Partenaires et organisations internationales
(Anglais)
|
Associação para o Desolvimento da Aerodinâmica Industrial - ADAI, Project Coordinator (P), Dep. of Geography, University of Alcala de Henares (E), Algosystems (G), Agroselviter, University of Turin (I), Dep. of Environment and Land Management, University of Aveiro (P), Dep. of Geography, University of Aix-Marseille (F), Dep. Ingenieria de Sistemas y Automática, University of Sevilla (E), CEREN (F), WSL (CH)
|
Résumé des résultats (Abstract)
(Anglais)
|
During the last year of the project the Spatial Data Handling Division focused on methodological aspects concerning wildfire occurrence and post fire risk analysis:
Spatio-Temporal Analysis of Wildfire Occurrence
The goal of this study was to find temporal and spatial patterns in wildfire occurrence in order to support forest fire management and prevention. To the INFLAME project, the methodological aspects concerning spatial and statistical analysis were of special interest. The study area was the whole canton of Grison (CH). Together with the Forest Service of the canton of Grison (CH) a digital database was created based on wildfires occurring between 1980 and 1997. Till today, 224 events with sufficient data quality could be processed. For every wildfire the ignition location, cause, date, time, type of wildfire, damage and burned area were assembled. Additionally, spatial attributes (slope, exposition, distance to infrastructure facilities) were derived for each ignition location using Geographical Information Systems (ARC/INFO and ArcView). Descriptive and especially explanatory statistics (chi-square-test, logistic regression analysis) were applied to derive fire occurrence probability for human as well as natural fires. GIS maps were calculated for high, medium and low fire seasons. 74 % of all fires are human caused (e.g., negligence, army, arson), 26 % are natural fires caused by lightning or in rare cases rock fall. During the entire analysis human and natural fires were strictly separated as they show distinct differences in their spatio-temporal behaviour. High season for human fires is the spring (March, April), which is also typical for the Alps where the main fire season is during late winter/early spring. Most human fires occur between 10 a.m. and 6 p.m. within a range of 200 meters to roads and hiking trails, below 1250 meters above sea level and on S, SE or W facing slopes. On weekends, increased fire ignition rates could be observed. Human fires are more frequent in the northern and central study area, whereas natural fires are concentrated to the South and Southeast. As expected, natural fires do not show a preference for certain weekdays, but occur in late afternoon (16.00 - 20.00) during the summer months (June, July, August). They show a slight preference for steep slopes (> 30°) and higher elevations (1250 - 2000 m.a.s.l.) but none for aspect. Although the data base was small, it was considered that the investigated methods provided reliable results and the derived GIS maps may be input to Spatial Decision Support Systems (SDSS) in wildfire management.
Fire Occurrence and Post Fire Risk Analysis
The (natural) hazard wildfire creates primary and secondary effects on the area where it occurs. To assess and validate the manifold role of wildfires in an area of interest both issues are important. In Switzerland, the loss in timber value due to fires can be neglected, whereas the destruction or partial damage of the protective functions of the affected forests may have major economic consequences as the probability for secondary effects such as erosion, debris flows, avalanches, etc. increases. For the assessment of wildfires a GIS-based framework for fire risk analysis was implemented. It is based on risk engineering principles where risk represents the product of the probability that an event takes place (occurrence) and the outcome of the event, usually but not necessarily referred to as damage. The framework enables the calculation of fire occurrence (e.g., based on the spatio-temporal pattern of historic fires in an area of interest, fire weather indices, fuel properties, etc.) and the possible outcome (damage) impacting objects of interest such as any human and/or natural values. The calculation of the damage potential is based on fire behaviour modelling assigning the probability for an object to be reached by a fire and the valuation of the objects. The valuation can be accomplished by economical, ecological and/or any societal criteria. In the case of post fire risk analysis the chain of possible secondary outcomes are of particular concern as their impact on the objects of interest may be even greater then the fire itself. This is especially true for the study area (Ticino) that on one side represents the wide spread chestnut belt of the Southern Alps and on the other side stands for areas with rugged terrain prone to topography bound natural hazards such as debris flows, rock falls or avalanches in higher altitudes. In the Canton of Ticino special consideration was given to the observation and analysis of erosion processes in burnt areas. There is strong evidence that heavy rainfalls provoke devastating debris flows affecting not only the burnt surface but also objects of interest in the closer or farther vicinity. The present study showed how risk analysis can be used to determine patterns where, when and why fires are most likely to start and how secondary effects, especially post fire run-off and erosion and consequently debris flows can occur and affect human or natural values.
|
