Top-down estimate of anthropogenic emission inventories and their interannual variability in Houston using a mesoscale inverse modeling technique

Bottom-up CH4 emission inventories, which have been developed from statistical analyses of activity data and country specific emission factors (EFs), have high uncertainty in terms of the estimations, according to results from top-down inverse model studies. This study aimed to determine the causes of overestimation in CH4 bottom-up emission inventories across China by applying parameter variability uncertainty analysis to three sets of CH4 emission inventories titled PENG, GAINS, and EDGAR. The top three major sources of CH4 emissions in China during the years 1990–2010, namely, coal mining, livestock, and rice cultivation, were selected for the investigation. The results of this study confirm the concerns raised by inverse modeling results in which we found significantly higher bottom-up emissions for the rice cultivation and coal mining sectors. The largest uncertainties were detected in the rice cultivation estimates and were caused by variations in the proportions of rice cultivation ecosystems and EFs; specifically, higher rates for both parameters were used in EDGAR. The coal mining sector was associated with the second highest level of uncertainty, and this was caused by variations in mining types and EFs, for which rather consistent parameters were used in EDGAR and GAINS, but values were slightly higher than those used in PENG. Insignificant differences were detected among the three sets of inventories for the livestock sector.

Download Full-text

Inverse modeling of black carbon emissions over China using ensemble data assimilation

Atmospheric Chemistry and Physics ◽

10.5194/acp-16-989-2016 ◽

2016 ◽

Vol 16 (2) ◽

pp. 989-1002 ◽

Cited By ~ 8

Author(s):

P. Wang ◽

H. Wang ◽

Y. Q. Wang ◽

X. Y. Zhang ◽

S. L. Gong ◽

...

Keyword(s):

Data Assimilation ◽

Black Carbon ◽

Atmospheric Chemistry ◽

Inverse Modeling ◽

Optimal Interpolation ◽

Top Down ◽

Bottom Up ◽

Error Matrix ◽

Ensemble Data Assimilation ◽

Ensemble Data

Abstract. Emissions inventories of black carbon (BC), which are traditionally constructed using a bottom-up approach based on activity data and emissions factors, are considered to contain a large level of uncertainty. In this paper, an ensemble optimal interpolation (EnOI) data assimilation technique is used to investigate the possibility of optimally recovering the spatially resolved emissions bias of BC. An inverse modeling system for emissions is established for an atmospheric chemistry aerosol model and two key problems related to ensemble data assimilation in the top-down emissions estimation are discussed: (1) how to obtain reasonable ensembles of prior emissions and (2) establishing a scheme to localize the background-error matrix. An experiment involving 1-year-long simulation cycle with EnOI inversion of BC emissions is performed for 2008. The bias of the BC emissions intensity in China at each grid point is corrected by this inverse system. The inverse emission over China in January is 240.1 Gg, and annual emission is about 2539.3 Gg, which is about 1.8 times of bottom-up emission inventory. The results show that, even though only monthly mean BC measurements are employed to inverse the emissions, the accuracy of the daily model simulation improves. Using top-down emissions, the average root mean square error of simulated daily BC is decreased by nearly 30 %. These results are valuable and promising for a better understanding of aerosol emissions and distributions, as well as aerosol forecasting.

Download Full-text

Trends and inter-annual variability of methane emissions derived from 1979-1993 global CTM simulations

Atmospheric Chemistry and Physics ◽

10.5194/acp-3-73-2003 ◽

2003 ◽

Vol 3 (1) ◽

pp. 73-88 ◽

Cited By ~ 68

Author(s):

F. Dentener ◽

M. van Weele ◽

M. Krol ◽

S. Houweling ◽

P. van Velthoven

Keyword(s):

Transport Model ◽

Meteorological Data ◽

Stratospheric Ozone ◽

Methane Emissions ◽

Anthropogenic Emission ◽

Top Down ◽

Bottom Up ◽

Annual Variability ◽

The Difference ◽

Ozone Precursor

Abstract. The trend and interannual variability of methane sources are derived from multi-annual simulations of tropospheric photochemistry using a 3-D global chemistry-transport model. Our semi-inverse analysis uses the fifteen years (1979--1993) re-analysis of ECMWF meteorological data and annually varying emissions including photo-chemistry, in conjunction with observed CH4 concentration distributions and trends derived from the NOAA-CMDL surface stations. Dividing the world in four zonal regions (45--90 N, 0--45 N, 0--45 S, 45--90 S) we find good agreement in each region between (top-down) calculated emission trends from model simulations and (bottom-up) estimated anthropogenic emission trends based on the EDGAR global anthropogenic emission database, which amounts for the period 1979--1993 2.7 Tg CH4 yr-1. Also the top-down determined total global methane emission compares well with the total of the bottom-up estimates. We use the difference between the bottom-up and top-down determined emission trends to calculate residual emissions. These residual emissions represent the inter-annual variability of the methane emissions. Simulations have been performed in which the year-to-year meteorology, the emissions of ozone precursor gases, and the stratospheric ozone column distribution are either varied, or kept constant. In studies of methane trends it is most important to include the trends and variability of the oxidant fields. The analyses reveals that the variability of the emissions is of the order of 8Tg CH4 yr-1, and likely related to wetland emissions and/or biomass burning.

Download Full-text

Insight Into Integrated Reservoir Management Using a Top-Down Intelligent-Reservoir-Modeling Technique: Application to a Giant and Complex Oil Field

Journal of Petroleum Technology ◽

10.2118/1210-0063-jpt ◽

2010 ◽

Vol 62 (12) ◽

pp. 63-64

Author(s):

Dennis Denney

Keyword(s):

Reservoir Management ◽

Oil Field ◽

Modeling Technique ◽

Reservoir Modeling ◽

Top Down ◽

Insight Into

Download Full-text

Modeling seed-origin oak regeneration in the central Appalachians

Canadian Journal of Forest Research ◽

10.1139/x05-291 ◽

2006 ◽

Vol 36 (4) ◽

pp. 833-844 ◽

Cited By ~ 10

Author(s):

P J Gould ◽

K C Steiner ◽

M E McDill ◽

J C Finley

Keyword(s):

Inverse Modeling ◽

Oak Regeneration ◽

Top Down ◽

Data Set ◽

The Third ◽

Advanced Regeneration ◽

Modeling Approach ◽

Seed Origin ◽

Level Model ◽

Term Data

We describe the development of a model to quantify seed-origin oak regeneration potential in advance of complete overstory removal in central Appalachian oak stands. The model was developed using a "top-down" modeling approach that differs significantly from the approaches used to develop similar models for other regions. The modeling approach was designed to take advantage of the best data available for the region. A stand-level model was first fit using a long-term data set from Pennsylvania that was developed, in part, from operational data collected through the course of timber sales. The stand-level model describes the relationship between oak advanced regeneration distribution (the percentage of 4 m2 sample plots that contained at least one oak seedling before harvest) and third-decade seed-origin oak stocking (the percentage of growing space occupied by seed-origin oaks in the third decade after harvest). Inverse modeling was used to fit a plot-level model using a highly detailed short-term data set collected as part of an ongoing study of regeneration development in Pennsylvania. A negative exponential function (1 eαx) was used for the plot-level model to simplify the calculation of multiple seedling success probabilities. The plot-level model predicts the probability that a 4 m2 plot will be occupied by an oak during the third decade after harvest based on the sum of the heights of oak advanced regeneration (aggregate height). The top-down inverse modeling approach used here proved to be a feasible alternative to the more common individual seedling modeling approach, which requires more specialized data that are often difficult to obtain.

Download Full-text