Detection of Fire-Prone Areas Using the PROMETHEE Decision-Making Method (Case Study: Watershed Basin of Shourdareh, Golestan Province, Iran)

Amirreza Mesbah ¹ ( PhD Student, Department of Natural Resource, Nour Branch, Islamic Azad University, Nour, Iran )
Khadijeh Mahdavi ² ( Assistant Prof., Department of Natural Resource, Nour Branch, Islamic Azad University, Nour, Iran )
Mahshid Souri ³ ( Assistant Prof., Rangeland Research Division, Research Institute of Forests and Rangelands, Agricultural Research, Education and Extension Organization (AREEO), Tehran, Iran )
Mohammad Reza Javadi ⁴ ( Assistant Prof., Department of Natural Resource, Nour Branch, Islamic Azad University, Nour, Iran )

تاريخ الإرسال : 28 الإثنين , ذو القعدة, 1443 تاريخ التأكيد : 30 السبت , شوال, 1444

الکلمات المفتاحية: Iran, prioritization, Fire, Golestan province, Shannon Weighting, Shourdareh,

ملخص المقالة :

Decision makers in fire management are faced with many alternatives and criteria. In decision making about the event of fire, various criteria including technical, economic, social and environmental criteria have to be considered simultaneously. Management to prevent and control fires in forests and rangelands will be effective if fire-prone areas and management identify and focus on these critical areas. Therefore, the present study was conducted in 2022 to identify fire-prone areas using PROMETHEE decision-making method in the watershed basin of Shourdareh, Golestan province, Iran. In the present study, according to fire expert’s opinion, 29 different environmental and social criteria were used to detect of fire-prone areas. In this regard, The Shannon entropy method was used to weigh the criteria. Then, according to the weight and value of each criterion for each sub-basin, the data were analyzed using the PROMETHEE II technique. Based on the results of PROMETHEE II technique, sub-basins of Gh3, Gh8 and Gh1 with Phi values of 0.335, 0.148 and 0.239, respectively, were in high susceptibility to fire class. While rangelands of sub-basins Gh2, Gh5, Gh6 and Gh7 with Phi values of -0.220, -0.117, -0.136 and -0.241 were in the low susceptibility to fire class. Sub-basins of Gh9, Gh10 and Gh11 with Phi values of 0.114, -0.078 and 0.025 were in the moderate susceptibility to fire class. To evaluate the method, the results of this study were compared with results of actual fire areas that prepared by the department of natural resources of Golestan province, Iran. According to the obtained kappa coefficient with the value of 0.82, the method had good and acceptable accuracy. Therefore, since the proposed method was a reliable screening method to identify areas at risk of fire, it can help the authorities in carrying out preventive activities.

نص كامل:



 

Detection of Fire-Prone Areas using the PROMETHEE l decision-making method (Case study: watershed basin of Shourdareh, Golestan province, Iran)

Abstract. Decision makers in fire management are faced with many alternatives and criteria. In order to make correct decision about the event of fire, various criteria, including; technical, economic, social and environmental criteria, have to be considered simultaneously. Management to prevent and control fires in forests and rangelands will be effective if fire-prone areas identify and management focus on these critical areas. Therefore, the present study was conducted to identify fire-prone areas to properly manage of these areas by PROMETHEE decision-making method in watershed basin of Shourdareh, Golestan province, Iran. In the present study, 29 different environmental and social criteria were used to detect of fire-prone areas. In this regard, Shannon entropy method was used to weight the criteria. Then, according to the weight and value of each criterion for each sub-basin, the data were analyzed using the PROMETHEE II technique. Based on the clustering results, Gh3, Gh8 and Gh1 sub-basins with Phi values; 0.335, 0.148 and 0.239, respectively, were in high susceptibility to fire class. While, rangelands of sub-basins Gh2, Gh5, Gh6 and Gh7 with Phi values; -0.220, -0.117, -0.136 and -0.241, respectively were in the low susceptibility to fire class. Gh9, Gh10 and Gh11 with Phi values; 0.114, -0.078 and 0.025, respectively, were in the moderate susceptibility to fire classs.. To evaluate the method, the results of this study were compared with results of fire areas prepared by the department of natural resources of Golestan Province, Iran. According to the obtained kappa coefficient, which was equal to 0.82; this method has a good and acceptable accuracy. Therefore, the proposed method is a reliable screening method to identify areas at risk of fire, it can help the authorities in carrying out preventive activities.

 

Key words: Fire, Prioritization, Shannon Weighting, Shourdareh, Golestan Province, Iran.

 

Introduction

Fire causes extensive damage to rangelands’ ecosystems in arid and semi-arid regions (Asadian et al., 2022). Fires in forests and rangelands, as one of the most important and main criteria in the destruction of natural resources have devastating economic, social, environmental and human effects (Zhang et al., 2016; Akinola and Adegoke, 2019). Fires may destroy e soil structure, so reduce nutrient and water availability for plants (Ahmadi, 2017). Fires reduce the winter forage of livestock and increase land grabbing (Pournemati et al., 2021). In addition to human causes, these fires include lightning, global warming and climate change, improper management, insufficient precipitation, hot winds, bed buildup and friction between dry beds are some of the criteria that can cause natural fires in rangelands and forests (Chuvieco et al., 2012; Ganteaume et al., 2013; Rahimi et al., 2020).

Fire hazard is an important concept that is significantly formed in fire management planning (Gai et al., 2011). Therefore, having an effective prevention strategy to deal with recurrent and destructive fires is essential (Sakellariou et al., 2019). Due to Iran's location in the dry belt of the Earth and the high-pressure subtropical zone, atmospheric conditions are provided for fires in forests and rangelands. On the other hand, human criteria or deliberate fires to convert forest and rangeland lands into agriculture cause fires in fo

