Particulate nitrate (pNO3–) has often been found to be the major component of fine particles in urban air-sheds in China, the United States, and Europe during winter haze episodes in recent years. However, there is a lack of knowledge regarding the experimentally determined contribution of different chemical pathways to the formation of pNO3–. Here, for the first time, we combine ground and tall-tower observations to quantify the chemical formation of pNO3– using observationally constrained model approach based on direct observations of OH and N2O5 for the urban air-shed. We find that the gas-phase oxidation pathway (OH+NO2) during the daytime is the dominant channel over the nocturnal uptake of N2O5 during pollution episodes, with percentages of 74% in urban areas and 76% in suburban areas. This is quite different from previous studies in some regions of the US, in which the uptake of N2O5 was concluded to account for a larger contribution in winter. These results indicate that the driving factor of nitrate pollution in Beijing and different regions of the US is different, as are the mitigation strategies for particulate nitrate.
A low-pressure reactor (LPR) was developed for the measurement of ambient organic peroxy (RO2) radicals with the use of the laser-induced fluorescence (LIF) instrument. The reactor converts all the ROx (= RO2 + HO2 + RO + OH) radicals into HO2 radicals. It can conduct different measurement modes through altering the reagent gases, achieving the speciated measurement of RO2 and RO2# (RO2 radicals derived from the long-chain alkane, alkene and aromatic hydrocarbon). An example of field measurement results was given, with a maximum concentration of 1.88 x 10(8) molecule/cm(3) for RO2 and 1.18 x 10(8) molecule/cm(3) for RO2#. Also, this instrument quantifies the local ozone production rates directly, which can help to deduce the regional ozone control strategy from an experimental perspective. The new device can serve as a potent tool for both the exploration of frontier chemistry and the diagnosis of the control strategies. (C) 2020 Chinese Chemical Society and Institute of Materia Medica, Chinese Academy of Medical Sciences. Published by Elsevier B.V. All rights reserved.
Feature-based time series representations have attracted substantial attention in a wide range of time series analysis methods. Recently, the use of time series features for forecast model averaging has been an emerging research focus in the forecasting community. Nonetheless, most of the existing approaches depend on the manual choice of an appropriate set of features. Exploiting machine learning methods to extract features from time series automatically becomes crucial in state-of-the-art time series analysis. In this paper, we introduce an automated approach to extract time series features based on time series imaging. We first transform time series into recurrence plots, from which local features can be extracted using computer vision algorithms. The extracted features are used for forecast model averaging. Our experiments show that forecasting based on automatically extracted features, with less human intervention and a more comprehensive view of the raw time series data, yields highly comparable performances with the best methods in the largest forecasting competition dataset (M4) and outperforms the top methods in the Tourism forecasting competition dataset.