Analysis On Spatial Distribution And Inuencing Factors Of China National Forest Villages Based on GIS

16 As a carrier to promote rural greening and beautification and further implement the rural 17 revitalization strategy, it is of great significance to study the spatial distribution characteristics of 18 China national forest villages and its influencing factors. Taking 7586 China national forest 19 villages as examples, the spatial distribution characteristics and influencing factors of China 20 national forest villages were studied by using such methods as nearest neighbor index, Tyson 21 polygon, cold hot spot analysis, standard deviation ellipse and kernel density index. The results 22 showed that: (1) the overall clustering distribution characteristics of China national forest villages 23 were significant, and the distribution type was agglomeration. There was no breakthrough in the 24 Hu Huanyong Line, and the southeast of the line was the main concentration area. (2) From the 25 perspective of spatial clustering, it shows the distribution characteristics of "hot spots in the south 26 and cold spots in the north". Hot spots are mostly located in the south, represented by Sichuan, 27 Guangdong, Hunan, etc., while cold spots are mostly in the north, mainly in Xinjiang, Xizang, 28 etc.(3) From the perspective of spatial distribution direction, the standard deviation ellipse 29 coincidence degree of the two batches is relatively high. The two batches are distributed in a dense 30 direction from northeast to southwest with Suizhou city, Hubei as the geometric center. The 31 concentration degree of the second batch increases on the basis of the first batch, showing a trend 32 of migration and distribution in the southwest.(4) From the perspective of the distribution 33 characteristics of kernel density, the distribution of kernel density has a strong correlation with 34 two factors, one is the forest vegetation coverage, the other is the distribution location of urban 35 agglomeration;(5) Elevation, slope direction, river basin, traffic are important factors affecting the 36 distribution of China national forest villages, which show the spatial distribution characteristics of 37 "low altitude, positive, near water and convenient transportation". Based on the spatial distribution 38 characteristics and influencing factors, the paper puts forward policy Suggestions for the 39 evaluation and construction of China national forest villages in the future.


43
China national forest village refers to the administrative village which is organized and guided by 44 the National Forestry and Grassland Administration and meets the evaluation standard through 45 comprehensive evaluation through the application of certain quantitative indicators, evaluation 46 methods and evaluation standards [1]. It is the model in the countryside, in the greening degree, 47 ecological civilization, environmental protection industry, quality of life and other aspects of 48 outstanding performance. In the Strategic Plan for Rural Revitalization, it is clearly proposed that 49 the construction of China national forest villages should be taken as a key action, the building of 50 green, beautiful and livable villages should be taken as a key content, and the green coverage rate 51 should be taken as the main index of assessment to continuously and steadily promote rural 52 greening and beautification [2].In response to the central strategic deployment, the National 53 Forestry and Grassland Administration has published the Rural Greening and Beautification 54 Action Plan. According to the plan, China plans to build 20,000 unique and ecologically livable 55 national forest villages and local forest villages by 2020.With the construction of forest villages as 56 the carrier, the natural ecological features of villages can be effectively protected, people's living 57 environment can be improved, green coverage rate can be significantly increased, and green 58 industry can develop rapidly [3].Up to now, China has published the first and second batch of list 59 of national forest villages. Take the lead of national forest villages, gradually develop local forest 60 villages and national model counties for rural greening and beautification, and finally drive all 61 villages to jointly promote rural greening and beautification, so as to realize rural revitalization 62 strategy. In view of this, this paper has two batches of national forest country geographical spatial 63 distribution, the influence mechanism and study about whether its distribution is breakthrough Hu 64 Huanyong Line, not only to the rich geography research object and the content is of great 65 significance for further research, but also to improve the popularity and the evaluation of China 66 national forest country and construction to provide the reference value. 67 Scholars at home and abroad to a lot of research on rural afforestation theme, foreign scholars 68 mainly involves internal social form in the country [ The existing exploration on various aspects of rural revitalization has laid a solid foundation 96 for the study of this paper. However, China national forest village is a new term emerging in 97 response to the development of The Times. In the academic world, it is a new field waiting to be 98 explored and analyzed by scholars. Few scholars begin to explore national forest village. In view 99 of this, this article possible innovation point is that fill a vacancy to China national forest village 100 as the research object, and the integrated use of spatial analysis methods and tools, to determine 101 the spatial distribution features, and clarify the distribution pattern of the various factors, as well 102 as to whether the breakthrough Hu Huanyong Line problem answer, subsequent evaluation and 103 development of forest country will provide theoretical basis and reference. village in the first batch, the demonstration will drive the rural greening and beautification step by 111 step and contribute the forestry power to the rural revitalization. By studying its spatial 112 distribution and making a superposition analysis with Hu Huanyong Line, we can explore the 113 correlation between its distribution and Hu Huanyong Line, and further excavate the formation 114 causes and influencing factors.  Therefore, the question is raised: can the spatial distribution of China national forest villages 124 break through the Hu Huanyong Line? 125 According to its spatial distribution characteristics, this paper analyzes the problem, discusses 126 the conclusions, and finally provides reasonable policy Suggestions for the construction and 127 evaluation of China national forest villages in the future. be divided into three types: uniform, random and condensed. The distribution characteristics of 137 points can be analyzed by using the nearest point index R, which is a geographical indicator of the 138 proximity between points [28].Its formula is: 139 Where, r 1 represents the closest distance in practice, r E represents the closest distance in 142 theory, n represents the total number of research objects, and A represents the area of the research 143 area. When R>1, the distribution is uniform. When R=1, the distribution is random; When R<1, 144 the distribution is a condensed distribution. 145