rest and rangeland areas and cause irreparable damage to ecosystems and ecological areas every year. Decision makers face many options and criteria in rangeland management and planning. One of their most important challenges is choosing the best and most appropriate option and prioritizing the options according to the defined criteria. In this regard, multi-criteria decision-making techniques can be a good solution to solve such problems. In order to efficiently manage and decide on natural resource projects, various indicators, including technical, economic, social and environmental indicators, must be considered simultaneously. One of the strongest and most effective multi-criteria decision-making methods is the PROMETHEE¹ II method. The PROMETHEE method is easily able to apply criteria with different measurement scales and defines six separate functions in proportion to the information and standard scale, so in multi-criteria decisions where the criteria usually have different scales, it is a suitable method for decision making (Chou et al., 2004). PROMETHEE is compatible and efficient in situations where many options have to be evaluated based on several quantitative, qualitative and often contradictory criteria (Albadvi et al., 2007). The PROMETHEE method is able to use criteria with different measurement scales without the need to scale the criteria. The PROMETHEE method has been used successfully in a wide range of real-world applications such as water resources management, health center prioritization, wastewater facility location, and watershed vulnerability (Banias, 2010; Huang and Tsai, 2010; Asghryzadeh and Nasrallahy, 2007). In Ghana, Darkwah et al., 2012 used the PROMETHEE method to rank the performance of their corporate operators. In their study, 5 criteria were ranked in the form of 4 options using the PROMETHEE method. The results showed that the PROMETHEE decision-making method is an efficient method in solving classification problems (Darkwah et al., 2012). The PROMETHEE method has been described as one of the most efficient MCDM² supersonic techniques for selecting the optimal flood reduction design in Athens (Maragoudaki and Tsakiris, 2005). In a study to plan and manage water resources in Romania, Anagnostopoulos et al., 2005 used the PROMETHEE method. The results showed that this method is an efficient method in water resources management (Anagnostopoulos et al., 2005). Eskandari (2013) used two methods of fuzzy hierarchical analysis and correlation in order to model the risk of fire and prepare a potential fire hazard map in part of the forests of northern Iran. The parameters used included four main criteria (topography, biological, climatic and human) and 12 sub-criteria. After preparing maps of all criteria and determining the weight of all of them by the two methods, fire risk models were obtained. Then, the maps of all effective criteria overlapped by considering their weight according to the fuzzy hierarchical analysis method. Enoh et al., (2021) for preparing the final forest fire risk zone map using different parameters such as land cover, aspect, elevation, slope and proximities to roads and settlements in the ArcGIS environment. Dehghan (2017) Comparing different multi-criteria decision making methods to determine suitable areas for implementing some water and soil protection activities in Gonabad watershed, Khorasan Razavi province, Iran. He stated that results of hierarchical analysis, network analysis and PROMETHEE II are close to each other.

Decision makers in fire management and planning are faced with many options and criteria. One of their most important challenges is to identify the most vulnerable areas and prioritize areas according to defined criteria. In this regard, multi-criteria decision-making techniques can be a good solution to solve such problems. In order to effectively manage and make the right decision in the event of a fire, various criteria, including technical, economic, social and environmental criteria, must be considered simultaneously. Management operations to prevent and control fires in forests and rangelands are effective when fire-prone areas are identified and remedial and management measures are focused on these areas. Therefore, the present study was conducted with the aim of identifying fire-prone rangelands in order to properly manage these areas by using PROMETHEE decision making method in Shourdareh basin of Golestan province, Iran.

 

Materials and Methods

Study area

The study area is known as Shourdareh and is located in Golestan province, Iran. In terms of political divisions of the country, the study area lies between 55°27' to 55°40' E and 36°56' to 37°5'N (Akhzari et al., 2013). The eastern part of the basin is located in Maraveh Tappeh city and the rest of it, is located in Kalaleh city. The villages of Qarnaq and Aq Chatal are located in the Shourdareh basin. It has 11 sub-basins including; Gh1 to Gh11. The area of the basin is equal to 120.74 km2 (Fig.1).

 

Fig.1. Geographical location of Shourdareh watershed in Iran and Golestan province, Iran

 

 

Fig.2.The present study was carried out by taking the following steps

 

Research Method

The current research was carried out as following steps

Step 1: Determine the effective criteria for rangeland fire

Based on the expert opinion, 10 experts from the department of natural resources with more than 10 years of work experience and 15 university professors in the fields of rangeland and forestry, the following criteria were examined as effective criteria in rangeland fires due to the high coefficient of variation in the basin:

Average annual and seasonal precipitation:

In order to estimate the amount of precipitation in Shourdareh basin, 24-hour precipitation information of Kachik station was used.

Slope and altitude:

To determine the slope and elevation, the DEM³ information layer of the region was used in GIS software environment and the slope and elevation classes of the region were determined.

Annual and seasonal temperature:

 As a result of the study performed on the data of evaporation stations in the region, it was found that there was a relationship between altitude and average temperature. This relationship is as follows (Alizadeh et al, 2022):

Equation 1:

 

Where: A and B: constant Coefficient of the equation, H: average altitude (m), T: temperature (degrees Celsius)

Relative humidity:

To obtain the relative humidity parameter in the study area, the relationship between the relative humidity parameters and temperature in the Kachik climatology station was used.

Evaporation: Using the evaporation pan statistics and using the following equation, the amount of evaporation from the free surface of water was estimated (Alizadeh et al, 2022):

Equation 2:

 

Where: E=amount of free water surface (mm), Epan= amount of evaporation from the pan, K= coefficient of evaporation pan

Type of climate:

The type of climate was determined by the Amberge method (Alizadeh et al, 2022):

Equation 3:

 

Where: Q2: Climate coefficient of Amberge, M: Average maximum temperatures in the hottest month of the year (Kelvin), m: Average minimum temperatures in the coldest month of the year (Kelvin), P: Annual precipitation (mm)

Plant types:

The vegetation type of Shourdareh watershed was classified by physiognomy-floristic method method based on two or three dominant (permanent) dominant species (Mesdagi, 2003).

Rangeland production:

In this study, field sampling was performed based on random-systematic method. The sampling units were plots located along linear transects. For this purpose, according to the conditions of the region, in each plant type, four 100 m transects in the slope direction and two 100 m transects perpendicular to the slope direction were established in the representative area of each type. Then 10 plots of 2 m2 were installed on each transect (Arzani, 1997). Then the amount of production in each plot was measured by cutting and weighing method (Mesdagi, 2003).

Rangeland condition:

4-factor method was used to determine the condition of rangeland types. Rangeland trend: To determine the rangeland trend, the scales method was used (Mesdagi, 2003).

Topographic Wetness Index:

Topographic Moisture Index is another topographic factor that was prepared and used based on the following equation:

Equation 4:

 

Where: α is the area of the drained area and β is the slope (tan) angle in degrees.

Topographic Wetness index plays an important role in soil moisture and slope stability. Then the TWI map was prepared using a digital elevation model map in SAGA GIS software with 3 classes.

