Satellite-derived Mountain Wave Turbulence Interest Field Detection

Satellite-derived Mountain Wave Turbulence Interest Field Detection Wayne F. Feltz, Jason Otkin, Kristopher Bedka, and Anthony Wimmers Cooperative Institute for Meteorological Satellite Studies (CIMSS), University of Wisconsin-Madison Madison, Wisconsin, U.S.A. Robert Sharman Research Applications Laboratory, National Center for Atmospheric Research Boulder, Colorado, U.S.A. ABSTRACT Mountain, or lee waves originate when air flows over mountain ridges within a stably stratified atmosphere. Turbulence generated by mountain waves can be an aviation hazard due to strong vertical motions generated by small scale vorticity maxima in large scale atmospheric rotors (Doyle, 2002). Satellite water vapor imagery has indicated the presence of wave structures over and downwind of mountain ranges for many years. Research has been conducted to categorize the wave structure by using frequency of pilot turbulence reports, often occurring within a cloudfree atmosphere (Uhlenbrock et al., 2007). The reports of severe turbulence were associated with complex water vapor patterns, the appearance of interference or crossing wave fronts that extended downwind from the mountain ridge for a significant distance. The events that were less turbulent, as reported by pilots, had wave signatures with simpler patterns such as straight/parallel features orthogonal to the wind flow. To better understand the origin of the water vapor features, synthetic MODIS 6.7 um water vapor imagery was derived from Weather Research Forecast (WRF) model thermodynamic fields to examine temporally high resolution NWP-simulated satellite imagery fidelity versus polar orbiting MODIS water vapor imagery. The presentation will show water vapor structure classification results, NWP simulated water vapor comparisons, and objective quantification techniques for detecting water vapor structures coincident with observations of mountain wave turbulence. 1. INTRODUCTION Atmospheric turbulence is a major aviation hazard responsible for 609 fatalities, 823 injuries, and an estimated loss of $134M from 1983-1997 (Eichenbaum, 2000). Turbulence can originate in several different ways such as rapid convective development, upper tropospheric folding (Endlich, 1964), topographically-induced waves, and rotor phenomena. All turbulence generated by these mechanisms can be clear-air or in-cloud. This paper specifically focuses on mountain wave turbulence features identified within MODIS 6.7 micron water vapor imagery. Uhlenbrock et al. (2007) indicated 6.7 um MODIS water vapor interference or crossing wave patterns which were more likely to contain pilot reports of mountain wave turbulence while other simpler linear features orthogonal to the wind patterns indicated little or no pilot reports of turbulence. While MODIS imagery provided 1-km spatial resolution detail to resolve the water vapor features, only discrete polar passes were available, providing no temporal aspect to wave

movement (i.e. standing waves versus progressive wave patterns). GOES-12 imager water vapor imagery provides only 4 km resolution drastically reducing the ability to resolve shorter wavelengths. In anticipation of 2 km water vapor imagery from the GOES-R Advanced Baseline Imager, the WRF-ARW NWP model was used to provide a 1km spatial resolution and 30-min temporal resolution temperature and water vapor structure on 06 March 2004 case study. The 6 March 2004 case study was selected because it exceeded the mean number of daily 2004 pilot turbulence reports by over six standard deviations (Uhlenbrock et al. 2007). The WRF-ARW NWP profiles were then used to simulate high temporal resolution 6.7 um water vapor imagery using an infrared radiative transfer model. The NWP output also provided an opportunity to understand the cause of the herring-bone interference pattern indicated within the imagery both temporally and spatially. It also provides a look toward the GOES-R Advanced Baseline Imager (Schmidt et al. 2005) capability to monitor mountain generated water vapor wave patterns using WRF-ARW simulated 6.7 um water vapor imagery which may provide objectively derived turbulence confidence fields of interest. This paper presents a unique methodology to use satellite infrared radiances to validate dynamic numerical weather prediction structure and model output to direct utility of derived satellite products. 2. DATA AND METHODOLOGY The wave patterns seen in 6.7 um water vapor imagery were correlated with pilot reports of turbulence. Approximately 90% of the severely turbulent days had wave signatures in the water vapor imagery. The wave signatures on these turbulent exhibited complex wave patterns with apparent interference and crossing wave fronts that extended downwind from the mountains for a significant distance. The days that were less turbulent showed wave signatures that were simpler, that is, patterns that were linear in orientation and shape. Pilot reports (pireps) were analyzed to show the number of turbulence occurrences and intensities for each day to identify the most turbulent days. Those days were then compared to the satellite imagery to look for correlation with the wave signatures. Figure 1 shows examples of this analysis. Wave patterns were correlated with the PIREP reports of turbulence. Figure 2 shows MODIS scans over Colorado that correspond to days that were very turbulent. Wave signatures are clearly present, and appear to be complex in nature. Often when wave signatures were present in the water vapor channel, they were not present in the visible or infrared window channels. Figure 2 also shows three visible and one thermal IR image from the same scans as the MODIS water vapor images in demonstrating the a lack of evidence of the mountain waves as seen above. 3. SIMULATION To better understand the primary cause for the interference pattern and to provide a simulation to show future GOES-R ABI water vapor radiance detection capability of these feature an NWP model simulation was performed. The Weather Research and Forecasting (WRF) model is a sophisticated numerical weather prediction model that solves the compressible non-hydrostatic Euler equations cast in flux form on a mass-based terrain-following vertical coordinate system. Prognostic variables include the horizontal and vertical wind components, various microphysical quantities, and the perturbation potential temperature, geopotential, and surface pressure of dry air. High-resolution global datasets are used to initialize the model topography and other static surface fields. A complete description of the WRF modeling system is contained in Skamarock et al. (2005). In order to properly analyze the fine-scale moisture and thermodynamic structures evident in the MODIS imagery (Fig. 3), a high-resolution model simulation was performed for this case using

