AD-A237 024

Defense Nuclear Agency IAlexandria, VA 22310-3398

DNA-TR-90-71

Target Area Operatirng ConditionsDust Lofting from Natural Surfaces

R. A. GajR. D. SmallPacific-Sierra Research Corporation12340 Santa Monica BoulevardLos Angeles, CA 90025-2587

OTICJune 1991 ELECTE

Technical Report

CONTRACT No. DNA 001-87-C-0298

Approved for public release;distribution is unlimited.

91-02232 I!1111111LI111111ii!11!!!i!t11111

Destroy this report when it is no longer needed. Do notreturn to sender.

PLEASE NOTIFY THE DEFENSE NUCLEAR AGENCY,ATTN: CSTI, 6801 TELEGRAPH ROAD, ALEXANDRIA, VA22310-3398, IF YOUR ADDRESS IS INCORRECT, IF YOUWISH IT DELETED FROM THE DISTRIBUTION LIST, ORIF THE ADDRESSEE IS NO LONGER EMPLOYED BY YOURORGANIZATION.

0N4

/,LjiK q

DISTRIBUTION LIST UPDATE

This mailer is provided to enable DNA to maintain current distribution lists for reports. We wouldI appreciate your providing the requested information.

0J Add the individual listed to your distribution list. Please return te:iiglblfo0 Deetethe ite oraniztio/indvidal.the document so that any additions,

I ~ elee te ctedorgaizaionindvidalchanges, corrections or deletionsI can be made more easily.

0 Change of address.______________

INAME:

IORGANIZATION:______________________________

OLD ADDRESS CURRENT ADDRESS

I TELEPHONE NUMBER: )

ZISUBJECT AREA(s) OF INTEREST:

ZI

IDNA OR OTHER GOVERNMENT CONTRACT NUMBER'-____________

ICERTIFICATION OF NEED-TO-KNOW BY GOVERNMENT SPONSOR (if other than DNA):

I SPONSORING ORGANIZATION: _________________

I CONTRACTING OFFICER OR REPRESENTATIVE -______________ __

I SIGNATURE __ __ _ _ _ _ _ __ _ _ _ -- _ _ _ _ _ _ _ _ -

DirectorDefense Nuclear AgencyATTN: TITLWashington, DC 20305-1000

DirectorDefense Nuclear AgencyATTN: TITLWashington, DC 20305-1000

ROIForm ApprovedREPORT DOCUMENTATION PAGE OMB No. 0704-0188Public reporting burden for this collection of information is estimated to 'verAge I hour per response including the time for reviewing instructions. searching existing data sources.gathering and maintalning the data needed, and complebng and revwrr, the collecton of information Send comments regarding this burden estimate or any other aspect of thiscollection of information, including suggestions for reducing this bu~den to Washington Headquarters Services Directorate fo information Operations and Reports. 1215 JeffersonDavis Highway Suite 1204. Arlington. VA 22202-4302, and to the Office of Management and Budget. Paperwork Reduction Project (0704-0188). Washington. DC 20503

1 AGENCY USE ONLY (Leave blank) 2 REPORT DATE 3. REPORT TYPE AND DATES COVERED

1 910601 ITechnical 870929 - 9004294. TITLE AND SUBTITLE 5. FUNDING NUMBERS

Target Area Operating Conditions C - DNA 001-87-C-0298Dust Lofting from Natural Surfaces PE - 62715H

6. AUTHOR(S) PR - RATA - RG

R. A. Gaj, R. D. Small WU - DH044920

7 PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8 DERFORMING ORGANIZATION

REPORT NUMBER

Pacific-3terra Research Corporation12340 Santa Monica Boulevard PSR Report 1817Los Angeles, CA 90025-2587

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSORING/MONITORING

A C NAGENCY REPORT NUMBER

Defense Nuclear Agency DNA-TR-90-716801 Telegraph RoadAlexandria, VA 22310-3398SPWE/Gallaway

I11.SUPPLEMENTARY NOTESThis work was sponsored by the Defense Nuclear Agency under RDT&E RMC CodesB4662D RA RG 00122 SPAS 3420A 25904D and B4662D RA RG 10122 SPWE 4400A25904D.

12a DISTRIBUTION/AVAILABILITY STATEMENT 12b. DISTRIBUTION CODE

Approved for public release; distribution is unlimited.

13 ABSTRACT (Maximum 200 words)

We explore how variations in soil type, vegetation cover, and climatic conditions influence thesweepup mass in target regions. A simple dust sweepup/suspension model, appropriate forthe high wind speeds associated with nuclear blast waves, is developed to depend explicitlyon the threshold shear velocity required to initiate dust lofting. Given an analytic driver forthe positive phase free stream wind speed versus ground range and time, sweepup massesfor a wide range of surface types and conditions are calculated. We find that for an airburstat SHOB = 500 ft/KTI / 3, the sweepup mass can be reduced to near zero If the surface is cov-ered with tall grass or a mature small grain crop. For bursts over loose, unvegetated sand,sweepup efficiencies are nearly six times greater than for a typical Nevada Test Site surface.For a lower altitude alrburst (SHOB = 50 ft/KTI /3), a somewhat smaller variation betweenthese extremes is predicted (e.g., scouring is possible even over grass or cropland). The yielddependence of sweepup mass and the surface area scoured by the blast winds is also ex-plored. The results indicate that the net dust injection from a nuclear laydown can vary sig-nificantly within individual target areas and may be a strong function of season-especiallyin agricultural regions.