Land use map:

In order to prepare the land use map of the basin and identify and separate the boundaries of arable lands from rangelands, reference images of satellite images from Google Earth software were used along with Landsat images. Then, this map was modified based on the geomorphological map and field visit and the final map was prepared.

Village and population density:

In the study of the number of population and households in the villages of basin was determined, the results of the Golestan health network and the statistics of health houses in the villages in the basin in 2017 were used.

 

Step 2: Determine the weight of each of the criteria

In the second step of the research, the weight of each criterion was calculated based on Shannon entropy method as follows:

Decision super matrix

First, the decision super matrix was formed with degree m × n. This super matrix includes m rows (11 sub-basins of Shourdareh watershed) and n columns (slope, evaporation, vegetation, etc.). Then, the weight of the indices was calculated by using the Shannon entropy method (Zhi-hong et al., 2006).

                

Step 3: Implement the PROMETHEE method

In the third step of the research, based on the weights obtained from Shannon entropy method, PROMETHEE method was implemented by using Visual PROMETHEE software (Kuncova & Seknickova, 2022).

 

Step 4: Classify the rankings

In the fourth step of the research, the classification of rankings was performed using the K-means clustering method (Chahoki Zare, 2012). The obtained rankings according to PROMETHEE technique for each sub-basin were classified using SPSS18 software package.

 

Step 5: Model evaluation

Kappa statistical coefficient was used to evaluate and validate the map of fire-prone areas of the Shourdareh watershed in Golestan province, Iran, using the PROMETHEE II technique (cohen, 1960).

l. 

Results

Weighting criteria

The results of weighting of each criterion based on the Shannon entropy method are presented (Table 1). Based on the results, the variables of annual temperature and annual precipitation with a weight of 0.1488 and 0.1019, respectively, had the highest weight and the criteria of rangeland condition with weight of 0.00001 had the lowest weight.

 

Table1. Criteria weight matrix based on Shannon entropy method

Variables	The entropy of each Criteria	Degree of deviation	Normalized  weight	RANK
	Ej	dj	Wj
Elevation (m)	0.9916	0.00838	0.00125	19
T mean (summer) (C°)	0.9999	0.00009	0.00001	28
T mean (year) (C°)	0.0038	0.99616	0.14886	2
Precipitation (summer) (mm)	0.9998	0.00022	0.00003	21
Precipitation (Spring) (mm)	0.9998	0.00022	0.00003	23
Precipitation (year) (mm)	0.3177	0.68231	0.10196	3
Evaporation (mm)	0.9998	0.00005	0.00001	29
T max (summer) (C°)	0.9998	0.00022	0.00003	22
T max (year) (C°)	0.9999	0.00012	0.00002	27
Relative humidity	0.9998	0.00018	0.00003	25
Climate	0.9171	0.08292	0.01239	15
Dry farming (ha)	0.9629	0.03706	0.00554	16
Residential (ha)	0.4266	0.57344	0.08569	7
Length of road- (km)	0.7381	0.26185	0.03913	9
Dam(ha)	0.9998	0.0002	0.00003	24
Vegetation type(I) %	0.9139	0.0861	0.01287	14
Vegetation type(II) %	0.4724	0.52761	0.07884	8
Vegetation type(III) %	0.3282	0.67179	0.10039	4
Grasses%	0.9112	0.08882	0.01327	13
Forbs%	0.8733	0.12668	0.01893	11
Shrubs%	0.9049	0.09515	0.01422	12
Bushy tree%	0.7899	0.21013	0.0314	10
Poor range condition (ha)	0.9999	0.00004	0.00001	30
Population density (n)	0.4002	0.59979	0.08963	5
Educated people (n)	0.4003	0.5997	0.08962	6
Slope%	0.9951	0.00493	0.00074	20
Stream length (km)	0.9816	0.01837	0.00274	18
South aspect %	0.9999	0.00013	0.00002	26
TWI	0.9808	0.01924	0.00287	17

 

Weighting matrix

Using the data of Table 1, scale less weighted matrix was formed for 11 sub-basins of the Shourdareh region of Golestan province, Iran which is presented in Table 2.

 

Table2. Scale weighted super matrix of Shourdareh watershed sub-basins

	Gh1	Gh2	Gh'3	Gh4	Gh5	Gh'6	Gh7	Gh'8	Gh'9	Gh10	Gh'11
Elevation	0.108	0.109	0.086	0.111	0.109	0.091	0.099	0.081	0.073	0.081	0.053
T mean (summer)	0.089	0.089	0.091	0.089	0.089	0.091	0.090	0.092	0.093	0.092	0.095
T mean (year)	0.088	0.088	0.092	0.088	0.088	0.091	0.090	0.093	0.094	0.093	0.097
Precipitation (summer)	0.088	0.091	0.087	0.092	0.094	0.091	0.097	0.090	0.089	0.094	0.087
Precipitation (spring)	0.088	0.091	0.087	0.092	0.094	0.091	0.097	0.090	0.089	0.094	0.087
Precipitation (year)	0.088	0.091	0.087	0.092	0.094	0.091	0.097	0.090	0.089	0.094	0.087
Evaporation	0.090	0.090	0.091	0.089	0.090	0.091	0.090	0.092	0.092	0.092	0.094
T max (summer)	0.090	0.090	0.091	0.089	0.090	0.091	0.090	0.092	0.092	0.092	0.094
T max (year)	0.089	0.089	0.092	0.088	0.089	0.091	0.090	0.092	0.093	0.092	0.096
Relative humidity	0.093	0.093	0.090	0.091	0.093	0.092	0.093	0.086	0.091	0.092	0.086
Climate	0.111	0.111	0.000	0.111	0.111	0.111	0.111	0.111	0.111	0.111	0.000
Dry farming	0.067	0.036	0.151	0.062	0.113	0.060	0.108	0.090	0.163	0.096	0.053
Residential	0.364	0.002	0.152	0.000	0.000	0.000	0.000	0.482	0.000	0.000	0.000
Length of road (Km)	0.139	0.045	0.187	0.000	0.000	0.052	0.000	0.109	0.135	0.082	0.060
Dam	0.000	0.000	0.000	0.000	0.000	0.000	0.000	1.000	0.000	0.000	0.000
Vegetation type (I)	0.167	0.044	0.241	0.097	0.119	0.075	0.029	0.059	0.093	0.036	0.040
Vegetation type (II)	0.005	0.000	0.288	0.464	0.005	0.000	0.000	0.000	0.235	0.004	0.000
Vegetation type (III)	0.585	0.000	0.027	0.000	0.000	0.000	0.000	0.388	0.000	0.000	0.000
Grasses	0.191	0.042	0.229	0.093	0.112	0.070	0.027	0.078	0.088	0.034	0.037
Forbs	0.270	0.032	0.188	0.081	0.086	0.054	0.021	0.141	0.073	0.026	0.029
Shrubs	0.206	0.038	0.220	0.100	0.102	0.064	0.025	0.092	0.088	0.031	0.034
Bushy tree	0.310	0.014	0.160	0.144	0.040	0.024	0.009	0.188	0.087	0.013	0.013
Poor range condition	0.399	0.000	0.018	0.000	0.000	0.000	0.000	0.583	0.000	0.000	0.000
Population density	0.490	0.000	0.105	0.000	0.000	0.000	0.000	0.405	0.000	0.000	0.000
Educated people	0.519	0.000	0.111	0.000	0.000	0.000	0.000	0.370	0.000	0.000	0.000
Slope	0.112	0.098	0.106	0.109	0.093	0.089	0.065	0.086	0.084	0.074	0.085
Stream l length	0.128	0.053	0.138	0.111	0.111	0.072	0.063	0.083	0.101	0.067	0.072
South slope	0.089	0.056	0.117	0.085	0.066	0.061	0.099	0.080	0.146	0.131	0.070
TWI	0.088	0.088	0.090	0.088	0.091	0.091	0.092	0.093	0.093	0.090	0.096

