Current Mesoscale Meteorological Research in the United States (1981)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

2 Survey of Current Research in Mesoscale Meteorology In recent years, there has been much research on mesoscale meteorological processes. In this report we classify this research into four basic types: (1) observational studies and field programs, (2) theoretical studies, (3) numerical modeling simulations and laboratory simulations, and (4) technological ad- vances in measurement systems. Although some research overlaps two or more of these categories, most projects tend to belong predominantly in one of them. Most progress in meteorological understanding has begun from observa- tions made possible by technological advances in observing systems. Once a phenomenon is identified, case-study analyses help to define the phenomenon and suggest hypotheses in terms of fundamental processes. Theoretical studies seek to explain the phenomenon in terms of the evolution of the structure from an initial or basic state as a function of the boundary conditions and internal forcing mechanisms. A technique of growing importance in mesoscale meteorology is numerical simulation by computer models, used as a form of theory and as a partial replacement for laboratory experiments. Ultimately the numerical models may be applied for practical prediction problems. In complicated meteorological problems, there is often feedback and inter- action between different areas of research. Thus theories suggest how to model or parameterize physical processes in numerical models. Numerical models may show behavior requiring theoretical explanation, provide hy- potheses for observational verification, or demand certain kinds of observa- tions that only new technology can provide. New observing systems provide different viewpoints of phenomena that help to elucidate physical principles and suggest hypotheses for further study.

Survey of Current Research in Mesoscale Meteorology 7 Much past and recent theoretical work has helped to clarify our under- standing of the fundamental processes in mesoscale meteorology. The basic mechanisms and sequences of buoyant and shearing instability and of stable wave dynamics are understood. It is dangerous to make predictions about fundamental research, however, and entirely new concepts may emerge that will change our basic approach to what were thought to be fairly well-understood phenomena. Some important areas of current research (see also Section 2.4) include the evolution and maintenance of rotating convective storms and in- tense vortices, the role of moisture and planetary boundary-layer processes in fronts, entrainment into the tops of stratiform and cumuliform clouds, and the interaction of gravity waves with turbulence and mean flows. Case-study analyses tend to respond rapidly to the existence of new ob- servational and analysis techniques, such as increasing availability of quanti- fied satellite and Doppler radar data as well as continued improvement of computational facilities. Various medium-sized field projects, summarized be- low, are specifically aimed at providing data sources for such research. If ade- quately supported, these and other programs should provide the scientific basis for improvements in forecast capability. In addition to the importance and value of small- and medium-sized proj- ects, there is increasing realization among scientists that occasional major, coordinated research efforts involving scientists from all four areas of research are necessary to make major progress in understanding important atmospheric phenomena and to apply the knowledge gained to national needs. Such proj- ects involve large computer simulation studies to help design the project; tightly organized field projects utilizing modern remote-sensing techniques as well as conventional data sources; and extensive data processing, analysis, and numer- ical modeling efforts after the field program to extract maximum information from the data. Some scientists, especially at universities, are reluctant to par- ticipate strongly in such work, expressing doubt as to the efficiency of "big- science" projects in comparison with the more traditional modes of individual and small-group research. However, there is a growing recognition that, in many cases, there is no alternative to occasional big projects if we are to make scientific headway in the complex problems that continue to confront opera- tional meteorologists. For example, tropical meteorologists participated in the GARP Atlantic Tropical Experiment (GATE) project, and probably much of their best work is yet to come. The same process is starting to occur around the Severe Environmental Storms and Mesoscale Experiment (SESAME). Once mesoscale meteorological phenomena are understood and a predictive or simulative capability has been demonstrated, it is important to make this capability available to those who can benefit from it-so-called technology transfer. Often this transfer is as big or a bigger problem than the original

