|
|
| Community Newsletter |
   |
|
Table of Contents
You, Too, Can Write a Grant Proposalby Charles Murphy, Kean College of New JerseyColleges and universities with programs in meteorology and atmospheric science are facing a significant challenge. The modernization of the National Weather Service involves an impressive array of new instrumentation: NEXRAD, profilers, GOES NEXT, automated observing stations. These advanced systems provide enhanced capabilities, increased sampling rates, and in some cases, totally new types of information. Equally impressive is the quantity of data that they generate, requiring ever more powerful workstations for data processing and storage. Can the academic community keep up? Integrating these new observational methods and data sources into the training of meteorology students is a necessary goal for the community. Failure to incorporate them into our curriculum jeopardizes the quality of the education that we provide our students. Keeping up is essential, and keeping up is getting hard to do. The current pace of technological progress is so swift that new generations of systems appear about every three years. The need to keep our academic data systems up to date involves both software and hardware. In order to be useful in an academic setting, software must be capable of acquiring, managing, and analyzing meteorological data while, at the same time, supporting curricular needs. These are specialized needs for a small market not met by off-the-shelf applications or by software developed for operational meteorology. The hardware requirements in academia are dictated to a large extent by the software packages and in general require ever faster and more powerful systems. Unfortunately, particularly for smaller colleges and universities, resources for new hardware platforms are scarce or nonexistent--a situation, given the current economic conditions, that will likely continue for some time to come. The Unidata Program was designed to help ameliorate some of these problems. Funded by the National Science Foundation, Unidata has developed into the major resource for software in the academic atmospheric sciences community. As you all know, Unidata's freely available software packages--the LDM (for data acquisition and management) and the analysis packages that Unidata supports (GEMPAK, McIDAS, WXP, and Ynot)-- provide the academic community with a range of software capabilities. What is less well known, however, is that NSF's Unidata Program also provides some help with hardware acquisitions. Each year, NSF's Division of Atmospheric Sciences sets aside a portion of its Unidata funds for the equipment grants program. This program is designed specifically to help colleges and universities acquire the computer equipment needed to run Unidata software. Tips for Preparing an Equipment Grant ProposalThere is no magic formula for writing a winning proposal. However, there are a few pitfalls you can avoid when writing your first proposal, and I have some hints for how to strengthen what you propose.
What Happens Nextby Sally Bates, Unidata Program CenterYou've completed your proposal and mailed the original and nine copies to NSF (with all the attendant forms), and now you wait. To find out what happens to your proposal, I interviewed Clifford Jacobs, the Unidata program director at NSF. "The proposal process is not very mysterious," Jacobs told me. "Essentially, we depend on reviews by peers to make the selections, and the criteria they use in making their evaluations are those listed in the program announcement." The method Jacobs uses is a panel of reviewers drawn from the Unidata community. Jacobs himself selects the reviewers: "In creating the panel, I try to get good representation of the Unidata community. I look for reviewers who are technically competent in UNIX and OS/2, who are familiar with Unidata activities, and I ensure that both small and large schools are represented on the panel," he explained. Once the panel members are selected, Jacobs assigns them proposals depending on the reviewers' areas of expertise. "Each proposal is read by at least three people," Jacobs stated, "and I try to ensure that the reviewers are familiar with the environment in which the PI [principal investigator] operates. For example, a reviewer from a large research institution might not have sufficient experience to judge a proposal from a small teaching college, so I avoid situations like that wherever possible." Members of the review panel read and evaluate their assigned proposals individually. The entire panel then meets once. Their task during the meeting is to rank all the proposals. "The final ranking process is a result of open discussion," Jacobs said. "The discussion covers a lot of territory. For example, the reviewers might feel that one proposal would be greatly improved with some technical modifications; they might suggest avenues for cutting costs in another; some proposals might appear to ask for too much, others for too little. In