MLS Science

Brief Overview of MLS Scientific Results To Date

The list of MLS-related publications contains full citations for all references cited here. If you are interested in learning more about individual results, please visit the publications page.

High latitude ozone depletion, chemistry and dynamics

Global distribution and variations of ozone and other species

SO2 injected into stratosphere by volcanoes

Tropical Dynamics

Upper tropospheric water vapor

SO2 from the Pinatubo Volcano

Starting within 10 days of launch, and continuing for approximately 2 months, MLS observed the 3-dimensional distribution and decay of residual SO2 injected into the tropical stratosphere by the Mt. Pinatubo eruption which occurred about 3 months before launch of UARS. These observations [Read et al., 1993] showed the Pinatubo SO2 mixing ratio maximum to occur around 26 km altitude with abundances of ~15 ppbv on 21 September 1991. The observed SO2 decay had e-folding times of 29 days at 26 km and 41 days at 21 km, consistent with expectations that the primary destruction of SO2 is due to reaction with OH leading to formation of stratospheric sulfate aerosols. Projected backward to time of eruption, the total amount of SO2 injected by Pinatubo is estimated from MLS data to be 17 Mtons, consistent with estimates inferred from other measurements. On 21 and 22 April 1993 MLS detected SO2 injected into the stratosphere by the South American Lascar volcano [unpublished results].

Early results from UARS MLS also included the first global maps of stratospheric ClO, the predominant form of chemically-reactive chlorine involved in the destruction of stratospheric O3. The initial MLS results [Waters et al., 1993a,b; see also Chipperfield, 1993] showed the lower stratospheric Antarctic vortex to be filled with ClO in the region where O3 was depleted, confirming earlier conclusions from ground-based and aircraft instruments that chlorine chemistry is the cause of the Antarctic ozone hole. They showed, for the first time, that ClO in the Antarctic vortex can become enhanced by June, and that O3 destruction by ClO is masked in the early Antarctic winter by influx of O3 expected from diabatic descent. These results also showed that the Arctic winter lower stratospheric vortex can become filled with enhanced ClO, leading to calculated vortex-averaged O3 destruction rates of ~0.7%/day. Results from 3D models [Douglass et al., 1993; Geller et al., 1993; Lefevre et al., 1994], produced shortly after the MLS results were obtained, showed the observed distribution of enhanced Arctic ClO was consistent with chemical-transport model predictions. A clear relationship was found between predicted polar stratospheric cloud formation along back trajectories and enhanced Arctic ClO observed by MLS, and sporadic large values of ClO seen by MLS outside the vortex were shown to be consistent with that expected to be caused by instrument noise [Schoeberl et al., 1993].

Definitive loss of Arctic ozone due to chemistry associated with the enhanced ClO was determined from analyses of combined MLS and UARS CLAES data by Manney et al. [1994]. Bell et al. [1994] found the expected anticorrelation between enhanced Arctic ClO measured by MLS and HCl measured from the ground. Additional confirmation of the paradigm of chemical processing by polar stratospheric clouds leading to activation of stratospheric chlorine is shown in the analyses of northern hemisphere CLAES, MLS and HALOE data by Geller et al. [1995], and in southern hemisphere MLS and CLAES data by Ricaud et al. [1995].

Differences between the Arctic and Antarctic winter vortex conditions as deduced from MLS observations are described by Santee et al. [1995], and deduced from combined MLS, CLAES and HALOE data by Douglass et al. [1995]. Mackenzie et al. [1996] compare lower stratospheric vortex ozone destruction calculated from the MLS ClO with the MLS-observed change in O3 for the northern winter of 1992-93 and southern winter of 1993. Additional comparisons between MLS observations and model results for polar chemistry are given by Ekman et al. [1995], Chipperfield et al. [1996] and Santee et al. [1996a]. Schoeberl et al. [1996a] use MLS, HALOE and CLAES data in an analysis of the development of the Antarctic ozone hole.

