Search Results

You are looking at 1 - 10 of 30 items for :

  • Author or Editor: A. S. Dennis x
  • Refine by Access: All Content x
Clear All Modify Search
A. S. Dennis

Abstract

Examination of radar records shows that snow trails frequently occur around the tops of showers. Evidence has been sought that showers are occasionally initiated by the seeding of supercooled cumulus clouds by snow. Certain observed radar patterns suggest strongly that showers were produced in this manner. As a check, plan diagrams with height contours have been synthesized and used to organize vertical sections obtained on different bearings. The constructed plan diagrams show showers occurring in regions of snow trails more frequently than elsewhere.

Full access
Martin R. Schock
and
A. S. Dennis

Abstract

Conditions within 16 cumulus congestus clouds were measured on aircraft penetrations near the −5C level and precipitation developments during the subsequent 45 min were noted. Significant correlations exist between the in-cloud conditions and the time of onset and intensity of precipitation. The degree of spatial correlation among the fields of temperature, vertical motion and water content within a given cloud proved more important to subsequent developments than did the mean values of these variables.

Full access
A. S. Dennis
and
M. R. Schock

Abstract

Silver iodide seeding of convective clouds in a number of experiments in South Dakota has yielded mixed results, including both increases and decreases in rainfall and, apparently, decreases in hail. The results vary with the synoptic situation and with location with respect to the silver iodide release point; there is tentative evidence that they can extend upwind. Two dynamic effects of seeding are postulated, namely, increased growth of seeded clouds and the suppression of neighboring clouds at distances of some tens of miles.

Full access
A. S. Dennis
and
Alexander Koscielski

Abstract

The height and temperature of first radar echoes observed on 49 test cases in a randomized cloud seeding project have been related to seeding treatment and to five cloud parameters, namely, cloud base height, cloud base temperature, maximum radar echo height, and the height and speed of maximum updraft as computed with a cloud model.

First echo height above cloud base in unseeded clouds averaged 11,000 ft and was strongly correlated with maximum updraft speed. First echo height in clouds seeded with silver iodide averaged 6900 ft. It was most closely correlated with cloud base temperature, suggesting that first echoes tend to appear at some fixed temperature. First echoes in salt seeded clouds appeared on an average 5300 ft above cloud base and showed no significant correlation with any of the cloud parameters mentioned.

It is concluded that both salt and silver iodide seeding can cause first echoes to appear closer to cloud base in unseeded clouds, with salt probably having a more pronounced effect in summertime convective clouds.

Full access
K. R. Biswas
and
A. S. Dennis

Abstract

Salt seeding took place below one end of a line of stratocumulus clouds with 350 lb of NaCl released. Cloud base was 9000 ft and cloud tops were at 15 to 18,000 ft above sea level. Cloud top temperature was near −2C and updraft speeds below the hue were near 3 m sec−1. The resulting shower was monitored by radar with the total rainfall being estimated at 280 acre feet. No rain fell from the unseeded portion of the cloud line or from any other clouds within 50 mi.

Full access
A. S. Dennis
and
F. G. Fernald

Abstract

The number per unit area of radar echoes from isolated showers, as a function of echo radius r, is found by analysis of data from several localities to he proportional to exp (−βr), where β varies with time and place inversely as the intensity of convective activity. Calculated values of β range from 0.8 (n mi)−1 for intense convection to 1.7 (n mi)−1 for light shower activity occurring in trade-wind cumuli.

Full access
A. S. Dennis
and
D. F. Kriege

Abstract

Ten seasons of commercial cloud seeding in Santa Clara County, California, are evaluated using target and control stations selected in 1955. Comparison of the linear regressions of target rainfall upon control rainfall, using seasonal totals, for the seeded winters and for ten immediately preceding ones gives evidence for a net increase in target precipitation. Examination of the rainfall distribution within the target area, which was not a part of the original evaluation plan, shows that seeding effects are concentrated downwind of the highest part of the Santa Cruz Mountains, which form the southwestern boundary of the target area, and suggests that they are limited to periods of convective instability.

Full access
A. S. Dennis
and
D. J. Musil

Abstract

A hailstone model is developed which simulates both wet and dry hailstone growth and partial or complete melting as the hailstone falls to the ground. The simulation is obtained by a consideration of the mass and heat budgets of the hailstone as a function of its size and of the environmental conditions. Competition among hailstone embryos is not treated in this model.

The hailstone model has been run in over 50 versions of a one-dimensional, time-dependent cloud model to derive estimates of maximum hailstone size from a wide variety of convective storms. The model suggests that hailstone diameter at the ground is determined closely by the strongest updraft encountered by the stone and the temperature at that level. In the model, most hailstones spend some time above the maximum updraft and significant growth occurs during descent from that level to the freezing level. A comparison of model predictions to hailstone observations in the Rapid City area during 1968 and 1969 shows fair agreement.

The model has been used to test the concept of hail suppression through artificial glaciation of cloud water and rainwater. It suggests that results would vary with temperature at the level of maximum updraft, but that artificial cloud glaciation would lead in a majority of cases to a reduction in maximum hailstone size at the ground.

Full access
A. S. Dennis
and
Alexander Koscielski

Abstract

A 3-year randomized crossover seeding experiment has been conducted in South Dakota to test effects of artificial nucleation upon supercooled convective clouds of spring and early summer. The associated rainfall observations have been analyzed by several statistical techniques. The principal conclusions are: 1) on days with isolated showers, rainfall has been heavier in the seeded target area than in the unseeded target area; 2) on days with widespread convective activity and southwesterly winds aloft, rainfall has been lighter in the seeded target area than in the unseeded target area, except for the regions 10–20 miles east of the Black Hills; and 3) on days with widespread convective activity and northwesterly winds aloft, rainfall has been lighter in the seeded than in the unseeded target area.

Full access
A. S. Dennis
,
Carol A. Schock
, and
Alexander Koscielski

Abstract

An extensive observational program has resulted in a description of the typical hailstorm of western South Dakota. The typical storm is cellular with new cells developing on its southwest side. The median hailstorm produces radar echoes to 12.2 km MSL and has a total up-air flux just over 1011 gm sec−1. Mean moisture inflow is around 109 gm sec−1, and precipitation efficiencies vary widely. Air and moisture fluxes in large storms exceed those in the median storm by a factor of 10. Maximum observed equivalent X-band radar reflectivity factor in the median hailstorm is 104.8 mm6 m−3 and occurs at ∼5.5 km MSL.

While there is evidence of occasional large steady-state cells characterized by very strong updrafts, hard hail and low precipitation efficiencies, most large storms pulsate as convective cells form and dissipate within them. Numerical models of hail-producing cells in the northern Great Plains should include time-dependent models capable of modeling the brake imposed on the updraft by precipitation.

Full access