1. Motivation and Background:
Knowledge regarding the distribution of water vapor is vitally important for most key areas of atmospheric sciences. For example, an accurate prediction of warm season precipitation amounts has remained an elusive goal for the atmospheric sciences (e.g., Uccellini et al. 1994) despite steady advances in the ability of numerical weather prediction models to forecast many other atmospheric variables. Some of this lack of skill is surely related to the fact that current prediction models must parameterize convection as a sub-grid scale event. However, evidence suggests that an accurate characterization of water vapor in the lower atmosphere is also a necessary condition for quantitative prediction of precipitation. For example, predicting the initiation of convection in a cloud-resolving model is highly dependent on very accurate estimates of water vapor within and just above the boundary layer (Crook 1996). In addition, once convection develops, the vertical profile of water vapor is of first order importance in the prediction of precipitation rates, since water vapor is directly involved in the calculation of various thermodynamic and microphysical processes. Hence, it is not surprising that quantitative precipitation forecasts and the development of techniques to accurately characterize water vapor profiles are included in the three U.S. Weather Research Program (USWRP) foci and are highlighted in the USWRP Prospective Development Team reports (e.g., Emanuel et al 1995; Dabberdt and Schlatter 1996).
The three-dimensional distribution of water vapor in the atmosphere is also of crucial importance to climate research, mainly because the global warming predicted by most general circulation models is dependent on a positive feedback between water vapor and surface temperature (Randel et al. 1996; Spencer and Braswell 1997). Water vapor distributions are also directly related to the accurate prediction of cloud and cloud-radiation feedbacks in climate models, which at the moment remains a source of major uncertainty in climate modeling (Ramanathan 1989; Browning 1994; Stokes and Schwartz 1994). Thus various climate programs, including the Atmospheric Radiation Measurement (ARM) project and the Global Energy and Water Cycle (GEWEX) Experiment, have focused in part on linking water vapor, clouds and radiative properties.
Clearly, addressing these broad goals of weather and climate research depends not only on the direct measurement of water vapor but also depends indirectly on many of the sub-disciplines in the atmospheric sciences. For example, if one is interested in predicting convective activity beyond a time-scale of a few hours for USWRP studies, then one must accurately treat boundary layer processes, such as the fluxes of latent heat and the entrainment of dry air into the boundary layer, from aloft. An accurate representation of the dynamics of the flow, particularly the vertical motions, would also be necessary for this prediction. Support for the role of mesoscale dynamics is evidenced by classic studies (e.g., Rhea 1966) that show convective activity over the southern Great Plains is often first triggered by motions associated with the dryline (Fig. 1) and fronts. Another example of the importance of related sub-disciplines is that if one is interested in upper-tropospheric water vapor for climate studies, then it seems likely that one must be able to predict the occurrence of mesoscale convective systems with their extensive cloud anvils, at least in a statistical sense. It can thus be argued that water vapor and water vapor processes are a central focal point and a key theme to the atmospheric sciences.
Unfortunately, despite the importance of water vapor to atmospheric studies, current observational techniques for measuring water vapor are lacking. For example, large humidity biases that often exceed 5% in the lower troposphere (Fig. 2) have been recently documented in radiosonde data (Cole and Miller 1998). Upper-troposphere biases also exist in this measurement (Soden and Lanzante 1996). In addition, while new operational systems, such as Next-Generation Weather Radar (NEXRAD), Aircraft Communications Addressing and Reporting System (ACARS) and the wind profiler demonstration network are operationally deployed within the central United States, there is no corresponding operational remote sensing system for obtaining accurate vertical profiles of water vapor with correspondingly high vertical and temporal resolution. Fortunately, instrumentation to accurately measure water vapor does exist in the research community. A review of the current state of technology to measure water vapor is described in Weckwerth et al. (1999) and includes water vapor Differential Absorption Lidar (DIAL) and Raman lidars, Fourier-Transform Infrared spectrometer (FTIR) spectrometers, radar refractivity techniques, Global Positioning System (GPS) techniques, the interferometric Synthetic Aperture Radar (SAR) techniques, the satellite Atmospheric Infrared Sounder (AIRS) measurements and combined sensor approaches. Despite this wealth of research instrumentation, to date there has not been a broad collection of ground-based and airborne water vapor remote sensors assembled to address the issues associated with three-dimensional variations of water vapor and its impact on clouds and warm season precipitation events. In this international experiment, we propose to utilize this research instrumentation to conduct a field project in the central U.S. to address a number of key issues associated with the three-dimensional distribution of water vapor in the troposphere.
