Dr. John W. (Jack) Glendening

30 April 2002   Copyright 2002

BLIPMAPs are a tool for planning cross-country soaring flights.  This article gives motivations for their use, a short summary of some parameters provided, how they are computed, and the future that might be anticipated for such forecasts.  It is an updated version of the article which appeared in the July 2002 issue of SOARING magazine, which is published by the Soaring Society of America.

Ten reasons to use a BLIPMAP
Easily interpreted, requiring little meteorological background
Relevant predictions, what a soaring pilot really wants to know
Supplies information not available on meteorological weather sites
Quickly evaluated using its graphical presentation map
Covers an entire region, including elevated terrain
Has been verified by actual pilot experience
Available on the evening prior to the flight
Uses the latest meteorological observations, updated through the night
To anticipate route legs having more difficult conditions and how the soaring day will begin and end
To avoid missing a great soaring day

Many pilots are interested in knowing the soaring conditions they will find upon arrival at the airfield - or whether it is even worth driving there.  Some already know that useful thermal predictions can be made by analyzing the vertical profiles of temperature called atmospheric soundings.  Such analyses have a long history, but producing them has traditionally required a large investment of time, special graph paper, and access to the required data.  More recently, with the advent of the Internet, such analyses can be performed more quickly due to the creation of automated tools ranging from Kevin Ford's venerable Thermal Index generator to the interactive GSD sounding webpage (all webpage locations are given in a sidebar below). 

Yet these have significant limitations:  they only indicate conditions at a single location, they don't provide many of the parameters that a soaring pilot needs, and they still require some investment of time.  These limitations matter to the cross-country pilot, who is interested in conditions over a wide area.  He might wish to make sounding analyses at multiple locations corresponding to possible routes, but this is usually not practical.  Furthermore, many sounding analyses are tied to specific sounding locations associated with airports, whereas the conditions of most interest may be far from an airport or in elevated terrain, at locations where the atmosphere differs significantly. 

This article briefly describes a solution to these forecast limitations:  using a computer to automatically analyze soundings over a wide region and calculating soaring prediction parameters which are displayed in a graphical format, akin to the weather maps familiar to TV watchers, and accessed via the Internet.  I call this product a "BLIPMAP", for "Boundary Layer Information Prediction MAP" (there is a separate product called a BLIPSPOT focusing on individual locations).  Here "Boundary Layer", abbreviated as "BL", refers to the scientific description for the turbulent atmospheric region mixed by thermals - so a climb to the top of a thermal nearly reaches the top of the BL, although the glider's sink rate does not allow a climb to the actual thermal top.  BLIPMAP predictions are currently available for all regions of the US.

BLIPMAP Predictions

The BLIPMAP program has been providing daily soaring predictions over the California-Nevada region since May 2001 and, based on pilot experience over the past thermal soaring season, has been found to be very useful, particularly to cross-country pilots.  What soaring parameters are provided?  Those are described in detail at the BLIPMAP webpage so below I will just discuss several of the more important ones.  Quite a few parameters are forecast but users do not need to understand or utilize them all - rather, they are there if needed.  Some parameters are applicable primarily in specialized circumstances, such as thunderstorm conditions, and some are trickier to use and understand than others.  I suggest starting with the two most important parameters, which answer the questions "how how high will I climb" and "what rates of climb can I expect", and considering others only after familiarity is gained or when one becomes motivated by encountering difficult conditions that could have been avoided. 

Thermal Heights

Since thermals create the BL, the thermal tops correspond to the "BL Top" and BLIPMAPs provide that parameter as a guide to the maximum thermalling altitude.  Over flat terrain the actual height reached by the glider will be somewhat lower, depending upon glider performance and pilot proficiency - but over complex terrain local small-scale topographic variations that are smoothed out of the model terrain grid can produce glider altitudes equaling the BLIPMAP predictions.  Fig.1 illustrates a BLIPMAP "BL Top" prediction at 21Z, or 1 PM local time, near the time of peak thermal activity.  The colors correspond to heights given in the color bar while black contours indicate the smoothed terrain elevations used by the model. 

