FSL Forum, Nita Fullerton Editor
LINK: Intro PageLINK: Dr. Adrian MarroquinLINK: Cycled Snow VariablesLINK: GFE at NWSLINK: Defining Observed FieldsLINK: WRF Model in NWSLINK: Value of Profiler DataLINK: Evaluating ForecastsLINK: BriefsLINK: Recent Publications  
By Edward J. Szoke, John Brown, and Brent Shaw



From 13 May to 25 June 2002, FSL scientists were involved in the International H2O Project (IHOP), an extensive field study covering the Southern Plains and based in Oklahoma. More than 100 people participated in the campaign to determine primarily how to improve the characterization of the four-dimensional distribution of water vapor and its application to better understand and predict convection. The four main components of the program included quantitative precipitation forecasts (QPFs), convective initiation (CI), atmospheric boundary layer processes, and instrumentation. FSL ran experimental versions of local and national scale models during IHOP to assist with nowcasting and short-range forecasting and to determine whether such models could provide useful forecast and nowcast guidance for convective weather.

FSL's Rapid Update Cycle (RUC) model, a national scale model, and the Local Analysis and Prediction System (LAPS), a smaller scale model, were used during IHOP. LAPS was designed to run onsite at a National Weather Service Forecast Office (WFO), using local analyses that take advantage of locally available data. LAPS is currently running in AWIPS at WFOs on an hourly cycle with a 10-km grid spacing. LAPS was run at a 12- and 4-km horizontal grid resolution and used to initialize some of the models that FSL ran during IHOP.

One goal of a short-range model is to provide better prediction of precipitation without a spin-up period. To aid in this goal a "Hot Start" scheme was developed using the LAPS cloud analysis to prescribe a vertical velocity profile where sufficiently deep clouds are present at initialization time. The three-dimensional dynamical relationship between mass and momentum is adjusted by the LAPS balance algorithm to force consistency with the diagnosed cloud vertical motions and allow for a smooth model start. During IHOP, a 12-km horizontal resolution MM5 hotstart initialized with LAPS was run, with a nested 4-km version covering the IHOP experimental domain (Figure 1). LAPS also was used to initialize a similar 12-km setup for the Weather Research and Forecast (WRF) model. In addition to these models initialized with LAPS, FSL ran a 10-km version of the RUC model. The RUC model employs a three-dimensional variational (3DVAR) analysis for the mass fields, and initial RUC hydrometeor fields are adjusted to correspond to base scan reflectivity patterns at the initial time, but without any modification of the initial vertical velocity field (in contrast to the hotstart method). The models run by FSL for IHOP are summarized in Table 1. (These experimental model runs are archived by UCAR at


Figure 1. The 12-km and inner 14-km IHOP domains for LAPS MM5 and WRF runs (points every 12 km).


Table 1. FSL Real-Time Models in IHOP (*except LAPS-WRF, available after IHOP)
x Hour
Convective Scheme
RUC/MM5 MixedPhase

Evaluation of Model Performance

One of our goals during IHOP was to compile a fairly extensive subjective evaluation of the various models in real time. Objective model evaluation was done within FSL (see Brent Shaw's article WRF). In order to perform subjective evaluation, an online evaluation form was designed that allowed the forecaster/nowcaster to document what the model was forecasting, the relationship of various forcing features to the subsequent convection forecast by the model, and the forecaster's confidence in the forecast. The forecasters also made many freeform comments that give insight into how the models performed with the various short-range forecast problems during IHOP. A summary of these comments follows:

