Electronic Data Submission for Paper 2003JD003481
OH column abundance over Table Mountain Facility, California:
Intraannual variations and comparisons to model predictions for 1997-2001
Franklin P. Mills,{1,2}
Richard P. Cageao,{1}
Stanley P. Sander,{1,3,4}
Mark Allen,{1,4}
Yuk L. Yung,{4}
Ellis E. Remsberg,{5}
James M. Russell, III,{6}
Ulf Richter{3}
{1}{Jet Propulsion Laboratory, California Institute of Technology,
Pasadena, California, USA}
{2}{Now at Research School of Physical Sciences and Engineering,
Australian National University, Canberra, ACT, Australia}
{3}{Division of Engineering and Applied Science,
California Institute of Technology, Pasadena, California, USA}
{4}{Division of Geological and Planetary Sciences,
California Institute of Technology, Pasadena, California, USA}
{5}{NASA Langley Research Center, Hampton, Virginia, USA}
{6}{Center for Atmospheric Sciences, Hampton University,
Hampton, Virginia, USA}
J. Geophys. Res., vol. 108, doi: 10.1029/2003JD003481, 2003.
INTRODUCTION
The auxiliary material in this electronic supplement provides additional
information on the OH column measurements made over the Jet Propulsion
Laboratory's Table Mountain Facility, California, their analysis, and their
interpretation via modeling. All of the auxiliary material is in the form
of figures for which the captions appear below.
2003JD003481-A1.eps
Figure A1.
Solid line is the smallest solar zenith angle (SZA) observable from TMF.
Each diamond indicates the SZA for a morning OH column measurement.
2003JD003481-A2.eps
Figure A2.
Same as Figure A1 but for afternoon measurements.
Solid line is the smallest solar zenith angle (SZA) observable from TMF.
Each diamond indicates the SZA for an afternoon OH column measurement.
2003JD003481-A3.eps
Figure A3.
Morning P1(1) OH column measurements as a function of SZA over TMF for
Jul 1997 to Dec 2001. Square is for measurements in Nov, Dec, and Jan.
Plus sign is for measurements in Feb, Mar, and Apr. Triangle is for
measurements in May, Jun, and Jul. "X" is for measurements in Aug,
Sep, and Oct. Vertical bars at each end illustrate the typical 2-sigma_r
random uncertainty for an OH column measurement at 10-20 and 65-75 degree
SZA, respectively. The smallest OH column abundances at 25-65 degree SZA
in this figure were measured after a solar storm in Apr 2000.
2003JD003481-A4.eps
Figure A4.
Same as Figure A3 but for afternoon measurements.
Afternoon P1(1) OH column measurements as a function of SZA over TMF for
Jul 1997 to Dec 2001. Square is for measurements in Nov, Dec, and Jan.
Plus sign is for measurements in Feb, Mar, and Apr. Triangle is for
measurements in May, Jun, and Jul. "X" is for measurements in Aug,
Sep, and Oct. Vertical bars at each end illustrate the typical 2-sigma_r
random uncertainty for an OH column measurement at 10-20 and 65-75 degree
SZA, respectively. The smallest OH column abundances at 25-65 degree SZA
in this figure were measured after a solar storm in Apr 2000.
2003JD003481-A5.eps
Figure A5.
Histogram of 2-sigma_{sf} spectral fit uncertainty ((Delta-column / column)
expressed as a percentage) for the P1(1) OH line observations over TMF
between Jul 1997 and Dec 2001. The total number of OH line observations
included in the histogram is 5511. An additional 152 OH line observations
are not included in the histogram because they had a 2-sigma_{sf} > 40%.
2003JD003481-A6.eps
Figure A6.
Same as Figure A5 but for Q1(2).
Histogram of 2-sigma_{sf} spectral fit uncertainty ((Delta-column / column)
expressed as a percentage) for the Q1(2) OH line observations over TMF
between Jul 1997 and Dec 2001. The total number of OH line observations
included in the histogram is 5495. An additional 168 OH line observations
are not included in the histogram because they had a 2-sigma_{sf} > 45%.
2003JD003481-A7.eps
Figure A7.
Same as Figure 1 but for afternoon measurements.