2. Specific Experimental Goals
We will conduct a research experiment over the central United States in May/June 2001. Our experimental goals are listed below.
A. Improved understanding of the role of water vapor three-dimensional variations in the initiation and maintenance of convection.
Observational studies (e.g. Weckwerth et al. 1996; Parsons 1999a; Parsons et al. 1999b) have shown that there are frequently wide variations in water vapor concentration in the pre-convective environments (Figs. 3 and 4) so that that the accuracy of water vapor measurements by rawinsondes often suffer from large sampling errors. We intend to use high-resolution measurements of water vapor to determine the accuracy, spatial scales and spatial and temporal resolution necessary for successful operational predictions of convection. We will also explore the interactions between the dynamics of low-level jets, dry-lines, gust fronts and cold fronts with the distribution of water vapor. We will also study the relationship between atmospheric water vapor, surface topography and surface hydrology, as it relates to quantitative forecasts of precipitation.
B. Improved understanding of how three-dimensional variations in water vapor and water vapor processes impact the prediction of clouds and radiation in coarse grid models.
The prediction of cloud fields in general circulation models and some numerical weather prediction models is particularly difficult due to the small horizontal and sometimes vertical extent of clouds relative to the size of the model grid. The ARM project has been focusing on using single column models (a single column of a general circulation model driven by observed boundary conditions) to test cloud and radiation parameterizations (Randall et al. 1998). Although this approach provides useful tests of these parameterizations, application of this technique to middle latitude cloud fields is difficult due to the need for accurate and detailed lateral boundary conditions (e.g., Mace and Ackerman 1996; Parsons and Dudhia 1996; Petch and Dudhia 1998). Obtaining lateral boundary conditions for water vapor is particularly difficult due to the lack of operational remote sensors to measure this quantity. In addition, the three-dimensional distribution of water vapor over the region characterized by the single column is typically relatively unknown. This experiment will aid the interpretation of tests of cloud and radiation parameterizations in single column models by adding understanding to the variation of the lateral boundary conditions for water vapor and by specifically characterizing the three dimensional water vapor field and its relationship to cloud fields. Single column models for this period can be run with lateral boundary conditions based on observations, products of data assimilation models and/or output from mesoscale models.
C. Evaluate operational measurement strategies aimed at improving estimates of water vapor through objective analysis and data assimilation.
Variational data assimilation techniques offer great promise for many atmospheric science applications due to their ability to use measurements that are not predictive variables in forecast models. Recently Guo et al. (1999) found encouraging improvements in model initial conditions through assimilating different data combinations: surface rainfall, wind profiler measurements, ground-based GPS measurements, and surface dew point measurements within a four-dimensional variational version of the mesoscale model MM5. The temperature improvements were far greater than the improvements in humidity. This experiment had shortcomings in that the spatial coverage of the surface rainfall and GPS measurements was limited. The experiment was also not designed to study the impact of a number of other potential observations, such as humidity sensors on commercial aircraft, the combined sensor approach of Stankov (1998), cloud information, space-borne water vapor lidars or supplementing the wind profiler demonstration network with lidar systems. Through our experimental approach outlined below, we hope to address the issues and to create a water vapor data set that would allow better testing of the impact of different measurement strategies on the accuracy of water vapor fields obtained by assimilation.
D. Test the performance of operational and research models to accurately predict the distribution of water vapor and water vapor related processes.
It is often difficult to validate water vapor fields against numerical models due to a general lack of accurate water vapor measurements on spatial scales that are compatible with the vertical and horizontal resolution of the model. For example, as mentioned earlier, rawinsonde measurements often have a bias and satellite measurements can have poor vertical resolution, especially in the lower levels. Through collaboration with other investigators, we will compare the observations against model estimates of water vapor from operational models and higher resolution research models such as several current research models that are run on quasi-operational basis. Routine surface flux measurements made by ARM and special observations that can be made with remote sensing techniques (i.e., Davis et al. 1997; Kiemle et al. 1997; Giez et al. 1998; Wulfmeyer et al. 1998, Wulfmeyer et al. 1999) will also be evaluated against model estimates. If there is a Southern Great Plains (SGP) 2001 soil moisture project within this time frame, then more in-depth boundary layer comparisons can be undertaken.
E. Explore the impact of improved water vapor measurements on quantitative precipitation forecasts for nowcasting and for cloud resolving and mesoscale models.