14 SUB.'ECT TERMS 15. NUMBER OF PAGESSweepup Soil Target Areas 38Scouring Threshold Velocity Vegetation 16. PRICE CODEDust Lofting Boundary Layers

17 SECURITY CLASSIFICATION 18 SECURITY CLASSIFICATION 19, SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACTOF REPORT OF THIS PAGE OF ABSTRACT

UNCLASSIFIED UNCLASSIFIED UNCLASSIFIED SAR

NSN 7540-280-550) Standard Form 298 (Rev 2-89)Piescibed by ANSI Slt 239.18298 102

UNCLASSIFIEDSECURITY CLASSIFICATION OF THIS PAGE

CLASSIFIED BY:

N/A since UnclassifiedDECLASSIFY ON;

N/A since Unclassified

SECURITY CLASSIFICATION OF THIS PAGE

UNCLASSIFIED

PREFACE

Fratricide probabilities are derived from model predictions of nuclear clouds. Exper-imental data are sparse and alternative validations are needed. Pacific-Sierra ResearchCorporation (PSR) has examined several key issues where uncertainties are large andrecommended validations for three such areas. They include: the influence of surfaceconditions on the sweepup mass of nuclear clouds; fireball quenching by entrainedmass; and long range cloud transport.

In Vol. 1 of this report series, smoke plumes and obscurations above target areas wereconsidered. Volume 2 considered long range dust transport and Saharan dust eventsas an analog for dispersion of nuclear clouds. Volume 3 recommends high energy ex-periments to simulate fireball-particle interactions.

In this volume, we develop an analysis for the mass entrained by nuclear clouds. Realsoil moisture and vegetation cover are accounted for and it is demonstrated thatsweepup mass in some target areas is considerably less than currently estimated.

This research was performed under contract DNA 001-87-C-0298 and monitored byDr. Charles R. Gallaway, Shock Physics Weapon Effects, Defense Nuclear Agency.

NT~Ij GRA&I

DTIC TABUnannounced E3j u s i i c a t i o n ---

- - -

ByDistributioc/ de-AvailabilitY Codes.

ii

CONVERSION TABLE

Conversion factors for U.S. customary to metric (SI) units of measurement

To Convert From To Multiply

angstrom meters (m 1.000 000 X E-10

atmosphere (normal) kilo pascal (kPal 1.013 25 X E+2

bar kilo pascal (kPal 1.000 000 X E+2

barn meter 2 (m 2 ) 1.000 000 X E-28

British Thermal unit (thermochemical) joule (J 1.054 350 X E+3

calorie (thermochemical) joule (J) 4.184 000

cal (thermochemical)/cm 2 mega Joule/m 2(Mj/m 2) 4.184 000 X E-2

curie giga becquerel (GBq)* 3.700 000 X E+ I

degree (angle) radian (rad) 1.745 329 X E-2

degree Fahrenheit degree kelvin (K) tK=(tof + 459.67)/1.8

electron volt Joule (J 1.602 19 X E-19erg joule (J 1.000 000 X E-7

erg/second watt (WI 1.000 000 X E-7

foot meter (ml 3.048 000 X E-l

foot-pound-force joule (J) 1.355 818

gallon (U.S. liquid) meter (m3) 3.785 412 X E-3inch meter (m 2.540 000 X E-2

jerk joule (J) 1.000 000 X E+9

joule/kilogram (J/Kg) (radiation doseabsorbed) Gray (Gy) 1.000 000

kilotons terajoules 4.183

kip (1000 lbf) newton (N) 4 448 222 X E43

kip/Inch 2 (ksil) kilo pascal (kPa) 6.894 757 X E+3

ktap newton-second/m 2 (N-s/m 2) 1.000 000 X E+2

micron meter (ml 1.000 000 X E-6

mil meter (ml 2.540 000 X E-5

mile (international) meter (ml 1.609 344 X E+3

ounce kilogram (kg) 2.834 952 X E-2

pound-force (lbf avoirdupois) newton (N) 4.448 222

pound-force inch newton-meter (N-m) 1. 129 848 X E-Ipound-force/Inch newton/meter (N/rm) 1.751 268 X E+2

pound-force/foot 2 kilo pascal (kPa) 4.788 026 X E-2

pound-force/inch 2 (psi) kilo pascal (kPa) 6.894 757

pound-mass (Ibm avofidupois) kilogram (kg) 4.535 924 X E-Ipound-mass-foot 2 (moment of inertia) kilogram-meter 2 (kg m 2) 4.214 'I I X E-2

pound-mass/foot 3 kilogram/meter 3 (kg/m 3 ) 1.601 846 X E+ l

rad (radiation dose absorbed) Gray (Gy)° 1.000 000 X E-2

roentgen coulomb/kilogram (C/kg) 2.579 760 X E-4

shake second (s) 1.000 000 X E-8

slug kilogram (kg) 1.459 390 X E+ I

torr (mm Hg. OC) kilo pascal (kPa) 1,333 22 X E-I

'The becquetel (Bq) is the SI unit of radioactivity; Bp = I event/s."The Gray (Gy) is the Si unit of absorbed radiation.