  • The MM5 models using LAPS for the initial state did an excellent job of initializing ongoing convection, but often this convection was lost in the first hour of simulation. Adjustments were made to the Hot Start scheme for a set of post-IHOP reruns of both MM5 and the WRF model, and our preliminary evaluation of some of these reruns indicates some improvement with this problem. The most easily "lost" convection involved elevated storms (nonsurface-based convection), while very strong individual storms and lines were much better retained from the initialization.
  • Outflows tended to be easily produced from convection in the MM5 model, especially so in the 4-km run, whereas the RUC tended to be more conservative in producing outflows but was able to do so.
  • The most difficult storms to forecast were elevated convection, which usually formed in the very early morning (presunrise) hours and could persist for up to 6 hours after sunrise. Coincidentally, this type of convection is also among the most challenging for forecasters, as it can occur without any obvious surface forcing feature present. Though seldom producing severe storms (at least during IHOP), elevated convective events were often of the "surprise" category. There were often indications in the model of possible activity, for example in the form of midlevel echo but without precipitation reaching the surface, so an underforecasting of the convection. Convection associated with a warm front (on the cool side of the warm front) also tended to be an area where the models were deficient. This type of convection often was not surface-based, sharing that characteristic with the elevated storms noted above.
  • Some of the forecasts of convective initiation along drylines were quite good. For a few cases the model beat the forecasters, particularly when temperatures both at the surface and aloft were quite hot. In these cases, forecasters overestimated the time it would take to break the cap, while the model more correctly forecast convective initiation earlier.
  • Other good forecasts occurred with well-defined surface-based forcing features, such as cold fronts.
  • As noted earlier, there was some skill in the model's ability to forecast storm type and evolution, with several events during IHOP that featured upscale growth into organized lines that often accelerated much faster than indicated in the precipitation fields from conventional models (for example, the Eta model).

Selected Case Studies From IHOP

Two cases are examined using both model runs during IHOP and the reruns that occurred after IHOP. The reruns provided a series of model runs from the WRF model, which was actually run during IHOP but not able to display in real time. As a result of the real-time experiences during IHOP, we decided to apply improvements to the hotstart method to the reruns. A significant improvement involved removal of a warm bias that existed in the LAPS initialization, and was likely responsible for overprediction of convective precipitation during IHOP from the MM5.

2 June 2002: Dryline Case – On this day the western half of the IHOP domain was dominated by very hot temperatures, reaching the low 100s (oF) during the afternoon. A well-defined dryline was not seen initially, but was sharpened in the early afternoon, as shown by the LAPS analysis of wind and dewpoint along with low-level reflectivity in Figure 2. This sharpening first appeared as a surge of westerly surface flow that emerged out of eastern Colorado that then pushed into Kansas. On this day, the IHOP forecasters predicted that a dryline would become better defined during the afternoon (somewhat later than what occurred) but they thought convective initiation along it would occur fairly late in the afternoon as they waited for the dryline to sharpen and temperatures to break the significant cap that was in place. As it turned out, the presence of the very hot surface temperatures and a somewhat stronger and earlier dryline push than expected allowed the cap to be broken and convective initiation to occur over 2 hours ahead of the IHOP forecasters' prediction. It was noted in IHOP that the MM5 model did a good job indicating this convection earlier than expected, particularly the run initialized at 1500 UTC. Some of the runs from that day are examined and the somewhat different forecasts are contrasted for this relatively "tricky" case. With the expense of some of the resources in IHOP, in particular the aircraft, timing of convective initiation was a critical forecast issue. In this case, even a forecast error of 2 hours for convective initiation from a midmorning forecast was critical and resulted in an aborted mission, since convection was well underway before the aircraft (leaving Norman, Oklahoma) could reach the dryline target.


Figure 2. LAPS analysis of surface wind, dewpoint, and low-level reflectivity at 2100 UTC 2 June. The western portion of Kansas is in the center.