the course of these discussions, the relative strengths and weaknesses of each proposal are agreed upon. By the end of the day, the panel usually has agreed on a relative ranking and recommended modifications and appropriate level of support for each proposal." Jacobs takes the ranking arrived at by the panel and, keeping in mind the funds allocated to the grants program and based on the panel discussions, begins negotiating with the principal investigators. Can a PI accomplish the same goals following a slightly different path? Would the PI consider adding another component? "The final award amount and specifics are the outcome of negotiations between myself and the PI," Jacobs remarked, "and these are strongly guided by comments from the review panel and the PI's present situation. The environment in which we operate is much too fluid; budgets can't be written in stone," he added. "Things change, a PI might not have considered a certain angle, some items can be dropped, others added. Settling on a budget takes time and sometimes several iterations." Depending on the outcomes of these negotiations and the level of funding available, NSF determines the final cutoff on grants. "I'm really concerned that faculty from smaller schools understand that applying for an equipment grant does not require that your institution have a strong research program," Jacobs stated. "In fact, smaller schools often have more significant proportions of minority, women, or handicapped students than the larger schools. Exposing these students to science and technology is of significant interest to NSF. If appropriate to your institution, this information should be made explicit in your proposal. "But there is another reason I'd like to urge small schools in particular to submit grant proposals," he continued. "Participation in the Unidata community is a two-way street, and experience has shown that small schools make valuable contributions to the Unidata community. Small schools should not underestimate their value to Unidata or their importance to NSF." NSF has several publications you should look at before submitting a proposal; in particular, see NSF publication NSF 92-89 Grants for Research and Education in Science and Engineering: An Application Guide, which is available on the NSF Gopher server (stis.nsf.gov) under NSF Publications, General Publications. You might also read Research Experiences for Undergraduates (Program Announcement; NSF 92-92) and Research for Undergraduate Institutions (NSF 89-60). These are available from: National Technical Information Service (NTIS) For information about the Unidata equipment grants, contact Clifford Jacobs, Division of Atmospheric Sciences, National Science Foundation. For information about Unidata and the equipment it supports, contact the Unidata Program Center: support@unidata.ucar.edu. Community's First Regional Workshop a Successby Russ Desouza, Millersville UniversityThis summer, the Unidata Users Committee realized one of its long-held goals: a regional workshop for Unidata users. The First Northeast Regional Workshop, hosted by SUNY-Brockport, August 17-19, was attended by 28 people from 16 different universities and colleges. Most participants were from the Northeastern part of the United States, although representatives from two Canadian sites and from as far away as North Carolina and Illinois were also present. The Unidata Program Center provided substantial support by underwriting the presence of Peggy Bruehl and Tom Yoksas from the UPC and Dan Vietor from Purdue; the UPC also helped with procuring a large-screen projection system. This workshop was organized and managed by Greg Byrd (SUNY-Brockport and a Users Committee member), Charlie Murphy (Kean College of New Jersey and a Users Committee member), Bill Fingerhut (Lyndon State College), and myself (Policy Committee member). Aside from one general meeting in Boulder (after a Users Committee meeting) and a conference call, all planning was done via email. The criteria we used in choosing a location were availability of hardware, access to the Internet, and institutional support. SUNY-Brockport met all these, particularly since they had just received NSF support for acquiring nine UNIX-based workstations to add to their existing network of OS/2-based PCs. This range of hardware allowed concurrent access to all Unidata-supported software, including McIDAS-OS2, McIDAS-X, GEMPAK, and WXP. Internet access made it possible to have hands-on demonstrations of Unidata's Gopher server, University of Michigan's Weather Underground, and North Carolina State's WXP GUI. The first day of the workshop consisted of presentations by selected members of the community on how they were using software; the presentations were followed by extended lab sessions where Bruehl, Yoksas, and Vietor held either question-and-answer sessions or tutorials. Several attendees brought their own PCs and gave their own demonstrations. The second day consisted of roundtable discussions focused on topics such as: