Research Activities
 Kenneth P. Bowman

This page describes some of my current research activities.

Dynamics and Chemistry of the Upper Troposphere and Lower Stratosphere

The first official research flight of the new NCAR Gulfstream V aircraft, named HIAPER, took place December 1, 2005 as part of the Progressive Science Mission. On this flight HIAPER made multiple cross-sections through a large tropopause fold over the central United States. This movie shows the tropopause in yellow, the PV=2 surface in purple, and the HIAPER flight track in color, indicating the in situ ozone amount.

Click on the image to view a large movie of this surface.

The figure below shows the backward and forward history of the air parcels within the fold observed by HIAPER computed using NCEP GFS analyzed winds. The parcels are color coded according to their history. Two separate branches of the mid-latitude jet merge just upstream of the fold. Air coming across the Pacific Ocean is colored blue, while air moving southward from high latitudes is colored green. This air mixed within the tropopause fold as it intruded into the middle and lower troposphere.

Air parcels colored red were lifted from the lower troposphere in warm conveyor belts associated with cyclonic systems located upstream over the Pacific Ocean. This air did not mix with the air in the fold, but moved eastward across the Atlantic in the main jet just below the tropopause.

HIAPER provided detailed in situ chemical measurements of this air mass that are revealing much about the dynamical processes that control the chemical composition of the upper troposphere and lower stratosphere.

Climate Dynamics

Current research involves the use of data from the Tropical Rainfall Measuring Mission (TRMM) satellite to understand climatic and dynamical processes in the tropical atmosphere.

Unlike sun-synchronous meteorological satellites, the TRMM satellite orbit precesses with respect to the Sun. Therefore, over the course of its 6-week precession period, TRMM observes the Earth at different local times throughout the day. Using this information it is possible to estimate the climatological diurnal cycle of precipitation at locations observed by TRMM. This information can be used to evaluate how well global climate models simulate the diurnal cycle. The QuickTime movie below (9 MB) shows the diurnal harmonic of precipitation throughout the tropics from TRMM observations and from simulations by the NCAR Community Climate Model (Version 3). The yellow semi-circle in the lower map indicates the position of the sun at local noon.

By focusing on a specific region, such as Africa, it is possible to see that the amplitude of the simulated diurnal cycle is too large and the peak rainfall occurs to soon. This is typical of simulations over tropical land masses. Additional details, including an analysis of the diurnal cycle over the oceans, can be found in Collier and Bowman, 2004 and Bowman, 2005b.

The TRMM satellite provides infrequent (about once per day), nearly instantaneous, area-averaged precipitation measurements. Rain gauges provide nearly continuous measurements at a point. We are studying the best ways to average TRMM and surface rain gauge data in order to compare the two different types of observations. It is helpful to time average the rain guage data around the time of the TRMM overpasses. This figure shows the correlation between the satellite retrievals and the rain gauge measurements as a function of gauge averaging period and satellite averaging area. (Bowman, 2005a). The surface data is from rain gauges on the NOAA/PMEL TAO/TRITON buoy array in the tropical Pacific. The buoy gauges have the advantage that they are not influenced by being located on islands, where local surface heating or topography may change the precipitation relative to the surrouding ocean.

The inherent space and time correlation scales of tropical rainfall give a maximum correlation for gauge averageing periods between 5 and 10 hours and spatial 1 degree x 1 degree spatial averages. As the averaging area increases, the optimal averaging time also increases, but the overall correlation goes down as the the gauge becomes less able to sample the precipitation in the larger boxes.

Using a near-optimal averaging time of 12 hours, we look for biases between the TRMM retrievals and the rain gauge data by correlating across the set of available gauge data. The results for 1998-2001 are shown below for the 25 TAO/TRITON gauges that were in service during at least part of that time (Bowman et al., 2003).

As the figure shows, there is very little apparent bias between the TRMM Microwave Imager (TMI) and the gauges. The TRMM Precipitation Radar (PR) is biased about 30% low relative to the gauges. These gauge data have not been corrected for wind effects. Serra and McPhaden (2003) have estimated that wind affects might cause significant underestimates of the actual rain rate by the gauges.

We are also using TRMM data to study the properties of tropical wave modes and oscillations. The time-latitude section of TRMM outgoing longwave radiation (OLR) and precipitation in the figure below show evidence of propagating coherent signals.

Cho and Bowman, 2004 used space-time spectral analysis to identify tropical wave modes in the OLR and precipitation data. The results are shown below.
The gray shading indicates spectral power density. The solid lines are theoretical estimates of the dispersion relation for different equatorial wave modes. There is a distinct low-frequency peak that corresponds to the Madden-Julian oscillation. To the best of our knowledge, this is the first time that equatorial wave modes have been identified in actual precipitation data.

Atmospheric transport and mixing

My research in this are aims to develop a theory of the large-scale transport of atmospheric constituents, especially trace species such as ozone and water vapor, and of dynamical properties properties like potential vorticity. Currently we are using Green's functions methods to diagnose the transport circulation of the troposphere.

This figure shows the climatological zonal-mean dispersion of air parcels in the latitude-altitude plane from selected latitude-altitude points (indicated by the intersecting horizontal and vertical lines) (Bowman and Erukhimova, 2004).

From these and other results, it is possible to develop the following schematic of the mean-meridional transport pathways for long-lived atmospheric trace substances (Bowman and Erukhimova, 2004).

Please report broken links to k-bowman@tamu.edu.