We first examine some forecasts initialized at 1200 UTC since reruns of both MM5 and WRF are available at this time. A forecast from the MM5 run initialized at 1200 UTC and valid at 2100 UTC is shown in Figure 3. In this and subsequent figures, when no contours are present, the model is forecasting reflectivity aloft with no precipitation reaching the surface. For the most part, the reflectivity values for the image in Figure 3 are 30 dBZ or less. Some stronger cells are forecast for which these surface reflectivity contours are depicted (e.g., in extreme northeastern Nebraska and along the Iowa/Illinois border). The insert shows a composite low-level reflectivity image overlaid with a visible satellite image (white areas) over a region centered on western Kansas. This indicates that storms were really producing rain by 2100 UTC, with maximum reflectivities exceeding 50 dBZ. Thus, the MM5 12-km run was forecasting high-based convection that would not produce precipitation, so it correctly indicated that storms would be produced along the dryline, but was underforecasting development. For comparison, the post-IHOP rerun of the MM5 12-km model for the same time is shown in Figure 4, and for the WRF 12-km model in Figure 5. The MM5 rerun is very similar to the original MM5 12-km run during IHOP, and the WRF forecast from the 1200 UTC run valid at the same time is also very similar. All the runs indicate convective development with reasonable timing but only forecast virga-producing storms. The forecasts from these same runs valid 3 hours later at 0000 UTC on 3 June (WRF is shown in Figure 6) were very similar to the 2100 UTC forecasts, in that there continued to be no indication of storms that would produce precipitation. In reality, the line of broken storms advanced slowly to the east, and by 0000 UTC extended all the way from north-central Kansas south-southwest to the far western Texas Panhandle.


Figure 3. MM5 12-km IHOP run of 9-hour forecast valid at 2100 ITC 2 June 2002. Image is a composite reflectivity, with contours indicating model surface reflectivity. Inset shows a composite low-level radar image at this time over western Kansas.

Figrue 4. Same as Figure 3 except for a 9-hour forecast from the MM5 12-km rerun valid at 2100 UTC.


Figure 5. Same as Figure 3 except for a 9-hour forecast from the WRF 12-km rerun valid at 2100 UTC.


Figure 6. As in Figure 5 except a 12-hour forecast from WRF 12-km rerun valid 0000 UTC on 3 June. The insert depicts the actual radar reflectivity at this time.

The MM5 4-km run initialized at 1200 UTC produced higher values of composite reflectivity, but still no surface reflectivity (and therefore no precipitation reaching the ground). On the other hand, the MM5 4-km run initialized 3 hours later at 1500 UTC did produce well-defined surface storms, although slower than what actually occurred and by 0000 UTC with a line of storms not far enough east (Figure 7). The MM5 12-km run for this same time made during IHOP (Figure 8) was not as bodacious with storm development as the 4-km run, but it did indicate a surface echo in the Oklahoma Panhandle, and another much farther north along the line. The actual reflectivity at 0000 UTC is shown in Figure 9. Though the forecasts (especially the 4-km runs) that were initialized at 1500 UTC were better than the 1200 UTC ones, for some reason this improving trend did not continue with the 1800 UTC runs. The 6-hour forecasts from the various models (MM5 4-km, MM5 12-km IHOP run, MM5 12-km rerun, and WRF 12-km rerun) are shown in Figure10 (a–d). The MM5 4-km run still correctly produces a surface echo, but there is less an of echo than was in the forecast from the 1500 UTC run, and the line of echoes is even farther west. The MM5 12-km runs (Figures 10b and c) are quite similar to each other, and neither predicts any surface echo. Recall that the 1500 UTC 12-km IHOP run (Figure 8) actually did predict an echo reaching the surface by 0000 UTC, so the forecast initialized 3 hours later is not as good, similar to the behavior of the 4-km run. Note that the WRF 12-km rerun (Figure 10d) is actually a little drier than the MM5 12-km rerun and similar to the WRF 12-km rerun from 1200 UTC.


Figure 7. MM5 4-km 1500 UTC IHOP run 9-hour reflectivity forecast (as in Figure 3) valid 0000 UTC 3 June.

Figure 8. As in Figure 7 except for the MM5 12-km 1500 UTC IHOP run 9-hour reflectivity forecast valid 0000 UTC 3 June.

Figurre 9. Observed low-level radar reflectivity at 0000 UTC 3 June.