The three-dimensional time-dependent fields of water vapor and other meteorological variables obtained from this experiment will be well suited to test the impact of improved water vapor measurements on quantitative precipitation forecasts of convective events. Through the efforts of this team and other investigators, we anticipate that forecast impact studies will be undertaken on a variety of scales through mesoscale and cloud-resolving numerical models, and using nowcasting techniques. Through these studies we hope to quantify how uncertainties in water vapor measurements impact quantitative precipitation forecasts and estimate how this impact compares to other uncertainties, such as those inherent in estimates of surface fluxes in the wind field (particularly convergence and vertical shear).
F. Evaluate and compare different measurement techniques to better establish accuracy and precision.
Remote sensing observations will be compared against reference radiosonde measurements taken by a mobile system and against in-situ aircraft measurements. In addition, airborne remote sensing measurements will be compared against other surface-based systems. These intercomparison and validation studies will allow us to establish accuracy, precision, range, resolution and performance limitations of the remote sensors involved in the experiment. If the experiment develops as planned, many of the instruments for water vapor profiling that currently exist around the world will be involved in the field measurement phase. The Cloud and Radiation Measurement (CART) site, with its Raman lidar for moisture profiling, tower measurement, balloon launches, and GPS precipitable water measurements, offers an excellent location around which to center intercomparison studies. Some of these intercomparisons will be undertaken in a co-located mode during the ARM water vapor IOP in October 1999. It is anticipated that fly-by calibrations will be made against the ARM CART Raman lidar system allowing comparisons between the Raman system, our high quality aircraft in-situ sensors, and the airborne DIAL system(s). Due to the relative uniqueness of these calibrations, we plan to make these comparisons over the depth of the troposphere and not just the lower levels. Results of these investigations will provide the needed information on measurement accuracy and statistical properties for ingest of the observational data into four-dimensional variational models, as well as establishing performance limitations for the various measurement techniques.
G. Obtain data on the operating characteristics of existing remote sensors to help design new sensors.
We will also use the observations from remote sensors present at the experiment to provide realistic information on operating characteristics and scaling of results for design of future instruments. A key observational need for quantitative precipitation forecasting is the horizontal variability of water vapor and its relationship to convective initiation. Preliminary design studies underway at NCAR indicate that measurements of water vapor structure to ranges beyond 10 km are feasible using optical remote sensing techniques. Observations from the moisture profiling instruments on-site during the experiment will be used to verify and improve models of system performance employed to develop the design specifications for the proposed new instrument. General operating characteristics of the instruments on airborne and ground-based platforms will also provide useful information for design of new instruments.
3. Preliminary Experimental Design
A. Location and Timing
We propose that the southern Great Plains would be the optimal location for this study, due to both the available existing experimental facilities, such as the ARM CART, as well as the existing operational facilities, such as NOAA’s Wind Profiler Demonstration network and numerous NEXRAD Doppler radars. In addition, it is well known that this region is characterized by strong gradients in moisture associated with the dry-line so the general meteorological situation is well established from previous field experiments. It is our intention to concentrate on warm season processes in this region where improvements in quantitative precipitation forecasts are more closely linked to convection. Due to the close link in nature and cloud resolving models between the water vapor fields in the lower troposphere and the initiation of convection, we believe that quantitative precipitation forecasts are more likely to be improved during the warm season than winter. After considering such factors as the seasonal variation in dry-line strength, convective intensity, and severe weather activity, we propose that the experiment be in May-June. The proposed year of operation is in 2001.
The general layout of the facilities will be such that the frequency of water vapor measurement will be the largest in the vicinity of the CART domain. Thus, on the outer portions of the wind profiler demonstration network, the primary source of water vapor measurements will be limited to serial rawinsonde ascents at supplementary sites. Lidars and other remote sensors for water vapor will be distributed over a three-dimensional grid closer to the CART domain. This inner area will also be the location for aircraft operations. In view of the goals of the experiment, measurements of the wind field and thermodynamic measurements of fields other than water vapor are clearly necessary.
