基于NPP-VIIRS的胶东经济圈GDP时空演变及空间化模拟研究

Spatio-temporal Evolution and Spatial Simulation of GDP in Jiaodong Economic Circle Based on NPP-VIIRS Data

为研究胶东经济圈经济发展的空间差异,以2012-2022年NPP-VIIRS夜间灯光数据和GDP数据为数据源,运用探索性空间分析方法探索GDP时空演变特征,采用指数平滑、空间建模和逐像元线性矫正等方法进行GDP空间化模拟和预测.结果表明:1)胶东经济圈最佳夜间灯光指数和GDP高度相关,有相似的空间分布和演变规律;2)经济欠活跃型区域逐步缩减,经济一般型区域空间分布由带状逐渐转为片状,GDP的空间聚集性趋于减弱;3)GDP热点区主要在青岛的黄岛区和胶州市,冷点区主要在潍坊的部分县区且逐渐收缩;4)指数平滑后的夜间灯光数据与GDP相关性显著增强,更适合进行拟合;5)GDP空间化模拟相对误差小于1%,模拟效果较好;6)高密度GDP主要分布在沿海县区和城市中心城区,并在各县区的交通线节点上零星出现.

To explore the spatial differences of the economy in Jiaodong economic circle,NPP-VIIRS night light data and GDP data from 2012 to 2022 were used as data sources,and the spatio-temporal evolution characteristics of GDP were analyzed by exploratory spatial analysis method.Meanwhile,spatial simulation and prediction of GDP were obtained by using exponential smoothing,spatial modeling and pixel-by-pixel linear correction.The results show that:1)The optimal night light index is highly correlated with GDP,and the two data have similar spatial distribution and evolution rules.2)The areas with less economic activity gradually shrink,while the spatial distribution of general economic areas changes from a strip to a sheet-like pattern,and the spatial clustering of GDP tends to weaken.3)The GDP hot spot areas are mainly in Huangdao District and Jiaozhou City of Qingdao,while the cold spot areas are mainly in some counties and districts of Weifang and gradually shrink.4)The correlation between the night light data after exponential smoothing and GDP is significantly enhanced,which is more suitable for fitting.5)The relative error between the simulated and statistical values of GDP is less than 1%,indicating that the simulation effect is good.6)High density GDP is mainly distributed in coastal counties and central urban areas,and appears sporadically at the transportation nodes of counties.