8 CURRENT MESOSCALE METEOROLOGICAL RESEARCH scientific one, but its solution is vital if the promised economic and safety benefits are to accrue. 2.1 OVERVIEW OF MESOSCALE RESEARCH During its two meetings, representatives from the federal agencies listed in Appendix A provided the panel with considerable information on current and planned mesoscale meteorological research. Much of the material could not be included in a brief report because of its length and complexity. Therefore, we decided to include only the major areas of current research. A summary cf this research, presented in Table 1, reflects the panel's perception of cur- rent, major efforts, although it may omit some important activities. Never- theless, we believe that the summary in Table 1, and the more detailed re- views of the four areas of mesoscale research to follow, provides a representa- tive overview of current mesoscale research activities in the United States. Two conclusions may be drawn from Table 1. First, a large number of groups, supported by different federal agencies, are working on similar prob- lems; with a few exceptions, this research is conducted without coordination with other groups working on similar problems. As discussed in Chapter 3, we believe that progress could be enhanced by increased cooperation and coor- dination between various groups. Second, much research is being conducted in the mesoscale aspects of extratropical cyclones and the precipitation pro- cesses embedded within the cyclones. This emphasis is in agreement with the conclusions reached in the recent reports listed in Chapter 1 and reflects the importance of precipitation to the nation. 2.2 OBSERVATIONAL STUDIES AND FIELD PROGRAMS The study of mesoscale meteorological processes has been hampered by lack of observations. Unlike case studies of large-scale meteorological systems, which have benefited greatly by operational data bases and analyses, meso- scale observational studies generally require special measurement programs to obtain the high spatial and temporal resolution data necessary to describe the evolution and structure of the mesoscale phenomena. Most of the recent mesoscale field programs (Table 2) have been centered around phenomena belonging to the lower end (meso-7) of the mesoscale spectrum (horizontal scales ranging from 2.5 to 25 km). These studies include cumulus convection, thunderstorms, frontal rainbands, flow around small hills, and urban meteorological processes. Because of their much higher cost and more extensive logistical requirements, observational studies of meso-a

si « Jsl Z XX XXX X X X 3 • 0. If o X XX X X 14 SI U o E H X X §1 tfl u•| O XX X X XXX W r " 7 ^ a ft> >i a. XXX X X XXXX II OS c u U S w *• XX X X XXX ll Z s| u. XX X XX X E•g "'"j'S 3 s < 1 X X X XXXXX X § "3 O 5^5 a > ^^ H XXXX , XX X BJJ 3 §| W i! U Z X XXX XX X 3 >> 3 0 Ss ^ S *" 3 3 J'S. U XX X XX X X a. •SB- U XX XX X XX X XX XX X 'S, a a a. '§S8 U ~I v. u >. X XXXX XX X X XXX X Qb B B *c5 U U'S.2 c •r o. _o * o f 0 ..0. •5 3=S 88 ? £ •pj C 1 ll&l us 1 I* Izl g^OT o « N 1* *ll si *« £§« *»***"** ^s^ ^.s 'S £~il « £? 0 |5|3 ZZ Z QuQQQ

E •3 i u. •<• u w eo w < [t, ^< O e« o z•z z Q 'O z z z u, < pa <*s S< 2 ° u5 t* •w (d H H-; < z H^ 50 e ^ « s ^ c' b co *' E 2 111 > C T3 < j» u w III >, ^ •a <o O Q z o •a j S « z w ppler radar, elec- trical, rawinsonde aircraft measure- <§ 10

CO ^ s" ^ 2 i-l CO <2 1£ «/ s 3= u o z_ ««3 as O <T <I p 33 Z < o -co u. pj- o z CO '—J O CO CO z Z o Z Z « 2 IS hi ^! ^ )__^ N*> O >•».' §£ SO J5 1 * " W) H 4> '{3 "o c J5 u »-• 2 | Q> •C 2 c/3 aS BQ U d> O > ^ CO • D ^ 5 fl OS M O9 OH' 03 d N{ 4> CO X 4-» 4-» « '5 o o sc 1 • i CO o r« •o £ *| * » 00 ON CO ed ^— > C *-T « ±i 6 o\ Ok .2•3 r-H A si 111 II fi 1 1 3 S 2i2 < 's •a 2 " •3 «"_- e g £o eo ™ II ill b & S 6 O*-H |1 "3 2 £.2 '>? S 2^ .2 -1 2 E •* e S £ H I* o ^ cT and to assess associated hazards to aircraft To improve short-term prediction of hurricane landfall and intensity Description and under- standing of the meso- scale structure of winte cyclones in the Pacific Northwest Assessment of potential t enhance winter/spring orographic precipitatioi in American River Basil includes mesoscale o"- servations Determine nature and origin of storms over Lake Michigan caused by cold-air outbreaks from "anada Understanding and pre- dictability of lee cyclo- genesis and other large and mesoscale effects of an airflow over large mountain massifs ibieviations to organizations. ,U ed - 0 w M | i i M S K § « t3 B s.s. *-» 09 •c « to S X h B9 v § s.- .3 •o u g *o w ^^ w "ff •> x X e 5 « 'Q H-J fc <« 8. 8 M J u tu tU ON 0. 11 _u O 5 2 i P 8 ^S. u V) J3 5 11

