|
1. Definition
|
Name
|
FUEL
MODELS
|
|
Brief
definition
|
Fuel,
in the context of wildland or agricultural fire, refers
to all combustible material available to burn. Fuels
are different according to load, size and presence/absence
of live tissues. Given their variability, a model
is a synthetic way of expressing and describing fuel
properties. A model can be defined as a collection,
a data-base of fuel properties which are significant
for rating fire danger and predicting fire behaviour.
|
|
Unit
of measure
|
Total
Fuel loading in tons/hectares, distinguished per load
of the different size classes (lag-time classes) and
presence/absence of live fuels with different moisture
content
|
|
Spatial
scale
|
|
|
Temporal
scale
|
|
2.
Position within the logical framework DPSIR
|
Type
of Indicator
|
Driving
Force/State
|
3.
Target and political pertinence
|
Objective
|
The
indicator contributes to the definition of the fuel
load and consequently the potential behaviour of fires.
It affects the intensity and impact of fires on ecosystems
in a territory affected by desertification.
|
|
Importance
with respect to desertification
|
The
prediction of fire behaviour is one of the most important
tasks required to operate fuel and fire management
decision support systems and dynamic vegetation models.
Knowledge of fuel bed characteristics, through the
analysis and description of spatial fuel property
layers, is crucial to fire managers and is becoming
increasingly important to ecologists, air quality
managers, and carbon balance modelers. As the source
of all fire behaviour and fire effects, fuel bed must
be characterized and mapped before any calculation
of fire potential can be made. Fuel mapping, hazard
assessment, evaluation of fuel treatment options and
sequences and monitoring of fire effects all require
a consistent and scientifically applied fuel classification
system. Thirteen stylised fuel models (Anderson, 1982)
were developed to provide standardized numerical fuel
bed descriptions in order to generate reasonable and
accurate fire behaviour, predictions using spread
models, such as the classical BEHAVE. Each model is
a database of about 30 fuel bed properties that determine
its fire behaviour potential. The models were conceived
of as a set of standardized and stylized inputs for
use in the spread model across the range of fire behaviour.
Fuel load and depth are significant fuel properties
for predicting whether a fire will be ignited, its
rate of spread, and its intensity.
|
|
International
Conventions and agreements
|
The
UNCCD recognised, among the particular conditions
of the Mediterranean, that the causes of desertification
include the extensive forest coverage losses due to
frequent wildfires (Convention text as of September
1994 and as of September 2001).
|
|
Secondary
objectives of the indicator
|
This
indicator represents the potential impact of fire
on land and is a good predictor of its expected effects
on forest and rural ecosystem. Information about the
expected severity of fire at a topographic scale can
help in addressing measures to recognise the greatest
fire hazard areas, to organise an efficient fire fighting
system to cope with fire, and indirectly to reduce
desertification.
|
4.
Methodological description and basic definitions
|
Definitions
and basic concepts
|
Fuel
models can be defined as numerical arrays, i.e. standardized
descriptions of physical properties of fuel heaps,
ranging from total load to surface/volume ratio, to
amount of fuel per size classes, to fuel bed depth,
to extinction moisture. Fuels have been classified
into four groups (grasses, brush, timber and slash)
and grouped into 13 models, with a numerical code
from 1 to 13, which roughly correspond to vegetation
types, namely:
- Grasses (1, 2,
3)
- Brushes (4, 5,
6, 7)
- Timber (8, 9, 10)
- Slash (11, 12,
13)
|
|
Benchmarks
Indication of the values/ranges of value
|
The
potential impact of fire on land is evaluated through
the use of:
- Combustibility
rate: i.e. the facility to burn (Velez, 2000) and
the
- Ignition rate,
i.e. the facility to start burning (Rodriguez y
Silva, 2000).
|
Fuel model
|
Combustibility rate
|
Ignition rate
|
|
1
|
10
|
1
|
|
2
|
10
|
1
|
|
3
|
10
|
0.9
|
|
7
|
10
|
0.7
|
|
|
|
|
|
4
|
10
|
0.6
|
|
5
|
10
|
0.2
|
|
6
|
10
|
0.6
|
|
|
|
|
|
8
|
5
|
0.5
|
|
9
|
5
|
0.4
|
|
10
|
5
|
0.2
|
|
|
|
|
|
11
|
1
|
0.2
|
|
12
|
1
|
0.1
|
|
13
|
1
|
0.1
|
according
tothe above table we have:
- class I - score
1: fuel models 11, 12, 13
- class II - score
1.33: fuel models 8, 9, 10
- class III - score
1.66: fuel models 4, 5, 6
- class IV - score
2: fuel models 1, 2, 3, 7
|
|
Methods
of measurement
|
There
are several way to estimate fuel models and therefore
biomass loading:
- Through the use
of the 13 standard fuel models formerly described
by Anderson (1982) with "Fuel Model Descriptions"
and typical photographic examples of American vegetated
landscapes corresponding to different models.