Tyson polygon 146
The criteria for defining the nearest point index used to determine the spatial distribution of point 147 elements are still controversial. In order to ensure the accuracy of determining the spatial 148 distribution types of China national forest villages, the coefficient of variation method for 149 calculating the area of Tyson polygon is used for testing. Since the area of Tyson polygon varies 150 with the degree of agglomeration of point elements, the coefficient of variation can be calculated 151 according to the area of Tyson polygon, which can be used to judge the spatial distribution type of 152 the research object. The variation coefficient formula is: 153 Where, R represents the standard deviation of the area of Tyson polygons, and S represents 156 the average value of the area of Tyson polygons. Duyckaerts pointed out that when CV=57% (including 33%-64%), the point elements were randomly distributed. When CV=92% 158 (including >64%), the point elements are clustered distribution. When CV=29% (including <33%), 159 the point element is uniformly distributed [29]. 160

Hot spot analysis 161
Hot spot analysis is a comprehensive analysis of the elements themselves and the surrounding 162 elements. The so-called hot spot refers to the element itself has high value and every element in 163 the surrounding element environment has high value at the same time. By calculating the statistics 164 of Getis-OrdG i * for each element in the data set, the spatial clustering location of high and low 165 value elements was determined based on the obtained Z score and P value [30].Its formula is: 166 Where, n represents the total number of elements, x j represents the attribute value of element 169 J, w ij represents the space weight between element i and j, and: 172 For a statistically significant positive Z score, the higher the Z score, the closer the clustering 173 of the high value (hot spot) will be. For negative Z-scores with significant statistical significance, 174 the lower the Z-score, the tighter the clustering of low values (cold points). 175

Standard deviation ellipse 176
Standard deviation ellipse is one of the commonly used methods to analyze the spatial distribution 177 direction characteristics. The elliptic area reflects the agglomeration degree of elements. The 178 major axis of the ellipse reflects the main distribution direction of the elements. The greater the 179 difference (bias) between the long axis and the short axis is, the stronger the directivity of element 180 distribution is. If the long axis and the short axis are equal, it means that they have a 181 non-directional distribution characteristic [31]. Its formula is: 182 Where, (x n , y n ) represents element coordinates; x、y respectively all the elements of x, y 186 axis average; θrepresents the direction of rotation; tanθ represents the standard deviation of the 187 ellipse; σ x 、σ y represent the maximum and minimum standard deviation distances, namely, the 188 length of the long axis and the short axis of the ellipse. 189

Kernel density analysis 190
The kernel density estimation method is used to reflect the condensate by the spatial distribution 191 of point density. Each grid node was taken as a center, and the circular area was searched with a 192 certain radius. The point elements falling into the circular area were counted, and different weights 193 were given to the point elements. The weight of the point elements closer to the center was greater, 194 and then the grid node density value was calculated [32].Its formula is: 195 In the formula, n represents the coordinate points, h represents the bandwidth, k represents  southwest China are in the top three, accounting for 28.71%, 17.06% and 15.03% respectively. 225 The gap between North China and South China is not large, while the gap between northwest 226 China and northeast China is at least 8.04% and 6.25% respectively. In terms of batches, the total 227 number of the first batch is 3947, and the number of the second batch is 3639. There is not much 228 difference between the two batches, and the spatial distribution characteristics of the two batches 229 are similar. From the perspective of Hu Huanyong Line, the number of the southeast area of the 230 line is 7, 167, accounting for up to 94.48%, while the northwest area only accounts for 5.52%. 231 There is a significant gap between the two areas divided by the line. 232

