Feasibility of a new-generation nighttime light data for estimating in-use steel stock of buildings and civil engineering infrastructures

Abstract

Regional scale estimation of in-use steel stock (IUSS) using satellite observation of nighttime lights (NTL) is essential for urban resource consumption management and resource recovery monitoring. S-NPP VIIRS (Suomi National Polar-orbiting Partnership’s Visible Infrared Imaging Radiometer Suite) NTL product is newly released in 2013, which has enhancements in both spatial and radiometric resolutions compared with the DMSP-OLS (Defense Meteorological Satellite Program’s Operational Linescan System) NTL products. To evaluate whether these enhancements can improve the estimations of IUSS of buildings (IUSSB) and civil engineering infrastructure (IUSSCE), the S-NPP VIIRS NTL products were used to estimate IUSSB and IUSSCE at both sub-national level of Japan and national level of world, and their performances were compared with those of using DMSP-OLS radiance calibrated (RC) NTL data. Our results indicate that the S-NPP VIIRS NTL data can offer more accurate estimates of IUSSB and IUSSCE (R2 values of 0.952 and 0.909) than those of DMSP-OLS RC NTL (R2 of 0.929 and 0.884) at prefectural level of Japan. Similar to DMSP-OLS RC NTL product, urban NTL of S-NPP VIIRS data has a stronger relationship with IUSSB, and IUSSCE was more closely related to total NTL. At national level, S-NPP VIIRS NTL also showed better estimations of IUSSB and IUSSCE. We confirmed that dividing the world into different regional groups is also required for estimating IUSSB and IUSSCE from S-NPP VIIRS NTL. For estimation of IUSSCE, two classifications including Asia region and Non-Asia region are appropriate when using S-NPP VIIRS NTL product, which is superior to DMSP-OLS RC NTL. For estimation of IUSSB, more classifications may be still required which is similar to DMSP-OLS RC NTL. Overall, our study indicated that S-NPP VIIRS data have greater capability in modeling IUSSB and IUSSCE than DMSP-OLS RC data. Our results are critical for ever-improvement of policy making on urban resource management.