Search

In this paper, we propose a hybrid model that combines wavelet transformation of the raw data and an artificial neural network with long short-term memory (LSTM). This model allows researchers to increase the accuracy of time series forecasting. The model is based on data from environmental monitoring of greenhouse gases on the Belyy Island of the Yamal-Nenets Autonomous Okrug, Russia. The raw data for building the proposed model were obtained in July–August 2016. The CH 4 concentration time series was decomposed using a discrete wavelet transform into four components - one approximating and three detailing. These components, along with a timed temperature series, were used to train six ANNs—two exogenous input autoregressive networks (NAR) and four LSTM networks. The forecast was calculated as the sum of forecasts for each of the components. Forecast accuracy was assessed using several indices (mean absolute error (MAE), root mean square error (RMSE), mean square relative error (RMSRE), Willmott's agreement index (IA1, IA2)) and a Taylor diagram. The hybrid model based on LSTM showed the best accuracy. The hybrid model based on LSTM showed the best accuracy. MAE, RMSE, and RMSRE errors decreased by more than 70%. In terms of IA1 and IA2, the models improved by 11% and 30% (when comparing the best hybrid models LSTM and NAR. For the hybrid LSTM model errors RMSE, RMSRE, and MAE are more than 20% less than for the base LSTM model for training in which data without wavelet transform were used. Willmott agreement indices (IA1, IA2) rose from 32% to 59%, depending on the models being compared. So, the increase in accuracy after applying the described approach was up to 79% for models based on LSTM (more precisely, from 20% to 79%, depending on the indicator). Applying wavelet transform data to train the NAR-based model reduced MAE, RMSE, and RMSRE errors by 27%, 30%, and 31%, respectively. The Willmott agreement indices (IA1, IA2) rose from 41% to 45% respectively. The Taylor diagram also shows the advantage of the proposed approach. • Measurements of the methane concentration were done. • Six models to predict the dynamics of were built. • Hybrid model with data wavelet transformation and LSTM showed the best accuracy. [ABSTRACT FROM AUTHOR]

Abstract Despite that Heavy-Duty Vehicles (HDVs) represent a small part of the overall vehicle population they have been identified as one of the most important contributors to air pollution. This is one of the reasons why HDV emissions regulations are becoming more and more stringent worldwide. Following this trend, Europe introduced the Euro VI standard which includes more stringent emission limits for hydrocarbons, PM and NO x , while for the first time a limit for solid PN emissions was set. At the same time increased concern regarding greenhouse gas emissions led to a series of initiatives including HDVs CO 2 certification and monitoring with the latest being the submission of a proposal for CO 2 emission targets for 2025 and 2030. Despite the general concern, published data regarding Euro VI HDV emissions are scarce, while real-world emission factor measurements of regulated and unregulated pollutants are even more difficult to find. The main objective of this paper is to present real-world diesel Euro VI HDVs emissions of both gaseous pollutants and solid PN. For that reason five HDVs, including four trucks and one bus, were tested on-road under typical driving conditions. A breakdown of the emissions to low, medium, and high speed conditions was also performed with the aim investigating the performance of aftertreatment systems under different speed conditions. All tested vehicles performed better compared to older technology diesel HDVs, thus reflecting the technological improvements introduced over the last years. However, relatively high emissions were observed for some of the pollutants over low speed phases due to reduced effectiveness of the corresponding emission control systems. Calculated emission factors were also compared to existing emission inventories and good correlation was found for NO x , CO 2 and solid PN emissions. Highlights • Five Euro VI HDVs were tested on-road under typical driving conditions. • The vehicles demonstrated overall lower emissions compared to older technology HDVs. • Relatively high emissions were observed for some pollutants over low speed phases. • Emission inventories succeed in calculating real world EFs of Euro VI HDVs. [ABSTRACT FROM AUTHOR]

One of the benefits of the combined heat and power (CHP) systems is reducing greenhouse gases (GHG) in the national level. These systems are installed in the close vicinity of local communities and their pollution can directly affect the local environment. This paper studies the adverse environmental effects of a 2 MW gas-burning CHP (GB-CHP) system in a residential community in Mashhad, Iran. We compare the amount of C O 2 and N O x emissions from the GB-CHP with the conventional choice of supplying electricity from the central power plants and heat from the locally installed boilers. The results show that the GB-CHP system increases local C O 2 and N O x emissions by approximately two and six times, respectively. The N O x emission is increased from 3.1 tonnes to 18.7 tonnes. The C O 2 emission increases from 3623 tonnes to 7238 tonnes. Furthermore, we use Gaussian Dispersion model to estimate the concentration of these emissions in the local area using the weather conditions of the most polluted times of the previous year. The results show that for a 3-meter stack, the maximum concentration of C O 2 and N O x are 14.8 m g / m 3 and 38.1 μ g / m 3 , respectively. Our analysis shows that increasing the height of the GB-CHP stack from 3 meters to 10 meters reduces the concentration by 40%. [ABSTRACT FROM AUTHOR]