Figure 10. a, above left) MM5 4-km 1800 UTC IHOP run 6-hour reflectivity forecast (as in Figure 7) valid 0000 UTC 3 June. b, above right) As in 10a except for MM5 12-km run. c, below left) As in 10a except for MM5 12-km rerun. d, below right) As in 10a except for WRF 12-km rerun.

In summary, for this case the models showed a dryline moving into western Kansas more or less as occurred. The main message from the model runs is that convection would be initiated by the dryline, but the storms would be weak without any surface rain, typical of high-based, mostly dry convection that might occur on such a hot day with marginal moisture. The 4-km MM5 runs accurately indicated that more substantial storms could occur that would produce surface precipitation, and in particular the rerun initialized at 1500 UTC was the closest to reality. In real time during IHOP forecasters saw the 4-km run but doubted that such echoes could develop with the environment that appeared to be in place, opting for a forecast of later and weaker storm development than indicated by the model (or than actually occurred). We are not certain at this stage of our research why the runs initialized at 1800 UTC did not perform as well as those initialized 3 hours earlier.

15 June 2002: Complex Case – All forms of convection were evident in the 15 June 2002 IHOP case. This single day began with early morning elevated storms, evolving into a supercell storm that eventually produced a tornado, moving to upscale growth of strong cells into an organized line that bowed and accelerated southward out of the domain. The actual focus of IHOP operations on this day concerned where convective initiation would occur along a dryline feature, which, like the rest of this case, was a fairly complex feature with a double structure.

In examining the performance of the various models, we will concentrate here on the convective types that occurred rather than specifics of the dryline. Widespread development of elevated convection over the Texas Panhandle between 0800–0900 UTC was a forecast issue for an early IHOP flight to investigate a low-level jet. The storms eventually exceeded reflectivities of 50 dBZ at low levels, and persisted well into the daytime hours (until around 1600 UTC). Fortunately, as forecast by the Storm Prediction Center, the storms did dissipate, allowing the rest of the day to become a very interesting IHOP case. However, the development of the storms was not anticipated by IHOP forecasters, and as is typical in cases of elevated nighttime convection, was a difficult forecast problem. A radar overview of the storms is presented in Figure 11. The model simulations from the 0600 UTC runs are depicted in Figure 12.


Figure 11. Low-level reflectivity composites. Two strong surface-based storms are shown at the south end of the area (top left and right, 0900 UTC), as elevated convection develops to the north over the Texas Panhandle (bottom 1500 UTC).

Figure 12. Model simulations from the 0600 UTC 15 June runs of the MM5 12- km (top row), MM5 4-km (middle), and WRF 12-km (bottom) for the elevated convection in the Texas Panhandle. Reflectivity is displayed as before for the MM5 models, while the WRF contours show a composite and image surface reflectivity.

At 0600 UTC (not shown) a couple of surface-based storms formed at the southern end of the Texas Panhandle, developing earlier in New Mexico and moving eastward. These storms (shown in Figure 11) continue to move southward with time. It is interesting that the Hot Start method nicely initialized the LAPS runs (MM5 and WRF) correctly with an echo at 0600 UTC, but the echo was mostly lost within the first hour. Although loss of the initial echo was a problem for other 0900 UTC cases during IHOP and is the subject of ongoing work with the Hot Start procedure, it appears particularly acute in situations like this, where nighttime surface conditions would not support surface-based storms. The area of elevated storms developed north of the longer lived echoes, and remained more or less in the same area, peaking around 1200 UTC and diminishing rapidly after 1600 UTC. All three of the model simulations shown in Figure 12 that were initialized at 0600 UTC do develop some mid-level echo, but for the most part, it is not in the Texas Panhandle specifically, and is certainly slow to develop (for example, note the lack of any echoes at 0900 UTC). There are contrasting forecasts among the three models, though, and apparent attempts at forecasting the elevated storms. The MM5 12-km run shows some significant surface echoes, though the southernmost storm moved out of New Mexico apparently in the same manner as the earlier strong storms. The more northern cells develop in southeast Colorado and may well be the models' forecast of elevated type storms. The MM5 4- km run also shows these more northern storms extending in a broken line from northwest to southeast. It is not certain what forced this line, but it could be more of a development along a warm frontal type boundary that was actually positioned somewhat farther east and north. The WRF 12-km model appears to come closest to positioning the mid-level echo correctly in the Texas Panhandle, although it underpredicts the strength of the storms with only limited surface reflectivity. (For the WRF model the image shows surface reflectivity of 20 dBZ and above, with the white contours showing values below 20 dBZ.)