12 CURRENT MESOSCALE METEOROLOGICAL RESEARCH (horizontal scales 250 to 2500 km) and meso-0 (horizontal scales 25 to 250 km) phenomena, such as tropical cyclones, fronts, organized systems of con- vection, and the environment of severe local storms, have been less frequent. Project SESAME, which is beginning to provide important results, is an ex- ample of a larger field program involving many groups in a coordinated, co- operative study. Additional studies of this magnitude are necessary to fill in the gaps in observations (and our understanding) of these scales of motion that link the synoptic scale to the cloud scale. 2.3 MESOSCALE COMPUTER SIMULATION MODELS An extremely important area of research is the field of numerical simulation, and a large fraction of the theoretical talent in meteorology works in that field. Much progress is being made in developing and applying new approaches to the problem of artificial lateral and upper-boundary constraints (an aspect of atmospheric indivisibility) and the closely related problem of grid-nesting. Most modelers are becoming sensitive to the need to test a model's sensitivity to uncertainties in the physics and numerical formulations, so that simulation is becoming a more fully respected tool of theoretical analysis. There is a tremendous variety of so-called mesoscale models, partly be- cause the term mesoscale covers such a wide range of scales and partly be- cause so many different meteorological phenomena belong, at least in part, in the mesoscale. Here we will arbitrarily classify mesoscale models into two basic types: (1) those that deal with the detailed cloud physics and dynamics of individual clouds, and hence explicitly resolve scales of motion less than the meso-7 scale (2.5 km < L < 25 km), which we will call "cloud-scale mod- els," and (2) those that resolve scales of motion larger than those associated with individual clouds and therefore have typical horizontal grid sizes of 2.5 to 100 km. In these models, the effects of clouds are usually not explicitly represented but are parameterized in terms of resolvable-scale variables. For simplicity we will call these "regional-scale models." 2.3.1 Cloud-Scale Models Cloud models involve many physical processes that take place on scales small- er than the meso-7 scale but that influence the precipitation and dynamic processes on the 7 scale. In addition, some of the models treat electrical, chemical, and radiative processes that may influence processes on the meso- scale, such as thunderstorms, acid rain, and radiative cooling of stratus-filled atmospheric layers.

Survey of Current Research in Mesoscale Meteorology 13 As opposed to regional-scale models, cloud-scale models attempt to simu- late the details of precipitation evolution and air motion over grid intervals of a few hundred meters (or less) to one kilometer, and over domains a few kilometers on a side to several 10's of kilometers on a side. Atmospheric depths up to 20 km are often modeled, and the models are generally nonhydrostatic. Cloud models in one, two and three space dimensions and either time depen- dent or steady state are being used in various aspects of cloud-physics research and operations. The simpler one-dimensional cloud models are coupled with some mesoscale models to yield predictions of precipitation over the meso- scale. The more dynamically complex three-dimensional cloud models simu- late many of the characteristics of severe local storms. Cloud-scale models vary in their microphysical complexity as well as in their dynamic complexity. Simpler, highly parameterized microphysical mod- els require one tenth or less of the computer resources (time and core storage) needed by the detailed particle spectra modeling methods. More microphysi- cal complexity is added as cloud electrification, cloud chemistry, and cloud radiative processes are included in the models. In addition, the treatment of both ice and liquid phases in clouds, necessary for snow and hail predictions (and rain also in highly convective situations), makes the cloud models more complicated and requires more computer power for their solution. The two- dimensional cloud models are being used most often to attack these cloud- physics problems. The various types of cloud models are summarized in Table 3. 2.3.2 Regional-Scale Models Regional-scale models as defined here include models with horizontal resolu- tions (As) ranging from about 2.5 to 100 km. Since typical horizontal grids consist of 40 x 40 points, the domains of these models range from 100 km x 100 km to 4000 km x 4000 km. Because regional-scale models simulate phe- nomena in which the horizontal scales are much greater than the vertical scales, they are usually hydrostatic. In principle, the basic dynamical frame- work of these models is very general. By altering the parameterization of the physical processes (e.g., shortwave and long-wave radiation, change of phase of water vapor, sensible and latent heat fluxes from the surface, turbulent mixing, cumulus convection, and terrain effects) and by varying the initial conditions, an amazing variety of mesoscale phenomena can be simulated and predicted. For example, models with rather similar basic characteristics have been used to simulate the following: • Flow around individual hills