iv

TABLE OF CONTENTS

Section Page

PREFACE ....................................................... iii

CONVERSION TABLE ........................................... iv

FIG U RES ....................................................... vi

I INTRODUCTION ................................................ I

2 CLASSIFICATION OF SURFACE TYPES .......................... 2

2.1 THRESHOLD SHEAR VELOCITY ............................. 2

2.2 OBSERVED VARIATIONS .................................... 3

2.3 SEASONAL VARIATIONS .................................... 4

2.4 NUCLEAR INFLUENCES .................................... 6

3 SW EEPUP MODEL .............................................. 8

3.1 BACKGROUND .............................................. 8

3.2 HIGH SPEED MASS FLUX MODEL .......................... 8

4 SWEEPUP CALCULATIONS ..................................... 11

4.1 METHODOLOGY ............................................ 11

4.2 VARIATIONS IN SWEEPUP MASS .......................... 11

4.3 VARIATIONS IN SWEEPUP RADIUS . ...................... 13

4.4 NEVADA TEST SITE SOIL ..................... ............. 17

4.5 NORMALIZED SWEEPUP MASS ............................. 21

5 IMPLICATIONS FOR TARGET AREAS ............................... 23

6 CO NCLUSIONS .................................................... 25

7 LIST OF REFERENCES ......................................... 26

v

FIGURES

Figure Page

1 Threshold shear velocity as function of season for U.S. agriculturalarea- loam soil ..................................................... 5

2 Threshold shear velocity as function of season for U.S. agricultural

area- sandy soil ..................................................... 7

3 Blowing parameter as function of normalized shear stress ............ 10

4 Net sweepup mass as function of threshold shear velocity ............ 12

5 Net sweepup mass as function of burst altitude ........................ 14

6 Net sweepup mass as function of yield ............................... 15

7 Effective sweepup radius as function of threshold shear velocity ...... 16

8 Effective sweepup radius as function of burst altitude ................ 18

9 Effective sweepup radius as function of yield ........................ 19

10 Sweepup mass relative to airburst over Yucca Valley as functionof threshold shear velocity .......................................... 22

11 Soils in the Minot AFB, North Dakota ICBM silo area ................. 24

vi

SECTION 1

INTRODUCTION

The amount of dust and debris which sured threshold shear velocities, and (2)can be scoured and lofted into a nuclear a theoretical mass flux model appropri-cloud depends strongly upon surface ate for the high surface wind speeds as-conditions in the target area. Regional sociated with nuclear blast waves. Theseand seasonal variations in land use, two components are combined such thatvegetative cover, soil moisture, and soil the vertical dust flux has an explicit de-texture all influence the sweepup mass pendence upon the threshold shear ve-source and lead to wide differences in locity. Time- and range-dependentthe mass injection. The dependencies nuclear blast winds are prescribed usingare poorly understood and are not at analytic positive phase dynamic pres-present accounted for in any DNA sure formulae. Although our model issweepup model. As a result, broad un- highly idealized (it cannot, for example,certainties are introduced any time account for precursor flows or unevensweepup models are applied to target terrain), the results nevertheless demon-areas outside the region for which they strate the seriousness of ignoring region-were originally developed (namely, dry, al and seasonal variations in target areasandy deserts). This is a serious prob- conditions. Moreover, they indicate alem, especially considering the geograph- clear direction for more detailed theoreti-ic variety of strategic targets areas found cal and experimental work.in the Soviet Union (e.g., subarctic taigaforests, black soil farmland, wooded rivervalleys, and semiarid steppes). Sweepup The report is divided into six sections. Infrom those regions can hardly be ex- Sec. 2, we describe the surface classifi-pected to resemble sweepup from a des- cation scheme and discuss how thresh-ert; but currently, it is impossible to old velocities can vary with soil type anddistinguish the differences. season., Next, in Sec. 3, we describe the

sweepup model., This model is thenIn this report, we calculate swecpup applied in Sec. 4 to calculate the totalmasses (without accounting for material sweepup mass and the "effective scour-fallback to the surface) for a wide range ing radius" for various surfaces, heightsof surface conditions. Our approach has of burst, and yields. The implications oftwo main components: (1) A surface these results are discussed in Sec. 5,classification scheme based on mea- and conclusions presented in Sec. 6.

SECTION 2

CLASSIFICATION OF SURFACE TYPES

The blast winds generated by low alti- regulated airflow [e.g., Gillette, 1978. Astude nuclear detonations are strong the wind speed in the tunnel is in-enough to scour material from virtually creased from zero, the shear velocityany natural surface. Dirt, pebbles, bush-es-even rocks and trees-can be carried r IP/2 dUaway when subjected to winds in excess -. /P )of several hundred meters per second. dnZ

