This is the unofficial, WWW version of the VORTEX-95 Operations Plan. In particular, this chapter likely WILL change a few times prior to 15 March!!! It may differ from the published operations plan which is available by 15 March from the National Severe Storms Laboratory.
Casting objectives in this form is a difficult but rewarding process. Hypotheses were informally reviewed and debated by the Principal Investigators, and thereby refined. This has been very beneficial in that it has served to keep the scope of the experiment narrow, and helped us design field strategies we hope are sufficient to gather the data needed to refute the hypotheses (if refutation is possible). We should note that it is always possible to turn broad objectives into hypotheses to little advantage; the benefit to our approach lies in the statements of testability and refutability.
In the period between the field phases of VORTEX-94 and VORTEX-95, much attention has been given to making sure that the field data gathering techniques are adequate for evaluating the hypotheses. In so doing, some hypotheses have been modified or discarded, and new hypotheses have been developed based on important preliminary results or the availability of new sensing systems. The numbering system used below has been retained from VORTEX-94; gaps indicate that an hypothesis has been discarded.
Hypothesis a.1Horizontal inhomogeneities (including inhomogeneities in the vertical derivatives) in the environmental water vapor, mass and wind fields are important influences in simulated and observed supercell structure and evolution. [Wicker]
Test Evaluate inhomogeneities through surface mesonet observations, mobile soundings, profilers, and aircraft wind and state variable data. Compare the evolution of, and severe weather produced by various storms on a given day. Determine if certain types of inhomogeneities are commonly associated with specific types of evolution and severe weather.
Refute No significant differences are noted between simulated storm evolution in environments with and without inhomogeneities. Observed types of inhomogeneities are not associated with particular types of storm evolution and severe weather.
Hypothesis a.2 In high-helicity environments, the elapsed time from storm inception to initial tornado formation and the duration of the tornadic storm phase both increase with the magnitude of the mid-level storm-relative wind. [Brooks]
Test Use prestorm environmental soundings, profilers, and NWS soundings to evaluate mid-level storm-relative wind strength and helicity. The elapsed time to initial tornado formation can be objectively measured [e.g. time from first radar echo to first tornadic circulation at the surface; or the onset of mesocyclone-strength vorticity (1x10**-2/s) at cloud base]. The duration of the tornadic phase can be similarly ascertained.
Refute Poor correlation in a sufficient sample.
Hypothesis a.3 Given the occurrence or expectation of strong thunderstorms, the regional time tendency of helicity is an important tool for refining forecasts of the onset, duration, and areal domain of tornado outbreaks. [Davies-Jones]
Test Analyze the tendency of helicity utilizing serial prestorm soundings from MCLASS, profiler, and surface mesonet data.
Refute Regional time tendency has no predictive value in a sufficient sample.
Hypothesis a.4 The intensity of supercell storms and their probability of developing tornadic vorticity, are related not only to deep-layer convective available potential energy (CAPE) and wind variations, but also to the JOINT VERTICAL DISTRIBUTIONS of CAPE and environmental vertical shear. The patterns most favorable to maximum overturning efficiency and vertical vorticity production are those where the buoyant layer is not significantly deeper than the layer containing favorable shear and helicity. [McCaul]
Test Obtain prestorm and in-storm updraft soundings, along with Doppler radar data or any other data that would measure storm updraft velocity profiles, for as many storms as possible. Examine CAPE, helicity, and Bulk Richardson Number compared with intensity of storm updrafts and rotation, and see whether differences in the matching of the vertical profiles of buoyancy and shear/helicity, if present, influence storm maximum updraft and rotation intensity.
Refute The sample of well-observed storms is sufficiently large and diverse, but storm updraft and rotation intensity show no dependence on how the buoyancy and shear profiles are matched in altitude.
Hypothesis a.5 The maximum horizontal vorticity in the lowest 1000 m of a supercell is a function of CAPE and DCAPE as measured in environmental soundings. DCAPE is defined as the integral with height of the downdraft buoyancy, assuming the downdraft has the wet bulb potential temperature of the minimum in the layer above the parcel, and that the ambient air has the temperature profile of the environment. [Rasmussen]
Test Estimate horizontal vorticity through pseudo-dual Doppler windfield analyses with additional data provided by soundings straddling the baroclinic regions, and mobile mesonet data. Using environmental soundings from all available cases, determine if there is a reasonably simple function expressing the relationship between horizontal vorticity, CAPE, and DCAPE. Determine if this function is satisfactory for all available cases.