^ •5 E g 4> "— > 3 0. o 'ti fe E o •a N S '3 E g US fl sed to predict variates in wei modification ] 1 O 4-> O <« » •- •o ^ o g * =3 cS 3 S •*-» U 1 t> 1 4-* C •S «• 0 « .3 o 'S u S u S? l|i "if '1 •a ^ « =3 5 M U u U Q, 0) O U V > U 01 rt - M •T3 t - « e. .M> i *"• 8ia > (£ 1 2 2 « 2 § ? I •o 1 1 1 1| i fill § E o, o E a i u 3oE. .s 0 0 > '*-» o •8 /— S s g. ^. t» •a o o 1 o •a I g B g, 1 4) 1 .s ¥ ¥ i 2 .*'»! M ^ ^ o a ° O o O fS O Q 0 "o ;\ & «. 1 a O i § Vj 1 CO s j at <u w_ u 2 1 " H 5" < O Q "u Q U 1 « S D D' T3 co Q H S O « «a z U & "S u, • ,. ^§ g•'-g •SI Eg. 0 0 > ^^ E & g >- .§ a """S 55 I ? n S § 3 co co o O .5 s o c/a •a <a .E H C Q X X E I 1 I I S g •* 0 X E 14

CO H •o " •i ^ *- U * u S g u Slab symmetry Some detailed ic microphysics, symmetry Axial symmetry Axial symmetry Axial symmetry Liquid bulk wat microphysics Some ice bulk w microphysics Liquid bulk wat microphysics Liquid bulk wat microphysics Nested in 3-D cl model Liquid bulk wat microphysics I 8 | g 2 M i o" 2 • •a a x C! *> *4J c 3 CO 1 •a i DO S o Pi e3 & S 1 O 3 i a o •«-• o * 1 3 etailed cloud micro- electrification, cloud modification etailed microphysics, hailstone growth, cloud electrification ropical ", ice bulk water microphysics etailed microphysics, particularly ice phase M O ropical " , mountain " ropical "u, detailed liquid microphysics physics, ice and liquid processes ailstorms, cloud :vere storms :vere storms ailstorms Drnado genesis a 1 | Q as Q H Q on H V >_, H H H CO BH £ 1 >o x x X X X <N g g g o E E 1 1 a x X g M | CN o M OO VI OO O O M Tf >fl u X X X x E x E X g X E x E * S* reviation g M E X 1 E •* E X X M 3 ^ •* x r- oo rt <n rt ^ \.£t k^ g E to O o •* ro 00 "* O — ' o "* f* (S MD ^ •<!• in \o en A •3 •9 i o i I 1 o U 0 Q Q § Q CO Q b I 5^ O ci 5 '""'•' f\ a_ O O_ 8- X H ¥ 0 1 i .a d « A 05 t ^' < <? O9 > 2 ^ hi ^ o •<j <j •<^ £ •< 0 Q CO <f] • 'C 5 U CO O CJ U a O O 9 z s M » K. 2 2 Z 33 2 « 15