China is the largest emitter of greenhouse gases, and urban clusters play an essential role in China's carbon emissions because of the country's unbalanced economic development. In this paper, the discrete air sample measurements of atmospheric CO 2 , CH 4 , and CO during the period between 2007 and 2013 from Shangdianzi (SDZ) station in Jing-Jin-Ji (JJJ) region and Linan (LAN) station in Yangtze River Delta (YRD) region, respectively, are presented and characterized. By backward trajectory cluster analysis, the low background CO 2 mole fraction and similar long-term trends with Waliguan global station (WLG) were observed at SDZ because most of the air masses originate mostly from remote clean terrestrial areas in the north with limited information of anthropogenic emissions. But the background CH 4 and CO are influenced mainly by ruminants and rice field emission from the surrounding regions and urban cluster emission in the JJJ region, which lead to the higher CH 4 and CO in summer at SDZ. Whereas, the CO 2 , CH 4 and CO background values observed at LAN are higher by the combined effects of long-distance atmospheric transportation from Northern and Central China and urban clusters emission from the YRD region. The observed polluted data at SDZ and LAN are all strongly affected by regional and local anthropogenic emissions, which lead to the higher GHGs concentrations compared to the background data. However, because the air mass transportation from the north is weak in winter and no winter heating is required and vegetation photosynthesis is still stronger in the south of China, the CO 2 concentration at LAN is lower than that at SDZ in autumn and winter. The polluted CO value at SDZ is found considerably higher than that at LAN, especially in spring and winter. By analyzing the correlations of observed regional enhanced values of CO 2 , CH 4 and CO at the two stations, we find the significant correlation between them, which means they may share the common-source of anthropogenic emissions from the regional fossil fuel and biomass burning in cold season. Moreover, the high coal-consumption proportion with the low CO 2 emissions factors, and much biomass burning with low combustion efficiency in JJJ region lead to the higher slope values of △CO 2 and △CO in spring and winter at SDZ and considerably higher than that at LAN. The results can provide a further understanding of regional/local features of atmospheric GHGs under the influence of human activities in the urban clusters in China. [ABSTRACT FROM AUTHOR]

The change of O 2 content in soil caused by aerated irrigation (AI) must inevitably affect the production and emissions of CO 2 and N 2 O from soils. This paper described in-situ observation of CO 2 and N 2 O emissions from AI soils with static chamber-GC technique, in order to reveal the effects of AI on CO 2 and N 2 O emissions from soils of greenhouse tomato fields in autumn-winter season. CO 2 and N 2 O emissions from AI soils mainly concentrated in the blooming and fruit setting period compared to other periods. AI increased cumulative emissions of CO 2 and N 2 O by 11.8% (p = 0.394) and 10.0% (p = 0.480), respectively, compared to the control. The integrative global warming potential of CO 2 and N 2 O on a 100-year horizon for the AI treatment was 6430.60 kg ha −1 , increased by 11.7% compared with that for the control (p = 0.356). Both the emissions of CO 2 and N 2 O from AI soils had the exponential positive correlation with soil water-filled pore space (WFPS). The highest peak of CO 2 and N 2 O fluxes from AI soils was observed at 46.7% and 47.5% WFPS, with WFPS ranging from 43.3% to 51.5% and from 45.6% to 52.3% during the whole growth stage, respectively. In addition, the average yield for the AI treatment (34.52 t ha −1 ) was significantly greater (17.4%) compared with that of the control (p = 0.018). These results suggest that AI do not significantly increase the integrative greenhouse effect caused by CO 2 and N 2 O from soils of greenhouse tomato fields, but significantly increase the tomato yield. The research results provide certain theoretical foundation and scientific basis for accurately evaluating the farmland ecological effect of AI technique. [ABSTRACT FROM AUTHOR]

This paper presents airborne measurements of multiple atmospheric trace constituents including greenhouse gases (such as CO 2 , CH 4 , O 3 ) and biomass burning tracers (such as CO, CH 3 CN) downwind of an exceptionally large wildfire. In summer 2013, the Rim wildfire, ignited just west of the Yosemite National Park, California, and burned over 250,000 acres of the forest during the 2-month period (17 August to 24 October) before it was extinguished. The Rim wildfire plume was intercepted by flights carried out by the NASA Ames Alpha Jet Atmospheric eXperiment (AJAX) on 29 August and the NASA DC-8, as part of SEAC 4 RS (Studies of Emissions, Atmospheric Composition, Clouds and Climate Coupling by Regional Surveys), on 26 and 27 August during its intense, primary burning period. AJAX revisited the wildfire on 10 September when the conditions were increasingly smoldering, with slower growth. The more extensive payload of the DC-8 helped to bridge key measurements that were not available as part of AJAX (e. g. CO). Data analyses are presented in terms of emission ratios (ER), emission factors (EF) and combustion efficiency and are compared with previous wildfire studies. ERs were 8.0 ppb CH 4 (ppm CO 2 ) −1 on 26 August, 6.5 ppb CH 4 (ppm CO 2 ) −1 on 29 August and 18.3 ppb CH 4 (ppm CO 2 ) −1 on 10 September 2013. The increase in CH 4 ER from 6.5 to 8.0 ppb CH 4 (ppm CO 2 ) −1 during the primary burning period to 18.3 ppb CH 4 (ppm CO 2 ) −1 during the fire's slower growth period likely indicates enhanced CH 4 emissions from increased smoldering combustion relative to flaming combustion. Given the magnitude of the Rim wildfire, the impacts it had on regional air quality and the limited sampling of wildfire emissions in the western United States to date, this study provides a valuable dataset to support forestry and regional air quality management, including observations of ERs of a wide number of species from the Rim wildfire. [ABSTRACT FROM AUTHOR]