How do you secure a systems administrator? How do you get the students to use the software? Is this technology making a difference? In the afternoon, the results of the discussions were reported back to the entire group; unfortunately, there were few "good" solutions. Again, there were lab sessions and demonstrations in the evening (in fact, both nights the labs were in use until after 10:00 p.m.). On the third day, workshop organizers gave participants an opportunity to provide feedback to Unidata in a session entitled "What do you want next from Unidata?" According to an exit interview, the workshop was a success. It provided users with the opportunity to meet, mingle, and share with each other in a relaxed and informative environment and to speak to the people most knowledgeable about Unidata. Most participants felt that 2-1/2 days was a good length for the workshop, but that the presentations took up a little too much of that time. Despite the rather substantial amount of hardware available, several complaints were voiced that there were not enough workstations. (Since all UNIX machines were in one room, we could not demonstrate or provide hands-on experience with both GEMPAK and WXP at the same time.) Everyone felt that such a workshop should be held at least every two years. The organizers feel that other areas of the country should seriously consider offering workshops, and we would gladly share our experience as organizers with anyone interested in hosting one. A copy of the workshop's program and a summary of the exit survey results can be viewed on the Unidata Gopher. The Move to UNIX: Hiring the Help You Needby Sally Bates, Undiata Program CenterThe move toward the UNIX operating system within the Unidata community has brought some sites face to face with the problem of formal "site administration." Maintaining and upgrading a series of networked UNIX workstations (or even one) is an enormous step in complexity above maintaining and upgrading one or two personal computers (PCs). The university community is not alone in being overwhelmed by this transition--many businesses are facing the same issues. Within the academic community the problem surfaces in two separate arenas: recognizing the need for devoting resources to systems administration and knowing how to hire someone to meet this need. Some departments may embark on the transition to UNIX assuming that no extra human resources are required. If faculty members and students have been keeping PCs running as part of their official routines, or unofficially in their "spare time," university administrators may assume that these resources will still suffice for the new systems. Unfortunately, UNIX workstations are not yet as user-friendly as personal computers, and maintaining and upgrading these systems requires a degree of experience beyond that of many faculty and students. We suggest that departments considering this transition plan or budget for at least some part-time help with systems administration. But what is a systems administrator and how do you hire one? Not knowing how to hire a systems administrator is such a pervasive problem that system administrators themselves have banded together to attempt to codify what it is that they do. The resulting system administrators group (which refers to itself as SAGE, a loose acronym for Systems Administrators Guild) is currently working on developing job description templates and checklists to help employers select systems administrators. As one SAGE discussion paper noted, since many employers are not familiar with what these jobs require, they "fall back into (mis)using the job classifications with which they are familiar" namely, for programmers, software engineers, or computer operators, all of which miss the mark. The templates and checklists have not yet been formally adopted by the SAGE organization, but members of the working group are fairly sure that there won't be substantial changes at this point. In any event, this will give you an idea of what kinds of knowledge and experience are needed in administering UNIX workstations. Not all Unidata sites will need full-time systems administrators. How much help you'll need depends on a number of factors, including how many workstations you're planning to have and whether your campus has a computing center that will help you maintain them. Since some Unidata sites have already made the transition from PCs to workstations (or to some combination of the two), the community is gaining a base of experience on what some of the hurdles can be; try posing your questions to: support@unidata.ucar.edu. Internet Data Distribution: The Field Test Beginsby Ben Domenico, Unidata Program ManagerAs you have read in earlier Newsletter articles, Unidata is moving toward an Internet Data Distribution (IDD) system that will eventually deliver raw data from a variety of sources in near real time. A prototype system has been running within the UCAR/NCAR community for many months, with some data also being delivered to a