16 CURRENT MESOSCALE METEOROLOGICAL RESEARCH • Mountain waves • Orographic precipitation • Air-quality simulations • Mountain-valley breezes • Sea breezes • Extratropical cyclones (mesoscale structure embedded within) • Tropical cyclones • Frontal circulations • Squall lines • Mesoscale convective complexes • Dry lines • Jet streak circulations • Coastal effects • Rainbands • Urban circulations The mesoscale models active as of 1980 are summarized in Table 4. 2.4 THEORETICAL STUDIES Although much of the research involving the numerical simulation models summarized in the previous section can be considered theoretical, the uncer- tainties in the numerical formulations (such as the finite-difference equations or boundary conditions) set this method of theoretical analysis apart from analytic studies in which closed mathematical solutions are obtained (usually for a physical problem that is greatly simplified compared with the ones in- vestigated with numerical models). In this section we discuss a few of the cur- rent theoretical problems that are related to mesoscale meteorology. 2.4.1 Gravity Waves and Wave Interactions The analysis of small-amplitude gravity waves appears to be rather well estab- lished, but much work is proceeding on various large-amplitude nonlinear ef- fects and on the interaction of waves with their mean flow and turbulent en- vironments, from the boundary layer to the upper atmosphere. A recent surge of interest has appeared in three-dimensional mountain wave theory and in the analysis of mountain waves with condensation. 2.4.2 Cloud Dynamics The three-dimensional simulations of convective clouds and storms noted in Table 3 are leading to much better understanding of the dynamics of convec-

Survey of Current Research in Mesoscale Meteorology 17 tion in an environment with vertical shear. In particular, a sequence of events leading to tornado formation now seems to be definable from the models, a result that should further stimulate observational studies in this area. Another important element of cloud dynamics, the entrainment process, is being re- considered theoretically on the basis of evidence that much more of it occurs from the top than had been generally assumed. This work may also lead to a useful understanding of the strong downdrafts (downbursts) often observed in thunderstorms. 2.4.3 Nocturnal Boundary Layers Theoretical understanding of the stable, nocturnal boundary layer has lagged behind understanding of the unstable, convectively driven boundary layer. The nocturnal boundary layer involves strong vertical stratification, strong vertical wind shear, and internal gravity waves. It is affected greatly by com- plex terrain, with blocking of low-level flow by complex terrain an important process. In addition, low-level jets, which arise from an imbalance of forces during the transition from an unstable to a stable boundary layer, are impor- tant phenomena in thunderstorms and squall-line generation. These jets are important for transporting heat and moisture and for producing mesoscale patterns of convergence/divergence. 2.4.4 Frontogenesis and Frontal Circulations Substantial progress in understanding the dynamics of surface and upper- tropospheric frontal systems has resulted from the use of the semigeostrophic equations in the study of straight frontal zones forming in adiabatic, inviscid flows containing confluence or shear. The semigeostrophic equations have the advantage of possessing the same mathematical form as the analytically trac- table quasi-geostrophic equations, but, on account of a transformation of horizontal coordinates, contain the nonlinear physical effects of advection by the divergent, ageostrophic circulation in the cross-frontal, vertical plane. In the case of the surface frontal zones, the analytic model results exhibit realistic slopes, develop from large-scale patterns in a reasonable time period of several days, and are strongest at the ground, as observed in nature. Recent studies into diagnosing the vertical circulations in upper-level frontal-zone jet-stream systems suggest that the thermally indirect circulation associated with upper- tropospheric frontogenesis is a result of the combination of confluence and shearing advection at tropopause level in the absence of curvature. Further progress in the theory of frontogenesis requires explaining the differences in the structure of cold and warm fronts, assessing the influence of curvature on frontal circulations, and including diabatic and viscous effects in an analytic

X u o S BJ OH •-•S 60 8^5 s 1 s s « •s 1 J3 J2 >. B 3 0 i 2 SV ^J s 1 |i 1 15 Ja 1 '8 11 •^ i •a ^* S S '•3 5 .y +2 9 ^3 'H " 0. & G O .t5 '^ &, g 2 8 2 ^ £ "^ _ Hi H a g ^ >, <o 'C J H > 0 *— t »— < t> •< a 1 3 i>- <« Q ll il •O O •a c •H c | s«| a 'S, "c a 2 (U x o C 5 CO H BC 0 Q 1 to CO C 4) •al O OS *s • S u « g m m & Is cs" r>T E a C M O •M ** o M — M •o ^ Q ^ •< O iu S 2 0 Q B o •2 ~ . 'G S o o ~ 0 S g g°°ol . 0 c £ 18

