ABSTRACT
In this article, the Defence Meteorological Satellite Program Operational Linescan System (DMSP-OLS) night-time light remotely sensed data on a small scale is proposed to evaluate port economic comprehensive scores (PECS) for the ports of the major cities, including Shanghai, Nanjing, Wuhan, and Chongqing, in the Yangtze River Valley, China. First, the concept and calculation method of port night-time light intensity (PNLI) are proposed. Second, an estimation method of PECS is proposed using factor analysis. Third, two regression models (i.e. first- and second-order polynomials) between PNLI and PECS are built and tested. The goodness-of-fit of the two models are compared, both with and without outliers. Finally, experimental results show that the proposed methods for evaluating PECS in the Yangtze River Valley, China are feasible. Wuhan has the strongest correlation (R 2 = 0.925) and passes the F-test, F t = 55.429 >4.26 = F t (0.05) (2, 9); Shanghai has the weakest correlation (R 2: 0.688) but still passes as well, F t = 9.944 >4.26 = F t (0.05) (2, 9). Factors confounding the correlation for Shanghai are discussed. Overall, this study not only proposes a new set of methods to evaluate the performance of port economies but also provides a feasible way to use DMSP-OLS data to study the geographic problems of small scales.