Yet, while it is clear some sweep up will is monitored until saltation (indicated byprobably always occur (especially at particles rolling or bouncing along theground ranges where dynamic pressui es surface) begins.' Here r. is the shear

maximize), it is equally clear that theamount of material swept up is not the stress, p is the air density, k -- 0.4 is thevon Karmann coefficient, and U Is the

same for all surfaces. Vegetated land, for horizontal wind speed at some height Zexample, yields less sweepup mass thanbarren ground; and moist (or frozen) soil above the ground. The value of U, atless than dry. The key issue is the ability which saltation first occurs defines theof a surface to resist the shear stresses threshold shear velocity U,.exerted by a nuclear blast wave and itsaccompanying winds. Any factor whicheither (a) lessens the stress applied di- The threshold shear velocity Uth canrectly to the soil (such as a plant cover), also be used to define the point at whichor (b) increases the cohesiveness of par- dust enters Into suspension. This regime,ticulates lying on the surface (such as a defined by the condition (Owen, 1964high moisture content) tends to lessenthe susceptibility of the soil to sweepup U,and thus reduce the net mass lofted. An . (2)explicit quantification of all those factors Uhas not been developed; however, the begins when the turbulent drag forcethreshold shear velocity can at least upn I n prt iles rst exceroughly categorize their net effect for a upon individual particles first exceedsgiven surface, their gravitational fall speed. It is at thispoint that the particles become lofted2.1 THRESHOLD SHEAR VELOCITY. and can be carried to high altitude. Sus-

pension Is the mechanism which createsThe threshold shear velocity is a mea- sand and dust storms in arid regions. Itsure of the minimum shear stress re- is also responsible for the dense sweep-quired to initiate particle motion along up layers generated by nuclear air-the ground. It is experimentally deter- bursts. Equation (2) therefore forms amined in the field using a portable, op- base for the sweepup model described inen-bottomed wind tunnel with a Sec. 3.

I. The relation between shear stress and vertical wind shear given by Eq. (1) is strictly valid only whenthe atmospheric surface layer Is statically neutral (well mixed). Corrections to this formula for otherstability profiles are available isee Businger, 19731.

2

2.2 OBSERVED VARIATIONS. example, tend to increase U~th in clay

The concept of using U. as an index for soils by strengthening their crustal

strength [Gillette, 19821.sweepup potential was originally devel-oped for application to "normal" (non- All of the soils listed in Table I were dry.nuclear) conditions such as dust erosion Technically, this means that the water

from drought-stricken agricultural land content in the first few inches below the

and deserts. Thus the only systematic surface was at or below the wilting pointmeasurements made to date have been for most crops. (The wilting point corre-

over dry, barren, or sparsely vegetated sponds to a water to soil volume ratio ofground, since such surfaces are most 0.05 for sandy loams, 0.08 for loam, andlikely to be subject to significant low- 0.17 for clay [Alderfer, 19771.) It has

wind speed sweepup. Nevertheless, even been found, however, that once the wa-

for this restricted range of surface types, ter content of a soil begins to exceed the

wide variations in U h have been found. wilting point (due to rainfall or irriga-tion), U th increases substantially aboveTable 1 summarizes some of these

varia-

tions. In general, soils which are loose the values listed in Table 1, although it(e.g., plowed or otherwise disturbed) or is not clear how much (Chepil, 1956.sandy are the most easily eroded (and Saturated or frozen soils are likely to be

thus have the lowest Uth ); loamy soils, even more stable, but, again, no mea-surements are available.

soils with a large number f large ele-ments such as pebbles and those with a Precipitation can modify the thresholdcrusted or cloddy surface are the least velocity in other, although less obvious,erodibie. In most cases, mineral content ways. Sandy soils which are initially dryalso influences soil erosion potential. and cloddy, for example, often becomeHigh concentrations of exchangeable so- more erodable several days to weeks af-dium or calcium carbonate (CaCO3), for ter a heavy rainfall. This is due to the

Table 1. Mean threshold velocities for unvegetated soils.

Threshold Percent MassShear Velocity in Particles

fm/s) >1 mm diameter Modulus ofCrustal Rupture

Soil Type Loose Cloddy Crusted Loose Cloddy (bars)

Sand 0.28 0.75 0.66 3 60 0.03Sandy loam 0.29 1.05 2.90 30 64 0.42Loamy sand 0.34 0.85 1.03 26 47 0.5Clay 0.54 >1.50 >2.00 42 94 0.75Silty Clay 0.56 ---- ---- 6 ..Silty clay 0.64 .... > 1.50 18 ..

loamClay loam 0.68 >1.09 1.20 28 81 0.38Loam 0.78 >1.50 >1.50 49 89 0.66Silt loam 1.08 >2.00 > 1.50 77 89 0.8Source: Gillette [1988.

3

tendency for rainfall to "melt" (disaggre- upper bound on U. above which sweepupgate) the clods lying on the surface. commences. This critical value likely de-Once the soil dries, it returns to a pends not only on the type of plant andsmooth, loose, and thus more easily ps stand density (plants per unit surfaceeroded state. On the other hand, if the rea), but also on the plant height, Itssoil is Initially loos, wetting and drying ter content, stem thickness, and aero-may ultimately lead to the formation of a dynamic cross section. However, for verycrust, thereby increasing U. Timing is dense plant covers such as small grains

an important issue here: If the surface is or grasses, we expect that the plantssubjected to strong winds very soon after would either be lodged (i.e., knockeda rainfall (or snowmelt), significant dry- over), broken, or uprooted long beforeing of the upper few millimeters may oc- this critical shear stress is reached. Thecur before the crust can form. Sweepup surface would no longer be as well pro-can then follow, even though the soil a tected and the critical shear velocitycentimeter or two below the surface is would drop to a value more representa-still water-soaked. This effect has been tive of bare soils. (Similar modificationobserved many times in loam and clay would result if the plant cover wereloam soils [Gillette, 19881. In a nuclear burned due to pre-shock thermal irradi-environment, thermal irradiation of the ation.) Unfortunately, no experimentalsurface may also dry the upper soil hori- data exists which relates high speedzon, possibly negating the stabilizing in- sweepup to shear streso over vegetatedfluencc of soil moisture. (This influence ground, although some measurements ofcould however be mitigated by a smoke, the critical bending moment for wheatdust, or steam layer just above the sur- and barley stem breakage have beenface which would tend to shield the made [Oda, Suzuki, and Odagawa,ground.) 19661. Lacking such data, we arbitrarily