The next feature of interest is whether the models could predict a long-lived echo that formed from a group of small cells east of Denver (near Limon) at 1500 UTC that gradually grew as they moved east, with more or less one main storm by 1800 UTC that then turned to the right as it moved into western Kansas (see the reflectivity images in Figure 13). This storm became supercellular but did not produce a tornado until 2100 UTC after it intersected a pre-existing north-south dryline (the IHOP focus) and then moved southward along it. Although the resolution of the model runs at 12 km (and to a lesser extent, 4 km) would appear to be too coarse to successfully forecast an individual storm, some surprisingly excellent forecasts have been made with a 10-km version of the MM5, so we were interested to examine the models for this event.

For this case the storm formed beyond the domain of the MM5 4-km model, and shown in Figure 13 are forecasts from the MM5 and WRF 12-km models rerun after IHOP using some improvements to the hotstart method. The runs are both initialized at 0600 UTC so the forecasts shown begin 15 hours into the run. Both runs seem to develop storms by 1500 UTC in the correct location in eastern Colorado, strengthening the storm and moving it at about the right speed to near the Colorado/Kansas border by 1800 UTC. The model then continues to strengthen the echo and turns it to the right, in pretty good agreement with the actual behavior. The MM5 rerun tends to have a more concentrated and stronger surface echo then the WRF, but both have fairly impressive forecasts considering the one valid at 2100 UTC is a 21-hour forecast. The IHOP MM5 12-km run (which extended to 12 hours) from 0600 UTC was not as successful as the MM5 rerun, but had a weaker echo in about the same location. For unknown reasons, the forecasts from the IHOP runs initialized for 0900, 1200, and 1500 UTC were not very good in forecasting this long-lived system. Even the IHOP MM5 12-km 1800 UTC run, with the storm already in progress, did not have a good forecast as it tended to lose the initialized storm for the most part by 1 hour into the forecast.


Figure 13. Comparison of the 12-km resolution MM5 and WRF model reruns initialized at 0600 UTC on 15 June with composite low-level reflectivity (top row). Model images and white contours are surface reflectivity (image is 20 dBZ and greater), with dimmer contours composite reflectivity.


In summary, results are mixed for this aspect of 15 June; on one hand the 0600 UTC runs indicate some fairly impressive predictability, but inability to repeat this predictability for the IHOP runs closer to the event is curious. We hope to compare MM5 and WRF reruns from 1200 UTC to the 0600 UTC reruns to see if the storm was still forecast for these later model runs.

The final portion of the 15 June 2002 case that is examined involves the organization of three areas of convection into a squall line by 0000 UTC on 16 June that then accelerates southward out of the IHOP domain by 0600 UTC. A radar overview of this evolution is shown in Figure 14. At 1800 UTC the organized storm discussed earlier is just crossing into western Kansas, and at 2100 UTC is at the western end of the line segment located in southwestern Kansas. By 0000 UTC a line extends from northern Oklahoma west-southwest into the Texas Panhandle, with the eastern portion developed from the area of cells that moved south out of Nebraska. After 0000 UTC, the line organizes and accelerates as it bows over western to south-central Oklahoma. The model forecasts from the 0600 UTC reruns of the MM5 12-km are shown in Figure 15 and the WRF 12-km forecasts are shown in Figure 16. The MM5 organizes a group of cells in Kansas at 2100 UTC into a line segment close to where it is actually found at 0000 UTC, then accelerates the line southward. Although the actual line moves faster than the forecast, the track is similar and the model forecast includes a bowing line as observed. Considering that the later period of this forecast is an 18–24 hour forecast, it is fairly impressive, with the model doing a very good job of forecasting the organization into an accelerating, bowing line. This line forms in about the right place even though the MM5 essentially misses all of the storms that around 0900 UTC began to form in a north-northwest to south-southeast line from central Kansas to west-central Nebraska. These storms continued to expand in about the same place, and appear to have been, at least initially, somewhat elevated type storms that developed just ahead of a warm frontal boundary. The earlier times of this MM5 forecast never included anything but some mid-level reflectivity, and even then it was west of where the line actually occurred. The difficulty in handling convection that may not have been surface-based or forced by a distinct low-level boundary is similar to the problems that all of the models had with the elevated convection in the Texas Panhandle discussed earlier.


