RESUMO
Understanding the effects of climate change on crops is vital for food security. We aimed to characterise the coincidence of yield variations with weather variable for potato in northern China using long-term datasets. Daily climate variables obtained from 607 meteorological stations from 1961 to 2014, detailed field experimental data for the period of 1982 to 2012 in northern China, and multivariate linear statistical model were used in this study. In particular, the first difference method was used to disentangle the contributions of climate change to potato yield. We concluded that during the potato growing, the average daily, maximum and minimum temperatures significantly increased by 0.23°C per decade, 0.20°C per decade and 0.36°C per decade from 1961 to 2014 in northern China, respectively. However, average total radiation, total annual precipitation and potential evapotranspiration from April to September all exhibited downward trends, but the variation of evapotranspiration (-9.99mm per decade) was greater than that of precipitation (-2.65mm per decade). The key climatic factors limiting potato yields in northern China over the past 30years at a regional scale were diurnal temperature range, precipitation, radiation and ET0. The potato yield in northern China was the most sensitive to variation of the diurnal temperature range followed by radiation, precipitation and reference crop evapotranspiration (ET0). Specifically, when the diurnal temperature range decreased 1°C, the potato yield increased 543.9kg·ha-1. When the total radiation decreased 1MJ·m2, the potato yield increased 63.8kg·ha-1. When the ET0 decreased 1mm, the potato yield increased 62.7kg·ha-1. When the precipitation increased 1mm, the potato yield increased 62.9kg·ha-1. A regression model describing the combined effects of different climate variables on potato yield in northern China was established.