Afternoon OH column measurements with spectral fit uncertainty smaller
than 36% as a function of SZA over TMF for Jul 1997 to Dec 2001 for
the P1(1) OH line. Points and vertical bars are the same as in
Figure A4, but excluding points with large spectral fit uncertainties.
Solid black curve is the best second-order fit to the data.
Solid red line is the best linear fit to the data. Short dashed red
lines are the 2-sigma uncertainties on the best linear fit. All fits
and the uncertainties on the fits were calculated via a weighted least
squares Singular Value Decomposition technique [Press et al., 1989
(Chapter 14)]. The second-order and linear fits are almost
indistinguishable for most SZA. The spectral fit uncertainties for
the OH column abundances measured after the Apr 2000 solar storm
(Figure A4) are generally larger than the 36% limit used to select
data for inclusion in this figure.
2003JD003481-A8.eps
Figure A8.
Same as Figure 1 but for Q1(2) measurements.
Morning OH column measurements with spectral fit uncertainty smaller
than 41% as a function of SZA over TMF for Jul 1997 to Dec 2001 for
the Q1(2) OH line. Solid black curve is the best second-order fit to
the data. Solid red line is the best linear fit to the data. Short
dashed red lines are the 2-sigma uncertainties on the best linear fit.
All fits and the uncertainties on the fits were calculated via a weighted
least squares Singular Value Decomposition technique [Press et al., 1989
(Chapter 14)]. The second-order and linear fits are almost
indistinguishable for most SZA.
2003JD003481-A9.eps
Figure A9.
Same as Figure A8 but for afternoon measurements.
Afternoon OH column measurements with spectral fit uncertainty smaller
than 41% as a function of SZA over TMF for Jul 1997 to Dec 2001 for
the Q1(2) OH line. Solid black curve is the best second-order fit to the
data. Solid red line is the best linear fit to the data. Short dashed
red lines are the 2-sigma uncertainties on the best linear fit. All fits
and the uncertainties on the fits were calculated via a weighted least
squares Singular Value Decomposition technique [Press et al., 1989 (Chapter 14)].
The second-order and linear fits are almost indistinguishable for most SZA.
2003JD003481-A10.eps
Figure A10.
Average for each day of normalized OH column measurements over TMF from
(a) morning and (b) afternoon for the P1(1) OH line at 10-65 degree SZA.
Uncertainties are 2-sigma_r uncertainties on the mean for each day
[Bevington, 1969 (chapter 5)].
2003JD003481-A11.eps
Figure A11.
Difference between daily average normalized TMF OH column measurements
at 10-65 degree SZA and one day average for normalized morning TMF OH
column measurements at 50-65 degree SZA. Uncertainties are 2-sigma_r
for the difference between the averages. No temporal pattern is apparent
in the difference so these differences are believed to be due to random
measurement uncertainties.
2003JD003481-A12.eps
Figure A12.
Same as Figure 2 but for the Q1(2) OH line measurements.
Horizontal axis is the average over one day of the normalized morning OH
column abundance for the Q1(2) OH line data collected at 10-65 degree
SZA between Jul 1997 and Dec 2001. Vertical axis is the average over
one day of the normalized afternoon OH column abundance for the same
conditions. Uncertainties are 2-sigma_r uncertainties on the mean for
each day [Bevington, 1969 (chapter 5)]. The weighted linear least-squares
fit through the measurements [Press et al., 1989 (Chapter 14)]
is described by a_{0n} = (0.80 +/- 0.07) * m_{0n} + (0.00 +/- 0.01)
where a_{0n} is the normalized afternoon OH column and m_{0n} is the
normalized morning OH column.
2003JD003481-A13.eps
Figure A13.
Same as Figure 2 but comparing the daily average normalized OH column
abundances from the Q1(2) and P1(1) OH lines for data collected at 10-65
degree SZA between Jul 1997 and Dec 2001. The weighted linear
least-squares fit through the measurements is described by
d_{0n} = (0.82 +/- 0.04) * d_{1n} + (0.00 +/- 0.01)
where d_{0n} is the normalized Q1(2) daily OH column and d_{1n} is
the normalized P1(1) daily OH column. This implies that deviations
from the mean for the Q1(2) OH line are typically 82% of the deviations
for the P1(1) OH line.