Figure 14. Series of composite low-level reflectivity images showing the organization of cells into a fast-moving squall line on 15 June 2002. Images are every 3 h beginning at 1800 UTC, with the reflectivity scale as in Fig. 9.

Figure 15a

Figure 15b

Figure 15c

Figure 15d

Figure 15e

(Figure 15a, top; 15b, second from top; 15c, middle, 15d, second from bottom; 15e, bottom. Times correspond to the radar images.)
A similar set of forecasts from the WRF 12-km rerun is shown in Figure 16. The WRF forecasts have a little more surface reflectivity than the MM5 forecasts, but like the MM5 run also misses the warm frontal convection discussed earlier. By 0000 UTC (compare Figures 15c and 16c), the forecast for the developing line segment is similar to the MM5 and about in the same location, though the WRF continues to produce far more echo presence (and hence precipitation), with a large diffuse surface echo extending to the east-northeast of the line. This is a better forecast than the MM5 for the extent of echo if one compares it to the observed echo at 0000 UTC that shows an extensive area of moderate-strength surface echo in about the same position as the WRF forecast. After 0000 UTC the forecast is not quite as good as the MM5 run with a smaller line that is located a bit too far east. However, like the MM5 rerun, it is impressive that the WRF model was able to predict the upscale growth to a bowing line in about the right place and about when it occurred.

Figure 16 a–e. As in Figure 15 but for the WRF 12-km rerun initialized at 0600 UTC on 15 June.

The MM5 runs during IHOP extended out only to 12 hours, compared to 24 hours for the reruns, so for comparison, runs beginning at 1500 UTC for the 12-km MM5 and at 1800 UTC for the 4-km MM5 will be shown. The 1500 UTC MM5 12-km IHOP run is shown in Figure 17. Note how the initialization from LAPS nicely captures the ongoing convection at 1500 UTC (Figure 17a), although the storms are quickly lost, mostly in the first hour. This occurred at times with MM5 during IHOP, as noted earlier, and for this case may have been exaggerated somewhat because much of this convection near an apparent warm front may not have been surface-based. Unfortunately there is such a loss of echo that by the 3-hour forecast (valid at 1800 UTC), a composite echo is forecast but none is predicted to reach the surface. Right after 1800 UTC, however, the mid-level echo shown entering northwest Kansas in the 1800 UTC forecast strengthens rapidly, then expands to form the line segment shown in the forecast valid at 2100 UTC. This line segment then moves southward and strengthens and expands, bowing somewhat at 0000 UTC but then becoming more of a straight line by 0200 UTC. The line in this forecast does not accelerate as fast as in the 0600 UTC forecasts from the WRF and MM5 reruns shown earlier.


