Study On Grazing Intensity And Early Warning Mechanism Of Sunite Grassland Based On Trajectory Data

The Sunite Grassland is located in the northwestern part of the Xilin Gol Prairie.The grassland is based on grazing and the animal husbandry is one of the major industries in the region.In recent years,the significant degradation of grassland resources has become a major problem in the economic development of animal husbandry and the stability of grassland ecosystems.One of the main causes of grassland degradation is overgrazing.Therefore,estimating the actual grazing intensity of pastures and improving the grazing early warning mechanism to guide the herdsmen’s reasonable grazing is of great significance to the balance of grass and livestock.With the development of animal husbandry informationization and modernization,pastoralists use the positioning equipment to track and manage livestock,which not only greatly saves labor costs,but also accumulates a large amount of trajectory data,which contains rich and useful information,such as livestock.Sports status,behavioral patterns,etc.Based on the above background,after investigating the trajectory data processing technology and the grazing intensity estimation method related literature,using the grazing spatiotemporal trajectory data,based on the grazing characteristics of Sunite grassland,a trajectory mining technique combined with the definition of grazing intensity is proposed.Estimation method of grazing intensity.The method firstly performs cluster analysis on the grazing trajectory data to obtain different foraging and eating areas of the livestock;Then,the clustering results are counted and calculated to obtain the grazing intensity in different areas of the grassland.Then carry out the research and system design of the grazing early warning mechanism,first judge the grazing level by the relationship between the carrying capacity and the grazing intensity;Then,the distributed grazing intensity estimation model and the grazing level judgment method are encapsulated,and the final calculation result is provided externally.Based on the Flask framework to develop the API interface,the user only needs to send a request to obtain relevant data,and then the results of the grazing intensity estimation by FineReport and The grazing level judgment results are visually displayed to realize the over-grazing and grazing state monitoring and early warning in the pasture area.Finally,based on the actual trajectory data generated during the grazing process,the grazing intensity is estimated,and the field landslide intensity of A,B grassland was selected in the range of 0¹.0 head?hm^-2,1.0².0 head?hm^-2 and 2.0³.0head?hm^-2 to conduct field survey experiments to verify the proposed The rationality of the estimation method of distributed grazing intensity and the correctness of the research results.The grazing intensity range of A and B grasses in Sunite Zuoqi is 0.6³.0 head?hm^-2,0.6².4 head?hm^-2,all of which are overgrazing.The experimental results show that the pastures with high grazing intensity have poor potential,and the two are negatively correlated,consistent with the theory;grassland A pasture growth was significantly weaker than grassland B,which was consistent with the calculation results.The results strongly prove that the proposed method can correctly estimate different foraging areas,eating and grazing intensity,and can qualitatively distinguish the degree of overgrazing in different pastoral areas.