Spatiotemporal change and non-stationarity of air temperature in China from 1959 to 2018

Abstract

Since 20th century, urban warming has become the major global change concern, affecting sustainability of cities. Based on observation data of air temperature from 1728 meteorological stations, we constructed 612 effective grid cells with a spatial resolution of 1°×1° by using the gridding method, analyzed their spatiotemporal variability and non-stationarity by using the temperature anomaly and autocorrelation test model. Results showed that China's accelerated warming trend, with an increase of 0.995℃ from 1959 to 2018. The air temperature changes of grid cells show significant spatial variability. Grid cells with air temperature rise more than 1℃, accounted for 42.81%, mainly distributed in the northwest, east coast, northeast areas. China's warming has become more worrying in the past 30 years, with an increase of 0.797℃. The annual average temperature increased by 1.124℃, 1.029℃, 1.048℃ respectively in Beijing-Tianjin-Hebei, Yangtze River Delta, Pearl River Delta Urban Agglomerations from 1959 to 2018, which were higher than the country increased value. The temperature autocorrelation coefficients of 49.35%-96.08% grid cells were greater than the critical value of 0.05 significance level within lag 1-13 periods. The 30-year (1959-1988 or 1989-2018) mean the temperature s of 98.04% grid cells were outside the 90% confidence threshold of 60-year mean temperature. Therefore, temperature changes exhibited non-stationarity over China, providing scientific supports for future temperature prediction, global change response planning, and sustainable development.