Figure 17 a–e. As in Figure 15 except for the MM5 12-km forecast made during IHOP and initialized at 1500 UTC. Note that the 12-hour forecast was not available, so the 11-hour forecast valid at 0200 UTC is in Figure 17e.
A comparison of the MM5 IHOP 12-km and 4-km runs initialized at 1800 UTC is shown in Figure 18. The two runs did capture the evolution to a line that accelerated and bowed with time. Organization into a stronger system with more bowing happens in the 4-km run ahead of the 12 km run, with the 4 km likely able to capture storm outflows better with its higher resolution. The 4-km run by 0500 UTC is still slower than reality with the position of the line, but not by much. A similar set of model runs for the 2100 UTC initialization, when the convection was beginning to organize more, is shown in Figure 19. Although there is some loss of the system in the first hour of the forecast after a good job of initialization (Figures 19a and 19e), more is retained than in some of the other runs because of the presence of a stronger echo at 2100 UTC. The MM5 12-km run is similar to the run initialized at 1800 UTC, although it develops a line sooner (by 0000 UTC) and ends up with a line position by 0600 UTC closer to reality and similar to the 0600 UTC MM5 rerun shown earlier (Figure 15e). Interestingly, the MM5 4-km run from 2100 UTC does not organize the convection into a line as fast as it did with the 1800 UTC run, and even at 0300 UTC has more of a broken line (not as good a forecast). By 0600 UTC it organizes the line more and accelerates it south of the 4- km domain, similar to the movement that was observed.


b) g)
c) h)
d) i)
e) j)

Figure 18 a–j, right. Comparison of MM5 4-km (left column) and MM5 12-km (right) forecasts from the 1800 UTC runs. Reflectivity is shown, as in previous figures. Note that the 11-hour forecast, not the 12-hour, is shown in 18e and 18j.
a) e)
b) f)
c) g)
d) h)

Figure 19 a–h (from top left to bottom right). As in Figure 18 except for the runs initialized at 2100 UTC 15 June 2002.

It is apparent that all the models were able to predict the upscale growth and organization of the convection into a line with good location and timing for the most part. There was good agreement between the different models and usually between the different initialization times. A consensus forecast from an ensemble viewpoint of the various runs would have been a good one. The dprog/dt method did not necessarily verify as well, however, especially for the MM5 4-km runs, with the 2100 UTC run not as good a forecast as earlier runs.

Summary and Future Work

The model forecasts examined here from two IHOP days encompass a variety of forecast problems typically encountered especially east of the Rocky Mountains, including convective initiation along a dryline, prediction of supercells, upscale growth and organization of storms into a squall line, and the very tricky forecast of overnight elevated convection. The forecasts presented well represent the behavior of the models used during the IHOP period, and indicate that there is potential for such models to offer forecast guidance that can be valuable to forecasters trying to predict convection. The model was most successful when the convective initiation was forced by a well-defined surface boundary, as in the 2 June 2002 case, and had the most difficult time with storms forced by more subtle boundaries (like the warm front on 15 June) or by no apparent surface boundary, like the elevated storms in the Texas Panhandle on 15 June. Some of the forecasts of supercell formation and movement, as well as upscale growth that occurred on 15 June were impressive, and there was even skill shown for such developments beyond the typical 6–12-hour limit that one might suspect for convective forecasts.

During IHOP, the RUC and MM5 special model runs by FSL were extensively used to help make short-range forecasts, with the models displayed on the FSL FX-Net workstation. Partial examination of an extensive real-time questionnaire completed by the forecasters for as many model runs as possible during IHOP has yielded constructive insight into various model issues that occurred, as well as how much the forecasters trusted some of the predictions.

Often these predictions carried far more detail as well as forecast precipitation (convection) than would be indicated by the operational models (Eta or GFS), and in some cases, some forecasters needed a spin-up time to understand whether the forecasts could be believed and how best to use them. For this study, we will continue completion of the analysis and questionnaires. We also want to examine model performance over a broader spectrum of IHOP days, not in the detail as was done for 15 June but more by phenomenon, such as the different convective types discussed for this same case.

Editor's Note: A complete list of references and more information on this and related topics are available at the main FSL Website, by clicking on "Publications" and "Research Articles."

(Edward Szoke is a meteorologist in the Forecast Research Division headed by Dr. John McGinley. He can be reached at Edward.J., or at 303-497-7395.)