Analysis on evolution stages of agricultural landscape in metropolitan based on variable structure co-integration test

Abstract: The economic growth and the change of agricultural landscape resources quantity have significant temporal variability and evolution stages. Researching the relationship between the 2 parties and recognizing the stage characteristics have always been a hot topic in the region of resource economics. This study tried to find the point of structural mutation, divided the evolution stages of urban agricultural landscape, and analyzed the reasons. It used time series data of the economic growth and the change of agricultural landscape resources quantity in the year from 1978 to 2015. Meanwhile, it also used the Gregory-Hansen integration test with structural break to analyze the relationship between them in the western suburbs of Qingpu District in Shanghai City. The results showed that: 1) The co-integration test of long period without considering the structural break point was not suitable for the Qingpu District which had been in the transitional period; the co-integration test with structural break analysis could reflect the change of economic and social structure in long time, and reflect the long-term equilibrium relationship between the economic and social system. It was difficult to capture the sudden change of economic sequence in a long time with the static integration analysis without considering the structural change. Meanwhile, it was also difficult to describe the changes of economic relations and economic structure. However, variable structure co-integration analysis could accurately and timely capture the mutation time of economic time series, and reveal the true relationship between economic variables. 2) The evolution of agricultural landscape could be divided into 3 stages in Qingpu District during the period of 1978-2015: Low level synergy stage (1978-1995), reverse push stage (1995-2007) and economic pull stage (2007-2015). The characteristics of low level synergy stage were slow economic growth, slow change in agricultural landscape area and low level coordination of agricultural landscape land elements to economic growth. In the reverse push stage, the characteristics were the sharp decrease of agricultural landscape resources. Agricultural landscape resources, which were the important economic stimulus resources, supported the rapid economic development, and its fast consumption had become the cost of economic growth. And in the economic pull stage, the characteristics were the reduction of the agricultural landscape area, and the input of agricultural landscape resources conformed to the requirement of economic growth. The land elements of agricultural landscape resources played an important role as one of the elements that participated in economic growth. 3) The economic growth was realized at the expense of agricultural landscape resources in Qingpu District. The analysis of the long-term equilibrium parameters of the error correction model showed that the values in the 3 stages were negative (-0.438, -0.919 and -0.223). From the analysis of variable structure causality test, the decrease of the number of agricultural land landscapes in the 3 stages was the Granger reason for economic growth. These results could well prove that during the period of the year from 1978 to 2015, the growth of economic in Qingpu District inevitably led to the reduction of the number of agricultural landscapes. The results are helpful to coordinate the economic development in that region and the protection of agricultural landscape resources. They also can provide scientific references for the less developed regions to better handle with the relationship between economic development and protection of agricultural landscape resources.