The original motivation for this experiment was related to the goals of the USWRP in that we wanted to improve the characterization of water vapor in the atmosphere and evaluate its relative impact on quantitative forecasts of precipitation. In order to accomplish this goal and the others mentioned in the previous section, we propose the following instrumentation.
i. Rawinsonde systems. There is not sufficient remote sensing capability to cover the domain of the wind profiler demonstration network in the vicinity of the CART site with remote sensors capable of measuring water vapor. For this reason, we propose that supplementary rawinsonde sites also be employed on the outer edges of this domain. The relative poor temporal frequency of sondes dictates that they be used on the outer portions of the domain or for verification purposes. The proposed rawinsonde sites will be necessary to evaluate the improvement on quantitative precipitation forecasts on the mesoscale, for the data assimilation tests, and to better characterize the regional weather pattern. The results of Guo et al. (1999) suggest that the overall improvement in the assimilation depend on the area over which the assimilation is undertaken. This result is expected due to the dependence of high-resolution simulations on lateral boundary conditions. It is also proposed that these rawinsonde systems include a NCAR reference rawinsonde system operated as a mobile system. As discussed earlier, a mobile reference system will allow for some inter-comparison and evaluation of the different types of sensors and techniques characterizing the water vapor field.
ii. Ground-based remote sensing systems. The Raman lidar at the CART site will serve as one of the systems providing near continuous measurement of water vapor. In addition there are several other ground-based systems that could be operated in the experiment (i.e., Machol et al. 1996; Melfi et al. 1998; Wulfmeyer and Boesenberg 1998; and other systems under development such as the NOAA/ETL mini-MOPA system). Distributing these systems over the central portion of the wind profiler demonstration area would allow for a high resolution, three-dimensional picture of water vapor over convective environments that has been not been obtained previously. When possible we propose that these systems be used in an integrated mode (e.g., Parsons et al. 1994). For example, one should consider co-locating the lidars with wind profilers in the demonstration network with Radio Acoustic Sounding System (RASS) and GPS receivers so that the variational approach of Stankov (1998) and other methods that rely on profiler measurements can be evaluated. Since convective initiation is most sensitive to humidity within and just above the boundary layer, we also propose that some of the lidar systems be co-located in an integrated mode with 915 MHz wind profilers in order to better resolve the convective boundary layer. Cloud radars at some sites would allow for an accurate determination of cloud layers, to define the relative humidity in layers where the lidars are unable to obtain measurements. Thus, this extension of the combined sensor approach is appealing, since it produces an all-weather measurement. Collaborators may wish to collect data from inexpensive GPS receivers in high-resolution arrays to attempt to use interferometric techniques to try to recover water vapor profiles. If these collaborators participate, then this array would be ideally located near scanning Doppler radar(s) so that assimilation tests with radar wind fields and GPS interferometric water vapor measurements can be undertaken. The radar technique recently proposed by Fabry at McGill University can also be evaluated. All these high-resolution measurements will complement the aircraft measurements discussed in the following subsection.
iii. Airborne systems. We are proposing that IHOP 2001 include several aircraft. We have had initial discussions with several of the groups with airborne lidar expertise and hope to encourage their participation in this project. A request will be made this fall to the NSF Observing Facility Panel for NCAR’s Electra aircraft. If this plane can be secured through the NSF deployment pool, we propose that the aircraft be instrumented with the ELDORA Doppler radar, the SABL high resolution lidar, an airborne water vapor DIAL lidar and in-situ measurement packages. It is necessary, of course, that the in-situ water vapor measurements are of very high quality (i.e., cryo-chilled mirror and laser diode systems, only). The water vapor DIAL most likely to be placed upon the Electra is from the lidar group at U of Paris/CNRS. A request has also been made to NOAA for the use of their P-3 aircraft instrumented in a similar manner. Efforts are underway to secure a DIAL system from other collaborators. Flight plans will be developed to meet the various objectives listed earlier. In addition, we will strive to involve a third research aircraft instrumented with DIAL capabilities. Possible candidate platforms for the DIAL systems include the National Aeronautics and Space Administration (NASA) DC-8 or the Deutsche Forschungsanstalt fuer Luft-und Raumfahrt (DLR) Falcon. The airborne measurements will extend our studies to convective scale features and allow the study of moving features, such as the dryline, fronts and outflow boundaries.
