Participating CIRA Scientists: Steve Albers, Chris Anderson, and Isidora Jankov
1. Long-term Research Objectives and Specific Plans to Achieve Them:
The Local Analysis and Prediction System (LAPS) integrates data from virtually every meteorological observation system into a very high-resolution gridded framework centered on a weather forecast office's domain of responsibility. Thus, the data from local mesonetworksof surface observing systems, Doppler radars, satellites, wind and temperature (RASS) profilers (404 and boundary-layer 915 MHz), radiometric profilers, as well as aircraft are incorporated every hour into a three-dimensional grid covering a 1040km by 1240km area. LAPS has analysis and prediction components. The prediction component is being configured using the RAMS, MM5, WRF, and ETA models. Any or all of these models, usually being initialized with LAPS analyses, are run to provide short-term forecasts. Ensemble forecasts using multiple models and initialization methods with verification are also produced.
LAPS is run in real-time at ESRL/GSD for a domain centered on the Denver, CO Weather Forecast Office. LAPS has also been ported to many locations (~150 worldwide), including universities such as Univ. of Oklahoma ("OLAPS") and Univ. of North Dakota. LAPS is running on-site at each National Weather Service Forecast Office (WFO) as an integral part of AWIPS. LAPS software is also being implemented at various U.S. government agencies such as Federal Highways Administration (MDSS), Range Standardization and Automation (RSA) at the U.S. Space Centers, National Ocean Service, U.S. Forest Service, and for international government weather bureaus such as China, Italy, Taiwan, Thailand, and Korea.
Research objectives related to LAPS continue to be the improvement and enhancement of the system in providing real-time, three-dimensional, local-scale analyses and short-range forecasts for domestic and foreign operational weather offices, facilities, and aviation and other field operations.
It is worth noting that LAPS and WRF improvements frequently have cross-cutting benefits that leverage towards many of the supported research projects (both within and external to NOAA) described later in this report. Funding has materialized for certain projects since the Statement of Work was formulated; LAPS improvements benefiting these projects are included in this section.
2. Research Accomplishments/Highlights:
LAPS Observational Datasets
Improvements were made in LAPS to analyze observations from new types of instruments and new data formats, thus expanding the envelope of meteorological data environments that we can operate in with our ever-growing set of users. These improvements are detailed below for surface and upper air observations.
Surface observation ingest quality control was improved. Duplicate station checks were made more versatile and otherwise streamlined. Default memory allocation was increased so more observations can be accommodated.
Upper Air Observations
The dropsonde ingest should now support the NIMBUS NetCDF format. This opens the way to do research on the usefulness of special sondes dropped into Pacific atmospheric rivers or Atlantic hurricanes to name two examples. Dropsonde support was also added to our case data retrieval script. Aircraft observation quality control was improved.
For cloud-drift winds, traditional IR data plus newer datasets for visible, water vapor, and sounder instruments/channels are being included. Support was added for the ASCII format we see with newer satellites such as GOES11 (and GOES12), as well as support for water vapor (WV) based measurements. The default list of satellites was also changed to GOES 10 and 11.
The handling of observation error was streamlined in the software. Temperature QC thresholds are now applied independently over land and water areas to help in the global analysis of sea surface or "ground" temperature. Surface verification bias statistics are now included in the data summary "what-got-in" log file.
We have worked towards more efficiency and other functional improvements for radar remapping and mosaicing. We now handle either /public or NCDC formats for archived wideband radar data. Error checking and debugging info was added for remap lookup table generation. The radar mosaicing program has improved warning messages about parameter consistency.
A new 1-D weighting array was added to improve looping efficiency with a 20% overall speedup noted in the wind analysis. A new K looping strategy was implemented that further speeds up the wind analysis somewhat. The maximum number of wind observations can now be set dynamically as a runtime parameter for better flexibility with machines of different memory capacity. Some other wind analysis mods were made to optimize the use of memory on the IBM. This helped pave the way to increase the vertical resolution of our operational runs.
The vertical radius of influence for surface wind observations was reduced to half that of other observations to help improve wind analysis accuracy in the boundary layer. Data structures were updated (for consistency with the wind analysis) in support of specifying temperature observation locations in between vertical grid levels.
