Latent heating rate profiles at different tropical cyclone stages during 2008 Tropical Cyclone Structure experiment: Comparison of ELDORA and TRMM PR retrievals

The NRL P3 with Electra Doppler Radar (ELDORA) operated during the TCS08 experiment and collected unique in situ observations and retrievals of the dynamics and thermodynamics of mesoscale convective systems (MCS) within pre-tropical cyclone (TC) disturbances. The ELDORA retrievals provide a chance to test hypotheses for mesoscale contributions to tropical cyclone formation. Two hypotheses have been proposed as to how the incipient TC surface vortex forms from the tropical disturbances via an interaction with a Mesoscale Convective System (MCS). The focus in the Top-down hypothesis is on the maximum potential vorticity (PV) in the middle levels involving a large stratiform region within the MCS. The stratiform region cloud leads to cooling below the melting layer due to melting and evaporation of precipitation, and the convective updraft region provides condensation heating. The focus in the Bottom-up hypothesis is on a maximum PV in lower levels in collaboration with a strong convective updraft region. Accordingly, one distinctive feature of the simulations of the Bottom-up process is a maximum in the latent heating profile that is lower in the troposphere in conjunction with large vertical velocities that favor spinup at lower levels.

The first space-borne radar (PR—Precipitation Radar) on the Tropical Rainfall Measuring Mission satellite (TRMM) has been retrieving the vertical profiles of precipitation and latent heating of many pre-TC disturbances over the past eleven years. The recently developed TRMM Spectral Latent Heating (SLH) algorithm provides latent heating rate (LHR) estimates from a cloud-resolving model look-up table based on the observed rain profile, the convective and stratiform classification, and the surface rain rate. The TRMM PR LHR observations are obtained from http://www.eorc.jaxa.jp/TRMM/lh/. Examination of the PR-derived LHR over numerous pre-TC tropical disturbances would greatly enhance the sample sizes for TC formation studies. However, the satellite-derived technique must be validated with in situ aircraft observations where possible.

The objective of this study is to validate the PR-derived LHR using the thermodynamic retrievals from the NRL P-3 ELDORA observations during the TCS-08 experiment. We will focus on five cases in which the PR overpasses were well correlated with the ELDORA observations: TY Jangmi (Case I), pre-TD Nuri (Case II), pre-TD Jangmi (Case III), pre-Sinlaku (Case IV), and TCS 25 (Case V). These cases include the mature, pre-developing and developing phases of tropical cyclones, and a non-developing system. The ELDORA dataset requires a number of sequential steps that retrieve dynamics and microphysics. The ELDORA data are first corrected for navigation errors and automatically edited using the NCAR SOLO II software package. The edited reflectivity and Doppler velocity are interpolated to a Cartesian grid, and synthesized using a three-dimensional variational approach. Using the derived three-dimensional winds, the horizontal and vertical gradient of pressure and temperature perturbations are calculated from the momentum and thermodynamic equations. Here the temperature perturbation field is calculated relative to the nearest dropsonde sounding from the US Air Force C-130 or the Taiwan DOTSTAR. The retrieved temperature field is used with the derived vertical motion field to calculate the LHR profile for comparison with the PR LHR profile.

In the first validation result for TY Jangmi (Case I), the Contoured Frequency Diagram by Altitude (CFAD) of PR LHR reasonably reveals two heating maxima near 4 km and 7.5 km that are also present in the ELDORA LHR. However, the CFAD of PR LHR does not have cooling maxima near 3 km and 7 km that are derived from ELDORA LHR. Overall, the PR LHR values are slightly larger than the ELDORA LHR in the range of positive LHR, and generally fail to represent negative LHR via evaporation cooling that are apparent in the ELDORA LHR. This presentation will include discussion on the possible factors leading to the discrepancies. The first discrepancy factor is an inherent observational sensitivity difference between the two sensors. That is, PR is only sensitive to large-ice particles (graupel) within the convective region due to its sensor frequency (13.8 GHz), and is not able to detect small particles mostly within the stratiform region and anvil of the MCS. Another possibility is that the look-up table based on cloud-resolving model simulations over the TOGA COARE region reasonably well represents cloud microphysics in the convective region of the MCS, but fails to simulate the stratiform and anvil region in strong vertical wind shear conditions. That is, such strong shear effects are not considered in the CRM simulations. In summary, this is first attempt to validate the satellite-derived latent heating of MCSs within tropical cyclones and pre-tropical cyclones using the possibly more accurate ELDORA retrievals, which may then contribute to future TRMM algorithm improvements.