few external sites with special needs (e.g., with data-reception problems). Now the Unidata community is starting an important test phase to determine how well the LDM4 will perform in delivering data to universities via the Internet. The field test starts during September with a number of relay sites connecting their LDMs to other LDMs at data-source sites. In turn, the relay sites will send the data on to end-user sites (a.k.a. leaf nodes). Each of the sites will be instrumented with monitoring software that will measure the reliability of the IDD system and, wherever possible, the performance of the underlying network. Several sites will be capturing data from both the IDD system and the Alden satellite delivery system, so comparison between the two delivery systems will be possible. Initially the test will focus on delivery of the Unidata/Wisconsin channel and the Family of Services data, which consists of Domestic Data Plus (DD+), International Data Service (IDS), and the High-resolution Data Service (HDS), which is replacing the Numerical Products Service. The University of Illinois is already supplying the DD+ data to a number of relay sites and an LDM running at the University of Wisconsin-Madison Space Science and Engineering Center (SSEC) is acting as the source for the Unidata/Wisconsin data. The Merit Network people at the University of Michigan are working as a team with the Department of Atmospheric, Oceanic, and Space Sciences there to set up an LDM near the NSFnet backbone. This LDM will serve as a source for the HDS and IDS data. In addition to these data-source sites, the University of Washington, the University of Arizona, North Carolina State University, and the State University of New York at Albany have volunteered to act as relay nodes. A number of universities who were having reception problems participated as leaf-node sites in the prototype experiment. At the moment, it appears that there will be a need for additional leaf-node sites, so we would appreciate hearing from you if you would like to participate. Potential sites need to be aware of several important points. First of all, participation in the IDD test does not provide special data-access rights. The ultimate source of the Family of Services data for the initial test will still be Alden dishes at Illinois and Michigan. Consequently, any site that will capture and use the data along the distribution chain will have to have an agreement with Alden for that particular data service. The second important item is that we expect sites to be willing and able to install, maintain, and actively monitor new releases of the LDM and the monitoring software and to reconfigure the distribution system in a timely fashion. It will also be important for all sites to have adequate network bandwidth into their LDM systems. For example, one of the criteria for relay nodes is that they have at least T1 (approximately 1.5-Mbits/second) connections into the Internet. End-user sites should have at least 56-Kbits/second connections and the connections should not already be saturated. We also require that your system be set up with time synchronization (NTP) and Domain Name Service monitoring. Our plan is to set up the initial test and monitor its performance over a period of weeks. If problems with the LDM are uncovered, fixes will be installed until the system is running reasonably smoothly. Our objective is to do as much of this testing as possible by mid-December. We will then analyze the results of the test to that point, publish a report, and decide where to go from there. Once all the test sites get up and running, the topology for the distribution of the DD+ data will look something like the following: Similar topologies are being set up for the other Family of Services data streams and the Unidata/Wisconsin channel. While this experiment is progressing, we will be working on more limited tests of the LDM with experimental data sets from the National Meteorological Center (NMC) and Forecast Systems Laboratories (FSL). Depending on how the full-blown IDD test goes and the experience gained with the experimental data sets, we will begin working the experimental data into the full IDD test in a controlled fashion early next year. The overall objective is to have the IDD sufficiently tested by next summer so that we can begin deployment to the Unidata community. Profiler Updateby Linda Miller, External Programs Coordinator The Interdepartmental Radio Advisory Committee (IRAC) has tentatively decided to grant an extension of the renewal of the 404.37-MHz license for NOAA's Profiler Demonstration Network. In addition, a working group will be formed to produce an operational plan agreeable to the US Coast Guard, NWS, OAR, and the SARSAT Program Office. IRAC is gaining an improved understanding of the value of the profiler information for education and research, partly due to the letters of support collected from the university community. They were forwarded to Mr. Richard Barth, Department of Commerce, Office