(1) System cluster analysis method

Cluster analysis is a multivariate statistical analysis method that classifies samples or variables according to their degree of similarity in nature. This method is to define the distance between classes on the basis of sample distance. First, n samples are classified into a class of their own, and then the two classes with the smallest distance are merged each time. After merging, the class and class are recalculated. distance between them, this process continues until all samples are classified into one category, and this process is made into a clustering pedigree diagram.

Systematic clustering methods include the shortest distance method, the longest distance method, the intermediate distance method, the center of gravity method, the class average method, the variable class average method, the variable method, and the sum of squares of deviations method. Although there are many methods of system clustering analysis, the steps of classification are basically the same. The only difference is that the distance between classes has different definition methods, resulting in different formulas for calculating distance. This study uses the longest distance method, and the β parameter value is 0. Use dij to represent the distance between samples Xi and Xj, and use Dij to represent the distance between class Gi and class Gj.

(2) Cluster analysis of basic agricultural inputs in the Poyang Lake region

Based on the total power of agricultural machinery (X1) and the total amount of chemical fertilizer use (X2) among the agricultural input factors in each region Cluster analysis was conducted on 6 indicators including the total amount of pesticide use (X3), the total amount of mulching film used (X4), the effective irrigation area (X5) and the annual average number of people engaged in agriculture, forestry, animal husbandry and fishery (X6) (Figure 3-5). The transformation results are listed in Table 3-2, and the clustering results are listed in Table 3-3.

Figure 3-5 Cluster pedigree diagram of basic agricultural inputs in Poyang Lake region

Table 3-2 Standardized data of cluster pedigree of basic agricultural inputs in Poyang Lake region

Table 3-3 Table of 29 regional classifications and average agricultural basic inputs

According to the system clustering pedigree diagram, the 29 counties in the Poyang Lake area can be classified (Cities, districts) are divided into 3 categories. The purpose is to further analyze the status of agricultural basic inputs in various regions and calculate the average of various basic agricultural inputs in each type of region.

The first type of areas includes: Donghu District, Xihu District, Qingyunpu District, Wanli District, Qingshan Lake District, Anyi County, Lushan District, Xunyang District, Jiujiang County, De'an County, Xingzi County, Hukou County, Pengze County, Yujiang County, Fengxin County, Wannian County. Among the basic agricultural input factors, the total power of agricultural machinery (X1), the total amount of chemical fertilizers used (X2), the total amount of pesticides used (X3), the total amount of mulching film used (X4), the effective irrigation area (X5) and the employees in agriculture, forestry, animal husbandry and fishery The average status of equal inputs is the lowest, indicating that agricultural inputs in these areas are relatively insufficient and need to be further strengthened.

The second type of areas includes: Xinjian County, Jinxian County, Leping City, Yongxiu County, Duchang County, Zhangshu City, Dongxiang County, and Yugan County, with various agricultural basic inputs ranking medium, indicating that Agricultural inputs are relatively modest in these areas.

The third category of areas includes: Fengcheng City, Gaoan City, Linchuan District, and Poyang County, with relatively high agricultural basic investment.

(3) Clustering according to the area of ??planting agricultural products in each region

Conduct cluster analysis based on 6 indicators of the average rice sowing area in each region. The clustering pedigree is shown in Figure 3-6, the standardized data table is shown in Table 3-4, and the clustering results are shown in Figure 3-6 and Table 3-5.

Figure 3-6 Area clustering pedigree diagram of agricultural products in the planting industry

Table 3-4 Standardized data of area clustering pedigree of agricultural products in the planting industry

Table 3-5 Table of classification and average sown area of ??major crops in 29 counties (cities, districts)

Continued table

It can be seen from the clustering results Table 3-5:

The first category of areas: Anyi County, Lushan District, Qingyunpu District, Xingzi County, Yujiang County, Fengxin County, Dongxiang County, and Wannian County are typical large-area rice planting areas, and the soybean area is also large. The planting areas of wheat and corn are relatively small, and the planting areas of cash crops such as oil crops, sugar and vegetables are also relatively small in these areas.

The second type of areas: Duchang County, Zhangshu City, Gao'an City, Linchuan District, Yugan County, Nanchang County, Xinjian County, and Leping City are typical large-area planting areas for rice and vegetables. The sown area is the largest, and the sown area for oil crops is also large. However, the sown areas for wheat, corn, soybeans, cotton and sugar in these areas are very small.

The third category of areas: Pengze County, Hukou County, Jiujiang County, Yongxiu County, and De'an County are typical large-area cotton planting areas, with a certain rice planting area, and corn planting area. The area planted with wheat is larger and the area planted with oil crops is larger, while the area planted with soybeans is smaller, the area with sugar is the smallest and the area planted with vegetables is smaller.

The fourth category of areas: Poyang County, Jinxian County, and Fengcheng City are typical areas with the largest rice planting area. Wheat, corn and oil crops have the largest planting area in the region, and vegetable planting area and soybean area are the largest. Larger, cotton and sugar have the smallest area.

(4) Cluster analysis based on the output of various forestry, animal husbandry, and fishery products

Cluster analysis based on 5 indicators of the average fruit yield in each region, clustering The pedigree diagram is shown in Figure 3-7, the data standardization table is shown in Table 3-6, and the cluster analysis results are shown in Figure 3-7 and Table 3-7.

Figure 3-7 Cluster pedigree diagram of forestry, animal husbandry and fishery product yields

Table 3-6 Cluster standardized data of forestry, animal husbandry and fishery product yields

Continued table

Table 3-7 Classification of 29 counties (cities, districts) and average output of major forestry, animal husbandry and fishery products

It can be seen from Table 3-7 :

1) Most of the areas in the Poyang Lake region belong to the first category. In the first category, these places basically produce fruits, and the average yield of aquatic products is relatively high. Only A very few cities do not produce fruits, and the average output of meat, milk, and eggs is relatively low.

2) In the second category of four regions: Yongxiu County, Yugan County, Fengxin County, and Linchuan District, the average output of fruits and meat is the highest, and the output of milk is the lowest. For the average production of poultry eggs and aquatic products, it is at a medium level.

3) The third type of area has only one Qingyun spectrum area, which has the highest milk production. In other words, Nanchang City’s production is basically supplied by it.

4) There is only one fourth category area, Nanchang County, which has the highest average meat output.