Convective-scale Data Assimilation of High-resolution Wind and Thermodynamic Observations during PECAN and VORTEX2

The accuracy of model-forecasted deep moist convection partly is limited by uncertainties in details of mesoscale environments associated with convection initiation and evolution to maturity. Ensemble Kalman filter data assimilation has been recognized as a powerful tool for producing dynamically consistent analyses of mesoscale environments and convective-scale phenomena by expanding the influence of limited observations to areas of the analysis domain not directly observed. Ongoing research by the coauthors and collaborators is exploring improvements in mesoscale details of shear and instability triggering mechanisms surrounding forming and maturing deep convection that is achievable by assimilating research observations collected during two recent field projects, the Plains Elevated Convection at Night (PECAN) experiment and the second Verification of the Origins of Rotation in Tornadoes Experiment (VORTEX2), into high-resolution model ensembles.

Research observations from Doppler on Wheels and other mobile radars, high-frequency soundings, profilers, and in situ surface and aircraft instruments collected during PECAN convection initiation missions are assimilated into a multi-scale (3-km and 1-km grid spacing) WRF ensemble. Initial analysis will be performed on the 24 June 2015 case in eastern Nebraska. The goal of this project is to provide the most detailed set of four-dimensional gridded kinematic and thermodynamic analyses possible for examination of processes instigating nocturnal convection, including localized details of the stability and shear in the surrounding environment. We will focus on understanding the roles of gravity waves (e.g., bores), surface boundaries, and a low-level jet in triggering precise locations of nocturnal convection initiation within broad areas of elevated mesoscale convergence. Although this work is in its early stages, preliminary mesoscale analyses will be shown.

 A second research project seeks to examine the value of assimilating super rapid-scan GOES-derived atmospheric motion vectors (AMVs) into convective-scale (< 3 km grid spacing) WRF ensembles to improve mesoscale details of shear surrounding developing tornadic storms. Assimilation of such AMVs typically is reserved for simulations of tropical storms over oceans (where few in situ observations are available) and to track the early development of potentially severe thunderstorms in their initiation stages. One goal of this is research is to understand the impact of AMVs over land relative to routine radiosondes, commercial aircraft, and radar wind observations. Preliminary results of the impact of the GOES AMVs collected on 5 June 2009, before and during the Goshen, CO, WY, VORTEX2 supercell, will be presented. 

Collaborators: Bob Rabin (NOAA/NSSL and UW-Madison/CIMSS), Tammy Weckwerth (NCAR), Kristen Rasmussen (NCAR), and Jim Wilson (NCAR)