Lance E. Deaver

Validation of hydrogen fluoride measurements made by the Halogen Occultation Experiment from the UARS platform

The Halogen Occultation Experiment (HALOE) on UARS uses the method of solar occultation limb sounding to measure the composition and structure of the stratosphere and mesosphere. One of the HALOE channels is spectrally centered at 3.4 μm to measure the vertical profile and global distribution of hydrogen chloride. The mean difference between HALOE and 14 balloon correlative underflight measurements ranges from 8% to 19% throughout most of the stratosphere. This difference is within the limits of error bar overlap for the two data sets. The mean differences between HALOE and HCl data from ATMOS flights on the space shuttle is of the order of 15 to 20% for the 1992 flight and 10% for the 1993 flight. Generally, HALOE results tend to be low in these comparisons. Also, comparisons with two-dimensional model calculations and HALOE data are in good qualitative agreement regarding vertical profile shapes and features in a pressure versus latitude cross section. HCl values increase from ∼0.3 parts per billion by volume (ppbv) to 1 ppbv in the lower stratosphere to 2.6 ppbv to 3.3 ppbv just above the stratopause which is the upper limit of HALOE single-profile measurements. There is a dependence of HCl results on the angle between the orbit plane and the Earth-Sun vector with HCl varying by ±9% in the upper stratosphere. This variation appears to be altitude dependent and it is not discernible in the data below about 10 mbar.

Halogen Occultation Experiment confirmation of stratospheric chlorine decreases in accordance with the Montreal Protocol

Near-global time series of Halogen Occultation Experiment (HALOE) derived total Cl, F, and the Cl/F ratio were evaluated and compared with data from two ground-based CFC measurement programs and United Nations Environment Programme (UNEP) best-case scenarios for CFC emissions. The HALOE Cl profiles confirm that the chlorine burden near the stratopause is decreasing and this decrease is driven by tropospheric decreases of methyl chloroform. The HALOE F profiles show that the fluorine burden near the stratopause is primarily influenced by CFC-12. The Cl/F ratio, which is insensitive to transport effects at 55 km, shows that the atmosphere near the stratopause is becoming more fluorinated since the inception of the HALOE mission and less chlorinated since the start of 1997. The temporal changes in the derived Cl, F, and Cl/F ratio at 55 km are in accord with UNEP projections and in situ observations. These combined results demonstrate that the Montreal Protocol and subsequent strengthening amendments are currently having the desired effect of reducing stratospheric chlorine.

Electric fields and current densities under small Florida thunderstorms

The surface electric field E and Maxwell current density JM have been measured simultaneously under and near small Florida thunderstorms. These records show that the amplitude of JM is of the order of 1 nA/m2 or less in the absence of precipitation and that there are regular time variations in JM during the intervals between lightning discharges that tend to have the same shapes after different discharges in different storms. Negative cloud-to-ground (CG) lightning produces an abrupt negative change in E and a corresponding negative (or bipolar) transient in JM that is followed by a positive overshoot. Under a storm, this overshoot peaks about 1 nA/m2 above background and then decays in a quasi-exponential or linear fashion until the next discharge occurs. Nearby cloud discharges produce a lightning transient and then either a small change in JM or a negative change that subsequently relaxes back to the predischarge level in 5 to 20 s. CG flashes at a range of about 20 km produce a fast transient in JM and then a positive overshoot that subsequently relaxes back to the predischarge level in 5 to 20 s. Distant cloud discharges produce overshoots and subsequent decays that are very similar to CG flashes but of opposite (i.e., negative) polarity. We believe that the major causes of the aforementioned time variations in JM between lightning discharges are currents that flow in the finitely conducting atmosphere in response to the field changes rather than rapid time variations in the strength of the cloud current sources. The displacement current densities that are computed from the E records dominate JM except when there is precipitation, when E is large and steady, or when E is unusually noisy.

An overview of the halogen occultation experiment (HALOE) and preliminary results

The HALOE experiment will fly on the Upper Atmosphere Research Satellite (UARS) in the last quarter of 1991. The experiment uses the solar occultation limb sounding approach, in combination with gas filter and broadband radiometry to provide measurements of temperature profiles and key gases in the ClO(y), NO(y), and HO(y) chemical families of the middle atmosphere. The instrument has been characterized in great detail to determine gains, spectral response, noise, crosstalk, field-of-view, and thermal drift characteristics. A final end-to-end test using a gas cell to simulate the atmosphere demonstrated measurement repeatability to about 1 percent and agreement between measured and calculated signals to within about 1 percent to 3 percent. This latter agreement provides confidence in knowledge of both the hardware as well as the software.