.

 

Fit functions

The proportional function of most criteria is the V-shape function, because the V-shape functions are the best choice for most quantitative criteria, and it is a special case of the linear preference function, which covers even small differences. The proportional function for qualitative criteria is also the usual function (Nasiri et al., 2013). The status of each criterion and the corresponding functions were presented in Table 3. At this stage, the Max and Min functions were determined for each criterion. Thus, according to the purpose of the research, among the selected criteria, the criteria that prevent fire risk were selected as the Min function and the criteria that increase the fire risk were selected as the Max function.

 

Table3. Status and functions of the criteria used for the PROMETHEE II method

	Min/Max	Preference Fn.	Preference
T mean (year)	max	V-shape	1.34
T mean (summer)	max	V-shape	1.21
Precipitation(year)	min	V-shape	27.51
Precipitation(summer)	min	V-shape	5.57
Precipitation(spring)	max	V-shape	9.18
Evaporation	max	V-shape	46.76
T max (year)	max	V-shape	1.61
T max (summer)	max	V-shape	1.37
Relative humidity	min	V-shape	3.26
Climate	max	Usual	1.12
Dry farming	max	V-shape	402.93
Residential	max	V-shape	21.46
Length of road	max	V-shape	24.22
Dam	max	V-shape	1.14
Vegetation type(I)	max	V-shape	747.54
Vegetation type(II)	max	V-shape	70.54
Vegetation type(III)	max	V-shape	324.85
Grasses	max	V-shape	401.57
Forbs	max	V-shape	100.98
Shrubs	max	V-shape	378.2
Bushy tree	max	V-shape	68.26
Poor range condition	max	V-shape	33.33
Population density	max	V-shape	408.85
Educated people	min	V-shape	282.95
Slope	max	V-shape	11.25
Stream length	max	V-shape	13366.5
South slope	max	V-shape	15.43
Twi	min	V-shape	1.21

The amount of Phi

The Phi rate for each of the eleven sub-basins of the study area based on the criteria of that
sub-basin is presented in Table 4.

Negatively Phi indicates the weakness of one sub-basin compared to other sub-basins. The higher Phi value, determines that sub-basin is high in terms of the criteria under consideration. For example, regarding the elevation criterion, the sub-basin (Gh6) has the lowest amount of Phi (-0.55), therefore, it has the lowest fire risk in terms of the elevation criterion compared to other sub-basins. If the sub-basin (Gh5) with the highest amount of Phi (+0.90) has the highest fire risk compared to other sub-basins.

 

Table4. Amount of phi sub-basins of Shourdareh watershed by PROMETHEE II method

Variables	sub-basins phi
	Gh 1	Gh 2	Gh 3	Gh 4	Gh 5	Gh 6	Gh 7	Gh 8	Gh 9	Gh 10	Gh 11
Elevation	0.14	-0.46	0.31	0.53	0.90	-0.55	0.30	-0.48	0.00	-0.49	-0.21
T mean (year)	0.00	-0.10	0.00	0.00	0.50	-0.10	0.00	-0.10	0.00	-0.10	-0.10
T mean (summer)	0.17	0.17	0.17	0.17	0.88	-0.68	0.17	-0.68	0.17	0.17	-0.68
Precipitation (year)	0.54	0.45	0.20	0.30	0.65	-0.22	-0.51	-0.51	-0.02	-0.02	-0.85
Precipitation (summer)	0.54	0.41	0.25	0.25	0.54	-0.28	-0.44	-0.44	0.10	-0.09	-0.84
Precipitation (spring)	-0.52	-0.42	-0.20	-0.31	-0.70	0.26	0.48	0.48	0.02	0.02	0.88
Evaporation	0.14	-0.46	0.31	0.51	0.90	-0.55	0.31	-0.48	0.01	-0.48	-0.20
T max (year)	-0.10	-0.10	0.10	0.10	0.70	-0.40	0.10	-0.10	-0.10	-0.10	-0.10
T max (summer)	0.00	-0.10	0.00	0.00	0.60	-0.10	0.00	-0.10	-0.10	-0.10	-0.10
Relative humidity	0.18	-0.24	0.81	-0.13	0.81	-0.13	-0.13	-0.42	-0.24	-0.24	-0.24
Climate	0.90	-0.20	-0.20	-0.20	-0.20	-0.20	0.90	-0.20	-0.20	-0.20	-0.20
Dry farming	0.73	-0.30	-0.01	0.82	-0.47	-0.36	0.07	0.30	-0.38	-0.65	0.24
Residential	0.20	0.83	0.92	-0.24	-0.24	-0.24	-0.24	-0.24	-0.24	-0.24	-0.24
Length of road	0.89	0.57	0.21	0.53	-0.33	-0.26	-0.10	-0.18	-0.41	-0.48	-0.44
Dam	-0.10	-0.10	1.00	-0.10	-0.10	-0.10	-0.10	-0.10	-0.10	-0.10	-0.10
Vegetation type (I)	0.95	0.64	-0.23	0.03	-0.38	0.06	-0.41	0.24	-0.11	-0.34	-0.46
Vegetation type (II)	0.78	-0.18	-0.19	-0.19	-0.19	0.96	-0.19	-0.19	-0.19	-0.19	-0.19
Vegetation type (III)	-0.14	0.95	-0.21	0.82	-0.21	-0.21	-0.21	-0.21	-0.21	-0.21	-0.21
Grasses	0.92	0.76	-0.09	-0.02	-0.39	0.02	-0.41	0.18	-0.15	-0.36	-0.46
Forbs	0.67	0.93	0.35	-0.10	-0.38	-0.05	-0.39	-0.02	-0.22	-0.37	-0.42
Shrubs	0.89	0.82	0.00	-0.02	-0.40	0.08	-0.42	0.09	-0.20	-0.37	-0.46
Bushy tree	0.40	0.95	0.40	0.00	-0.37	0.31	-0.37	-0.25	-0.32	-0.37	-0.39
Poor range condition	-0.17	0.84	0.95	-0.20	-0.20	-0.20	-0.20	-0.20	-0.20	-0.20	-0.20
Population density	0.05	0.92	0.86	-0.23	-0.23	-0.23	-0.23	-0.23	-0.23	-0.23	-0.23
Educated people	-0.07	-0.94	-0.83	0.23	0.23	0.23	0.23	0.23	0.23	0.23	0.23
Slope	0.55	0.69	-0.25	-0.25	-0.25	0.62	-0.56	0.12	-0.08	0.20	-0.77
Stream length	0.78	0.67	-0.16	0.21	-0.36	0.39	-0.44	0.40	-0.35	-0.65	-0.49
South slope	0.40	-0.02	-0.20	0.91	-0.48	-0.20	0.54	-0.32	-0.44	-0.44	0.26
Twi	0.00	0.00	-0.10	-0.10	-0.10	0.30	0.00	0.00	0.00	0.00	0.00