Most apparent is that higher thermalling altitudes are found at higher terrain heights.  Of course one doesn't need a fancy weather prediction model to expect that relationship, but the model does provide specific predictions in elevated terrain and daily inspection of the BLIPMAPs reveals day-to-day changes and shifts in the locations of the highest thermalling heights.  For example, on this day thermalling altitudes in the coastal mountains north of San Francisco were predicted to be higher than those in the coastal mountains south of San Francisco - this is typically the case since elevations are lower in the southern range, but nevertheless on some days the higher predicted thermalling altitudes will be in the southern mountains. 

Fig.1:  "BL Top" BLIPMAP for the California-Nevada region, giving heights of thermal tops

Fig.1's date was chosen because conditions were then unusually strong for October, allowing two pilots to make record-setting flights from the coastal mountains west of California's Central Valley, north of San Francisco, to Minden Nevada, just east of Lake Tahoe at the knee of Nevada's western boundary.  A close-up of this region (Fig.2) and this can be compared to conditions predicted for the previous day (Fig.3), a day more typical of the fall season.  On October 1 thermal tops were about 5000 ft higher in the elevated terrain along which the flights occurred than on the previous day.  Conditions in the Central Valley itself were relatively weak on both days, with a maximum thermal height of roughly 3000 ft, and since flight altitudes were above that height, crossing the Central Valley required a long traversing glide.  These figures exemplify BLIPMAP's ability to make predictions at different locations, since conditions in the mountains differed greatly between the two days but conditions in the valley did not.
Fig.2:  Details of "BL Top" BLIPMAP for October 1 Fig.3:  Details of "BL Top" BLIPMAP for September 30

Thermal Strength

Predicting rate of climb is more complex, and here BLIPMAP provides what other sounding analyses cannot.  While empirically-obtained expressions exist which attempt to predict climb rate from a temperature sounding, their accuracy is greatly limited by the fact that they do not explicitly depend upon the physical mechanisms that actually drive the thermals.  Thermal strength fundamentally depends upon two factors:  the rate at which heat flows into the atmosphere from the ground and the depth of the BL.  The importance of the former should be intuitively clear - if the ground heats more, then thermals will tend to be stronger.  The BL depth is important because a thermal bubble accelerating vertically will reach a higher velocity if it can accelerate for a longer time.  A well-established theoretical formula relates these factors to thermal updraft strength and BLIPMAPs employ this formula for their prediction of W*, or "W star", the scientific abbreviation for the convective updraft scale which here represents the predicted average thermal vertical velocity at mid-BL.  To estimate the glider climb rate, its sink rate must be subtracted from W*.  It is gratifying to my sense of scientific correctness to find, based on pilot reports, that W* provides realistic predictions of glider climb rates without any need for empiricism.  This prediction is possible because surface heating is predicted by the model, but this heating is not available from traditional sounding analyses so they cannot predict W*. 

Fig.4:  "W*" BLIPMAP, predicting thermal updraft velocities

Fig.4 illustrates the W* predictions corresponding to the height predictions in Fig.1.  Strong updrafts were predicted over the elevated slopes on both sides of the northern Central Valley, along the "record setting" route - much stronger than for the previous day.  Looking elsewhere, one can see why soaring conditions in Nevada's Great Basin are legendary:  large updrafts are generally predicted over that region, with local clouds being responsible for a patch of low updraft velocities between Reno and Las Vegas and for a swath across southern California and northern Arizona (also, soaring over Utah's Great Salt Lake is not very promising!). 

Other Predictions

While the above parameters are those most often useful when predicting thermal lift, additional factors should often be considered.  As one example, if winds are strong then thermals can be torn apart by wind shear even when surface heating is strong.  BLIPMAPs provide an estimate of this effect by calculating a Buoyancy/Shear ratio, the wind shear effect being significant when this "B/S ratio" is 5 or less.  Since this depends upon the surface heating, it also is not predictable from a traditional sounding analysis.  The B/S ratio can often be neglected when making soaring forecasts since wind shear is often not a major factor - but ignoring it will lead to an incorrect forecast on those days when wind shear is a factor. 

All pilots know that thermal soaring is greatly affected by the presence or absence of clouds, but predictions are difficult because the effects are are not straightforward.  Clouds can improve thermal soaring, since their release of latent heat adds to the buoyancy aloft, but clouds can also limit thermal soaring through imposition of a cloud base below the thermal tops or by decreasing solar heating of the ground.  In addition, models find humidity difficult to accurately predict, so any cloud predictions will have much uncertainty.  Nevertheless the BLIPMAP program does attempt some BL cloud predictions, the more successful of which are its predictions of overdevelopment and of thunderstorm development.  More problematic is its prediction of smaller clouds, such as puffy BL cumulus or thin cloud layers aloft - this is a present forecast weakness. 