C. Cooperation between other planned efforts and institutions
The success of this project is highly dependent on attracting collaborations in a number of areas. First, the "entrainment" of some of the premier international groups working on measuring water vapor with DIAL and Raman systems is necessary. This therefore requires that some of these groups will need to show agencies in their own country that their participation is of sufficient scientific interest in order to warrant some level of funding. Second, the cooperation of the ARM project will be extremely helpful in a number of ways: the additional scientific expertise of their investigators working on water vapor, access to their data store for ARM and external data, and possible sounding support if our effort coincides with one of their planned IOP periods. Third, we hope other efforts with overlapping goals, such as Thunderstorm Initiation Mobile Experiment (TIMEX) and the SGP soil moisture project will cooperate and merge efforts where appropriate. The TIMEX project, in particular, has numerous overlapping goals and some of the same investigators. Since at last report the TIMEX project has been delayed, we hope that our convective initiation efforts contained in our broader goals will serve in some sense as a pre-TIMEX effort. Finally, we invite others working on improvements in water vapor instrumentation, assimilation of water vapor related measurements, and the impact of uncertainties in water vapor in convective precipitation efforts to join this study.
Browning, K.A., 1994: Survey of perceived priority issues in the parameterization of cloud-related processes in GCMs. Quart. J. Roy Meteor. Soc., 120, 483-487.
Cole H., and E. Miller, 1998: Correction and re-calculation of humidity data from TOGA COARE radiosondes and development of humidity correction algorithms for global radiosondes data. In press. WMO Proceedings on the TOAGA COARE 1998 Workshop.
Crook, N.A., 1996: Sensitivity of moist convection forced by boundary
layer processes to low-
level thermodynamic fields. Mon. Wea. Rev., 124, 1767-1785.
Dabberdt, W.F., and T.W. Schlatter, 1996: Research opportunities from
observing and modeling capabilities. Bull. Amer. Meteor. Soc., 77, 305-323.
Davies, K.J., D.H. Lenschow, S.P. Oncley, C. Kiemle, G. Ehret, A. Giez and J. Mann, 1997: The role of entrainment in surface-atmosphere interactions over the boreal forest. J. Geophys. Res., 102, 29, 219-29, 230.
Emanuel, K. and Co-authors, 1995: Report of first prospectus development
team of the U.S.
weather research program to NOAA and the NSF. Bull. Amer. Meteor. Soc., 76, 1194-1208.
Giez, A., G. Ehret, R.L. Schwiesow, K.J. Davis and D.H. Lenshow, 1998: Water vapor flux measurements from ground-based vertically-pointed water vapor differencial absorbtion and Doppler lidars. J. Atmos. Oceanic Technol., in press.
Guo, Y.-R., Y.-H. Kuo, J. Dudhia, D. Parsons, and C. Rocken 1999: Results of mesoscale 4DVAR assimilation tests over the southern Great Plains. To be submitted to Mon. Wea. Rev.
Kiemle, C., G. Ehret, K.J. Davies, D.H. Lenschow and S.P. Oncley, 1997:
Airborne water vapor
differential absorption lidar studies of the convective boundary layer. Bouyant Convection in Geophysical Flows, E.J. Plate, E.E. Fedorovich, D.X. Viegas and J.C. Wyngaard, Eds., Kluwer, 207-238.
Mace, G.G., and T.P. Ackerman, 1996: Assessment of error in synoptic-scale
from wind profiler and radiosonde network data. Mon. Wea. Rev. 124, 1521-1534.
Machol, J.L., R.M. Hardesty, B.J. Rye, and C.J. Grund 1996: Proposed compact, eye-safe lidar for measuring atmospheric water vapor, Advances in Atmospheric Remote Sensing with Lidar. A. Ansmann, R. Neuber, P. Rairoux and U. Wandinger, Eds, Springer-Verlag, Berlin.
Melfi, S.H., D.D. Turner, K.D. Evans, D.N. Whiteman, G.K. Schwemmer, and R. A. Ferrare 1998: Upper-tropospheric water vapor: a field campaign of two Raman lidars, airborne hygrometers and radiosondes. Preprints, 19th Intl. Laser Radar Conf., Annapolis, MD.
Parsons, D.B., 1999a: Measurements of a nocturnal dry-line, Submitted to Mon. Wea. Rev.
________, and co-authors 1999b: The possible impact of a network of surface GPS measurements of vertically integrated water vapor on nowcasting convection and on understanding the mesoscale environment. To be submitted to Mon. Wea. Rev.
________, and J. Dudhia, 1996: Observing systems simulation experiments and objective analysis tests in support of the goals of the Atmospheric Radiation Measurement Program. Amer. Meteor. Soc., 125, 2353-2381.
________, and co-authors, 1994: The integrated sounding system: description
observations from TOGA-COARE. Bull. Amer. Meteor. Soc., 75, 553-567.
