Δασικές πυρκαγιές σε Μεσογειακούς θαμνώνες, φρύγανα και χορτολίβαδα στην Ελλάδα: Σύγκριση της παρατηρηθείσας συμπεριφοράς πυρκαγιάς με τις προβλέψεις του BehavePlus - Wildfires in Mediterranean shrublands, phrygana and grasslands, in Greece: Comparisons of observed fire behaviour to BehavePlus predictions (In Greek with English abstract)

Abstract

This paper presents a comparison of ninety five (95) Rate Of Spread (ROSobserved) and seventy(70) Flame Length (FLobserved) observations of surface wildfire behavior in Greece with predictions from the BehavePlus fire behavior prediction system for tall and short Mediterranean shrublands (maquis), phryganic lands dominated by the small xeric shrub Sarcopoterium spinosum, and grass.
Four fuel models, which had been developed for Greece, were used to describe the four fuel types: a) “Evergreen-schlerophyllous shrublands (1.5 - 3 m)” for tall maquis, b) “Evergreen schlerophyllous shrublands (up to 1.5 m)” for short maquis, c) “Phrygana II (Sarcopoterium spinosum)” for phryganic areas dominated by Sarcopoterium spinosum and d) “Mediterranean grasslands” for grass. The pairs of ROSobserved values and BehavePlus ROSpredicted values, were correlated via linear regression for each of the data subsets. The resulting four linear regression equations, with ROSobserved as the dependent variable and ROSpredicted as the dependent, are statistically significant (p<0.001) and can be used for adjusting BehavePlus ROS predictions to “real world” ROS estimates.
More specifically, BehavePlus ROS predictions were close to ROSobserved so adjustment is not considered as necessary. On the other hand, in the case of grasslands, BehavePlus under-predicts ROS by approximately 50%. As the grass adjustment equation is statistically significant and its adjusted R2 value is high (R2adjusted = 0.847), it should be used for adjusting BehavePlus ROS predictions in these fine fuels.
The analysis also shows that the equations for short maquis and Sarcopoterium spinosum phrygana should also be used to adjust ROSpredicted values to the lower expected values, but this should be done with caution due to weaknesses of the equations.
In regard to flame length (FL) predictions from BehavePlus significant deviations were found for all four fuel types. The most important finding of this analysis was that BehavePlus consistently under predicted flame length for the Sarcopoterium spinosumdominated phrygana.The under prediction was significant and its importance is even greater because the underestimation takes place in a band of FL values that includes the threshold value of 1.2 m which is considered as the limit for direct attack on the flames with hand tools. In ten (10) out of N=26 cases, the prediction was for FL<1,2 m while the observed FL value was well above this threshold. This is an important result that can be very useful for the safety of firefighters and it should be seriously taken into consideration in operational firefighting in the country.

Full text

1

175
De Luis, M., Baeza, , -Hidalgo, J.C. 2004. Fuel characteristics and fire
behaviour in mature Mediterranean gorse shrublands. International Journal of Wildland Fire 13: 79-
87.
Dimitrakopoulos, A.P. 2002. Mediterranean fuel models and potential fire behavior in Greece.
International Journal of Wildland Fire 11: 127-130.
Mitsopoulos, I.D., Dimitrakopoulos, A.P. 2014. Estimation of canopy fuel characteristics of Aleppo
pine (Pinus halepensis Mill.) forests in Greece based on common stand parameters. European
Journal of Forest Research 133 (1): 73-79.
Mitsopoulos, I.D, Dimitrakopoulos, A.P. 2007. Canopy fuel characteristics and potential crown fire
behavior in Aleppo pine (Pinus halepensis Mill.) forests. Annals of Forest Science 64(3): 287-299.
Pausas, J. C., Llovet, J., Rodrigo, A., Vallejo, R. 2008. Are wildfires a disaster in the Mediterranean
basin? -- a review. International Journal of Wildland Fire 17:713-723.
Reinhardt, E., Lutes, D., Scott J. 2006. FuelCalc: A method for estimating fuel characteristics.
Proceedings of Fuels Management-How to Measure Success. 28 30 March 2006, Portland, Oregon
(US. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Proceedings
RMRS P-41, Fort Collins, Colorado), pp. 273 282.
Rothermel, R.C. 1972. A Mathematical model for predicting fire spread in wildland fuels. USDA
Forest Service Research Paper INT-115. Odgen, UT
Scott, J.H. 1999. NEXUS: a system for assessing crown fire hazard. Fire Management Notes. 59(2):
20-24.
Scott J. H., Burgan Robert E. 2005. Standard fire behavior fuel models: a comprehensive set for use
with Rothermel''s surface fire spread model. Gen. Tech. Rep. RMRS-GTR-153. Fort Collins, CO:
U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 72 p.
Scott, J.H., Reinhardt, E.D. 2001. Assessing crown fire potential by linking models of surface and
crown fire potential. USDA, Forest Service, Rocky Mountain Research Station, Research Paper
RMRS-29, Fort Collins, USA. 59p.
Stocks, B.J., Alexander, M.E., Wotton, B.M., Stefner, C.N., Flannigan, M.D., Taylor, S.W. 2004.
Crown fire behaviour in a northern jack pine - black spruce forest. Canadian Journal of Forest
Research 34: 1548-1560.
Werth, P., Potter, B., Clements, C., Finney, M., Goodrick, S., Alexander, M., Cruz, M., Forthofer, J.,
McAllister, S. 2011. Synthesis of knowledge of extreme fire behavior: volume I for fire managers.
General Technical Report. PNW-GTR-854. Portland, OR: U.S. Department of Agriculture, Forest
Service, Pacific Northwest Research Station. 144 p.
Xanthopoulos, G., Manasi, M. 2002. A practical methodology for the development of shrub fuel
models for fire behavior prediction. In Proc. 4th Int. Conf. on Forest Fire Research. November 18-
23, 2002. Luso-Coimbra, Portugal.
Behaveplus
, .1, .2
1
info@m-athanasiou.gr
2
, , gxnrtc@fria.gr

