Temporal and spatial changes in pig manure load in agricultural lands in China

Crop-livestock and poultry farming cycle is an important way of solving agricultural non-point source pollution, promoting chemical fertilizers reduction and enhancing sustainable development of agriculture. Rational utilization of manure in agricultural lands is the basis of regional crop-livestock and poultry farming cycle. In order to control the pollution of breeding pigs through crop and pig farming cycle, we used the emission coefficient method to calculate the load intensity and carrying capacity of pig manure in agricultural lands in 31 provinces (municipalities) in China for the period 2007-2016. The spatial statistical model in ArcGIS was used to analyze temporal and spatial patterns of the evolution of pig manure load intensity in farmlands during the investigated decade. Then the multivariate linear regression model was used to explore the main driving factors of temporal and spatial differences in load intensity of pig manure in farmlands. The results showed that the load intensity of pig manure in China's agricultural lands in 2007-2016 was low, which was 4.148 t·hm^-2 averagely with obvious inter-provincial variations. The load intensity generally increased first and decreased then. There were five evolution patterns of pig manure load intensity in farmlands in 31 provinces. They were continuous upward trend in Qinghai and Tibet; down-up-up trend in Xinjiang; continuous downward trend in 4 province of Ningxia, Sichuan, Guizhou and Hebei; up-up-down trend in 12 provinces including Shanghai; up-down-down trend in 6 provinces including Chongqing; and down-up-down trend in 6 provinces including Inner Mongolia. The load intensity of pig manure in farmland showed significant spatial autocorrelation on the whole, which initially decreased and then increased during the decade. The hotpot regions of load intensity of pig manure apparently shifted to the north and west from the east and south. The multivariate linear regression analyses showed that urban-rural income ratio, crop area per capita and proportion of sections with excellent water quality had negative effects, while profit rate in the last year, annual pork consumption per capita, population density, river network density had positive effects on pig manure load intensity in farmland. These factors affected load intensity mainly through market rules, supply-demand relationship, environmental carrying capacity, livestock and poultry farming regulations and government policies. Finally, policy recommendations were put forward based on the results, such as formulating regional plans and negative list policies based on provincial shift of load intensity hotspots and carrying capacity of pig manure in farmlands, establishing cycle-coupling industries of crop farming and livestock/poultry in a certain area.