Ozone changes in the lower stratosphere from the Halogen Occultation Experiment for 1991 through 1999

The Upper Atmospheric Research Satellite (UARS) Halogen Occultation Experiment (HALOE) Version 19 ozone profile data set was analyzed for both periodic and nonseasonal polynomial changes in lower stratospheric ozone for 1991-1999. The profile data were screened for cloud contamination at the lowest levels and were then integrated within six half-Umkehr (each ∼2.5 km thick) layers from 32 to 253 hPa. The column values were then binned and averaged into 10°-wide latitude regions but separated into sunrise and sunset measurements, yielding an effective time series of zonal-averaged ozone that spans 8 years. The results extend to near tropopause levels, 253 hPa in the extratropics and 127 hPa in the tropics. Multiple linear regression techniques were applied to the data of each latitude zone and half-Umkehr layer. We developed models that included annual, semiannual, and interannual periodic terms plus polynomial terms at 50°N, 30°N, equator, 30°S, and 50°S. The amplitudes of the three periodic terms vary with latitude and pressure altitude and do not maintain their same order of importance in all cases, emphasizing the changes and hemispheric asymmetry of the transport mechanisms in the lower stratosphere. We find no clear evidence for long-term change at most latitudes and layers. Steady declines in ozone of 0.5% yr -1 were found in the 32-45 hPa layer at 50°S, most likely due to transport of ozone-depleted air from higher latitudes. Our lowest tropical layer (90-127 hPa) shows an ozone decline in the early 1990s but an increase in the last half of the decade. Steady increases of ozone of 3-4% yr -1 were found at 50°N from 127 to 253 hPa and at 30°N from 179 to 253 hPa, probably in response to a changing net circulation for the Northern Hemisphere lower stratosphere. However, there is no highly significant trend at 50°N for the column ozone over the deeper layer from 8 to 253 hPa. For Northern Hemisphere middle latitudes it is concluded that the previously reported declines of zonal-average ozone in the lower stratosphere for the 1980s and early 1990s have not continued over the decade of the 1990s.

Analysis of near-global trends and variability in Halogen Occultation Experiment HF and HCl data in the middle atmosphere

We present an analysis of trends and variability in daily-averaged near-global Halogen Occultation Experiment (HALOE) retrievals of HF and HCl at 55 km. The goal of decomposing HALOE measurements into a series of components is to determine the temporal scales associated with variability and to estimate the trends in HF and HCl. To determine the significance of each component in the statistical model, partial correlations are calculated between the data and each model element using the rank-order correlation coefficient. We also account for the effects of serial correlation in our data (i.e., the lack of independence between consecutive points) in the determination of statistical significance. The results provide evidence of highly statistically significant slowdowns in the rates of accumulation of HF and HCl at 55 km. HF shows evidence of interannual and annual variability at 55 km that is not evident in HCl due to the short chemical lifetime of HCl.

HALOE observations of a slowdown in the rate of increase of HF in the lower mesosphere

A statistical analysis of monthly averaged HF measurements from HALOE shows definite signs of a slowdown of the increase in HF at 55 km. The data used in this study are monthly averaged HF data from a latitude band from 50N to 50S at an altitude of 55 km. We find that the combination of a third-order polynomial, an annual and semi-annual cycle, and a 900-day cycle provides a good model of the data. The changes in the HF trend are consistent with the trends of CFC-11 and CFC-12 in the troposphere in the early 1990s, adjusted by an estimated lag time to account for transport from the troposphere up to 55 km.

Empirical correction of thermal responses in the Solar Occultation for Ice Experiment nitric oxide measurements and initial data validation results

The Solar Occultation for Ice Experiment (SOFIE) makes broadband transmission measurements centered at 5.32 μm to determine the concentration profile of nitric oxide (NO). These measurements show a signal oscillation due to detector temperature variations that severely limit the accuracy of NO retrievals if corrections are not applied. An empirical correction was developed to remove this instrumental error. This paper describes the correction, its impact on the retrieval, and presents a comparison from 87 to 105 km versus coincident atmospheric chemistry experiment-Fourier transform spectrometer (ACE-FTS) measurements. The southern hemisphere (SH) shows excellent agreement between the datasets, with statistically insignificant differences. The northern hemisphere (NH) SOFIE measurements exhibit a low bias of -18.5% compared to ACE-FTS. NH measurements (sunrise observations) are still under study, and only SH NO data (sunset observations) are currently publicly available as of SOFIE data version 1.2.

High precision refraction measurements by solar imaging during occultation: results from SOFIE

A new method for measuring atmospheric refraction angles is presented, with in-orbit measurements demonstrating a precision of +/-0.02 arcsec (+/-0.1 microrad). Key advantages of the method are the following: (1) Simultaneous observation of two celestial points during occultation (i.e., top and bottom edges of the solar image) eliminates error from instrument attitude uncertainty. (2) The refraction angle is primarily a normalized difference measurement, causing only scale error, not absolute error. (3) A large number of detector pixels are used in the edge location by fitting to a known edge shape. The resulting refraction angle measurements allow temperature sounding up to the lower mesosphere.

On-orbit calibration of HALOE detector linearity

The Halogen Occultation Experiment (HALOE) conducted satellite solar occultation measurements for 14 years ending on 21 November 2005. HALOE contained a calibration wheel, which included three neutral density filters that were used to examine response linearity through a combination of ground and on-orbit measurements. Although measurement uncertainties preclude a confident assessment of the true extent of nonlinearity, the on-orbit data lead to the conclusion that any existing response nonlinearity has changed by less than 2% over the mission lifetime. This conclusion eliminates a potentially significant uncertainty when using HALOE data for studies of long-term atmospheric trends.