Ranking of Shourdareh watershed sub-basins of Golestan province

Results shows the ranking of sub-basins based on the Phi value for each sub-basin. According the results, sub-basin (Gh3) has the highest Phi value with a score of 100, so Gh3 has the highest fire potential compared to other sub-basins. While, the sub-basin (G7) has the lowest fire susceptibility in terms of fire compared to other sub-basins, Because of the lowest Phi and with a score of 30 (Fig 2). The low and negative value of Phi indicates the weakness of one sub-basin compared to other sub-basins in terms of fire susceptibility. While, the high and positive amount of Phi shows that sub-basin is more susceptible than the other sub-basins. Also, the vectors show the superiority of sub-basins to each other. The ranking results of Shourdareh watershed sub-basins are presented in Table 5.

s.

 

Fig.2. Ranking network by PROMETHEE II

 

Based on the results of K-means clustering presented in (Table 5), the rangelands of sub-basins Gh3, Gh8 and Gh1 with Phi rates of 0.335, 0.148 and 0.239, respectively, were in the high susceptibility to fire class, Sub-basins Gh2, Gh5, Gh6 and Gh7 with phi rate of -0.20, -0.117, -0.136 and -0.241 were in the medium susceptibility to fire class and rangelands of sub-basins Gh9, Gh10 and Gh11 with Phi 0.114, -0.078 and 0.025, respectively, were in the low susceptibility to fire class.

 

Table5. Ranking of Shourdareh watershed sub-basins by PROMETHEE II method

 

Results of the fire susceptibility map of Golestan province show in three classes e  including: low susceptibility to fire which is marked in green on the map (Fig.3), In these sub-basins, the amount of participation, topographic wetness index and relative humidity, which are the factors influencing the fire, are higher than other sub-basins. High susceptibility to fire which is marked in red on the map (Fig.3), It is noteworthy that in these sub-basins, criteria such as population density and the number of villages were higher than other sub-basins, which has increased the fire potential. Also, moderate susceptibility to fire which is marked in yellow on the map (Fig.3), in this class, the values of criteria are intermediate between the two classes of high susceptibility to fire and low susceptibility to fire.

 

Fig.3. Fire susceptibility map of Golestan province, Iran

Discussion

In the present study, various environmental and social criteria were used to determine the areas prone to rangeland fires in the Shourdareh region of Golestan. The main criteria which were used in this research included topographic criteria, biological criteria and climatic criteria that are consistent with Eskandari et al., (2013), research. Topographic criteria were included three important sub-criteria of elevation, slope and aspect; the climate was including four sub-criteria of temperature, evaporation, humidity and precipitation, which was consistent with Johar Jafar (2016). Also, in this research used different criteria such as land use, distance from the road, distance from residential areas and distance from the river criteria which is consistent with criteria used by Sitanggang et al., (2013) and Parajuli et al., (2020). The results by weighting criteria of present research based on Shannon entropy method showed that the criteria of annual temperature and annual precipitation with weight of 0.14 and 0.10, respectively, had the highest weight to determine the areas prone to rangeland fires in the Shourdareh region of Golestan. Other researchers such as Ariapour and Mohammad Shariff (2014), Abedi Gheshlaghi and Valizadeh Kamran (2018) also emphasized the importance of precipitation and temperature criteria in the field of fire occurrence. Based on the results of the present research, it was determined that the ranking of sub-basins based on the phi value for each sub-basin, sub-basin G3 with Phi rates of 0.335 has the highest Phi value and a score of 100 and has the highest fire potential compared to other sub-basins. The G7 sub-basin with Phi rates of -0.241 has the lowest fire susceptibility in terms of fire with the lowest Phi compared to other sub-basins. Based on the results of this research, the sub-basins that had higher ratings such as G3 that were more susceptible to fire than other sub-basins, have less amount of participation, relative humidity and topographic humidity causes the creation of suitable environmental conditions and therefore the fuel to be prepared for fires. In this regard, Janbazghobadi (2018) also confirmed the current research and stated that temperature is one of the most important criteria in the occurrence of fire as high temperature increases evapotranspiration and decreases humidity and therefore increases the fire. Precipitation also reduces the risk of fire by regulating soil uptake and moisture. These results are consistent with the results of Ariapour and Mohammad Sharif, (2014) who stated that high spring precipitating causes biomass growth (fuel factor) and high evaporation causes fuel drying and capability of flammability of most materials. It is also consistent with the results of the research of Dashti et al., (2021) that showed the highest probability of fires occurring in the period between June and September, when the temperature, evaporation, fuel content is high and humidity is low. The highest slope percentage of the G3 sub-basin is located in the southern aspect. Aspect plays a key role in fire behavior because in the northern hemisphere, the southern slopes are in a better position than the northern slopes in the event of a fire. The results of Musaybeigi and Mirzabeigi, (2016) showed that the highest incidence of fires occurred in areas with southern aspects due to receiving the most amount of sunlight. Vice versa, the sub-basins that have less potential, such as G7 have more participation, relative humidity, topographic humidity and less evaporation that is consistent with Eslami, et al., (2019) and Polat, et al., (2019) researches. Also, the results of the present study showed that the population density is high in the fire-prone sub-basins, the access way to these sub-basins, proximity to roads and settlements is more that is consistent with Johar Jafar (2016) and Ajin et al., (2017). The results of Dalir et al., (1400) research also showed that human criteria play a decisive role in the occurrence of forest and rangeland fires. Research Yang et al., (2007) showed that more than 90% of fires are caused by humans. In this regard, based on research findings Arndt et al., (2013), forest roads and population density have a significant role in the occurrence of fires. In other words, increasing access by road puts more pressure on forest and rangeland ecosystems, and this effect is exacerbated by population growth. Also, results of present research based on the Shannon entropy method illustrated that the criteria of annual temperature and annual precipitation with weight of 0.1488 and 0.1019, respectively, had the highest weight to determine the areas prone to rangeland fires in the Shourdareh region of Golestan that is consistent with Azizi and Yosephi, 2009.

