عنوان مقاله [English]
Siahkal forests in northern Iran experienced extended and devastating fire seasons in recent decades, in most cases, driven by dry, warm/hot, windy weather. Identifying forest fuel and wildfire risk dynamics is important for an integrated fire-forest management strategy. To understand how forest structure controls these dynamics, we quantified forest fuels and fire behaviour across different forest types in the study area. In this work, we present the results of an analysis of wildfire behaviour from historical fire records, vegetation type, weather, and fuel moisture. We used a modelling approach to estimate the physical parameters of surface fire behaviour: flame length and fire size; as well as the potential crown fire occurrence. We selected a large forest watershed area in Siahkal County of north Iran, under temperate broadleaf forests and mixed plantations with variable compositions. To run the model, we gathered data on forest structure and composition, and physical information. We detected moderate to high levels of fire size and flame length, ascribed to the high availability of very dry fine biomass. The crown fire potential varied greatly throughout the landscape. Low stands were more prone to crowning. The results show that crown fire potential in plantation stands especially conifers are higher compared to natural stands. In addition, the early stages of planted stands are more likely to experience crown fire due to their low mean height and low CBH. The findings can assist in the identification of priority areas where forest structure needs to be managed.
Agricultural land development, large-scale land acquisitions, past logging practices, excessive grazing by livestock, and silvicultural treatments have altered forest structure and changed the fire regime from low to mixed-severity to a high severity wildfire in the old-growth Hyrcanian forests of northern Iran in the past 3 decades. Furthermore, fast-growing tree species such as poplars, maples, Alnus, and Pinus in large-scale plantations were used under short-rotation forestry in the area. In this work, we described the fuel complexes created by the variable forest structure and used a fire simulation system to test for differences in surface and crown fire behaviour among the forest types. Surface and crown fires are modeled using FlamMap MTT simulation (Finney, 2006) and based on local fuel models and historical fire information in the study area. The simulation output maps in this study include estimates of the main characteristics of surface fire, i.e., flame length and fire size, as well as the crown fire probability, which is used to inform land managers about suitable places for silvicultural treatments to remove and reduce combustible fuels, and finally, it is used to design strategies for fire risk management.
The study was conducted in the watershed of 25 (Shenroud), which is located in the lowland and sub-mountain zones of the Hyrcanian forests of northern Iran. This forest watershed compromised 190 km2 and extends from the Shenrud River in the north to the Dorfak Peak in the south.
Field data were obtained from the forest inventory carried out in the framework of forestry plans in the study area. To measure surface and crown fuel characteristics, square samples with dimensions of 30 x 30 meters (130 samples in total) were used by stratified random sampling method in the summer and fall of 2020. All trees with a diameter at breast height (DBH) of 7.5 cm were measured in each plot. Tree height, the height of the live crown base (CBH), and crown width (in two directions at right angles to each other) were measured in a subsample of three trees per plot (the northernmost, southernmost, and one dominant tree). The individual tree measurements were used to estimate the following canopy fuel variables in each plot: canopy height (CH, m), canopy basal height (CBH, m), and canopy bulk density (CBD, 100kg m-3). In order to estimate the percentage of grass cover and litter on the forest floor in the center of each sample plot, a microplot with dimensions of 1 x 1 meter was taken. In addition, in these microplots, the diameter of dead and fallen wood pieces was measured.
For the study area, using FlamMap MTT (Finnet, 2006) 10,000 random fires were simulated based on ignition points of historical fires in the studied period. Fire history data for the study area showed 176 recorded fires during the period 2000-2020, almost entirely caused by human-caused fires. The total burned area for these fires is approximately 1524 ha.
Fire growth calculations were performed with outputs with a resolution of 30 meters. According to the historical information about the fire and the prevailing wind conditions in the study area, the duration of the fire was simulated for 6 hours with the direction of the west wind and the wind speed of 10 km/h. Finally, the analysis of the fire model outputs including flame length, fire size, and the crown fire probability was done on the scale of the landscape and also the main fuel types in the study area.
The results obtained in Table 1 show that conifer plantation has more fuel load in both live and dead biomass. In this model, the average load of live and dead fuel was estimated at 7.3 and 4 tons per hectare, respectively. In general, the fuel load of live biomass in the broadleaf forest is small, which plays an insignificant role in crown fires. While the amount of dead biomass fuel load is important in the case of surface fires in both conifer and broadleaf forests (Velizarova et al., 2014).
Regarding the characteristics of the forest canopy, Figure 1 presents the spatial distribution of the canopy cover, canopy height, canopy base height, and canopy bulk density for the study area. General statistics are also presented in Table 1. Among the woody fuel models, the natural beech forest has a relatively closed canopy with an average canopy cover of 80%, and after that, the mixed broadleaf forest has the highest canopy cover of 60%. According to the canopy cover, the highest canopy height (18 m) was recorded in the natural beech forest. The variability observed in the average values of the crown base height was lower compared to other crown fuel variables among the available fuel models. The average values of crown base height ranged from 2 m in the shrub fuel to 3.4 m in the natural beech forest. The average crown bulk density in the study area is equal to 15 (100 kg/m2). High crown bulk density values (>20 100 kg/m2) were obtained in the northern areas of the study area and in plantation fuel types. The highest average crown bulk density in the available fuel models was equal to 18 (100 kg/m2) in conifer plantation. The small and dense branches of the Loblolly pine and the Norway spruce have led to a higher density of crown fuel per unit area in conifer plantation. This result is in accordance with the result of the study by Fernández-Alonso et al. (2013).
The results of the spatial distribution of the flame length, fire size, and the crown fire probability in the landscape are presented in Figure 1. Regarding surface fire characteristics (flame length and fire size), medium to high levels of these components was detected, which was due to the high availability of very dry fine biomass in the study area. The size of the fires after a 6-hour simulation ranged from 0.1 ha to 2,300 ha, with an average of 430 ha. The flame length also varied from 0 m to 5 m, with an average of about 1 m. Regarding the crown fire potential, which is a function of the quantity and order of combustible fuels, the low level of this component was simulated in the study area. Based on the model results, the crown fire probability varied from 0 to 15% in the study area, with an average of 2%. Also, in Table 2, the differences in the average of these simulated fire behavior characteristics in different fuel models are presented.
In the study area, with the increase in tree density from the grass-fuel model to the wood-fuel model, a significant decrease in fire spread rate and fire intensity and as a result, the simulated fire size is observed. By creating a colder and wetter microclimate, tree cover increases the moisture content of fuel and reduces flammability (Newberry et al., 2020). This case has led to less flammability of wood-fuel models, especially in natural forests in the studied area. This result is in accordance with the studies of Ray et al. (2010) and Dodonov et al. (2013).
Based on the results of the probability of crown fire occurrence, short forest stands are more prone to crown fire, so fuel models with a higher height of the crown base (3 meters and more in natural forests) have lower fire intensity which is in accordance with the result of Jain and Graham (2007).