8 1 o ^ 3 .5 Nickerson Rosenthal, J< Willoughb) Hovermale Phillips Ballantine 1 The variation Fritsch £ 1 •O 1 > eb I P U •S1 '5 S V3 •'tn U i <D B o a c £ 2 §8 C [esoscale co: complexes errain effec rographic p tion "3 u i— H r* ension and in £ a 1 0 ,0 U •a 11 •a •8 1 o • in OT C 2 HI i Q, O g u .3 2 HO H H S 0 H § •3 3 c E 1 8 a EcS o g. >o o .5 E ,3 o o E E 0 0 £ ^N — ' — i co r™^ ^H 3 g s 1 0 O 0 0 <N O J=T3 00 x S X| X variation of 1 3d or specify X X sis Si^ ill E g >n 1 X a•9, SS•i E 3 Jo <•""5 2§s 00 OO ^ J-8J ts oo */^ (N fN CO \o w »— 1 f*"i ^O ro *« 4J 'y 5 JS U O •-^ •-• •S n "O .".a c u u £ ~ O c *?> 1EI « m r>f w m CO 5§I x » 2 :g E c OS D. O 'S S O ft >. S• ^•c M) oi X Z U Z O9 I l|| a: w Z a -0 O 19

20 CURRENT MESOSCALE METEOROLOGICAL RESEARCH framework. Spurred by recent observational studies, the theory of rainband formation is now an active area for research. 2.4.5 Small-Scale Convection Theoretical work on small-scale and moist convection is achieving some suc- cess through analytical studies, using limited spectral models, of the transi- tions between flow regimes. In laboratory experiments, a distinct sequence of flow types involving motionless conduction, steady flows, periodic flows, and turbulence is produced as the intensity of the forcing is increased. The spec- tral models are designed with limited degrees of freedom but include suffi- cient nonlinear interactions that make such transitions possible. These models have generated considerable interest in the mathematical community and are now producing physical results (orientation of cloud rolls in a shearing flow) that can be compared to observations of atmospheric convection. Investiga- tors working with these models argue that they reveal the essential physics of the situation, even though it is drastically simplified. 2.5 TECHNOLOGICAL ADVANCES IN MEASUREMENT SYSTEMS Technologies already demonstrated or at advanced stages of development are beginning to provide practical methods for observing mesoscale weather sys- tems and for collecting, processing, analyzing, interpreting, and disseminating mesoscale data and information products. Continuing rapid progress in these areas is expected. 2.5.1 Observations The smaller spatial scales and more rapid development and evolution of meso- scale weather systems require substantially greater spatial resolution and more frequent observations than are available from routine synoptic networks. A number of remote-sensing techniques have been developed that address this problem. The geostationary meteorological satellite imaging systems have been available for more than a decade, but fast and convenient methods for processing and interpreting the data have only recently become available at several institutions. The satellite systems provide, about every half-hour, quantitative visible and infrared imagery covering most of the globe with a resolution of a few kilometers. Wind estimates can be derived from cloud motions, and estimates of cloud and surface temperatures can be derived radiometrically. The recent addition of a multichannel atmospheric sounder (VAS) adds the capability to deduce vertical temperature and humidity soundings of a quality similar to those available from polar orbiting satellites