In order to investigate the compatibility of real fires with the fire prone areas resulting from this study, the results of burned areas prepared by the General Department of Natural Resources of Golestan Province, were used. Based on the kappa coefficient, which is equal to 0.82, according to the classification of kappa Coefficient, the model has a good and acceptable accuracy. This result is consistent with the research result of Al-Fugara et al., (2021), that stated areas where fires have occurred in the past need to be recorded and then compared with the fire potential areas resulting from the methods used to determine the relationship between them, correspond. Foody (2020) also used kappa coefficient to evaluate their model and stated that kappa coefficient greater than 0.80 indicates the high accuracy of the model. In general, the proposed method is a reliable screening tool to identify fire prone areas that it can help the authorities to take preventive measures. In order to increase the accuracy and precision in future research, efforts should be made to use as much as possible all the criteria affecting the fire phenomenon in determining critical and fire-prone areas.

Conclusion

In general, the proposed method is a reliable screening tool to identify fire prone basins that it can help the authorities to take preventive measures. In order to increase the accuracy and precision in future research, also, efforts should be made to use as much as possible all the criteria affecting the fire phenomenon in determining critical and fire-prone areas. It is recommended that more research be done on the use of other techniques such as data mining, neural networks and other decision-making techniques to identify more powerful techniques for sub-ranking to identify fire prone basins. Finally, it is suggested that suitable management measures be used for sub-basin with high potential for fire such as G3, as well as appropriate management measures to prevent the spread of fire to other sub-basins.

 

References

Abedi Gheshlaghi, H. and Valizadeh Kamran, K.H. 2018. Assessment and zoning of forest fire risk using multi-criteria decision making and GIS techniques. Journal of Environmental Hazards, 7 (15): 49-66. (In Persian).

Ahmadi, J., Farzam, M. and Lakzian, A., 2017. -Investigating Effects of a Prescribed Spring Fire on Symbiosis between Mycorrhiza Fungi and Range Plant Species. Journal of Rangeland Science, 7(2): 138-147.

Ajin, R.S., Loghin, A.M., Vinod, P.G. and Jacob, M.K. 2017. The Risk Analysis of Potential Forest Fires in a Wildlife Sanctuary in the Western Ghats (Southwest Indian Peninsula) Using Geospatial Techniques. International Journal of Health System and Disaster Management, 5(1): 18-23.

Akhzari, D., Eftekhari Ahandani, S., Attaeian, B. and Ildoromi, A., 2013. Prediction of Land Use Management Scenarios Impact on Water Erosion Risk in Kashidar Watershed, Azadshahr, Golestan Province. Journal of Rangeland Science, 3(2): 165-176.

Akinola, O.V. and Adegoke, J. 2019. Assessment of Forest Fire Vulnerability Zones in Missouri, United States of America. International Journal of Sustainable Development & World Ecology, 26(3): 251-257.

Al-Fugara, A., Mabdeh, A.N., Ahmadlou, M., Pourghasemi, H.R., Al-Adamat, R., Pradhan, B. and Al-Shabeeb, A.R 2021. Wildland Fire Susceptibility Mapping Using Support Vector Regression and Adaptive Neuro-Fuzzy Inference System-Based Whale Optimization Algorithm and Simulated Annealing. ISPRS International Journal of Geo-Information, 10(6): 1-28.

Alizadeh, A., Kamali, G.H.A., Mousavi, F. and Mousavi Baighi, M., 2022. Weather and Climatology, 17th edition, Ferdowsi University of Mashhad, 392 pages.

Anagnostopoulos, K.P., Petalas, C. and Pisinaras, V., 2005. Water Resources Planning Using the AHP and PROMETHEE Multicriteria Methods: The Case of NESTOS River –Greece, the 7th Balkan Conference on Operational Research ìBACOR 05î Constanta, Romania, 1-12.

Ariapour, A. and Mohammad Shariff, A.R.B. 2014. Rangeland Fire Risk Zonation Using Remote Sensing and Geographical Information System Technologies in Boroujerd Rangelands, Lorestan Province, Iran. Ecopersia, 2(4): 805-818.

Arndt, N., Vacik, H., Koch, V., Arpaci, A. and Gossow, H., 2013. Modeling Human-Caused Forest Fire Ignition for Assessing Forest Fire Danger in Austria. Forest-Biogeosciences and Forestry, 6(6): 315-325.

Arzani, H., 1997. Evaluation Initiative Directions Different Climatic Zones of Rangelands, Research Institute Forests and Rangelands, Tehran, Iran, 67 pages. (In Persian).

Asadian, G.H., Azimnejad, Z. and Bakhtiari, M. R., 2022. Effects of Fire on Vegetation cover and forage production of Solan Rangeland in Hamedan Province, Iran. Journal of Rangeland Science, Available Online DOI, 10.30495/RS.2022.1936787.1570.

