Study On The Spatial-Temporal Variability Of Cultivated Land Productivity And Its Main Control Factors

The cultivated land is the essence of the land resources, and the productivity is one of the most important functions for cultivated land resources. Protecting and improving the productivity of cultivated land is the basis and guarantee to ensure food security and has a very important significance for safeguarding the steady and continuous development of social economy. Nowadays, because of the food security problem has become increasingly prominent, how to grasp the characteristics of spatial-temporal variation of the cultivated land productivity, and identify the main controlling factors not only become the social problems to be solved urgently, but also become a hot spot of research in academic circles. The area of Heilonggang is one of the main grain-producing areas in Hebei province, therefore, improving and maintaining local cultivated land productivity has a great significance for ensuring food safety in Hebei province or even in the whole country.In view of the above-mentioned facts, taking Xiong county in the area of Heilonggang as the typical case, taking the scientific use of cultivated land and ensuring food safety as the goal, and relying on the cultivated land basic fertilities and questionnaires from local farmers, this paper analyzed the characteristics of the cultivated land natural attribute and management factors affecting cultivated land productivity using classical statistics and geo-statistics. The spatial-temporal variation characteristics of cultivated land productivity from two scales of county area and landscape were explored, and its main controlling factors were revealed using model of nonlinear parameter classification and regression trees (CART). Main conclusions of this research are as following:(1)This paper systematically analyzed the factors of cultivated land natural attribute and management influencing the cultivated land productivity from two scales of county area and landscape. So the variable character of cultivated land basic fertilities and the level of cultivated land managements were defined.(2)The cultivated land productivity was significantly different at county level, the minimum is 4.75t/hm2 and the maximum is 6.95t/hm2. The spatial distribution pattern of cultivated land productivity had the regularity higher in southwest and east and lower in southeast and parts of the middle. The rank (from high to low) of the five cultivated land productivity on different landscape types was the loam above and clay below slope >light loam homogeneous slope >clay loamy homogeneous bottom land >the sandy above and loam below hillock > the sandy loam homogeneous sustained hillock.(3)Cultivated land management factors were more persuasive than factors of cultivated land natural attribute to explain spatial variation characteristics of the cultivated land productivity at county level. Mainly controlling factors successively were probability of irrigation, nitrogen application amount, rapidly-available nitrogen and organic matter, and their total effects were 51%. Natural attributes of cultivated land were the main controlling factors among different landscape types, successively were cation exchange capacity, content of organic matter, rapidly-available potassium and nitrogen, and their total effects were 55%. Cultivated land productivity in different landscape scales was mainly controlled by management factors of cultivated land, the results indicated that the influence of human factors was more important.(4)The spatial distribution pattern of the temporal variability of cultivated land productivity at county level had the regularity higher in the southwest and northwestern, over 6%, and lower in the southwest, under 5%. The variation of landscape pattern maked no difference to the temporal variability of cultivated land productivity.(5) The temporal variability of cultivated land productivity at county level was mostly controlled by probability of irrigation, nitrogen application amount and rapidly-available potassium, of which total effects67.80%.The main control factors mainly relate to probability of irrigation, nitrogen application amount and potassium application amount at different landscape scales.