تعیین مناطق مستعد کشت علوفه ارزن در استان اردبیل با استفاده از روش‌های AHP, ANP, DEMATEL, Antropy Shanon and WLC,

Abstract
Millet fodder plant is one of the important plants in Iran and the world, and due to its abundance of nutrients and its resistance to drought, it has a special advantage over other forages.The purpose of this research is to identify the areas prone to millet fodder cultivation in Ardabil province with multi-criteria methods and using the data of climatic elements and factors; Rainfall, average temperature, minimum temperature, maximum temperature, altitude, slope and soil depth.AHP, ANP, DEMATEL, Shannon Antropy and WLC methods were used to determine the weights of criteria and options.By using the weighted linear combination method in the GIS environment, information layers were integrated and the final map of millet fodder cultivation was prepared.The results showed that about 25% are very suitable, 36% are suitable, 30% are slightly suitable and 4.05% are unsuitable for millet fodder cultivation in Ardabil province.The results of AHP, ANP, DEMATE models showed that rainfall and temperature criteria, and in the Shannon Antropy model, height and slope have the most influence during the growth period of millet fodder.The output classification of two WLC and Antropy Shanon models about options showed that millet fodder cultivation areas overlap in both methods and the northern and central regions of the province have suitable conditions for millet fodder cultivation.

Introduction
Annually,Millet is one of the traditional cereals in dry and semi-arid areas of tropical regions, which has a high tolerance to drought and salinity stress.Millet is the sixth most important cereal after wheat, rice, corn, barley and sorghum. Despite the agricultural importance of millet in the past, it also has a special place in agriculture. The adaptability and resistance of this plant to adverse environmental conditions has spread it in many tropical countries of the world, from West Africa to the Indian subcontinent, and has made this plant an important agricultural plant.In addition to being used as fodder for livestock and poultry, millet seed is also consumed by humans and today it forms the food of about 500 million people on the planet. This plant has an area of more than 19 million hectares in Africa and about 15 million hectares in Asia. Apart from this, this valuable plant is more or less cultivated in other parts of the world. Today, various plants are cultivated to produce fodder, among which millet fodder has a special place due to its adaptability to dry climate conditions. The statistics of the cultivated area of millet in 2017 is about 8.9 thousand hectares and its production is reported as 18.4 thousand tons.According to the latest statistics of FAO in 2017, the cultivated area of millet in the world was 31.2 million hectares, from which 28.4 million tons of millet were produced..Khajepour (2012) described the optimal climatic needs of millet fodder in a description; Plant millet is thermophilic. The minimum soil temperature for germination is 14 degrees Celsius. The ideal temperature for millet germination is 20 to 30 degrees Celsius. The base temperature for millet is 10°C and the maximum is 30°C. The growth of millet in the average day and night temperature is less than 15 °C and stops at 5 °C. Millet grows in most soils except coarse sandy soils. The threshold amount of water for millet production in rainy conditions is estimated to be about 150 mm.Considering that Ardabil province is one of the hubs of animal husbandry in the country, and every year, due to the lack of fodder, livestock farmers face a severe shortage of fodder. Therefore, the development of millet cultivation in most areas of the province due to low water requirements can solve the fodder shortage in the province and this study can also help in locating suitable areas for the cultivation of this product according to the research records.

Methodology
Ardabil province is located in the northwest of Iran. Its geographical location is in latitude 37°45' to 39°42' and longitude 47°3' to 48°55'. Its area is 17,953 square kilometers (Sebahani and Del Ara, 1402). In the cold seasons, it is influenced by the north and northwest migratory currents.
Study data: In this research, from the data of climatic elements; Annual precipitation in millimeters, temperature (average, minimum and maximum) in centigrade from synoptic and rain gauge stations during the statistical period (1371 to 1401) and climatic factors; Height in meters above sea level, slope in percentage and soil depth in centimeters were used. Hierarchical method: AHP implementation in three modes; 1- Creating a tree hierarchy. 2- Pairwise comparison of the criteria of each level compared to one of the high level criteria 3-Weights are calculated. Network analysis process: (ANP) This method is one of the multi-criteria decision making techniques and extended form (AHP) which is used to solve the criteria and options that are not independent from each other (Ataei, 2019). The steps of the ANP method are: 1-Building the analysis model (network). 2- Formation of pairwise comparison matrices and calculation of weight vectors. 3-Formation of the primary supermatrix. 4- Formation of weighted super matrix. 5- Forming the limit supermatrix. Shannon's entropy model: This model is mainly used to rank the development in different fields of science, and the greater the dispersion in the value of a specific index, the more important that index is (Shannon, 1988). It is used to calculate the weight of indicators. In this model, m is the number of criteria (indices), n is the number of options, and aij is the evaluation of the i-th option for the j-th index. DEMATEL method This technique only examines the relationships of indicators in a situation where they affect each other and are affected by each other. The only input of this technique is direct matrix. This matrix is always square and its dimensions are equal to the number of indices. Weighting method (WLC): The simple weighting method is based on the concept of weighted average. Then, by multiplying the relative weight by the value of that attribute, the final value of each alternative is determined, and the alternative that has the most weight will be the most suitable alternative for the intended purpose.
Conclusion
Each of the criteria was prepared based on the optimal climatic requirements of millet fodder cultivation with the WLC method of weighting and the location map of millet cultivation for each criterion. The results showed that based on the amount of rainfall, 18% of the province's area can be harvested only once a year, and irrigation is needed to harvest 50% of the province's area. There is no limit for millet cultivation in 39% of the province's average temperature criteria, and in 45%, the cultivation time must correspond to the change of hot months.Therefore, there is no need to spend enormous expenses in this area.In terms of minimum temperature, about 36% and maximum temperature is 58% of the area of the province suitable for millet cultivation. The results of topographical data analysis showed that about 49% based on height, 45% based on land slope and 36% based on soil depth are suitable for millet cultivation.The results of benchmark analysis based on WLC method show that; In areas where the soil depth is less than 20 cm, the slope is more than 13% and the altitude is 1600 meters above sea level, the minimum temperature is less than 7 and the maximum temperature is less than 13 degrees Celsius, and the rainfall is less than 250 mm, the conditions are not suitable for millet cultivation.In this research, for the location of millet fodder cultivation, AHP, ANP, DEMATEL, Antropy SHANON, WLC methods and also from the criteria; Precipitation, average temperature, minimum temperature, maximum temperature, altitude, slope and soil depth were used. Using Exper Choice and SuperDicision software, the weight of the criteria was calculated using the optimal climatic requirement of millet.The results showed that precipitation and average temperature have the most important effect during the growth period of millet fodder. The results of combining the criteria with the weighted linear combination method in the GIS environment showed that about 25% are very suitable, 36% are suitable, 30% are slightly suitable and 4.05% are unsuitable for millet cultivation in Ardabil province.