- Through the use
of the "ICONA fuel models", i.e. of the
13 USA standard fuel models adapted and described
in Spain by ICONA for Spanish forest regions, with
"Fuel Model Descriptions" and typical
photographic examples of Spain vegetated landscapes
corresponding to different models (Velez Munoz,
2000).
- It is also possible
to build ad hoc models through specific software
such as NWMDL.
|
|
Limits
of the indicator
|
This
indicator requires information on land use at a very
small scale; land cover and vegetation maps at a suittable
scale can help to identify the mosaic of fuel models.
|
|
Linkages
with other indicators
|
Fire
risk, Burned area,
Fire frequency.
|
5.
Evaluation of data needs and availability
|
Data
required to calculate the indicator
|
List
and description of canonical models. Parameters, as
input for NWMDL if ad hoc models must be built, such
as:
- Fuel load, ratio
of surface area to volume for each of the three
size classes (from 0,6cm, to 7,6 cm), depth of fuel
bed, live fuel moisture, dead fuel moisture, including
that at which fire will not spread, called extinction
moisture.
|
|
Data
sources
|
Vegetation
Maps. Direct survey on the ground. R.S. images can
also be used to directly simulate fuel maps.
|
|
Availability
of data from national and international sources
|
International
and National cartographic surveys (i.e. CORINE land
cover).
|
6.
Institutions that have participated in developing the indicator
|
Main
institutions responsible
|
University
of Basilicata, Italy
|
|
Other
contributing organizations
|
|
7.
Additional information
|
Bibliography
|
Andrews P.L., 1986 BEHAVE Fire behaviour
prediction and fuel modelling system. Burn subsystem. USDA,
Forest Service, Gen. Techn. Rep. INT- 195, Intermountain Forest
and Range Experiment Station, Ogden, Ut.
Anderson, H.E., 1982. Aids to Determining
Fuel Models For Estimating Fire Behaviour. USDA Forest Service,
Gen. Techn. Rep. INT-122. Intermountain Forest and Range Experiment
Station, Ogden, UT.
Rothermel, R.C., 1972. A Mathematical
model for Predicting Fire Spread in Wild Land Fuels. USDA
Forest Service, Intermountain Forest and Range Experiment
Station, Research Paper, INT-115, Ogden, UT.
Burgan R.E., Rothermel T.C., 1984 BEHAVE:
fire behaviour prediction and fuel modelling system. Fuel
subsystem. USDA, Forest Service, Gen. Techn. Rep. INT- 167,
Intermountain Forest and range Experiment Station, Ogden
Rodriguez y Silva F., 2000 Ejemplos de
planes de defensa. In: Velez R. (ed.), 2000 La defensa contra
incendios forestales: fundamentos y experiencias. McGraw Hill,
Madrid
Salazar L.A, 1985 - Sensitivity of fire
behavior simulations to fuel model variations. Res. Pap. PSW-178.
Berkeley, CA : USDA, For. Serv., Pacific Southwest For. Range
Exp.Stn.
Velez Munoz R., 2000 Combustibles forestales.
In: Velez R. (ed.), 2000 La defensa contra incendios forestales:
fundamentos y experiencias. McGraw Hill, Madrid
|
|
Other
references
|
Nardiello D., 1998 - Attivitą di prevenzione
contro gli incendi boschivi e cartografia di rischio:
applicazione alla Val d'Agri dell'analisi territoriale
multidisciplinare (Sistema A.F.S). Degree Thesis,
University of Basilicata, Academic Year 1997-98
Lovreglio R., 2001-Proposta per la realizzazione
di un piano antincendio boschivo per la Riserva Naturale di
Torre Guaceto (Br), unpublished manuscript
|
|
Contacts
Name and address
|
Prof.
Agostino Ferrara
University of Basilicata
Polo Universitario di Macchia Romana
85100 Potenza, Italy
e-mail: ferrara@unibas.it
|
|
