Demonstrating the value of targeted upper-air observations for assimilation into the TTU WRF ensemble using ensemble sensitivity analysis in convective scenarios

Journal Title
Journal ISSN
Volume Title

Ensemble Sensitivity Analysis (ESA) provides a framework for estimating the impact of assimilating observations in a numerical weather prediction (NWP) model. This, theoretically, allows for observation targeting, the gathering of atmospheric observations at a specific place and time in order to improve the forecast more than a static observational network would be able to. While ESA has been shown to be useful for identifying atmospheric features to which a forecast is sensitive, its utility as a targeting tool is still unknown. Part of the reason for this is the mixed results field campaigns using it have found. The results of targeting experiments using ESA may be improved by increasing the precision of the targeting technique, as to this point dropsondes and radiosondes used are generally assumed to take a vertical path through the atmosphere. To this end, a system for predicting the path a weather balloon launched from any point on the surface using varying winds from the TTU WRF ensemble (a convection-allowing NWP ensemble) and varying rates of ascent (ROAs) is developed and tested. The paths the balloons take through the atmosphere give a range of predicted ESA target values, which can be used to gauge the probability a balloon launched from any point will observe a target above any desired threshold. These hypothetical balloon trajectories are shown to be more accurate than vertical paths and provide reasonable targeting probabilities when verified against real balloons launched from the Amarillo National Weather Service forecast office, AMA. Winds and targeting cases are taken from an archive of TTU WRF ensemble runs from the Spring of 2016. In addition, a climatology of these targeting cases is developed to investigate the behavior of target values generally and how this may impact future targeting experiments.

Ensemble Sensitivity Analysis, ESA, Numerical Weather Prediction, NWP, Predictability, Convection, Thunderstorms, WRF