176
(ROSobserved
Sarcopoterium spinosum
BehavePlus.
BehavePlus
BehavePlus ROSobserved
BehavePlus
BehavePlus
BehavePlus
BehavePlus
:
BehavePlus
(Cruz et al. 2003a, Gould et al 2011
Albini 1976a
McKinion and Baker 1982, Mayer and Butler 1993)
Rothermel and Reinhardt
(Cruz et al. 2003, Gould et al. 2011
Albini 1976).
Albini et al
Alexander and Thomas 2003a,b

177
(Alexander and Cruz
Rothermel 1983, Norum and Miller 1984, Lawson and Armitage 2008).
BehavePlus (Andrews et al
Athanasiou and Xanthopoulos
II 3,0 m
Dimitrakopoulos et al Dimitrakopoulos
I m V II
VI
BehavePlus
I, II, V VI
Athanasiou and Xanthopoulos 2014).
ROSpredicted FLpredicted
BehavePlus.
Table 1. The values of the parameters of the four fuel models that were used as inputs for predicting surface fire
rate of spread (ROSpredicted) and flame length (FLpredicted) with BehavePlus.
I II V VI
1 HR (MTON/HA) 9.91 17.88 3.50 4.82
10 HR (MTON/HA) 6.80 13.30 1.02 0.49
100 HR (MTON/HA) 3.60 8.5 0.28 0
LIVE HERB (MTON/HA) 0 0 0 0
LIVE WOODY (MTON/HA) 7.70 10.60 0.85 0
1 HR S/V (1/CM) 55 55 65 78
LIVE HERB S/V (1/CM) - - - -
LIVE WOODY S/V (1/CM) 55 55 65 -
FUEL BED DEPTH (CM) 102.19 203.58 40.00 27.53
EXT MOISTURE (%) 34 34 20 14
HEAT CONTENT (J/G) 20000 20000 19054 18600
Athanasiou and Xanthopoulos (2010),
Athanasiou and Xanthopoulos
(ROSobserved) (FLobserved)
Alexander and Thomas (2003b), Clements et al Athanasiou and
Xanthopoulos (2014).

178
BehavePlus
Van Wagner
(Van Wagner
BehavePlus
Dimitrakopoulos Athanasiou and
Xanthopoulos, 2014).
BehavePlus
BehavePlus
ROSobserved
FLobserved),
NEWMDL BEHAVE (Burgan and
Rotrhermel
h litter
hr fuel bed depth
dead fuel moisture of extinction NEWMDL
ROSobserved FLobserved
ROSpredicted FLpredicted BehavePlus.
Table 2. ROSobserved and FLobserved observations per fuel type that were compared with the corresponding
BehavePlus predictions (ROSpredicted and FLpredicted values).
ROSobserved FLobserved
13 17
38 11
V 26 26
VI 18 16
h FDFMC
Rothermel
h
h (Andrews et al