For the cloud omega field, the single namelist parameter now switches all of CWB collaborator Adan Teng's changes in both the radar omega and the cloud omega routines. Therefore, we would run without Dr. Teng's flag for HMT (stratiform cases) and perhaps with the flag for convective or tropical storm cases. There may be ways to make the cloud omega code adjust in a more dynamic fashion between these regimes. In brief, perhaps the parabolic profile could be set to extend about halfway above the convective portion of a cloud into the overlying stratiform region.
Improvements were made to cloud analysis handling of radar echoes and cloud bases as a result of evaluating LAPS reruns for the Greensburg, KS tornado case day. Furthermore, the use of surface derived lifted condensation level (LCL) information is being considered as an aid in the decision tree to find cloud base height in the vicinity of radar echoes.
Ed Szoke collaborated with us to obtain cloud analysis output to perform comparisons with CloudSat cross-sections. These were presented at the CIRA Coffee Confab in November, 2007.
The precipitation analysis has a new gauge of only 1-hour precipitation analysis that is substituted for incremental precipitation areas where the radar analysis is missing. The background (e.g. GFS) precipitation is used as a first guess if it is available. This is being tested in real-time on the global LAPS run.
General Software Improvements & Portability
LAPS documentation, process logging, and CDL descriptions were improved. Software was streamlined and made more consistent helping to make it more understandable. Software was made more portable to different platforms and quality control checks were improved. LAPS was made more portable to work with the Intel 'ifort' compiler. LAPS build and ingest driver scripts were updated and improved. Scripts that report what data get into the analyses were improved. The scheduling script can now accept a command line argument that selects from a variety of combinations of runtime executables depending on the application.
Archive case rerunning scripts accessing mass store data were refined. The scripts for accessing archived observations for retrospective LAPS runs can now automatically untar the data files. Data purging scripts were improved.
We continue to run an hourly global LAPS analysis (GLAPS) on a 21km resolution domain. Real-time data and graphics are available on the LAPS website, an ESRL-wide web database, as well as Science On a Sphere (SOS). Improvements were made in the geometric re-projection of wind barbs when displaying GLAPS on SOS. The SOS display resolution was recently increased. Color tables and color bars were improved as well. GLAPS was featured in a recent town hall meeting given by the ESRL Director.
Fig. 1. Analyzed surface temperature and wind on the GLAPS 21-km global domain
A LAPS software build was set up on our newest supercomputer (wjet) for the purpose of testing and integrating STMAS 3-D software. For surface STMAS, MADIS QC flags were added for pressure variables (MSLP, altimeter, and station pressure). A QC check was also added for the MADIS subjective station reject list. In collaboration with NOAA colleagues, a journal article is being submitted.
Fig. 2. STMAS surface temperature and wind on the 5-km CONUS domain
The LAPS software distribution and the associated website continue to be maintained.
WWW LAPS Interface
A lag time parameter was added that will allow reading data up to 3 hours old without specifying a time when using the "on-the-fly" webpage. A test for missing model RH data was included to help get the best humidity field when plotting model forecasts. Image support was added for balanced divergence. Place holders were added for future high-resolution coastline (RANGS) data use.
The color scheme for surface observation plots now reflects the vertical location of the observations. Color tables were improved for Theta (e-sat) cross-sections and for precipitable water providing better global moisture depiction. Color bars were improved for high resolution plots. Wind barb density was increased for zoomed in upper air plots while satellite image plots now have improved zooming functionality. Changes were made so labels can be displayed with zoomed in fields.
Cloud and precipitation type cross-sections were optimized to work with the new high-resolution display strategy. Full station names can now be plotted for stations up to five characters. Updates from the "on-the-fly" page have been added to the LAPS software release.
Mesoscale NWP Model Initialization and Evaluation
Regarding software improvements to the WRF numerical model, Schultz microphysics has been added to the microphysical suite in the latest version 3 of WRF-ARW code.
3. Comparison of Objectives VS Actual Accomplishments for the Report Period:
Our accomplishments for this project compare favorably with the goals projected in the statement of work.