Petch, J. C., and J. Dudhia, 1998: The importance of horizontal advection
of hydrometers in a
single-column model. J. Climate, 11, 2437-2452.
Ramanathan, V., R.D. Cess, E.F. Harrison, P. Minnis, B.R. Barkstrom, E. Ahmad, and D. Hartman, 1989: Cloud-radiative forcing and climate: Results from the earth radiation budget experiment. Science, 246, 57-63.
Randall, D.A., K.-M. Zu, R.C.J. Summerville, and S. Iacobellis, 1996:
Single column models and
cloud ensemble models as links between observations and climate models. J. Climate, 9, 1683-1697.
Randell, D.L., T.H. Vonder Haar, M.A. Ringerud, G.L. Stephens, T.J. Greenwald and C.L. Combs, 1996: A new global water vapor dataset. Bull. Amer. Meteor. Soc., 77, 1233-1246.
Rhea, J.O., 1966: A study of thunderstorm formation along drylines, J. Appl. Meteor., 5, 58-63.
Soden, B.J., and R. Lanzante 1996: An assessment of satellite and radiosonde
upper-tropospheric water vapor. J. Climate, 9, 1235-1250.
Spencer, R.W., and W.D. Braswell, 1997: How dry is the tropic free troposhere
? Implications of
global warming theory. Bull. Amer. Meteor. Soc., 78, 1097-1106.
Stankov, B.B., 1998: Multisensor retrieval of atmospheric properties. Bull. Amer. Meteor. Soc., 79, 1835-1854.
Stokes, G.M. and S.E. Schwartz, 1994: The Atmospheric and Radiation Measurement (ARM) Project: Programatic background and design of the cloud and atmospheric radiation testbed. Bull. Amer. Meteor. Soc., 75, 1201-1221.
Uccellini, L. W., P. J. Kocin, and J. M. Sienkiewicz 1994: Advances
in forecasting extratropical cyclogenesis at the National Meteorological
Center, The Lifecycles of Extratropical
Cyclones, Proc. of An International Symp., vol. 1, ed. S. Gronas and M.A. Shaprio, 259-274.
Weckwerth, T.M., J.W. Wilson, R.M. Wakimoto, 1996: Thermodynamic variability within the convective boundary layer due to horizontal convective rolls. Mon. Wea. Rev., 124, 769-784.
_________, V. G. Wulfmeyer, R. M. Wakimoto, R. A. Banta, R. M. Hardesty, and J. W. Wilson, 1999: NCAR/NOAA lower-tropospheric water vapor workshop. To be submitted to Bull. Amer. Meteor. Soc.
Wulfmeyer, V., 1999: Investigations of humidity skewness and variance
profiles in the
convective boundary layer and comparison of the latter with large eddy simulation results. J. Atmos. Sci., accepted.
________, and J. Boesenberg 1998: Ground-based differential absorption lidar for water vapor profiling: assessment of accuracy, resolution, and meteorological applications. Appl. Optics, 37, 3825-3844.
________, F. Jansen, J. Boesenburg, L. Hirsch and G. Peters, 1999: Investigation of turbulent processes in the lower tropospere with water vapor DIAL and radar-RASS. J. Atmos. Sci., accepted.
Figure 1: Frequency of new radar echo formation relative to the surface position of the dryline for April, May, June 1950-1962. a) Isolated radar echo formation, and b) squall line formation. These figures clearly show a tendency for convective clouds to first develop along the dryline. Taken from Rhea (1966).
Figure 2: Average rawinsonde corrections for ISS TOGA COARE sites divided into mean corrections and standard deviations for day and night periods. This correction is based on over 4000 sondes according to the technique described in Cole and Miller (1998). This figure is provided by Junhong Wang (NCAR/ATD) as part of her ongoing research.
Figure 3: Large sub-synoptic temporal variations in the vertically integrated water vapor as measured by a microwave radiometer in west Texas. The large values corresponded to the formation of a convective squall line over the site. Adapted from Parsons (1999a).
Figure 4: Mobile CLASS sounding comparisons taken within a roll updraft region (thick lines) and between a roll updraft and downdraft (thin lines). **These balloon launch locations are shown schematically in the center relative to the boundary layer roll circultions. Schematic cloud indicates cloud penetration by the 1642 sounding.** The ascent rates for each sonde are shown on the left. Horizontal wind speeds are shown on the right (full barb - 5 m/s; half barb - 2.5 m/s).
Figure 5: Location of the DOE/ARM Southern Great Plains site. Figure courtesy of ARM.