of Radio Frequency Management. The profiler network experienced some downtime in mid-July for system upgrades, maintenance, and also because of lightning hits at two sites. The Haskell profiler has been moved about 10 miles to Morris, Oklahoma, but the identifier HKL02 will continue to be used. Unidata McIDAS: Latest Release Offers Many Enhancementsby Tom Yoksas and Matt Hicks, Unidata Program CenterThe UPC recently released new and improved versions of its Unidata McIDAS-OS2 and Unidata McIDAS-X software. The changes you will see in these releases affect not only specific routines, but also the ease with which the OS/2 and UNIX users at a site can share data. And much of the documentation was rewritten--the Unidata McIDAS manuals now include complete documentation of the most important test-code routines. How We're Making It BetterAlthough a complete list of changes is beyond the scope of this article, there are some important enhancements that we would like to point out. (More details on the current capabilities of McIDAS on the Unidata Gopher under:>Information about Unidata Software Packages and Products >Unidata McIDAS Support
Unidata McIDAS continues to be Unidata's most widely distributed display-and-analysis package: Unidata McIDAS-OS2 is distributed to 83 institutions, and McIDAS-X is distributed to 54 (some sites use both packages). Although the two packages are distinct, only one license is required to obtain either or both of them. Friendlier in the FutureAnd McIDAS itself continues to evolve. SSEC is currently working on implementing what they refer to as a Distributed Data Environment (DDE) for McIDAS. The DDE concept is one of clients (applications) getting their data from servers that may be resident on the same machine as the client or on machines accessible over a TCP/IP network. This distributed paradigm contains the promise of transparent access by McIDAS applications to data sets in such diverse formats as netCDF, HDF, and GRIB. Naturally, Unidata will be most involved with the development of the netCDF-data-server portion of this new environment.SSEC is also rethinking the separate routines that the McIDAS package comprises. There are numerous instances where several McIDAS routines do the same types of tasks. For example, in the current Unidata releases, there are at least six routines that can be used to load satellite imagery. While each routine does something that the others don't, the burden of remembering and learning how to use these routines is daunting. SSEC plans to reduce the number of routines that compose McIDAS, while at the same time providing equivalent or enhanced functionality to the remaining routines. This is a massive effort; we should see the first benefits of this work within a year or two. Finally, for McIDAS-X users, Unidata and SSEC are experimenting with a package for building a graphical user interface (GUI). The package, Tcl/Tk from John Ousterhout of UC Berkeley, is an attractive candidate for building flexible and extensible GUIs on top of command-line driven, display-and-analysis packages like McIDAS. It may also be useful for providing GUIs for other packages distributed by Unidata (i.e., WXP and, to a lesser extent, GEMPAK). We will report the results of the Unidata experiments at the January AMS meeting in Nashville, TN. Ynot Version 3.0 Has New Look, Improved Usabilityby Mitch Baltuch, Unidata Program CenterUnidata has released version 3.0 of Ynot, its spatial data analysis and visualization program. What's new? What isn't?
This is a binary-only release for Sun Microsystem SparcStations running SunOS 4.1.x and for SGI workstations running IRIX 4.0.5. A binary distribution for Solaris 2.x will follow shortly. Ynot, version 3.0, is available to licensed Unidata members via FTP from the ynotdist account on ftp.unidata.ucar.edu. If you do not have a Ynot license from Unidata and you are interested in obtaining Ynot, contact support@unidata.ucar.edu for assistance in licensing. If you have any questions regarding the new release of Ynot, send email to the support address above. Anyone wishing to be added to t Ynot mailing list, ynot-users, should send a subscription request to ynot-users-request@unidata.ucar.edu. Unidata/Wisconsin Data Stream Updateby Tom Yoksas, Unidata Program CenterThe Unidata/Wisconsin data channel is an evolving stream of information that, although aimed primarily at Unidata McIDAS users, is used by a wide selection of Unidata sites. Several times over the past two years, Unidata and SSEC have updated the contents of this channel and changed the manner in which the data products were broadcast. The following is a synopsis of the changes made to the channel in the past two years and a discussion of changes planned for the near future. In December 1991, the contents of and broadcast method for what was then the only gridded data product in the stream were changed. This change replaced fields that could be calculated locally on McIDAS workstations with fields at the 250-mb level and