version 2.2 of the WRF model. The simulation was initialized at 00 UTC on 06 March 2004 using 1º Global Data Assimilation System analyses and then integrated for 24 hours on two nested domains containing 5- and 1-km horizontal grid spacing, respectively. Simulated atmospheric fields from the outer 5-km domain provided high-resolution lateral boundary conditions for the higher resolution inner domain. The simulation contained 65 vertical levels with the model top set to 10 hpa. The vertical resolution decreased from < 100 m in the lowest 1-km to ~440 m from 7 km to the model top. Sub-grid scale processes were parameterized using the Thompson et al. (2006) mixed-phase cloud microphysics scheme, the Mellor-Yamada-Janjic planetary boundary layer scheme (Mellor and Yamada 1982), and the Dudhia (1989) shortwave and Rapid Radiative Transfer Model (RRTM) longwave radiation (Mlawer et al. 1997) schemes. Surface heat and moisture fluxes were calculated using the Noah land surface model. No cumulus parameterization scheme was used; therefore, all clouds were explicitly predicted by the microphysics scheme. Figure 1. Number of moderate of greater turbulence reports for March (top) and April 2004 with inlayed water vapor images for selected days. WRF model output, including the surface skin temperature, atmospheric temperature, water vapor mixing ratio, and the mixing ratio and effective particle diameters for each hydrometeor species, were ingested into the Successive Order of Interaction forward radiative transfer model (Heidinger et al. 2006) in order to generate simulated top of atmosphere brightness temperatures. Gas optical depths were calculated for each MODIS infrared band using the Community Radiative Transfer Model. Ice cloud absorption and scattering properties were obtained from Baum et al. (2006), whereas the liquid cloud properties were based on Lorenz-Mie calculations. The NWP forecast indicated that total precipitable water vapor structure was reorganized to mirror NWP vertical motion and MODIS 6.7 water vapor structure. This indicated presence of deep tropospheric waves that may be a proxy to increasing turbulence field of interest confidence.

Figure 2. Aqua MODIS channel 27 (6.7 um water vapor) and channel 1 (visible) at 1) 06 March at 1950Z, 2) 05 May at 1800Z, and 3) 30 December at 1930Z. All images are from 2004 and over Colorado, U.S. Animations of this simulation is available at http://cimss.ssec.wisc.edu/goes_r/awg/proxy/nwp/co_turb/index.html Some of the features to note within the NWP simulation are the wave structure is deeply embedded throughout the troposphere the total column precipitable water vapor has been reorganized into regular wave-like pattern inference pattern caused by interaction between orographically driven wave and propagation of cold front north to south over front range of Rockies