179
h Live Woody
ROSobserved
BehavePlus (ROSpredicted
SPSS (v
p-value <0.001) (Athanasiou and Xanthopoulos 2014):
ROSobserved = 0.165 + 0.886 * ROSpredicted, adjusted R2 (1)
ROSobserved = 0.127 + 0.709 * ROSpredicted, adjusted R2 (2)
ROSobserved = 0.101 + 0.783 * ROSpredicted, adjusted R2 (3)
ROSobserved = -0.023 + 1.562 * ROSpredicted, adjusted R2 (4)
p-
value p-
value1=0.052, p-value3=0.266 and p-value4
p-
value2
R2adjusted R2adjusted
Sarcopoterium spinosum
V
Cistus spp
(Flomis fruticosa).
FLobserved (FLpredicted BehavePlus
FLobserved FLpredicted
(FLobserved) FLpredicted
V BehavePlus. FLobserved FLpredicted
FLobserved
FLpredicted (Athanasiou and Xanthopoulos 2014).
ROSobserved ROSpredicted
(p R2adjusted
p-
value
Sarcopoterium spinosum
(Sarcopoterium spinosum
V

180
ROSobserved
ROSpredicted
II
Athanasiou and Xanthopoulos
BehavePlus (ROSpredicted
ROSobserved
I V BehavePlus
ROS
and (3),
N
BehavePlus
ROS
R2adjusted
VI ROS
BehavePlus
BehavePlus
ROSobserved).
ROSpredicted.
Table 3. Solution of equations (1)-(4) for a range of values of ROSpredicted.
ROSobserved (km/h)
ROSpredicted
(km/h)
0 0.165 0.127 0.101 -0.023
1 1.051 0.836 0.884 1.539
2 1.937 1.545 1.667 3.101
3 2.823 2.254 2.450 4.663
4 3.709 2.963 3.233 6.225
5 4.595 3.672 4.016 7.787
6 5.481 4.381 4.799 9.349
7 6.367 5.09 5.582 10.911
8 7.253 5.799 6.365 12.473
9 8.139 6.508 7.148 14.035
10 9.025 7.217 7.931 15.597
BehavePlus
BehavePlus ,
m
(Deeming et al. 1977, Hirsch and Martell FLpredicted
m FLobserved
Xanthopoulos
FL
Athanasiou and Xanthopoulos 2014).

181
BehavePlus
Live Woody
V BehavePlus Cistus
spp Live Woody
V
BehavePlus
V,
BehavePlus
BehavePlus
Athanasiou and Xanthopoulos
BehavePlus ,
V
(International Association of Wildland Fire
(Doctoral Student Scholarship Award
-
Wildfires in Mediterranean Shrublands, Phrygana, and Grasslands, in Greece:
Comparisons of Observed Fire Behaviour to Behaveplus Predictions
Athanasiou M.1and Xanthopoulos G.2
13673 Acharnes, Greece,
info@m-athanasiou.gr
2
Terma Alkmanos, 11528, Athens, Greece, gxnrtc@fria.gr
Abstract
This paper presents a comparison of ninety five (95) Rate Of Spread (ROSobserved) and seventy(70)
Flame Length (FLobserved) observations of surface wildfire behavior in Greece with predictions from the
BehavePlus fire behavior prediction system for tall and short Mediterranean shrublands (maquis),
phryganic lands dominated by the small xeric shrub Sarcopoterium spinosum, and grass.
Four fuel models, which had been developed for Greece, were used to describe the four fuel types: a)
-schlerophyllous shrublands (1.5 -

182
ROSobserved values and BehavePlus ROSpredicted values, were correlated via linear regression for each of
the data subsets. The resulting four linear regression equations, with ROSobserved as the dependent
variable and ROSpredicted as the dependent, are statistically significant (p<0.001) and can be used for
More specifically, BehavePlus ROS predictions were close to ROSobserved so adjustment is not
considered as necessary. On the other hand, in the case of grasslands, BehavePlus under-predicts ROS
by approximately 50%. As the grass adjustment equation is statistically significant and its adjusted R 2
value is high (R2adjusted = 0.847), it should be used for adjusting BehavePlus ROS predictions in these
fine fuels.
The analysis also shows that the equations for short maquis and Sarcopoterium spinosum phrygana
should also be used to adjust ROSpredic ted values to the lower expected values, but this should be done
with caution due to weaknesses of the equations.
In regard to flame length (FL) predictions from BehavePlus significant deviations were found for all
four fuel types. The most important finding of this analysis was that BehavePlus consistently under
predicted flame length for the Sarcopoterium spinosumdominated phrygana.The under prediction was
significant and its importance is even greater because the underestimation takes place in a band of FL
values that includes the threshold value of 1.2 m which is considered as the limit for direct attack on
the flames with hand tools. In ten (10) out of N=26 cases, the prediction was for FL<1,2 m while the
observed FL value was well above this threshold. This is an important result that can be very useful for
the safety of firefighters and it should be seriously taken into consideration in operational firefighting
in the country.
Albini, F.A., 1976. Estimating wildfire behavior an d effects. Gen. Tech. Rep. INT-30. Ogden, UT:
U.S. Department of Agriculture, Forest Service, Intermountain Forest and Range Experiment
Station. 92 p.
Albini, F.A., Alexander, M.E.; Cruz, M.G. 2012 A mathematical model for predicting the maximum
potential spotting distance from a crown fire.. International Journal of Wildland Fire 21(5):609-627.
Al
-8.
63 (4): 4-12.
Alexander M. E., Cruz M. G. 2013. Limitations on the accuracy in model predictions of wildland fire
behavior: A state-of-the-knowledge overview. The Forestry Chronicle Vol. 89, No 3 May/June
2013
Guide. General Technical Report RMRS-GTR-106WWW revised. Ogden, UT: U.S. Department of
Agriculture, Forest Service, Intermountain Forest and Range Experiment Station. 132 p.
- -4
Athanasiou, M. and G. Xanthopoulos 2010. Fire behaviour of the large fires of 2007 in Greece. In
proceedings of the 6th International Conference on Forest Fire Research, 15 -18 November 2010,
Coimbra, Portugal. D.G. Viegas, Editor. ADAI/CEIF, University of Coimbra, Portugal. Abstract p.
336, full text on CD.
- -9