some precipitation forecasts. The change in the broadcast method allows Unidata McIDAS sites to recover the files even when data reception problems cause loss of portions of the product. In November 1992, the method for broadcasting surface data was changed in order to allow access to up-to-date station information (e.g., lat, lon, u elev, ID, etc.) and eliminate stations that were no longer operational. Currently, the broadcast can contain up to 1750 stations from North America and Pacific US territories. The only downside to this change was an increase in storage requirements from approximately 2 to 3 MB per day. Also in November 1992, the format of the MDR (manually digitized radar) product was changed. The map outline that was previously "burned into" the image was removed and the presentation was altered to allow display of all 6 VIP levels from reporting radar sites. These changes made the MDR product more generally usable in McIDAS remapping and NORTELing (combined image display) applications, but required the user to learn a new method for displaying the image. At about the same time as the surface and MDR products were changed, Unidata and SSEC tested using the broadcast for delivering code changes to Unidata McIDAS-OS2 sites. This successful effort was made necessary by the great numbers of OS/2 users who had no access to Unidata by Internet FTP. Being able to send users the executable code needed to exploit data product changes without mailing diskettes has greatly diminished the impact that changes to the data stream would otherwise cause. Unidata will continue to deliver crucial changes to OS/2 users via this method. In June 1993, Unidata and SSEC added an LPATS lightning product to the data channel for a short evaluation period. The full LPATS data stream was provided to SSEC free of charge during the summer of 1993 so that SSEC could prepare an hourly summary display of cloud-to-ground lightning that would be usable by sites capable of displaying McIDAS AREA files. The end of the evaluation period is scheduled for September 30. By that time, you should have communicated your impression of the value of the product to the Unidata Users Committee, so that they can recommend continuation or cancellation of the product to the Unidata Policy Committee. The major issues in the decision are the scientific value of the graphical product and the cost to Unidata for access to the LPATS stream. Also added to the data stream in June of 1993 was an image product that is composed of remapped and combined GOES, METEOSAT, GMS, and polar-orbiter satellite data centered over Antarctica. These composite images, already being generated at SSEC for another project, were a fairly inexpensive addition to the data stream. As a compliment to the Antarctic composite images, a set of global, zero-hour (initialization) grids was also added to the stream. This product from NMC allows users to examine global patterns of temperature, geopotential, and winds at 10 vertical levels (1000, 850, 700, 500, 400, 300, 250, 200, 150, and 100 mb). Temperature, pressure, and geopotential information is available at the tropopause, while wind and pressure information is available at the level of maximum winds. In addition, relative humidity information is available at 5 vertical levels (850, 700, 500, 400, and 300 mb). The last field available is global mean sea-level pressure. Unidata continues in its efforts to secure three global products produced by the WetNet project of NASA: cloud liquid water, total precipitable water, and land-use classification. These products, produced from SSM/I data from DMSP satellites, should be added to the data stream in the very near future. As a precursor to the broadcast of these products, WetNet investigators will be releasing documentation on how the products are generated and what uses they are intended for. The current contents of the Unidata/Wisconsin data stream are listed below:
WXP Is Alive and WellCommunity responses indicate WXP is well used.by Mike Wright, Unidata Program CenterIn late April of this year, I sent out a questionnaire to licensed WXP sites asking whether they use the package and if so, how they use it. The percentage of sites responding was good and the results interesting. The SurveyThe questions were divided among three areas: questions about WXP use, questions about LDM use, and general questions about UNIX and systems administration experience. Some sites provided more than one response; multiple responses have been combined to keep the site count accurate.The ResultsWXPOf the 85 LDM-and-WXP-licensed sites, 47 responded (55% of licensed sites). It should be noted that WXP and LDM licenses have been given as a bundle (SDM license), so some sites that may only be interested in one of the two products are licensed for both. Of the 47 sites that replied,
Machines Being UsedThere were 36 replies indicating the types of machines used with WXP.
LDMI asked sites whether they use the LDM. Of the 85 licensed LDM sites, I had 47 replies.