Figure 3: WRF-ARW 500 hpa water vapor mixing ratio, 400 hpa vertical velocity, and total integrated precipitable water for 06 March 2004. Figure 4 shows the comparison between observed MODIS 6.7 um brightness temperature and synthetic 6.7 brightness temperature calculated from a 20-hr forecast using fast model and thermodynamic profiles from simulation. The qualitative agreement is impressive. The simulated image indicates a similar interference pattern observed within the MODIS image. This comparison provides confidence that the simulation can be used as a proxy data set to development objective pattern recognition methodologies for the Advanced Baseline Imager. Figure 4: Observed MODIS (left) and WRF-AWR NWP simulated (right) MODIS 6.7 um water vapor brightness temperature. The data set provides the following resource: Simulation allows synthetic GOES-R Advanced Baseline Imager radiance calculation to provide insight into what future water vapor imagery will look like with improved ABI horizontal and temporal resolution Provides a proxy data set to test objective detection techniques described below Provides physical insight into nature of wave interference pattern 4. OBJECTIVE DETECTION RESEARCH Preliminary pattern recognition techniques are being investigated to objectively isolate destructive interference patterns, including wave length, orientation, coherence, and gradient. Since mountain wave turbulence is directly connected to orography and wind field, the detection methodology can be limited to geographical areas prone to mountain wave water vapor pattern development. The end product would be a satellite-based pattern recognition methodology resulting in a degree-ofconfidence interest field of possible mountain wave turbulence.

Figure 5: Various pattern identification fields for objectively classifying areas in which mountain wave turbulence may occur. ACKNOWLEDGMENTS This research was funded by the NASA Applied Sciences Program through the NASA Advanced Satellite Aviation-weather Products (ASAP) project. 5. REFERENCES Ackerman, S. A., S. Bachmeier, K. Strabala, and M. Gunshor, 2005: A Satellite View of the Cold Air Outbreak of 13-14 January 2004. Wea. Forecasting, 20, 222-225. Baum, B. A., P. Yang, A. J. Heymsfield, S. Platnick, M. D. King, Y.-X. Hu, and S. T. Bedka, 2006: Bulk scattering properties for the remote sensing of ice clouds. Part II: Narrowband models. J. Appl. Meteor., 44, 1896-1911. Doyle, J. D. and D. R. Durran. 2002: The Dynamics of Mountain-Wave-Induced Rotors. Journal of the Atmospheric Sciences: 59, No. 2, pp. 186 201. Dudhia, J. 1989: Numerical Study of Convection Observed During the Winter Monsoon Experiment Using a Mesoscale Two-Dimensional Model. J. Atmos. Sci., 46, 3077-3107. Eichenbaum, H., 2000: Historical overview of turbulence accidents. MCR Federal, Inc. Report TR-7100/023-1. Heidinger, A. K., C. O Dell, R. Bennartz, and T. Greenwald, 2006: The succesive-order-of-interaction radiative transfer model. J. Appl. Meteor. Clim., 45, 1388-1402. Mellor, G. L., and T. Yamada, 1982: Development of a turbulence closure model for geophysical fluid problems. Rev. Geophys. Space Phys., 20, 851-875. Mlawer, E. J., S. J. Taubman, P. D. Brown, and M. J. Iacono, 1997: Radiative transfer for inhomogeneous atmospheres: RRTM, a validated correlated-k Model for the longwave. J. Geophys. Res., 102, 16663-16682.

Otkin, J. A., D. J. Posselt, E. R. Olson, H.-L. Huang, J. E. Davies, J. Li, and C. S. Velden, 2007: Mesoscale numerical weather prediction models used in support of infrared hyperspectral measurements simulation and product algorithm development. J. Atmospheric and Oceanic Tech., 24, 585-601. Schmit, T. J., Mathew M. Gunshor, W. Paul Menzel, James J. Gurka, J. Li, and Scott Bachmeier, 2005: Introducing the Next-generation Advanced Baseline Imager (ABI) on Geostationary Operational Environmental Satellites (GOES)-R, Bull. Amer. Meteorol. Soc. pp. 1079-1096. Skamarock, W. C., J. B. Klemp, J. Dudhia, D. O. Gill, D. M. Barker, W. Wang, and J. G. Powers, 2005: A description of the Advanced Research WRF Version 2. NCAR Tech. Note/TN-468+STR, 88 pp. Thompson, G., P. R. Field, W. D. Hall, and R. M. Rasmussen, 2006: A new bulk microphysical parameterization for WRF & MM5. Preprints, Seventh Weather Research and Forecasting User s Workshop, Boulder, CO, NCAR, CD-ROM. Uhlenbrock, N. L., K. M. Bedka, W. F. Feltz, and S. A. Ackerman, 2007: Mountain Wave Signatures in MODIS 6.7 um Imagery and Their Relation to Pilot Reports of Turbulence.. Wea. Forecasting, 22, 662-670.