2003JD003481-A14.eps
Figure A14.
OH sensitivity coefficients calculated as described in the text from the
analytic model for changes in H_2O concentrations at three times during a
day at spring equinox near 30 degree N latitude. The morning calculation
was done for ~ 60 degree SZA and the afternoon for ~ 65 degree SZA.
2003JD003481-A15.eps
Figure A15.
Same as Figure A14 but for O_3.
OH sensitivity coefficients calculated as described in the text from the
analytic model for changes in O_3 concentrations at three times of day
at spring equinox near 30 degree N latitude. The morning calculation was
done for ~ 60 degree SZA and the afternoon for ~ 65 degree SZA.
2003JD003481-A16.eps
Figure A16.
Same as Figure A14 but for "OverO_3."
OH sensitivity coefficients calculated as described in the text from the
analytic model for changes in "OverO_3" at three times of day
at spring equinox near 30 degree N latitude. The morning calculation was
done for ~ 60 degree SZA and the afternoon for ~ 65 degree SZA.
2003JD003481-A17.eps
Figure A17.
Profiles of the fraction of the total OH column ([OH]_j/(sum of [OH]_j)).
The long dashed line is the February 1992 model calculation from
Pickett and Peterson [1996]. The dotted line is the empirical
high sun OH profile from Canty et al. [2000]. The short dashed lines
are the OH profiles from the 1997 MAHRSI measurements [Conway et al. 2000].
The green short dashed line is the lower limit MAHRSI OH profile,
the blue short dashed line is the upper limit MAHRSI OH profile, and
the black short dashed line is the nominal MAHRSI OH profile.
The red curves are extensions of the published profiles to lower or higher
altitudes. The profiles were extended to lower and higher altitudes than was
reported in the source publications by scaling the standard noontime OH profile
from the Caltech/JPL photochemical model to match the OH concentrations near
the lower and upper altitude limits reported in the source publications.
Three of the extended OH profiles were used for calculating the weighting
functions as described in the text. The solid lines are the OH profiles
from the Caltech/JPL photochemical model that were used in the analytic
model for calculating the sensitivity coefficients.
The green solid line is the morning 60 degree SZA OH profile.
The blue solid line is the afternoon 65 degree SZA OH profile.
The black solid line is the noontime 30 degree SZA OH profile.
The OH concentrations as a function of altitude that were used to
calculate the fraction of the OH column at each altitude are shown in
Figure A18.
2003JD003481-A18.eps
Figure A18.
Profiles of OH concentration as a function of altitude.
The long dashed line is the February 1992 model calculation from
Pickett and Peterson [1996] and has a total OH column abundance of
5.1 * 10^13 cm^-2 with model-based extrapolations.
The dotted line is the empirical high sun OH profile from
Canty et al. [2000] and has a total OH column abundance of
6.7 * 10^13 cm^-2 with model-based extrapolations.
The short dashed lines are the OH profiles from the 1997 MAHRSI
measurements [Conway et al. 2000].
The green short dashed line is the lower limit MAHRSI OH profile
(total OH column 5.9 * 10^13 cm^-2 with model-based extrapolations),
the blue short dashed line is the upper limit MAHRSI OH profile
(total OH column 8.3 * 10^13 cm^-2 with model-based extrapolations),
and the black short dashed line is the nominal MAHRSI OH profile
(total OH column 7.1 * 10^13 cm^-2 with model-based extrapolations).
The red curves are extensions of the published profiles to lower or higher
altitudes. The profiles were extended to lower and higher altitudes than
was reported in the source publications by scaling the standard noontime
OH profile from the Caltech/JPL photochemical model to match the OH
concentrations near the lower and upper altitude limits reported in the
source publications. Three of the extended OH profiles were used for
calculating the weighting functions as described in the text.
The solid lines are the OH profiles from the Caltech/JPL photochemical
model that were used in the analytic model for calculating the
sensitivity coefficients.
The green solid line is the morning 60 degree SZA OH profile
(total OH column 4.8 * 10^13 cm^-2), the blue solid line is the
afternoon 65 degree SZA OH profile (total OH column 5.5 * 10^13 cm^-2),
and the black solid line is the noontime 30 degree SZA OH profile
(total OH column 7.2 * 10^13 cm^-2).