Survey of Current Research in Mesoscale Meteorology 21 in recent years, but much more frequently. Additional improvements to satel- lite observing systems, including possible wind sounding capabilities using Doppler lidar techniques, are in various stages of planning. Ground-based remote-sensing technologies also are evolving rapidly. Digital radar techniques have improved the recognition and interpretation of radar echoes from mesoscale structures and the estimation of quantitative precipita- tion. Doppler radar technology for measuring horizontal and vertical winds has been developed and demonstrated by several institutions. A national'pro- gram (NEXRAD) to upgrade the nation's weather radar network during the 1980's is under way with the support ofNOAA,DOD, and FA A. Specialized radar technologies also are under development for continuous ground-based observation of wind profiles throughout the troposphere and lower strato- sphere and for studying cloud formation, structure, and content (8-mm radar). In connection with the Prototype Regional Observing and Forecasting Ser- vice (PROFS) program, NOAA's Wave Propagation Laboratory is developing a continuous, remote-sensing profiler that determines wind profiles using the radars mentioned above and defines temperature and humidity profiles using microwave radiometers similar to those operating on satellites. Total water vapor and liquid-water content are observed using another microwave radiom- eter. Some components are now in field testing, and a complete system for observing all of these variables should be assembled within the year. Several institutions also have developed optical devices (e.g., lidars) to observe aero- sols, clouds, and other constituents, while advanced Doppler lidar instru- ments are being developed that will be able to measure profiles of winds and possibly temperature. Automation techniques are being applied increasingly to make and collect observations from mesoscale networks of conventional in situ surface instru- ments. A national program to automate the entire complex of surface synoptic and aviation observations, including ceiling, visibility, and current weather, has been initiated by FAA, DOD, and NOAA. Sophisticated in situ and remote-sensing instruments and data-processing systems also have been adapted to research aircraft to allow rapid, high-resolution observations of thermo- dynamic, cloud physical, and turbulent vertical flux variables, as well as pro- files below aircraft. Chemical sampling and analysis instruments for both ground and aircraft also are now available. Taken together, these technologies provide an impressive and comprehen- sive array of tools that allow the essential physical and dynamical mesoscale features to be observed. 2.5.2 Data Processing, Analysis, and Interpretation As noted above, advanced data-processing and display techniques have become an essential and integral part of most remote-sensing observation systems.

22 CURRENT MESOSCALE METEOROLOGICAL RESEARCH This is true both for image processing and manipulation and for quantita- tive computations (e.g., specifying the location, calibration, and mathematical "inversion" of multispectral radiometric or Doppler radar and lidar data). These techniques are well advanced and are now used routinely to process vast volumes of data in real time. At this stage in our conceptual and theoretical understanding of mesoscale weather systems and their structure and dynamics, it is difficult to analyze and interpret completely a comprehensive set of observational data. However, interactive data processing and display tools greatly simplify the task of syn- thesizing the information content of various meteorological fields and cloud and radar imagery. Reduction to common scales and formats, analysis and contouring of fields, overlay of fields and images (with color enhancements), time-lapse or "animated" displays, three-dimensional perspectives, and other processing techniques facilitate human interpretation of the data. Fast access to a comprehensive data base, coupled with versatile analysis and display soft- ware and hardware, allows adaptive exploration of physical hypotheses by in- teractive analyses. The capability to assimilate and interpret the massive array of observation- al data quickly is particularly critical to operational applications such as meso- scale weather warnings and forecasts. At present, NWS forecasts at most field offices are unable to extract routinely the mesoscale information content of even the existing data sources, especially satellite and radar data. However, an immediate improvement in warnings and short-range forecasts is possible using existing technologies. These improvements would be enhanced and made widely available by the establishment of reliable conceptual models for mesoscale weather phenomena. Ultimately, quantitative models of mesoscale weather will be needed if fundamental improvements are to be made in short-range weather forecasting. Extension of numerical-dynamical techniques to the mesoscale has been ac- complished in several research models, but this demands rather large comput- ing capacity. The technology exists and is becoming cheaper. Some extension to the larger end of the mesoscale spectrum will be possible with the new operational computers planned for NMC, and statistical models probably can be applied at individual weather stations using the local applications concept of the Automation of Field Operations and Services (AFOS). Additional computing power is needed for research modeling. 2.5.3 Dissemination For the public or specialized users to take action based on weather informa- tion to minimize loss of life or property, or to achieve economies in their activities, weather information must be made available in a timely, under-

Survey of Current Research in Mesoscale Meteorology 23 standable, and credible fashion. Mesoscale weather information places an extraordinary demand on dissemination systems because it must be locally detailed and specific and because it is applicable to short-lived, fast-changing phenomena. Fortunately, rapidly evolving technologies are available to address these problems. Fast, reliable communication systems that are capable of pro- viding detailed information, including graphics and imagery, are available. Ar- rangements to allow selective request-reply services and "targeted" broad- casting are under test by industry. It is up to the meteorological and hydro- logical communities to develop information products that are suited to these future dissemination systems.