Spatial distribution type 236
The mean nearest neighbor tool in ArcGIS10.3 was used to analyze the spatial distribution types 237 of China national forest villages, and the spatial distribution types were preliminarily determined 238 according to the nearest neighbor index obtained. The mean nearest neighbor results are shown in 239 In order to ensure the accuracy of the results, the tyson polygon method was used to verify 246 the results obtained by mean nearest neighbor. The Tyson polygon of China national forest village 247 was created by ArcGIS10.3 (Figure 2a), and the average area and area standard deviation of the 248 Tyson polygon were calculated. According to the results, R=8753.9342, S=1222.4104, so 249 CV=716.12%, much higher than 64%, indicating that the distribution of China national forest 250 villages is of agglomeration type, and the degree of agglomeration is obvious. Based on the mean 251 nearest neighbor results, the spatial distribution type of China national forest villages can be 252 determined as cohesive type. 253

Cold hot spot analysis 254
In ArcGIS10.3 tool, the China national forest village was firstly connected with the vector map of 255 China in space, and then the hot spot analysis was carried out in the space statistical tool to 256 produce the cold spot map of the spatial distribution of the China national forest village ( Figure  257 2b).The Z value is divided into four stages by natural break point classification method, which are 258 respectively represented as four intervals of hot spot area, sub-hot spot area, sub-cold spot area 259 and cold spot area. It can be seen from the figure that hot spots include nine provinces, namely 260 Sichuan hot spots is more than that of cold spots, but in terms of coverage area, cold spots account for 268 64.95% of the total area, accounting for more than half of the total area. From the figure, we can 269 also intuitively observe the distribution characteristics of "hot spots in the south and cold spots in 270 the north". Hot spots are mainly in the south, and most areas in the north are cold spots. 271

Spatial distribution direction characteristics 272
In the ArcGIS10.3 geographical distribution tool, the standard deviation ellipse was made for the 273 first and second batches of China national forest villages and the average center point location was 274 determined respectively (Figure 2c). It was found that the standard deviation ellipse and the 275 average center point of the two batches had a high degree of coincidence, indicating that the 276 spatial distribution direction characteristics and the degree of agglomeration of the two batches 277 were almost the same. Take the first group as an example for analysis. The standard deviation of 278 the ellipse is 1044273.14 on the long axis and 726345.12 on the short axis. The difference 279 (skewness) between the two is 317,928.02.The long axis is from northeast to southwest, indicating 280 that the China national forest villages are densely distributed from northeast to southwest. The 281 average center is located in Suizhou city of Hubei Province, indicating that the geometric center of 282 national forest and rural intensive areas is Suizhou City of Hubei Province. To sum up, the first 283 batch of China national forest villages showed a significant spatial distribution direction with 284 Suizhou city in Hubei province as the center and extending to the northeast to the southwest. 285 Second, compared with the first batch of ellipse area slightly smaller show that the second batch 286 of concentration increased, from the Angle of rotation 27.2989 ° and 27.7063 °, slightly to the 287 clockwise rotation Angle, the second batch of northeast to southwest is more significant, the 288 average center moved to the southwest of small distance, shows that the trend of the development 289 of the China national forest village has to southwest direction, but in general the second batch of 290 characteristics on the basis of the first changes not by much.

Kernel density distribution characteristics 293
The kernel density function in ArcGIS10.3 spatial analysis tool was used to study the spatial 294 distribution density of two batches of China national forest villages. The results are shown in 295 Figure 3.According to the figure, the spatial distribution density of China national forest villages 296 was significantly different, and the two batches of kernel density map had similar characteristics. 297 On the one hand, the density distribution of China national forest villages is highly correlated with 298 the coverage of vegetation. The density and high-value areas are almost all within the coverage of 299 forest vegetation, which also conforms to the selection conditions of China national forest villages. 300 On the other hand, the high density area can be found in accordance with the distribution of urban 301 agglomeration degree is high, the first formed the Yangtze river delta, the central plains, the Bohai 302 sea, the pearl river delta, the Yangtze river middle reaches high density area and 303 Chengdu-Chongqing density area, formed the second batch of the Yangtze river delta, the central 304 plains, Beijing-Tianjin-Hebei, Chengdu-Chongqing high density area and the pearl river delta, the 305 Yangtze river middle reaches density area. Urban agglomeration as the future economic 306 development pattern in the core of the most dynamic and potential region, is important and 307 optimized development zone, the strategy of development priority zones in urban agglomeration 308 based on country often is in the leading level in the afforestation, the establishment of China 309 national forest country, step by step to achieve rural afforestation all over the country.