Refute No functional relationship exists between horizontal vorticity, CAPE, and DCAPE.
Hypothesis a.6 In the region of supercell initiation, horizontal inhomogeneities in the warm sector environment are so insignificant that they would cause no appreciable difference in simulated supercell structure and evolution. [Weisman]
Test Perform simultaneous releases of four soundings to sample an area approximately the same size as a typical supercell simulation domain in the target storm initiation area prior to the development of deep convective clouds.
Refute Even without performing numerical simulations, based on established sensitivities of simulated supercell storms to sounding-derived parameters (e.g. CAPE, shear, helicity, etc.), it can be established that the pre-storm environment is not horizontally homogeneous to the degree described in the hypothesis. Perform numerical simulations as needed.
Hypothesis a.7 A supercell proximity sounding is one in the inflow sector that is not significantly modified by the storm and is thus useful for forecasting. There is a distance from the main updraft within which proximity soundings cannot be obtained, and outside which they can be obtained. [Weisman]
Test Using one mobile lab, serial soundings will be made in the inflow region of the storm environment to attempt to find a range outside which the soundings only have insignificant differences when compared to one another but within which soundings are significantly modified. On some storms, serial soundings will be performed by three teams simultaneously, positioned along a line extending from the storm outward through the inflow sector. The horizontal variation of conditions in the inflow sector will be assessed.
Refute No significant differences are found in the inflow sector, or significant horizontal variation extends to large distances and a range beyond which the storm is not modifying the environment cannot be determined.
Hypothesis a.8 Supercell updraft precipitation character is a function of the degree of updraft seeding by hydrometeors falling from the storm\qs anvil, or from the anvils of nearby storms. [Straka and Rasmussen]
Test Monitor the temporal development of the reflectivity fields in P-3, ELDORA, Cimarron (if available), and 88D (when available) at resolutions of 5 minutes or better. Analyze surface reports of hydrometeor fallout locations. Estimate the amount of anvil-produced precipitation re-entering the updraft (in a relative sense, comparing storms with different updraft precip characteristics). Measure the storm-relative flow in the storm anvils in the Doppler data. Utilize the T-28 aircraft (if/when available) to measure hydrometeor concentrations in the sub-anvil air immediately adjacent to the main updraft towers.
Refute The intensity of precipitation in the updraft region is not a function of the rate at which the updraft is seeded by ingesting hydrometeors falling through the near environment. The seeding rate is not inversely related to the storm-relative windspeed in the anvil.
Hypothesis a.9 The rear-flank downdraft in a supercell is forced by both cooling due to phase change (melting and evaporation), and perturbation high pressure aloft associated with the stagnation of storm-relative environmental flow impinging on the updraft/mesocyclone. It is not forced by perturbation low pressure existing at low-levels because of the presence of stronger rotation at low levels than aloft [Straka and Rasmussen]
Test Estimate the acceleration due to cooling. This could be done by defining a downdraft parcel process curve based on low-level equivalent potential temperature below the RFD, computing dCAPE and consequent downdraft acceleration. Estimate the vertical pressure gradient via pressure decomposition from the Doppler-derived wind field.
Refute Either of the forcing mechanisms (cooling or downward-directed PGF associated with the storm-induced perturbation in the mid-level flow) is found to be insignificant. The downward-directed perturbation PGF exists in the RFD and is due to stronger rotation below.