I will mention one final prediction because it provides something not usually forecast - the BL wind convergence.  Convergence is created by horizontal differences in wind speed which, since air cannot be greatly compressed, sqeezes air upward.  Pilots often refer to this phenomenon as a "shear line" which is permissible if one realizes this refers to horizontal wind shear but can be misleading to the unawary since generally "wind shear" refers to a vertical variation in wind speed and/or direction.  While this parameter's accuracy is limited by the model's coarse resolution of convergence features, pilot reports indicate that convergence predictions often do correspond to convergence lines that form in the northern end of California's Central Valley and can be utilized for lift, being particularly useful at the end of the day when thermal lift has died off.  While the observed locations are often not precisely as predicted, since the actual lines are very small-scale, the existence of such lines in predictions is correlated to their actually being found in the air. 

Fig.5:  "BL Convergence" BLIPMAP

Fig.5 illustrates strong convergence predicted at the northwest corner of California's Central Valley (depicted in white).  It also predicts a north-south convergence line near the center of the southern Central Valley (in red), a line which initially formed over the mountains to the west and moved eastward.  Many pilots are not accustomed to using convergence features, but having predictions may help pilots look for them.  However, such predictions can be "noisy" so must be viewed with caution - they are much more reliable when consistently found over several consecutive prediction times.

How Are These Predictions Obtained ?

BLIPMAP predictions are based upon computations by a numerical atmospheric weather model, the research version of the Rapid Update Cycle (RAP) model run by NOAA's Global Systems Division (GSD).  This model was chosen because it employs a forecast grid with a 20 km spacing - which over the CA-NV forecast domain depicted produces 3600 soundings!  Such fine-scale resolution is necessary to obtain adequate discrimination of the larger terrain features such as the Sierras and the coastal mountains, but many significant smaller-scale terrain features are still smoothed over - which should be remembered when evaluating model predictions.  Another reason for using this model is that its entire array of gridded predictions is available for analysis, including parameters such as the surface heating.  Since it is updated hourly using the latest observed data, its predictions contain a generous dose of reality - for example, clouds detected on satellite images are melded into its forecast predictions.  The model contains a sophisticated treatment of the ground surface, as is important for predicting thermal growth, and includes fine-scale specification of surface properties, assimilation of observed surface temperatures, and even subsoil transport of water and heat.  Additional benefits are that general meteorological variables predicted by the model are available at the GSD RAP forecast product webpage and that the GSD RAP sounding webpage can be used to interactively view the profiles produced by the model. 

The model predictions must be post-processed to provide parameters specifically of interest to soaring pilots.  A computer downloads the data for the entire US, selects regions of interest, computes the parameters, creates the maps, and posts them on the "drjack" website.  These computations are initiated the evening before and, if data is available, the first BLIPMAP is posted to the website prior to 0130Z, with updates being provided through the night.  For the CA-NV region predictions for multiple forecast times are available, which have been found useful in indicating when the thermal day will begin and how it will end.  Convergence features, for example, often occur in the later afternoon hours. 

The Future

Last year I produced BLIPMAPs only for the CA-NV region. where I fly.  But given the success of the BLIPMAP predictions over the last thermalling season, I recently expanded their coverage to include the rest of the US on an experimental basis (see sidebar for webpage locations).  Due to present bandwidth constraints, forecasts for these additional regions are for a single time only, either 18Z or 21Z, but that may change.  I expect these BLIPMAPs to be particularly useful in the western US where large-scale terrain is more significant, but also expect them to be useful in eastern regions.  Whether these products will be continued beyond the summer soaring season depends upon the amount of usage that occurs and upon pilot reports indicating that the BLIPMAPs are actually utilized for flight planning.

The ability to post-process high-resolution model output provides opportunities to forecast conditions which produce good soaring but are often overlooked.  For example, the convergence predictions illustrate how non-thermal phenomena can be forecast.  Other possibilities also exist.  For example, it might be possible to make cross-country ridge soaring forecasts by determining a small-scale ridge height and orientation for each 20 km grid box and then computing the perpendicular wind component at the ridge height over the model grid. 

