Working towards confident spaceborne monitoring of carbon emissions from cities using Orbiting Carbon Observatory-2

摘要

Cities are responsible for 70% of fossil fuel CO2 emissions (FFCO2), and these emissions are poorly monitored. As a result, the use of spaceborne observation as a tool for addressing urban FFCO2 emissions has intensified. This work aims to understand the suitability of modern spaceborne remote sensing for capturing CO2 urban enhancement and for monitoring FFCO2 emissions from cities. We have used the first four years of observations from NASA Orbiting Carbon Observatory 2 (OCO-2) to collect CO2 urban anomalies (XCO2ano) from large cities (> 500 km(2) area) by combining urban-to-rural gradient and statistical filtering approaches. Approximately half of all XCO2ano (44%) were significant and noise-free to meet the accuracy requirements for city-scale applications. We captured positive CO2 enhancement over urban areas compared to background at a global scale. Median XCO2ano estimates were positive at 1.07 +/- 0.80 ppm (worldwide), 1.05 +/- 0.80 ppm (Northern Hemisphere), and 0.96 +/- 0.80 ppm (Southern Hemisphere). Most monthly XCO2ano (similar to 83%) emerged in four regions of the Northern Hemisphere (East Asia, Europe, North America, and South Asia), where most cities with strong emissions are located. We report that the XCO2ano from numerous cities have a moderate linear relationship with city size (r = 0.54-0.65) and FFCO2 strength (r = 0.64). As expected, five out of the six strongest XCO2ano were found over the megacities ( > 10 million population) of Los Angeles (2.04 +/- 0.91 ppm), Tehran (1.94 +/- 1.54 ppm), Rhine-Main Metropolitan Area (1.51 +/- 0.59 ppm), Pearl River Delta (1.48 +/- 1.11 ppm), and Seoul (1.47 +/- 1.72 ppm), with Houston (1.50 +/- 0.72 ppm) as the only non-megacity. Overall, we have shown that spaceborne remote sensing of XCO2 is suitable for capturing CO2 enhancement over a wide range of cities (especially in the Northern Hemisphere). Though we registered the global-scale urban footprint in CO2 signal and demonstrated the sensitivity of XCO2ano to city size and FFCO2, spaceborne remote sensing is still limited in its provision of columnar CO2 enhancement which is fundamentally linked with underlying urban emissions. This linkage may be more effectively addressed when spaceborne remote sensing becomes increasingly optimized for city-scale applications.