Hypothesis a.10 Convective initiation at the dryline requires two factors, namely; (i) mesoscale lift providing sufficient energy to overcome CIN; (ii) minimized loss of CAPE (with attendant increase of LCL, LFC, and CIN) due to mixing of the parcel with warm, dry ambient air. Convective dryline environments are characterized by the joint occurrence of strong convergence and high relative humidities extending through the depth of the Convective Boundary Layer (CBL). The latter implies the existence of moisture plumes, possibly in conjunction with elevated moist layers, prior to and in the region of the first deep convection. [Ziegler]
Test Estimate the horizontal distribution of near-surface parcel CAPE, LCL, LFC, and CIN using M-CLASS soundings, low level aircraft traverses, mobile mesonet, and fixed mesonet and NWS surface observations. Use M-CLASS soundings and aircraft traverses to estimate profiles of stability parameters. Use aircraft stepped traverses and ELDORA clear air measurements to estimate vertical circulations and air parcel trajectories in the CBL. Use GOES-8, WSR-88D network, and P-3 surveillance radar data and visual cloud observations to identify significant deep convection. Stratify cases with or without dryline deep convection according to: (i) ratio of mesoscale updraft kinetic energy to CIN; (ii) lapse rates of stability parameters and length and extent of mixing along parcel trajectories.
Refute Based on a sufficient sample of dryline cases (both with without dryline convection), the hypothesis is refuted if deep convection: (i) occurs with weak convergence and/or strong stability parameters lapse rates; (ii) does not occur despite strong convergence and negligible stability parameters lapse rates.
Hypothesis a.11 The efficiency with which surface parcel CAPE is realized in the updrafts of early deep convection depends on the degree to which CAPE is reduced through mixing with less potentially-buoyant air prior to reaching the LFC. This reduction through mixing is primarily a function of trajectory length and the difference between surface mixing ratio and mixing ratios along the trajectory. (This applies regardless of the mechanism which is responsible for bringing the parcels to the LFC; e.g. dryline, warm front, cold front, etc.) [Rasmussen]
Test Use M-CLASS soundings and aircraft traverses to estimate profiles of CAPE derived from parcels at various levels. Use airborne dual-Doppler and echo height data to estimate updraft intensity in first half hour of cloud growth. Compare early updraft intensity to CAPE derived from surface conditions and CAPE using parcels near cloud base.
Refute Early updraft intensity is not related to CAPE of the parcels upon arriving at the LFC; CAPE depletion, derived from parcels at the LFC, is not a function of trajectory length and mixing ratio deficit.
Hypothesis a.12 Vertical east-west wind shear within the cooler air mass east of the dryline increases during the afternoon from the action of a thermal solenoid which induces a thermally direct, frontogenetic secondary circulation. This evolution leads to increased moisture convergence to assist storm initiation and increased hodograph curvature to increase probability of updraft rotation. [Ziegler]
Test Evaluate east-west component of thermal solenoid with aircraft stepped traverses, mobile soundings, and mobile mesonet traverses across the dryline. Evaluate evolution of shear from aircraft stepped traverses, mobile soundings, and wind profiler. Correlate changes of u-component shear to solenoid magnitude based on averages over periods of 0.5-1 hour duration.
Refute Either thermal solenoids are absent or locally increasing wind shear cannot be correlated via the horizontal vorticity equation in a sufficient number of cases.
Hypothesis b.1 Tornadogenesis is related to intensification of low-level vertical vorticity which pre-exists the updraft that causes the intensification through stretching. This implies that the only relationship between mesocyclones and tornadoes is that mesocyclones establish the conditions for long-lived storms that have the potential of producing low-level convergence boundaries which are regions of sufficient vertical vorticity for tornadogenesis. This also implies that most differences between tornadoes can be associated with the strength of the updraft that is stretching the vorticity (assuming little variation in the magnitude of pre-existing vorticity from one case to another). [Wakimoto]
Test Using mobile mesonet, surface mesonet, nearby WSR-88D data, aircraft Doppler data, and stereophotogrammetric wind estimates, document the presence or absence of surface boundaries near growing clouds and the vertical vorticity fields associated with these boundaries. Document the growth of vortices along the boundaries.
Refute This hypothesis would be refuted by documenting the occurrence of a tornado without a pre-existing region of low-level vertical vorticity associated with a boundary. It would also be refuted if it is shown that the low-level vertical vorticity being amplified is due to the presence of a mesocyclone and not an incidental pre-existing feature.