assume that a dense plant cover resultsVery few measurements have been made in a threshold shear velocity of at least 3of the effects of live vegetation upon m/s. Clearly, experiments must be per-sweepup. It has been found, howeve:, formed to refine this value.that any sort of grass or small grain cropcover (e.g., wheat, barley, rye) is usually 2.3 SEASONAL VARIATIONS.sufficient to preclude sweepup for all butthe strongest natural winds All of the effects discussed above-the(U. ; 2 m/s) influence of vegetation cover, soil mois-

,Gillette, 19881.2 Even a ture, land use, and soil texture-imply

sparse vegetation cover with a fractional that the poteatial for sweepup (andlateral cover (frontal silhouette area di- thus U.,h ) must vary seasonally. Figurevided by ground area) greater about 2 to 1, derived from a database used to pre-4 percent can reduce the shear stress dict soil erosion in U.S. agriculturalupon the surface to near zero under nat- areas [Gillette and Passi, 19881, illus-ural wind conditions [Marshall, 19711. trates the expected variation U forNevertheless, there presumably exists an th

2. One exception occurs when a planted field lies immediately downwind from a barren, sandy (lowU.,,) surface from which sweepup Is occurring. Material from the upwind source can then spreadInto the field, effectively "sandblasting" the plants and the underlying ground surface, thus initiat-ing sweepup from where it would otherwise not be expected [Gillette, 19881. Obviously, such down-wind sandblasting effect could also be impo.-tant in target areas containing a mixture of vegetatedand bare (e.g.. fallow) surfaces.

4

6- Loam soil,

spring-planted wheat/barley

5-

-

/Moist year,3- moist previous2- Dry year,

2 / dry previous

i IilI m~llJ I I I qiiil01r

J F M A M J J A S O N D J

Month

6

Loam soil, fallow5

4

EMoist year,3- 3moist previous

0 Dry year,~dry previous,-~ ~ / ,._,...i

J F M A M J J A S O N D J

Month

Figure 1. Threshold shear velocity as function of season for U.S. agricultural area-loam soil.

5

loam soil as function of time of year for have a crucial influence upon sweepup.two land use types (fallow land and a For example, thermal radiation incidentspring-planted crop) and for two "clima- upon the surface prior to arrival of thetic" classes (corresponding to a pro- blast wave can heat the soil to a pointlonged moist period and a prolonged where bound water is explosively re-drought, respectively). A similar plot for leased, thus causing "popcornlng" ofsandy soil is shown in Fig. 2. Obviously, particles from the surface [Versteegen,the most important influence is the pres- Rault, and Hillendahl, 1989]. This effect,ence and maturity of the crop (in this which is most prevalent for clay soils,case, assumed to be a small grain cere- may disrupt surface crusts and loweral), which greatly Increases the 3,tability the threshold velocity of the soil. On theof the soil from shortly after germination other hand, thermal irradiation can alsoin the spring until harvest in late sum- melt or "glaze" the surface, thereby sta-mer. On the other hand, in spring, when bilizing the soil by leaving a thin butthe ground is plowed, crusts and clods presumably strong crust. Sandy sur-are broken, the threshold velocity is re- faces are probably the most susceptibleduced, and the soil becomes more vul- to glazing, although it has only rarelynerable to wind erosion. After harvest, been observed, most notably followingthreshold velocities remain moderately the TRINITY test in 1945 [P. Verstecgen,high due to the common practice of leav- personal communication, 19891.ing a vegetative residue or stubblethrough the winter. In drought years, Blast effects can also modify the dusthowever, insufficient moisture is avail- producing potential of the surface. Air-able to support this residue and clodding blast loading of subsurface air pores canand crusting of the surface is inhibited, cause spalling and disruption of an un-Threshold velocities therefore remain broken surface with the passage of thelow. blast wave. The blast can also uproot

trees and bushes, which would break2.4 NUCLEAR INFLUENCES. the surface and reduce the threshold ve-

locity, as well as increasing the shearThreshold velocity measurements to date stress upon the ground. Models relatinghave focused only on conditions likely to blast effects to tree blowdown do existbe encountered under normal condi- [Morris, 19731; however, no quantifica-tions. In a nuclear environment, howev- tion of its effect upon soil lofting is yeter, several other factors may prove to available.

6

6Sandy soil,Spring plantedwheat/barley - -

4 /

_E - Moist year,moist previous

Dry year,dry previous

0/

J F M A M J J A S O N D J

Month

6-

Sandy soil, fallow5-

4

E Moist year,,3 - moist previous

2- rydry previous

J F M A M j J A S O N D J

Month

Figure 2. Threshold shear velocity as lunction of season for U.S. agricultural area-sandy soil.