System Administration and UNIXPart of the questionnaire asked about comfort with UNIX and UNIX workstation system administration. Systems AdministrationIn this area, I had a few non-LDM- and non-WXP-licensed sites reply in addition to the earlier 47. Of 52 replies,
UNIXThere were also 52 respondents to this question, again including five sites that were not licensed users of WXP or the LDM. Of the 52 replies, six sites said they have access to an on-site expert. When asked about the faculty/students' level of comfort with UNIX, the breakdown was as follows:
Email: An Important Source of Help and IdeasFeeling Listless? Unidata Email Lists Offer Something for Everyone.by Mike Wright, Unidata Program CenterUnidata maintains a number of email lists designed to facilitate direct contact between members of our user community. Each list relates to a specific package or service provided by Unidata. These lists are useful forums for discussions, data requests, and information about Unidata, Unidata systems, and other topics of interest to the community. To request to be added to or deleted from a list (except community), send email to list-request@unidata.ucar.edu (e.g., gembud-request@unidata.ucar.edu). If there are problems with email reaching you, contact support@unidata.ucar.edu so that we can try to find a solution. Archives of the lists are available through the Unidata Gopher. The archive is a database index that can be searched for topics, keywords, etc. Unidata General
Email Lists for Unidata Systems
Courtesy ListsUnidata does not support the products associated with these lists. We maintain the lists as a courtesy to the general network community.
A ReminderAs a final note, some sites may have experienced changes in their hardware or systems on campus during the summer. Such changes can cause your email address to change; we have been experiencing an increase in bounced messages, suggesting that this has, in fact, happened to some of our users. Please verify that your email address is still valid and notify Unidata support of any change.NSF Accepting Proposals for EquipmentThe Division of Atmospheric Sciences, National Science Foundation, is now accepting equipment proposals for Unidata-supported platforms from academic institutions engaged in teaching and research in the atmospheric and related sciences. (Assistance in defining these OS/2 and UNIX-based systems is available by contacting the Unidata Program Center.) Small schools and schools that have not previously submitted proposals are encouraged to apply (see "You, Too, Can Write a Grant Proposal"). It is NSF's intent to assist those institutions that presently do not have interactive meteorological computer capability to participate in the Unidata Program. (Certain variants of UNIX, including ULTRIX and AIX, are acceptable platforms for the Unidata software.) The target date for submitting proposals to be considered for funding in FY 1994 is December 31, 1993.The following criteria will be used to evaluate the proposals:
Unidata equipment awards will provide funds for hardware and the requisite software only. Significant cost sharing for the equipment by the institution is required. Site preparation and maintenance costs should be provided by the institution as well. Submission of Unidata ProposalsProposals should be clearly identified for consideration under Unidata. Ten copies should be sent to: Division of Atmospheric Sciences Proposals should follow the guidelines specified in Grants for Research and Education in Science and Engineering (NSF 92-89). In addition, proposals should describe:
Clifford Jacobs GEMPAK GIFs on Gopher!There is a new menu item called "Examples of graphics from GEMPAK" on the Unidata Gopher. We have compiled a group of sample GEMPAK graphics that are representative of the functionality (both graphical and meteorological) and flexibility of GEMPAK. You will find sample contour plots, cross sections, four-panel displays, meteograms, surface and upper-air maps, skew-T log-P charts, streamlines, vector plots, alternate maps, and time series. Periodically we will add new GEMPAK GIFs to this directory, so it's a good idea to check what's available from time to time. In particular, when contributions are made to the GEMPAK contrib directory, we will try to provide a sample GIF on Gopher. The item "Table of Contents" has a short description of each graphic, including the name of the data set and the GEMPAK program used to generate the graphic. You can find the GEMPAK-examples menu item in two places on the Unidata Gopher: >>Unidata Gopher Information Service and >>Unidata Gopher Information Service We've made these maps available to illustrate the products that universities can generate with their own Unidata systems. These maps are examples only and are not updated. NPS Subscribers: You Need LDM4Unidata sites who