Asghryzadeh, A. and Nasrallahy, M. 2007. Ranking Companies Based on PROMETHEE Method. Journal of Human Sciences, 11 (3): 84-59.

Azizi, G.H. and Yosefi, H., 2009. Grmbad and Forest Fires Burning in the Province of Mazandaran and Gilan 25-30 Dec 2005. Journal of Geographical Research. 92, 3-28. (In Persian).

Banias, G., 2010. Assessing Multiple Criteria for the Optimal Location of a Construction and Demolition Waste Management Facility. Journal of Building and Environment, 45:2317-2326.

Chahoki Zare, Mohammad Ali, 2012, Data analysis in natural resources research using SPSS Software. Tehran University Jihad Publications, Tehran, Iran, 320 Pp. (In Persian).

Chou, T.Y., Lin, W.T., Lin, C.Y., Chou, W.C. and Huang, P.H. 2004. Application of the PROMETHEE Technique to Determine Depression Outlet Location and Flow Direction in DEM, Journal of Hydrology volume, 287: 49-61.

Chuvieco, E., Aguado, I., Jurdao, S., Pettinari, M. L., Yebra, M., Salas, J. and Martínez-Vega, F. J., 2012. Integrating Geospatial Information into Fire Risk Assessment. International Journal of Wildland Fire, 23(5): 606-619.

Cohen, J., 1960. A Coefficient of Agreement for Nominal Scales. Educational and Psychological Measurement, 20(1):37–46.

Dalir, Z., Farajzadeh, Z. and Zibaei, M. 2021. Socio-Economic and Environmental Factors Affecting the Occurrence of Forest Fires in Iran and Coping Strategies. Journal of Agricultural Economics and Development. 29 (1): 113): 25-55. (In Persian).

Darkwah, K.F., Inusah, A. and Amponsah, S.K., 2012. Logistic Preference Function for Preference Ranking Organization Method for Enrichment Evaluation (PROMETHEE) Decision Analysis. African Journal of Mathematics and Computer Science Research, 5(6): 112-119.

Dashti, S.H., Amini, J., Ahmadi Thani, N. and Javanmard, A. 2021. Zoning of Fire-Prone areas in Forest Ecosystems of North Zagros (Case study: Sardasht forests in West Azerbaijan. Journal of Environmental Hazards, 10 (30): 105-126. (In Persian).

Dehghan, M., 2017. Comparison of Different Multi-Criteria Decision-Making Methods in Order to Determine Suitable Areas for Implementing Some Water and Soil Protection Measures. Master Thesis, Faculty of Natural Resources and Environment, Ferdowsi University of Mashhad. (In Persian).

Enoh, M.A., Uzoma, C.H.O. and Needam, Y.N. 2021. Identification and Modelling of Forest Fire Severity and Risk Zones in the Cross–Niger Transition Forest with Remotely Sensed Satellite Data. Egyptian Journal of Remote Sensing and Space Science. 24(3): 879-887.

Eskandari, S., Oladi Qadiklaei, J., Jellilund, H. and Serajian, M.R. 2013. Modeling and Forecasting Fire Risk in the Forests of Seh Neka-Zalmarud Section Using GIS. Iranian Journal of Forest and Poplar Research, 21 (2): 217-203. (In Persian).

Eslami, R., Azarnoosh, M.R., Kialashki, A. and Kazemnejad, F. 2019. Modeling the Spread of Forest Fire Based on GIS and Cellular Automation (Case Study: Babelroud watershed, Mazandaran). Journal of Geographical Sciences, 16 (32): 107-117.

Foody, G.M. 2020. Explaining the Unsuitability of the Kappa Coefficient in the Assessment and Comparison of the Accuracy of Thematic maps Obtained by Image Classification. Remote Sensing of Environment, 239: 1-11.

Gai, C., Weng, W. and Yuan, H. 2011. GIS-Based Forest Fire Risk Assessment and Mapping. In 2011 Fourth International Joint Conference on Computational Sciences and Optimization. pp. 1240–1244. IEEE. April 2011,Kunming, Yunnan, China. https://doi.org/ 10.1109/ CSO. 2011.140

Ganteaume, A., Camia, A., Jappiot, M., San-Miguel-Ayanz, J., Long-Fournel, M. and Lampin, C. 2013. A Review of the Main Driving Factors of Forest Fire Ignition over Europe. Environmental Management, 51(3): 651-662.

Huang, P. and Tsai, W. 2010. Using Multiple-Criteria Decision-Making Techniques for Ecoenvironmental Vulnerability Assessment: A Case Study on the Chi-Jia-Wan Stream Watershed, Taiwan, Journal of Environment Monitoring Assess, 168: 141–158.

Janbazghobadi, G.R. 2018. Investigation of Forest Fire Risk Areas in Golestan Province, Based on the Fire Risk Index (FRSI) Using GIS Technique. Journal of Spatial Analysis of Natural Hazards, 6(3): 102-89.

Johar Jafar, D., 2016. Fire Risk Zoning in Forest Areas in Kermanshah Province Using Experimental Models and Fuzzy Logic. Master Thesis in Geomorphology, Department of Natural Geography, Faculty of Literature and Humanities, Razi University, 99 p. (In Persian).

Kuncova, M. and Seknickova, J., 2022. Two-stage weighted PROMETHEE II with results’ visualization. Central European Journal of Operations Research, 30:547–571.

Maragoudaki, R. and Tsakiris, G. 2005. Flood Mitigation Planning Using PROMETHEE. Athens, 5-10.

Mesdagi, M., 2003. Pasture and rangeland in Iran. Astan Quds Press, Mashahd, Iran, 320 p.

Musaybeigi and, M. and Mirzabeigi, F., 2016. Forest Fire Risk Zoning Using Network Analysis Model and Geographic Information System (Case Study: Manshet and Qalarang Protected Area). Journal of Environmental Science, 14 (4): 175-188.

Nasiri, H., Darvishi Boloorani, A., Faraji Sabokbar, H.A., Jafari, H.R., Hamzeh, M. and Rafii, Y. 2013. Determining the Most Suitable Areas for Artificial Groundwater Recharge Via an Integrated PROMETHEE II-AHP Method in GIS Environment (Case Study: Garabaygan Basin, Iran). Journal of Environmental Monitoring and Assessment, 185(1): 707-718.