183
Athanasiou, M., and G. Xanthopoulos. 2014. Wildfires in Mediterranean shrubs and grasslands, in
Greece: In situ fire behaviour observations versus predictions. In proceedings of the 7th
International Conference on Forest Fire Research: Advances in Forest Fire Research, 17-20
November 2014, Coimbra, Portugal. D. G. Viegas, Editor. ADAI/CEIF, University of Coimbra,
Portugal. Abstract p. 99, full text on CD.
83.
Burgan, R.E., Rothermel, R.C. 1984. BEHAVE: fire behavior prediction and fuel modelling system
FUEL subsystem. Gen. Tech. Rep. INT-167. Ogden, UT: U.S. Department of Agriculture, Forest
Service, Intermountain Forest and Range Experiment Station. 126 p.
Clements, H., D. Ward, and C. Adkins. 1983. Measuring fire behaviour with photography.
Photogrammetric Engineering and Remote Sensing: Series II, vol. 49: 213-217.
Cruz, M.G., Alexander, M.E., and Wakimoto, R.H. 2003a. Definition of a fire behavior model
evaluation protocol: a case study application to crown fire behavior models. In Fire, Fuel
Treatments, and Ecological Restoration: Conference Proceedings, 16-18 April 2002, Fort Collins,
Colo. Technically edited by P.N. Omi and L.A. Joyce. USDA For. Serv. Proc. RMRS-P-29. pp. 49-
67.
Deeming, J.E.; Burgan, R.E.; Cohen, J.D. 1977. The National Fire-Danger Rating System -1978.
United States Department of Agriculture, Forest Service, General Technical Report INT -39,
Intermountain Forest and Range Experiment Station, Ogden Utah.
Mateeva, V
VI
12(3): 192-206.
Dimitrakopoulos, A.P., 2002.
International Journal of Wildland Fire 11(2) 127 130.
Gould, J., McCaw, L., Cruz, M., Anderson, W., 2011. How good are fire behavior models? Validation
of eucalypt forest fire spread. Paper Presented at WILDFIRE 2011 e 5th International Wildland Fire
Conference. http://www.wildfire2011. org/material/papers/Jim_Gould.pdf.
Hirsch KG, Martell DL (1996) A Review of Initial Attack Fire Crew Productivity and Effectiveness.
International Journal of Wildland Fire 6, 199 215.
Jolly W. M. 2005. Sensitivity of a fire behavior model to changes in live fuel moisture USDA Forest
Service, RMRS, Fire Sciences Laboratory, Missoula, MT Presented at the Sixth Symposium on Fire
and Forest Meteorology, Oct. 25-27, 2005, Canmore, AB, Canada.
Lawson, B.D. and O.B. Armitage. 2008. Weather guide for the Canadian Forest Fire Danger Rating
System. Nat. Resour. Can., Can. For. Serv., North. For. Cent., Edmonton, AB. 73 p.
Norum, R.A. and M. Miller. 1984. Measuring fuel moisture content in Alaska: Standard methods and
procedures. USDA For. Serv., Pac. Northwest For. Range Exp. Stn., Portland, OR. Gen. Tech. Rep.
PNW-171. 34 p.
Rothermel, R.C. 1983. How to predict the spread and intensity of forest and range fires. Gen. Tech.
Rep. INT-143. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Forest
and Range Experiment Station. 161 p.
Van Wagner, C. E. 1977. Conditions for the start and spread of crown fire. Canadian Journal of Forest
Research. 7: 23 34.
Xanthopoulos, G. 2007. Forest fire related deaths in Greece: confirming what we already know. p. 339.
-17, 2004, Seville,
Spain. Full paper on the CD accompanying the book of abstracts.
-
-9