Elevation factor 313
The difference of elevation means that climate, precipitation, temperature, soil, vegetation and 314 other conditions are different, and thus determine the factors that directly affect the spatial 315 distribution of China national forest villages, such as population distribution, economic 316 development, forest resources and so on. In ArcGIS10.3, China's topographic elevation map is 317 divided into five levels according to the classification method of natural discontinuity points, and 318 then superimposed with China national forest villages to obtain the topographic distribution map 319 of national forest villages (Figure 4a), and the number of them is counted (Table 2).From table 2,  320 with the increase of grade, the higher elevation country quantity is less, the elevation in rural -321 152-788 m between 5793, proportion is as high as 76.36%, followed by 789-2092 m between rural 322 number is 1572, accounting for 20.72, the first two level country accounted for 97.08%, only 323 2.92% of national forest village is located in elevation above 2093 m, and the elevation between 324 4669-8848 -m rural existence, has not been found country presents the obvious characteristics of 325 "low altitude" distribution. Low altitude tends to be characterized by small topographic relief, flat 326 terrain, developed transportation and convenient living conditions, which are conducive to 327 population concentration and residence. In addition, the humid airflow from the eastern ocean, due 328 to the low altitude, can directly enter the inland and bring abundant precipitation, which is 329 conducive to the cultivation and growth of vegetation.

Aspect factor 335
The slope direction is the direction projected by the normal line of the tangent plane of a point on 336 the slope surface on the horizontal plane. The direction is defined as the included Angle formed 337 between the projection and the positive north of the point, so the value range is 0~360°.Different 338 elevation has great influence on climate, precipitation, vegetation, temperature, etc., and the 339 geographical position of different slope direction at the same height also makes the difference 340 significantly. In order to explore the influence of slope direction on the distribution of China 341 national forest villages, the elevation map was introduced into ArcGIS for slope direction analysis, 342 and the obtained slope direction map was superimposed with China national forest villages, and 343 the number of different slope direction villages was counted, as shown in Table 3 supply. Taking the secondary river as the center, three buffer zones of different levels were made, 365 namely 30km, 60km and 90km. The China national forest village and river buffer zones were 366 superimposed (Figure 4b), and the corresponding number of buffer zones of each level was 367 counted respectively (Table 4).According to Table 4, the proportion of China national forest 368 villages within 30km of the river is 37.66%, 25.73% within 30-60km, 17.18% within 60-90km, 369 and 80.57% within 90km of the river. In addition, China's terrain is high in the west and low in the 370 east, and most rivers flow from the west to the east. In contrast, the central and eastern regions 371 have more water than the west. According to Figure 4b, it can be intuitively observed that most 372 villages are located in the middle and lower reaches of rivers. As can be seen from the above 373 information, most of the country's forest villages are located around rivers, and their location close 374 to water sources can provide a lot of help for residents' life and rural development. As rural 375 afforestation model of national forest village, river conditions are necessary, on the one hand, 376 water, rural greening coverage on the development of forestry plays an important role, on the other 377 hand, water is the rural residents the necessities of life, has a population of water source place to 378 gather, thus forming various unique country. In addition, the utilization of rivers can also be 379 further explored. River resources can often be turned into tourism resources, and promoting the 380 development of local tourism is an important direction to promote the development of rural 381 economy. To sum up, watershed factor is an important factor affecting the spatial distribution of 382 China national forest villages. 383 384

Traffic factor 387
Accessibility has an important impact on all aspects of rural areas, such as the dissemination and 388 development of rural traditional culture, the quality of life of residents, the influence of the outside 389 world on it, the degree of rural greening, and the level of economic development. Traffic 390 developed directly decided the country accessibility, which represented by main highway and 391 railway, respectively make 50 km, 100 km from the buffer, and compared with China national 392 forest village, national forest rural transportation network distribution (Figure 4c), and statistics of 393 the different distance within the buffer country number, as shown in table 5.According to statistics, 394 83.42% of China national forest villages are located within 100km of the main road, of which 395 58.87%, more than half, are within 50km.86.04% of China national forest villages are located 396 within 100km of the main railway, among which 61.84% are within 50km, accounting for a 397 relatively high proportion. According to Figure 4c, most of the China national forest villages are 398 located near the main roads and railways, indicating that the vast majority of the China national 399 forest villages are located near the dense transportation network, that is, the China national forest 400 villages have strong accessibility. Rural greening and beautification cannot be separated from the 401 support and help of the outside world. Strong accessibility provides conditions for the 402 communication within and outside the countryside. In addition, as the saying goes, "If you want to 403 get rich, build roads first", rural economic development must rely on convenient transportation 404 facilities. Rural revitalization requires rural economy to be improved and rural residents' living 405 standards to be improved. If you don't keep up with The Times, you will eventually fall behind 406 and be eliminated. 407