Parajuli, A., Gautam, A. P., Sharma, S. P., Bhujel, K. B., Sharma, G., Thapa, P. B. and Poudel, S. 2020. Forest Fire Risk Mapping Using GIS and Remote Sensing in Two Major Landscapes of Nepal. Geomatics, Natural Hazards and Risk, 11(1): 2569-2586.

Polat, S., Ghasemi Aghbash, F. and Mahdavi, A. 2019. Fire Risk Zoning in the Forests of Ilam District. Journal of Forest Research and Development 6 (1): 152-136.

Pournemati, A., Sepehry, A., Barani, H. and Sefidi, K. 2021. Indigenous Knowledge of Shepherds in Determining the Flammability of Vegetation: A Case Study of Khalkhal Semi-Steppic Rangelands of Iran. Journal of Rangeland Science, Available Online DOI,10.30495/rs.2022.685974.

Rahimi, I., Azeez, S. N. and Ahmed, I.H. 2020. Mapping Forest-Fire Potentiality Using Remote Sensing and GIS, Case Study: Kurdistan Region-Iraq. In Environmental Remote Sensing and GIS in Iraq .pp. 499-513. Springer, Cham.

Sakellariou, S., Tampekis, S., Samara, F., Flannigan, M., Jaeger, D., Christopoulou, O. and Sfougaris, A. 2019. Determination of Fire Risk to Assist Fire Management for Insular Areas: The Case of a Small Greek Island. Journal of Forestry Research, 30(2): 589-601.

Sitanggang, I. S., Yaakob, R., Mustapha, N. and Ainuddin, A. N. 2013. Predictive Models for Hotspots Occurrence Using Decision Tree Algorithms and Logistic Regression. Journal of Applied Sciences, 13(2): 252-261.

Yang, J., He, H.S., Shifley, S.R. and Gustafson, E.J. 2007. Spatial Patterns of Modern Period Human-Caused Fire Occurrence in the Missouri Ozark Highlands. Forest Science, 53(1): 1-15.

Zhang, Y., Lim, S. and Sharples, J. J. 2016. Modelling Spatial Patterns of Wildfire Occurrence in South-Eastern Australia. Geomatics, Natural Hazards and Risk, 7(6): 1800-1815.

Zhi-hong, Z., Yi, Y. and Jing-nan, S. 2006. Entropy Method for Determination of Weight of Evaluating in Fuzzy Synthetic Evaluation for Water Quality Assessment, Journal of Environ Mental Science, 18(5): 1020-1023.

 

تعیین مناطق مستعد آتش سوزی با استفاده از روش تصمیم گیری پرومته

 مطالعه موردی: حوزه آبخیز شوردره، استان گلستان

 

چکیده. تصمیم‌گیران در مدیریت و برنامه‌ریزی آتش‌سوزی با تعداد زیادی گزینه و معیار مواجه هستند. به منظور مدیریت کارآمد و تصمیم‌گیری صحیح در زمینه وقوع آتش‌سوزی، معیارهای مختلفی شامل معیارهای فنی، اقتصادی، اجتماعی و زیست‌محیطی بایستی توأمان در نظر گرفته شوند. عملیات مدیریتی برای پیشگیری و کنترل وقوع آتش‌سوزی جنگل‌ها و مراتع زمانی کارآمد هستند که مناطق مستعد آتش‌سوزی شناسایی و اقدامات اصلاحی و مدیریتی بر روی این مناطق متمرکز گردد. بنابراین انجام پژوهش حاضر، با هدف شناسایی مناطق مستعد وقوع آتش‌سوزی به منظور مدیریت و برنامه‌ریزی صحیح این مناطق و با کمک روش تصمیم‌گیری PROMETHEE در حوزه اّبخیز شوردره استان گلستان انجام گرفت. بدین منظور، معیارهای مختلف محیطی و اجتماعی مورد استفاده قرار گرفتند. سپس وزن‌دهی معیارها به روش آنتروپی شانون انجام شد. با توجه به وزن و مقدار معیارهای مختلف به دست آمده برای هر زیر حوزه، داده‌های مربوط به زیرحوز‌ها با استفاده از تکنیک پرومته II تحلیل شدند. بر اساس نتایج به دست آمده، مراتع زیرحوزه های 3 و 8 و 1 با میزان فی به ترتیب 335/0، 148/0 و 239/0 در کلاس مناطق با استعداد آتش سوزی زیاد، مراتع زیرحوزه های 2، 5، 6 و 7 با میزان فی به ترتیب 220/0-، 117/0-، 136/0- و 241/0- در کلاس مناطق با استعداد آتش سوزی متوسط و مراتع زیرحوزه های 9، 10 و 11 با میزان فی به ترتیب 114/0، 078/0- و 025/0 در کلاس مناطق با استعداد آتش سوزی کم قرار گرفتند. برای ارزشیابی روش، نتایج حاصل از این تحقیق، با نقشه مناطق آتش‌سوزی تهیه شده توسط اداره کل منابع طبیعی استان گلستان، مقایسه گردید. نتیجه نشان داد که بر اساس ضریب کاپای به دست آمده که برابر 82/0 می‌باشد، روش پرومته از دقت خوب و قابل قبولی برخوردار بود. بنابراین، روش پیشنهادی یک روش غربالگری قابل اعتماد برای شناسایی مناطق در معرض خطر آتش سوزی است که می تواند به مقامات در انجام اقدامات پیشگیرانه کمک کند.

کلمات کلیدی: آتش‌سوزی، تکنیک‌ فرا‌رتبه‌ای، وزن‌دهی شانون، اولویت‌بندی، شوردره استان گلستان.

 

 

15

 

---

[1]  Preference Ranking Organization Method For Enrichment Evaluation

[2]  Multi-Criteria Decision-Making

[3] Digital elevation model 

row	sub-basins	cluster				score	ranking
1	Gh9	1	0.244	0.130	0.114	62	4
2	Gh8	2	0.301	0.157	0.143	66	3
3	Gh7	3	0.073	0.314	-0.241	30	11
4	Gh6	3	0.077	0.214	-0.137	37	9
5	Gh5	3	0.115	0.233	-0.118	39	8
6	Gh4	1	0.160	0.223	-0.063	43	6
7	Gh3	2	0.421	0.086	0.336	100	1
8	Gh2	3	0.050	0.270	-0.221	31	10
9	Gh11	1	0.239	0.215	0.025	52	5
10	Gh10	1	0.138	0.217	-0.078	42	7
11	Gh1	2	0.389	0.149	0.240	81	2

---