Continuous evolution of biogeochemical models developed in the past decades makes possible the more and more accurate estimation of trace and greenhouse gas fluxes of soils. Due to the detailed meteorological, soil, biological and chemical processes the modeled fluxes are getting closer and closer to the real values. For appropriate evaluation models need large amount of input data. In this paper we have investigated how to build an easily accessible meteorological input data source for biogeochemical models, as it is one of the most important input data sets that is either missing or difficult to get from meteorological networks. The DNDC ecological model was used for testing the WRF numerical weather prediction system as a potential data source. The reference dataset was built by numerical interpolation based on measured data. The average differences between the modeled output data using WRF and observed meteorological data in 2009 and 2010 are less than 3.98 ± 1.6; 8.68 ± 6.72 and 6.5 ± 2.17 per cent for CO 2 , N 2 O and CH 4 , respectively, for the test years. Generalization of the results for other regions is restricted, however this work encourages others to examine the applicability of WRF data instead of observed climate parameters. [ABSTRACT FROM AUTHOR]

The fishing industry is heavily reliant on the use of fossil fuel and emits large quantities of greenhouse gases and other atmospheric pollutants. Methods used to calculate fishing vessel emissions inventories have traditionally utilised estimates of fuel efficiency per unit of catch. These methods have weaknesses because they do not easily allow temporal and geographical allocation of emissions. A large proportion of fishing and other small commercial vessels are also omitted from global shipping emissions inventories such as the International Maritime Organisation's Greenhouse Gas Studies. This paper demonstrates an activity-based methodology for the production of temporally- and spatially-resolved emissions inventories using data produced by Automatic Identification Systems (AIS). The methodology addresses the issue of how to use AIS data for fleets where not all vessels use AIS technology and how to assign engine load when vessels are towing trawling or dredging gear. The results of this are compared to a fuel-based methodology using publicly available European Commission fisheries data on fuel efficiency and annual catch. The results show relatively good agreement between the two methodologies, with an estimate of 295.7 kilotons of fuel used and 914.4 kilotons of carbon dioxide emitted between May 2012 and May 2013 using the activity-based methodology. Different methods of calculating speed using AIS data are also compared. The results indicate that using the speed data contained directly in the AIS data is preferable to calculating speed from the distance and time interval between consecutive AIS data points. [ABSTRACT FROM AUTHOR]

Flask air samples were collected weekly at the Akedala (AKD) regional atmosphere station in north Xinjiang, China, and atmospheric carbon dioxide (CO 2) and methane (CH 4) mole fractions were measured using cavity-ring down spectroscopy. AKD station is under a typical temperate continental semiarid and arid monsoon climate with westerly air flow. This paper presents the long-term trends and seasonal variations in atmospheric CO 2 and CH 4 concentrations for the first time and discusses the possible influencing mechanisms. The CO 2 and CH 4 background mixing ratios were 412.1 ± 0.36 ppm and 1968.9 ± 0.95 ppb, respectively, in 2018. The growth rates were 2.62 ppm y−1 and 9.67 ppb y−1 for CO 2 and CH 4 , respectively, during 2010–2018, which were higher than the surface measurements at Waliguan Baseline Observatory at similar latitudes. The averaged CO 2 seasonal cycle showed a minimum in summer and a maximum in winter, and the annual amplitude was high at 18.7 ppm. The CH 4 seasonal cycle showed a rapid recovery in summer during July and August. Back-trajectory analysis showed that the atmospheric information at AKD station can provide some implications for uptake from Kazakhstan and Russia, as well as information about natural and anthropogenic emissions in northern Xinjiang. This study revealed that the long-term upward trend of the CO 2 concentration in AKD might be attributed to emissions from energy consumption of heavy industry. The comprehensive effects of livestock, geological micro seepage and oil and gas production, and vertical transport also enhanced the atmospheric CH 4 concentrations. The results of this study may improve our basic understanding of observed atmospheric CO 2 and CH 4 in China and Central Asia and the underlying fluxes to be used in atmospheric inversion models. • Anthropogenic activities strongly influenced the CO 2 background strongly in winter. • CO 2 and CH 4 growth rates were 2.62 ppm y−1 and 9.67 ppb y−1, respectively. • Eastern Kazakhstan could be a potential source region. • Higher CH 4 was attributed to enhanced regional sources and vertical transport. • Peak slopes of ΔCO/ΔCO 2 and ΔCO/ΔCH 4 occurred in winter and spring respectively. [ABSTRACT FROM AUTHOR]