Close Window

Texas A & M University: "High-Impact Forecasting in the Southeast Texas Upper-Air Sparse Region."


This Outreach Partners project will examine the high-impact forecast consequences of the lack of upper-air rawinsonde network observations within the Houston/Galveston (HGX) NWS forecast office’s county warning area (CWA) and assess how existing tools (model soundings, satellite retrievals, ACARS, supplemental observations, etc.) can mitigate this shortfall. Currently, forecasters provide forecasts and warnings without knowledge of the representativeness of tools such as model analyses of the atmospheric temperature/moisture profile. This forecast problem is especially critical during high impact forecast/warning events involving winter precipitation type and onset and severe thunderstorms. In both cases, small deviations in the vertical distribution of temperature/moisture can have a significant bearing on whether or not a high-impact weather event occurs, as well as the magnitude, geographic extent, and timing of the event.

The researchers’ approach is for the HGX forecasters (or Storm Prediction Center forecasters) to identify high-impact events in real-time for which the academic PI and a team of student volunteers would launch special rawinsonde observations. In addition to assessing the value of the observation to the forecast/warning effort via a subjective rubric to be completed by HGX post-event, the sounding data will also be compared objectively to existing tools (model soundings, satellite retrievals, etc.) to characterize representativeness and best existing mitigating tools in the absence of any in-situ soundings. The supplemental soundings will be assimilated into the WRF EMS model utilizing the Local Analysis and Prediction System (LAPS) for an initial analysis of NWP impact potential. The data set and case studies collected during the year will then become the basis for follow-on numerical studies and will also be used to determine typical model forecast biases across southeast Texas.