Hypothesis b.2 Low-level mesocyclone intensification is due to near-ground horizontal vorticity, generated through forward-flank baroclinity, being reoriented into vertical vorticity and stretched in the low-level updraft. Tornadogenesis results from the abrupt tilting, in the boundary layer, of this horizontal vorticity due to the movement of the RFD-associated gust front into the area below the updraft. [Rasmussen and Straka]
Test Examine the vorticity dynamics of the region near the surface near the vortex as low-level mesocyclone intensification and tornadogenesis occurs using P-3 and ELDORA Doppler syntheses (and mobile Doppler radar if available), and mobile mesonet data. Establish the magnitude of the horizontal component of vorticity (through dual-Doppler syntheses) and the magnitude of low-level forward-flank baroclinity (through soundings and mobile mesonet observations). Establish that the RFD gust front does overtake and turn the FF horizontal vorticity into the vertical where it is stretched by the updraft.
Refute FF horizontal vorticity is weak or non-existent during tornadogenesis or low-level mesocyclone intensification. Tornadogenesis does not occur at the intersection of the RFD gust front and FF low-level horizontal vorticity zone.
Hypothesis b.3 Tornadoes are located in strong equivalent potential temperature gradients on the cool side of the storm outflow boundary; solenoidal generation of streamwise vorticity is significant for tornadogenesis. [Davies-Jones]
Test Evaluate buoyancy gradients and winds at surface using turtles, mobile mesonet, and fixed mesonet where possible. Use stereo photogrammetry to estimate wind fields near cloud base and where other tracers are available. Estimate the solenoidal generation rate. Evaluate the vorticity budget based on aircraft pseudo-dual Doppler data if available.
Refute No significant solenoidal generation is found, based on observations.
Hypothesis b.4 Tornadogenesis is inevitable if stretching near the ground is sufficiently large; this will occur when the "background" vertical vorticity becomes larger than ~2 x 10**-3/s in an area beneath the low-level updraft. [Wicker]
Test Use mobile mesonet, and surface and airborne Doppler radar observations to measure surface winds and determine horizontal shear along the forward flank and rear flank boundaries. Examine the levels of shear in the pre-tornadic and tornadic stages. Group cases according to the occurrence or non-occurrence of tornadogenesis.
Refute In some cases, tornadogenesis does not occur despite the condition being met.
Hypothesis b.5 A unique, significant updraft pulse (defined as an updraft between 1 and 4 km that exceeds 25% of the 30-minute average vertical velocity and which develops during a several-minute period) precedes the development of a tornado. [Wicker]
Test Dual- or pseudo-dual Doppler data and perhaps stereo photogrammetry is used to establish the evolution of the vertical velocity structure.
Refute Unique, significant updraft pulses not observed.
Hypotheses c.1 This is actually a set of straw man hypotheses. Since our knowledge of the dynamics of real tornadoes is very limited, carefully gathering data in the vicinity of tornadoes may refute some of these hypotheses and help lead to significant advances in our understanding. Tornado flow can be characterized as follows: [Lewellen, Davies-Jones]
Refute Observations do not agree with the specific features of the hypothesis.
Hypothesis c.2 Tornado intensity is a function of prestorm helicity and vertical vorticity of the environment. [Wicker]
Test Measure prestorm helicity and vertical vorticity using profilers, NWS rawinsonde network, mobile balloon launches prior to storm development, and mesonet data. Compare to observed tornado intensity.
Refute No correlation.
Hypothesis c.3 The middle-level TVS is associated with, and precedes, the near-ground tornado. [Wicker]
Test Aircraft or WSR-88D Doppler data are used to establish the evolution and location of a TVS, and these data are compared to triangulated field observations of tornado location.
Refute The TVS and tornado do not share a common axis in space and time.
Hypothesis d.1 The debris generated by a tornado and lofted high into a storm returns to earth in a repeatable pattern. [Snow]
Test Detailed debris fallout mapping shows repeatable patterns of debris deposition.
Refute There is no repeated pattern of debris deposition in a sufficient sample of fallout mappings.
Hypothesis d.2 The pattern of return of debris to surface, both in the near- and far-fields, can be predicted using knowledge of vortex/storm dynamics and the meso- and cloud-scale wind fields. [Snow]
Test Construct conceptual, physical, and numerical models of the debris deposition process based on the data gathered in the field program, including debris deposition data as well as detailed data regarding the mesoscale and storm scale flow. Test the models against the actual cases to determine prediction value.
Refute The lack of prediction capability of the models would indicate that it is not yet possible to predict debris deposition.