Search Results
Abstract
An automated cloud-edge detection algorithm was developed and extensively tested. The algorithm uses in situ cloud physics data measured by a research aircraft coupled with ground-based weather radar measurements to determine whether the aircraft is in or out of cloud. Cloud edges are determined when the in/out state changes, subject to a hysteresis constraint. The hysteresis constraint prevents isolated transient cloud puffs or data dropouts from being identified as cloud boundaries. The algorithm was verified by detailed manual examination of the dataset in comparison to the results from application of the automated algorithm.
Abstract
An automated cloud-edge detection algorithm was developed and extensively tested. The algorithm uses in situ cloud physics data measured by a research aircraft coupled with ground-based weather radar measurements to determine whether the aircraft is in or out of cloud. Cloud edges are determined when the in/out state changes, subject to a hysteresis constraint. The hysteresis constraint prevents isolated transient cloud puffs or data dropouts from being identified as cloud boundaries. The algorithm was verified by detailed manual examination of the dataset in comparison to the results from application of the automated algorithm.
Abstract
The accuracy and availability of data from a network of 915-MHz boundary layer wind profilers operated by the U.S. Air Force on the Eastern Range are assessed using an automated quality control (QC) algorithm developed by the authors. The accuracy and reliability of the automated algorithm is assessed using the results of an extensive manual examination of the same data used for the assessment of the instruments. The details of the automated algorithm and the manual screening process are provided.
Data were collected over a 647-day period from five profilers configured to produce one profile every 15 min, resulting in about 200 000 measurements. The results indicate that the instruments provide reliable, accurate data except when maintenance problems or heavy precipitation are present. Precipitation affected as much as 25% of the measurements in the dataset. The automated QC algorithm proved extremely effective in identifying unacceptable data. Only 0.03% of the data passing automated QC were identified as bad by manual review. While some valid data were identified as bad, the automated algorithm appears to provide exceptional performance for use in automated operational assimilation of boundary profiler data for model initialization and data visualization.
Abstract
The accuracy and availability of data from a network of 915-MHz boundary layer wind profilers operated by the U.S. Air Force on the Eastern Range are assessed using an automated quality control (QC) algorithm developed by the authors. The accuracy and reliability of the automated algorithm is assessed using the results of an extensive manual examination of the same data used for the assessment of the instruments. The details of the automated algorithm and the manual screening process are provided.
Data were collected over a 647-day period from five profilers configured to produce one profile every 15 min, resulting in about 200 000 measurements. The results indicate that the instruments provide reliable, accurate data except when maintenance problems or heavy precipitation are present. Precipitation affected as much as 25% of the measurements in the dataset. The automated QC algorithm proved extremely effective in identifying unacceptable data. Only 0.03% of the data passing automated QC were identified as bad by manual review. While some valid data were identified as bad, the automated algorithm appears to provide exceptional performance for use in automated operational assimilation of boundary profiler data for model initialization and data visualization.
Abstract
The performance of an improved signal-processing algorithm implemented on the NASA 50-MHz radar wind profiler at Kennedy Space Center is analyzed. In 1990, NASA began using a 50-MHz Doppler radar wind profiler to demonstrate the applicability of the technology to assessing launch wind conditions at Kennedy Space Center. To produce critical wind profiles in minimal time, NASA replaced the conventional signal-processing system delivered by the manufacturer with a more robust system. The new signal-processing system uses a median filter to remove spurious Doppler spectral data and constrains the search for the atmospheric signal by a first guess. The new system has been in nearly continuous operation since mid-1994. Over this period, the system performance was evaluated in varied weather conditions, and numerous comparisons with wind profiles from radar-tracked jimspheres were accomplished. The system is now integrated into the prelaunch wind evaluation structure. This paper discusses the details of the new signal-processing system and presents the results of the performance analysis.
Abstract
The performance of an improved signal-processing algorithm implemented on the NASA 50-MHz radar wind profiler at Kennedy Space Center is analyzed. In 1990, NASA began using a 50-MHz Doppler radar wind profiler to demonstrate the applicability of the technology to assessing launch wind conditions at Kennedy Space Center. To produce critical wind profiles in minimal time, NASA replaced the conventional signal-processing system delivered by the manufacturer with a more robust system. The new signal-processing system uses a median filter to remove spurious Doppler spectral data and constrains the search for the atmospheric signal by a first guess. The new system has been in nearly continuous operation since mid-1994. Over this period, the system performance was evaluated in varied weather conditions, and numerous comparisons with wind profiles from radar-tracked jimspheres were accomplished. The system is now integrated into the prelaunch wind evaluation structure. This paper discusses the details of the new signal-processing system and presents the results of the performance analysis.