7

SECTION 3

SWEEPUP MODEL

In the previous section, we reviewed very few measurements of F in this re-many of the links between surface condi- gime, especially at the very high windtions and the potential for dust lofting. A speeds characteristic of nuclear blastgeneral model would account for soil waves. In one experiment [Hartenbaum,type, texture, moisture, temperature and 19711, a value P = 1.0 was found overvegetation. In this first analysis we an uncohesive sand (mean particle ra-roughly account for all parameters usinga single quantitative measure-the dius 1251im; U~th 0.35 m/s ) surfacethreshold shear velocity. The sweepup for free stream wind speeds U betweenmodel we develop is clearly an approxl- 34 and 115 m/s (corresponding to shearmat ion, but nonetheless accounts for velocities in the range 5.0 __ U. -5 18.6real soils In calculating the amount ofdust leaving the surface., m/s). This result cannot be easily ex-

plained by simple dimensional analyses.3.1 BACKGROUND. A model which accounts for the effect of

dust loading on the airstream at verySweepup models relate the vertical mass high wind speeds is required.flux F of dust particles leaving the sur- 3.2 HIGH SPEED MASS FLUX MODEL.face to the shear stress T. = p U. 2 ex-

erted on the ground by a sheared For wind speeds In the regime(rotational) wind field. The general form U, ;> 10 U.th Ithe vertical mass flux isis related to the surface shear stress by

(Mirels, 1984):F = a ( U, - U .1h ) ( 3 )F

2 t n(1 +B)(4

where U, > U.t h and a and P3 are empir- Uo

ical coefficients. At low wind speeds, di- where Po and U0 are the density andmenslonal analysis suggests that F horizontal wind velocity In the freeshould vary with the kinetic energy de- stream (i.e., at the top of the dusty sur-livered to the surface [Gillette, 19801. face layer) and B is a dimensionlessThus we expect /3 = 3; this dependence blowing parameter which expresses the

effect of transverse particle injection onhas indeed been borne out by observa- the local shear stress In plane paralleltions of natural sweepup from desert flow. Assuming that the mass loading ef-surfaces [Shinn et al., 1976; Westphal, fectively reduces the shear stress to theToon, and Carlson, 19871. However, as threshold value required to maintainwind speeds rise arid dust concentra- dust lofting, B is given by the implicit re-tions increase, the lofted particles begin lationto act as a momentum sink on the air-flow. Thus the shear on the surface is 2reduced, the lofting ability of the wind Un(l + B) = B (5)diminished, and F is no longer propor-

tional to U. 3. Unfortunately, there are

8

which, In the suspension regime the "rough plate" formula [Schlichting,(U. > 10 U.th), can be approximated 19581:

U, = - [2.87 + 1.58 log(x/Ks)]-1 .25

kn(1 + B) = 3.92 (6) (7)U.th

where x is the distance behind the shockwave, K. . 0.05 m is the roughnessheight, and x > 100 Ks. Finally, substi-

The functional dependence of B on the tuting Eqs. (6) and (7) Into Eq. (4), we ob-shear stress ratior./r.th given by Eqs. (5) tain [see Eq. (8) below]

and (6) is illustrated in Fig. 3. The sweepup mass flux is therefore in-versely (but weakly) dependent upon

To prescribe U0 as a function of time, threshold velocity. This expression com-

groud prngeibel, and hfn ti of bu, pares well with the experimental result Fground range, yield, and height of burst, c Uo1 .' 44 derived by Hartenbaum [ 1971J]

we use the ideal airblast approximations deuetly u atebasi for

given by Brode (1987). These approxima- and subsequently used as the basis for

tions are valid only during the positive many nuclear sweepup models

overpressure phase of the blast; we [Schlamp, Schuckman, and Rosenblatt,

therefore assume all the sweepup occurs 1982; Bacon, Dunn and Sarma 1988.

during the positive phase. The free- However, unlike previous models, the de-

stream velocities are then used to com- pendence of sweepup on threshold veloc-ity (and hence on surface type) is explicit

pute the shear velocity U, according to in Eq. (8).

1.80 po 1.240.24 [2.87 + 1.58 log(x/Ks)] 2.80U *th

9

10~6

10 Q5

10 _~

E 0Analytic~ /

r- 102 /0

101 /

100

1O0 II100 101 102 1014 0

Normalized shear stress, U.2/Uth 2

Figure 3. Blowing parameter as function of normalized shear stress.

10

SECTION 4

SWEEPUP CALCULATIONS

We now use the mass flux model and no sweepup occurs (AMMAX = 0). Thusideal blast wave driver to compute the RMAx-i = RE defines the effective sweep-variation in net sweepup as a function of up radius. The total sweepup mass M isthreshold shear velocity. Two quantities then found by simply summing over theare of interest: (1) The integrated sweep- rings:up mass M; and (2) the effective sweepupradius RE, defined as the the maximum IMAXrange (from ground zero) at which M = AM . (10)sweepup can occur for a given yield, i=lHOB, and surface type.

Note that M is a measure of the initial4.1 METHODOLOGY. sweepup; it does not account for mass

fallback to the surface, nor does it in-We treat the surface as fiat, homoge- elude mass lofted during the negative

neous (i.e., no spatial variations in phase.

threshold velocity), and thermally ideal.