At present, BLIPMAP forecasts are available only for a single day since the GSD RAP model does not produce longer term forecasts.  Yet computer power progressively increases and just recently forecasts with twice the resolution of the RAP BLIPMAPs have became available from the National Weather Service's ETA model on an experimental basis.  These forecasts extend out to 84 hours, although they are not updated as frequently as are the GSD RAP forecasts, and could potentially provide BLIPMAP forecats up to 3 days in advance!  However, a practical barrier to using this data for improved predictions is that very large bandwidth is needed to transfer the half-gigabyte model output files to a local computer for the required post-processing, since the costs involved are significant. 

The Fine Print

There are many points that I have glossed over in the interest of readability, more details being available at the BLIPMAP webpage.  One important caveat is that predictions of the BL Top and W* do not presently consider the release of buoyancy aloft by cloud formation processes, so will underpredict these factors when convective clouds occur in the BL.  I'll also mention that the variability of both the atmosphere and the numerical models means that any one prediction will always have uncertainty.  Relative predictions are therefore considered more reliable than absolute predictions:  for example, the prediction that today will be a better day than yesterday or that soaring conditions are better to the north of the field than to the south are more likely to be accurate than the numerical values predicted. 

The automated BLIPMAP performs the rote work required for a soaring forecast and provides numbers which are meant to be immediately useful to soaring pilots.  But still better forecasts are possible when pilots use their own knowledge, intelligence, and previous experience to evaluate these predictions and create a "value added" interpreted forecast, particularly when model weaknesses are taken into account. 

BLIPMAP predictions are intended to provide thermal soaring predictions not available elsewhere.  BLIPMAPs do not indicate when non-thermal soaring conditions such as wave or ridge lift occur.  Predictions of other, more traditional, meteorological factors which affect soaring conditions, such as precipitation, will need to be obtained from other sources, such as the GSD RAP forecast product webpage. 

In addition to the BLIPMAPs I also produce a daily thermal soaring forecast product designed for individual locations called the Thermal Index Prediction (TIP).  Being the first soaring prediction product I created, TIPs are based upon more traditional sounding analysis techniques, though extended by utilizing an estimate of the surface heating.  I consider BLIPMAPs to be more accurate, but TIPs remain useful because they provide advance predictions, of up to two days from the current day, which BLIPMAPs do not.  Although TIPs can be used by pilots to evaluate soaring conditions expected for the current day at an individual location, I believe they best serve as an "early warning" of future conditions and that current day prognostication is best accomplished by examining either the BLIPMAPs or their companion product called BLIPSPOTs, which provide BLIPMAP forecast parameters in text format at hourly intervals for individual locations.

BLIPMAPs are continuing to evolve and GSD forecast availability is subject to change, so it is possible that the BLIPMAPs available at the time of your reading this differ slightly from what is described above.  Reading the provided website documentation will hopefully resolve any questions.
Web Links

Kevin Ford's Thermal Index Report Generator:

GSD RAP interactive soundings:

GSD RAP forecast products:

DrJack's Thermal Index Prediction (TIP):

DrJack's BLIPMAPs:



Special thanks are due to to Milt Hare, who flies his ASH-25 from the Williams Soaring Center, for providing many useful pilot reports - he is more knowledgeable than I regarding the day-to-day use of BLIPMAPs in making soaring forecasts.  Thanks also go to Dr. Stan Benjamin of NOAA's GSD for providing access to the daily GSD RAP model output, to soaring pilot Brian Choate of Webbnet Internet Solutions for providing the "drjack" website access, and to meteorologists and soaring pilots Dennis Eckert and Walt Rogers for reviewing this manuscript. 

About the author:

Jack Glendening, sometimes irreverently called "Dr.Jack" when meteorology is under discussion, is a research meteorologist presently at the Naval Research Laboratory in Monterey, CA.  His specialty is Boundary Layer Meteorology, studying the turbulent atmospheric layer created by thermals and other surface-based eddies.  He has performed extremely high-resolution simulations (5-20 meter grid spacings) of thermals and roll vortices, using research models which explicitly resolve the individual turbulent eddies within them, and has published peer-reviewed papers analyzing coherent turbulent structures found within the Boundary Layer.  The soaring predictions described here owe their existence to the frustrations engendered by a medical condition which for awhile left him unable to fly but still able to pound on a computer keyboard.