Northeast China (NEC), the most productive maize growing area in China, has experienced pronounced climate change. However, the impacts of historical climate changes on maize production and their spatial variations remain uncertain. In this study, we used yield statistics at prefecture scale over the past three decades, along with contemporary climate data, to explore the yield-climate relationship and its spatial variations. At the regional scale, maximum and minimum temperature changes had opposite impacts on maize yield, which increased by 10.07.7% in response to a 1°C increase in growing season mean daily minimum temperature ( Tmin), but decreased by 13.47.1% in response to a 1°C increase in growing season mean daily maximum temperature ( Tmax). Variations in precipitation seemed to have small impacts on the maize yield variations (-0.95.2%/100 mm). However, these responses of maize yield to climate variations were subject to large spatial differences in terms of both the sign and the magnitude. ~30% of the prefectures showed a positive response of maize yield to rising Tmax, which was in contrast to the negative response at the regional scale. Our results further indicate that the spatial variations in the yield response to climate change can be partly explained by variations in local climate conditions. The growing season mean temperature was significantly correlated with the response of maize yield to Tmax ( R=-0.67, P<0.01), which changes from positive to negative when the growing season mean temperature exceeds 17.90.2°C. Precipitation became the dominant climatic factor driving maize yield variations when growing season precipitation was lower than ~400 mm, but had a weaker influence than temperature over most of the study area. We conclude that, although NEC is a region spanning only more than one millions of kilometer squares, the divergence of the yield response to climatic variations highlights the need to analyze the yield-climate relationship at fine spatial scales.
