890 |
ISBN: | 978-9955-28-831-2 |
ISSN: | 2029-7106 print 2029-7092 online |
Appears in Collections: | Konferencijų straipsniai
|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
|