Estimates of carbon dioxide emissions based on incomplete condition information: a case study of liquefied natural gas in China.

Recent calculations of carbon dioxide (CO2) emissions have faced challenges because data consist of only partial information, which is called "incomplete information." According to the emission factor method, energy consumption and CO2 emission factors with incomplete information may lead to unmatched multiplication between themselves, which affects accuracy and increases uncertainties in emission results. To address a specific case of incomplete information that has not been fully explored, we studied the effects of incomplete condition information on the estimates of CO2 emissions from liquefied natural gas (LNG) in China. Based on Chinese LNG sampling data, we obtained the specific-country CO2 emission factor for LNG in China and calculated the corresponding CO2 emissions. By applying hypothesis testing, regression analysis, variance analysis, or Monte Carlo (MC) simulations, the effects of incomplete information on the uncertainty of CO2 emission calculations in three cases were analyzed. The results indicate that calorific values have more than a 9.8% impact on CO2 emission factors and CO2 emissions with incomplete sample information. Regarding incomplete statistical information, the impact of statistical temperature on CO2 emissions exceeds 5.5%. Regarding incomplete sample and statistical information, sample and statistical temperatures can individually increase estimate biases by more than 5.2%. Significantly, the impacts of sample temperature and statistical temperature may offset each other. Therefore, the incomplete condition information is quite important and cannot be ignored in the estimation of CO2 emissions from LNG and international fair comparison.