Assuming cylindrical symmetry about 4.2 VARIATIONS IN SWEEPUP MASS.the burst point, the incremental dustmass AMi lofted from a ring of width AR We have computed the variation in totaland radius Ri is sweepup mass M for threshold shear ve-

locities from 0.20 to 4.0 m/s, scaledIT + D heights of burst (SHOBs) from 20 toAMi = 2,T RiAR Fidt (9) 1000 ft/KT" /3 , and yields (W) from 500

to 1000 KT. 3 Figure 4 shows the depen-dence of M on U~t for a 500 KT detona-

where T is the blast wave time of arrival dth

at Ri, Du is the dynamic pressure posi- tion at three burst altitudes: 397 fttive phase duration, and F, is the local (SHOB = 50 ft/KTI 3), 1984 ft (SHOB =sweepup mass flux (Eq. 8). We set ARi, 250 ft/KT"/3), and 3969 ft (SHOB = 50060 m and use empirical formulae to de- ft/KT"13). We find that M decays expo-termine T and D,, as functions of height nentially with increasing U*,h , and isof burst, yield, and range [Brode, 1987). most sensitive to variations in thresholdTo simulate the low windspeed cutoff for velocity at low U.ih -i.e., over dry, loose,dust suspension, we further assume F, =0 for U, < 1U h. unvcgetated soils. In this regime, small

changes in surface type can produce

Equation (9) is integrated numerically large changes in sweepup. For example,

using the trapezoidal rule and a timestep we find that a 50 ft SHOB detonation

At = D/100 for= 1 toi= IMAX where over loose sand (U*,h = 0.28 m/s) raises

IMAX corresponds to the first ring where over two times as much mass as an

3. The burst altitude SHOB = 20 ft/KT"' 3 roughly corresponds to the minimum for non-cratering air-bursts [Rosenblatt, 19811.

II

cImCL

M E

-44

Co 0 0C) (n/0

U(I

LO -~ L

II

I::I!~ 3n) 000Q)4

l>.Cu *

0.5

-oo

000 ~00

(1)V 'ssew dndeemS

12

identical burst over loose loam (U. = tive to the results in Fig. 5 for allSHOB h 650 ft/KT1 3.0.78 m/s). If the burst occurs at

an even

higher altitude (e.g., 500 ft SHOB), the Sweepup also increases with yield. Fig-difference is more pronounced. For high- ure 6 shows the variation in M for fourly non-erodible soils (Uth > 2.0) or soils surface types and two burst altitudes (50

which are densely vegetated and 250 ft SHOB) for yields between 500KT and 1 MT. The calculations show a(U 3linear dependence on yield in this range,

pressed, especially at higher burst alti- with sandy surfaces being the most sen-tudes. Indeed, we find that when a 500 ft sitive to changes In W and desert pave-SHOB detonation occurs over a surface ment the least. Thus the sweepupwith U.th > 3.5 m/s (characteristic of an efficiency (defined as soil mass lofted perearly to mid-summer wheat crop-see megaton of weapon yield) is yield depen-Figs. t and 2), no sweepup occurs at all. dent. For example, a 500 KT burst at 50

ft SHOB produces sweepup efficiencies

ranging from 0.12 Tg/MT for desert

The strong dependence of sweepup mass pavement to 0.33 Tg/MT for sand. Rals-

on burst altitude Is Illustrated in Fig. 5. Ing the yield to I MT, however, nearly

Here we have plotted M as a function of doubles these values to between 0.23

SHOB for a 500 KT burst over four sur- and 0.64 Tg/MT. Slightly lower efficien-

face types ranging from loose sand to a cies result at all yields when the burst

desert pebble pavement (mass equivalent occurs at a higher altitude (250 ft SHOB)modal pebble diameter 1.0 cm) similar to over the two least erodible surfaces

that found in Yucca Valley at the Nevada (loam and desert pavement), while higher

Test Site (NTS)., We find that the depen- efficiencies are obtained over the most

dence of M on SHOB is highly nonlinear, erodible ones. This is consistent with

with the largest variations occurring over Fig. 5.

the most erodible soils. In particular, we 4.3 VARIATIONS IN SWEEPUPfind that sweepup mass at first decays RADIUSrapidly with increasing SHOB, reaching RADIUS.a relative minimum around 80 to 100 ft The variations in sweepup mass dis-SHOB. With further increases in burst cussed above can be attributed to a com-altitude, however, the sweepup mass be- bination of two factors: (1) Variations ingins to rise-indicative of the build-up in positive-phase dynamic impulse inte-blast dynamic pressure with Increasing grated over the sweepup area, and (2)SHOB for shocks waves in the Mach re- variations in the size of the sweepupflection regime. The positive dependence area itself. Sweepup area can be charac-of M on SHOB continues until an alti- terized by the effective sweepup radiustude is reached at which the Mach wave RE. Figure 7 shows the dependence of REbegins to weaken-typically between 300 on threshold shear velocity for threeand 400 ft SHOB. For bursts above this burst altitudes. Comparing these resultsaltitude, sweepup again diminishes, with Fig. 4, it is clear that the sweepupeventually dropping to zero. Over desert mass M depends strongly upon RE-bothpavement, the cutoff occurs at SHOB =760 ft/KT"' 3 (HOB = 6032 ft); over loam, decay exponentially with increasing U,th'it is at SHOB = 935 ft/KT 1 3 (HOB = and both display the greatest sensitivity7421 ft). These results are modified to threshold velocity over bare, loose soil.somewhat over non-ideal surfaces due to (Not all the dust to radius RE is lofted toenhanced velocities in the precursed stabilization; but all is raised from thewave. In particular, M is increased rela- surface and is directed toward the

13

300

W =500 KT

250

U-th = 0.28 rn/s(loose sand)

~.200

E 150C0.t 04 /

4)

W 100 1h=07ml

50 (desert pavement) N

01 11 I0 200 400 600 800 1000

SHOB (ft/KT/3)

Figure 5. Net sweepup mass as function of burst altitude.