are subscribers to the Numerical Product Service (NPS) data stream and who are still using LDM3 need to update their LDM software immediately. As of October 1, 1993, Alden will be replacing the NPS with a stream call HDS (High-resolution Data Stream). Unidata has modified only the LDM4 software to ingest this data stream; we will not be modifying early versions of the LDM for this stream. While it has been possible to run LDM3 at NPS sites, most of you should already have been using LDM4 for some time now.If you currently subscribe to NPS and do not use the LDM4 as your ingesting software, please contact Unidata support immediately: LPATS Lightning Data StatusOn June 15, Unidata added hourly summary LPATS lightning data to the Unidata/Wisconsin data.Under an agreement with Alden Electonics, Inc., the University of Wisconsin-Madison Space Science and Engineering Center was provided with the LPATS data stream for the purpose of providing a lightning product to the Unidata community for a short evaluation. The product consisted of a graphic depicting cumulative hourly cloud-to-ground lightning stroke information on a national scale. The evaluation ended September 30. The resumption of the product will depend upon the Unidata Users Committee's recommendation and availability of funding. NGM Data Available on CD-ROMAt the Unidata Users Workshop, held in Boulder in June 1991, participants endorsed the following initiative: Resolved:
Data Support Section Included on the disk is the GRADS display software produced by the University of Maryland, as well as a netCDF conversion program for use with WXP software. New Book Helps with Connectivity DecisionsThe UPC recently received an announcement of a new book called Connecting to the Internet: An O'Reilly Buyer's Guide that, the announcement says, shows users "how to assess their needs so that they can determine the kind of Internet service that is best for them...the book provides checklists for evaluating [their] Internet needs, choosing a network provider, and deciding on dialup and dedicated line connections." Author Susan Estrada has been involved with the Internet since 1985, when she was also involved with planning for the original NSFnet. She also founded and built an Internet service provider. If you aren't yet connected to the Internet, this book might help answer some of your questions or even suggest some that you had not thought of asking.The book is available from: O'Reilly and Associates You can also order the book by sending email to gopher.ora.com. Research Floater DiscontinuedBeginning October 1, 1993, the UPC discontinued its subsidy of a second floater image designed to support field programs. The so-called "research floater" has been a useful capability (see the discussion of its use in "Unidata/Wisconsin Data Stream Update'. The decision to discontinue was difficult for the UPC, but necessary given the program's funding for the next fiscal year.Unidata Welcomes Robb KambicIn August, Robb Kambic joined Unidata as a software engineer II. He spent the last two years working in the data management branch of FSL, where he was involved in shipping archival data to the National Climate Data Center.Originally from Pittsburgh, Robb has been a Colorado resident for almost 20 years. He received his Masters degree in computer science from the University of Colorado in Boulder. He is now providing support for the LDM; this is helping to free up other staff resources to develop the IDD system. Robb enjoys gardening and frequently treats the UPC staff to the fruits (and vegetables) of his labors. FOS Catalog of Products Now On-LineWe now have the current NWS Family of Services (FOS) catalog (domestic data and public services) and have made it available to the community on-line. If you look in the top level of the Unidata Gopher, you will see the entry>> Datastream Product Catalogs We converted the file to ASCII and created a WAIS index of the contents. This allows you to do a search on an unfamiliar WMO header and get further information without having to page through the entire catalog. Is Your Software Up to Date?
Distributed versions of Unidata supported software packages and their release dates (as of Fall 1993). UPC Supported Platforms
Computers and operating systems supported for Unidata-distributed software (as of Fall 1993). |
|
Please send comments to info@unidata.ucar.edu |
|
   |
And what does this cost? That depends on where you're located and how much experience you're looking for. UNIXWorld publishes an annual salary chart (image at left).
Some of these products are backed up and archived on the machine unirec.wisconsin.edu. For details on accesssing this data see the Unidata McIDAS Users Guide or look on the table lists the institutions and principal investigators who were granted access to the research floater in the past year.