MLS measurements of Arctic ClO and O3 for the five northern winters observed to date are described in the collective papers of Waters et al. [1993a,b; 1995], Manney et al. [1994; 1995a,b; 1996a,b,e], and Santee et al. [1995, 1996b]. Low ozone "pockets" in the middle stratospheric winter anticyclone have been observed in MLS data and analyzed by Manney et al. [1995c], who conclude these cannot be explained solely by transport. Morris et al. [1996] show that the pockets can be explained by chemistry operating on air masses which are isolated in the anticyclone.

High-latitude dynamics

MLS observations have been used in several studies to provide information on vortex and high-latitude dynamics. Early results showed the evolution of ozone in the 1992 late winter southern polar vortex [Manney et al. 1993; see also Manney et al. 1995a] with suggestions of poleward transport associated with episodes of strong planetary wave activity, intensification of waves in temperature and ozone when a 9-day eastward traveling wave 1 becomes in-phase with a stationary wave 1 with implications for transport [Fishbein et al.1993], and large tongues of air peeling off the Antarctic vortex edge and migrating to midlatitudes during the breakup of the Antarcic vortex [Harwood et al., 1993].

Lahoz et al. [1993] analyzed northern hemisphere mid-stratosphere vortex processed diagnosed from MLS H2O, CLAES N2O and potenital vorticity calculated from UKMO data and found descent in the vortex with little or no large scale mixing across the vortex edge in the middle stratosphere. The three-dimensional evolution of the northern hemisphere stratospheric water vapor distribution observed by MLS during October 1991 to July 1992 is documented by Lahoz et al. [1994], with these distributions showing clear signatures of the effects of diabetic descent through isentropic surfaces and quasi-horizontal transport along isentropic surfaces, and the organization of the large-scale winter flow by the interaction between the westerly polar vortex and the Aleutian high. The seasonal evolution of southern hemispheric water vapor observed by MLS is described by Lahoz et al. [1996b].

Manney et al.[1995d] simulate the transport of passive tracers observed by MLS and CLAES over a 20-30 day period using Lagrangian transport calculations, and find the agreement between calculated and observed fields is best inside the polar vortices, and better in the Arctic than in the Antarctic; however, MLS H2O observations show behavior that is inconsistent with the calculationss and with that expected for passive tracers inside the polar vortex in the middle-to-upper stratosphere. Although there is not always detailed agreement outside the vortex, the trajectory calculations still reproduce the average large-scale characteristics of passive tracer evolutions in midlatitudes. Similar analyses [Manney et al.1995e] show the evolution of ozone in the lower stratosphere during early winter to be dominated by dynamics in December 1992 in the Arctic, and that ~50% of the chemical destruction of Antarctic ozone in June 1992 may be masked by dynamical effects, mainly diabatic descent, which bring higher ozone into the lower stratosphere vortex; the analyses suggests that dynamical changes masked ~20-35% of chemical ozone loss during late February and early March 1993 in the Arctic.

Morris et al. [1995] use trajectory-mapping techniques with MLS, CLAES and HALOE data to analyze dynamical wave-breaking events. Orsolini et al. [1996] use MLS O3 data to initialize a high-resolution transport model and examine ozone laminae along the Arctic polar vortex edge.

Global distribution & variations of ozone and other species

An overview of zonal mean O3 results from the first two and one-half years of MLS operation is given by Froidevaux et al. [1994]; in addition to features observed in stratospheric O3, this work includes initial results of examining residual differences between the stratospheric O3 column from MLS and the total O3 column from TOMS -- with information on tropospheric ozone as the ultimate goal. Analyses by Ziemke et al. [1996] using these data sets have shown zonal asymmetries in southern hemisphere column ozone that have implications for biomass burning. Elson et al. [1994] describe large-scale variations observed in MLS O3 and Elson et al. [1996] show zonal and large-scale variations in MLS H2O.

