Area of influence (AOI) development : Fast generation of receptor-oriented sensitivity fields for use in regional air quality modeling
Journal
J. of Environmental Science and Technology
Type
journal article
Date Issued
2007
Author(s)
Habermacher, Florian
;
Napelenok, Sergey
;
Akhtar, Farhan
;
Hu, Yongtao
;
Russell, Armistead G.
Abstract
An approach is developed and tested to extend discrete,
source-based sensitivity results to provide a complete set of
information for source-air quality impacts, including
inversion of those results to develop receptor-oriented source-
impact sensitivities. First, the decoupled direct sensitivity
analysis method in 3D (DDM-3D) is used to calculate a finite
number of forward sensitivities from discrete points.
These results are then interpolated using tessellation to
provide complete fields of forward, emissions-based
sensitivities, i.e., how emissions in any one grid cell within
the domain impact any other cell. Receptor-oriented
sensitivities are then found by inverting the set of forward
sensitivities and can be used to identify the area of
influence (AOI). This economically provides results similar
to what would be found using an adjoint model. The
present approach is computationally less intensive than
adjoint modeling for a large number of receptors, and provides
both source-oriented and receptor-oriented pollutant
response fields that can be used for air quality management
and health impact analyses. The forward sensitivity
interpolation procedure, as well as the receptor-oriented
sensitivities, is evaluated using data withholding
source-based sensitivity results to provide a complete set of
information for source-air quality impacts, including
inversion of those results to develop receptor-oriented source-
impact sensitivities. First, the decoupled direct sensitivity
analysis method in 3D (DDM-3D) is used to calculate a finite
number of forward sensitivities from discrete points.
These results are then interpolated using tessellation to
provide complete fields of forward, emissions-based
sensitivities, i.e., how emissions in any one grid cell within
the domain impact any other cell. Receptor-oriented
sensitivities are then found by inverting the set of forward
sensitivities and can be used to identify the area of
influence (AOI). This economically provides results similar
to what would be found using an adjoint model. The
present approach is computationally less intensive than
adjoint modeling for a large number of receptors, and provides
both source-oriented and receptor-oriented pollutant
response fields that can be used for air quality management
and health impact analyses. The forward sensitivity
interpolation procedure, as well as the receptor-oriented
sensitivities, is evaluated using data withholding
Language
English
HSG Classification
contribution to scientific community
Refereed
Yes
Publisher
ACS
Volume
41
Number
11
Start page
3997
End page
4003
Pages
7
Subject(s)
Division(s)
Eprints ID
101824