Randel et al. [1995] include MLS and HALOE data in analyzing changes in stratospheric ozone following the Pinatubo eruption. Dessler et al. [1995; 1996a,b] used MLS ClO and O3 data, along with that of other UARS instruments, to provide information on various aspects of stratospheric chlorine chemistry. The latitudinal distribution of ClO in the upper stratosphere [Waters et al., 1996] shows a minimum in the tropics as expected from quenching by increased amounts of upper stratospheric CH4 in the tropics. Jackman et al. [1996] compare zonal mean ClO from MLS with model results, which clearly shows a discrepancy in the middle and upper stratosphere which has not yet been resolved and which has implications for problems in accurately calculating ozone in the middle and upper stratosphere.

Two-day waves in the stratosphere have been analyzed by Limpasuvan and Leovy [1995] using MLS H2O data, and by Wu et al. [1996] using MLS temperatures. Four-day waves observed in MLS ozone, temperature and geopotential height have been analyzed by Allen et al. [1996]. MLS data have been used in calculations of stratospheric residual circulation by Rosenlof [1995] and Eluszkiewicz et al. [1996]. Pumphrey and Harwood [1996] use MLS 183 GHz radiances to explore the distribution of water vapor and ozone in the mesosphere and provide information on mesospheric dynamics.

Tropical dynamics

Kelvin waves observed in MLS tropical data have been analyzed by Canziani et al. [1994, 1995] and Stone et al. [1995], and MLS observations of the semiannual oscillation by Ray et al. [1994]. Randel et al. [1993] describe CLAES and MLS observations of stratospheric transport from the tropics to middle latitudes by planetary wave mixing. Carr et al. [1995] performed initial analyses of MLS tropical stratospheric H2O data, and Mote et al. [1995] found variations in these data which could be related to the annual cycle in tropical tropopause temperatures. More extensive analyses by Mote et al. [1995], greatly aided by the use of HALOE H2O and CH4, confirmed that tropical air entering the stratosphere from below is marked by its water vapor mixing ratio and retains a distinct memory of tropical tropopause conditions for 18 months or more; this analysis implies that vertical mixing is weak and that subtropical stratospheric "transport barriers" are effective at inhibiting transport into the tropics. Schoeberl et al. [1996b] also use MLS and other UARS data to estimate the dynamical isolation of the tropical lower stratosphere.

Upper tropospheric water vapor

Although not designed specifically for this measurement, UARS MLS is sensitive to upper tropospheric water vapor when the field-of-view of its ClO spectral band is scanned down through the troposphere which happens once per minute on each limb scan. Important features of the MLS measurement technique for upper tropospheric water vapor are its ability to observe through cirrus and to determine vertical structure with more than 1300 profiles per day. Upper tropospheric water vapor measurements are important for determine feedback effects associated with increasing greenhouse gases and radiative forcing of climate.

Initial results [Read et al. 1995] have demonstrated the ability of MLS to sense upper tropospheric water vapor, including (1) its insensitivity to cirrus, (2) its ability to observe synoptic-scale features, and (3) its ability to see detrainment streams extending from tropical convective regions. The preliminary measurements from MLS show reasonably good agreement with near-coincident aircraft measurements [Newell et al., 1996a] and a distribution in the tropics that is consistent with the Walker circulation [Newell et al., 1996b]. Elson et al. [1996] show the zonal-mean and wave-variances in these measurements seen in the first three years of the UARS mission. Newell et al. [1997] found variations in MLS tropical upper tropospheric H2O over the 1991-1994 period to be closely related to sea surface temperature variations in the eastern tropical Pacific, including both seasonal and interannual components. Stone et al. [1996] have used MLS upper tropospheric H2O measurements to investigate the structure and evolution of eastward-traveling medium-scale wave features in the southern hemisphere summertime, and found results consistent with paradigms for the structure and evolution of baroclinic disturbances.

MLS measurements of upper tropospheric water vapor in the tropical Pacific show expected correlations with El Nino events . During the strong El Nino developing in September 1997 the MLS data show an anomalously moist upper troposphere over the eastern tropical Pacific, and an anomalously dry upper troposphere over the tropical western Pacific .