14

350 - SHOB = 50 ft/KT"13

-- -SHOB =250 ft/KT 1/3 U-t U.= 0.28 rn/s

300- 0

2 5 0 -.040U 'th = O .4 5 rn /s

UiCn

E

o150 U 'th= 0. 78 rn/s

100 - 0 0000 00U-th= 1.36 rn/s

500 600 700 800 900 1000

Yield, W (KT)

Figure 6. Net sweepup mass as function of yield.

15

00

Z U)

0

co o ' 0II - 0

0i

0 00 f00LO 0L00 IU

o CJ~t

Iw)3j sie dnem AIII3

I * 16

pedestal.) Unlike the sweepup mass, the soil has a two-layer structure: A thinhowever, there is no "crossover" with in- desert pavement veneer consisting ofcreasing threshold velocity; the sweepup large pebbles (tens of millimeters to aradius for a 500 ft SHOB burst is always centimeter or two in diameter), overlyinglarger than for detonations at lower alti- a deeper layer of finer material-typicallytudes (at least until the cutoff at U,, h - sand or sandy loam imbedded with grav-

el. This desert pavement is usually suffl-3.5 m/s is reached). This is as we would cintopentwdeosnofheieexpet fr aMachref-ctd shck.It im- cent to prevent wind erosion of the fine

expect for a Mach reflected shock. It sim- particles for all but the strongest naturalply indicates that as bursts occur at winds; in a nuclear environment, howev-greater heights above the surface, more er, it is easily removed, as has been ob-of the kinetic energy is converted into served following bursts over Yucca Valleyhorizontal dynamic pressure. The rela- [Lamar, 19621. It therefore seems likelytionship between RE on SHOB is illus- that nuclear sweepup over desert pave-trated in more detail in Fig. 8. Unlike the ments may proceed in two stages: A firstsweepup mass M (Fig. 5). we find that RE stage during which only the large par-varies linearly with SHOB. ticles comprising the top layer layer are

removed, followed by second, more vigor-Finally, in Fig. 9, we show the variation ous stage during which the now-exposedin RE as a function of yield. Again, the underlying soil is scoured. Thus, moredependence is linear, with the greatest mass may ultimately be lofted than ifsensitivity found for the most erodible only the heavy top layer were present.soils. However, when expressed in terms

of sweepup area (,-r RE 2) , this sensitivity To test whether this two-stage processcan significantly increase the net sweep-

disappears. For example. from Fig. 9, we up mass, we have modified our model tofind that a I-MT burst always sweeps an account for the presence of desert pave-area about 1.6 times larger than a ment over a fine soil. We assume that500-KT detonation at the same altitude, the surface is initially covered with aregardless of soil type. single layer of pebbles with mass modal

diameters Dm = 1 cm. To this layer we4.4 NEVADA TEST SITE SOIL. assign a threshold shear velocity U.ih =

1.36 m/s. This value is consistent withAll of the results presented so far have the empirical relation for dry, disturbedimplicitly assumed that the soil proper- desert soils (Gillette et al., 19801ties are independent of depth. Thus wehave assumed that the threshold shear U = 0.43 + 0.93 Dm (11)velocity remains constant throughout Uth 0 9the period sweepup is occurring. For where Dm is in centimeters. Assumingmany agricultural soils, which have a the particles are spherical and have adeep, well-mixed upper layer of topsoil, mass density 2.0 x 103 kg/iM3 , we calcu-this is an appropriate assumption. But late the areal density of the pebble layermany natural soils are not so homoge- to be 10.5 Ap kg/m 2, where 0 < A :5 1.0neously structured. The Yucca Valley re- is the area fraction of ground actuallygion of the NTS is an example. Soils in covered by pebbles. (We estimate Al l

most of th4s area are representative of"wind-stabilized" alluvium from which 0.3 for Yucca Valley.) Setting U.th = 1.36

most of the erodible (i.e., small) surface m/s in Eq. 8 and using Eq. 9. the time-elements have long since been removed dependent pebble mass removed fromby natural wind processes. As a result, each concentric ring around ground zero

17

10

0.28 (ose

0.h .5 o

Uw 4 0.78 (loose loam)

Ui U-th = 1.36 rn/s (desert pavement)

0

SHOE (ft/K -p,) 1 0

Figu r 8. ffective sw eePup radius as fun ct i0 , Of burst altitude .

10SHQ8 : 5 Oft/KT 1/3

U-th 0.28 fr/sS6

&L.M.N'N .U

UAMMM th 0.5~ rnLU

*O 0

W000

piur 9. ffectv sweePuP radlius~ as functonofYild

19~ecI

is calculated. If, before the sweepup (3) an "infinitely deep" pebble layer forphase ends, we find the total mass re- which U h remains fixed at 1.36 m/smoved has exceeded the critical density10.5 Ap kg/iM2, we assume that the throughout the calculation. The results

pebble layer has been completely re- are presented in Table 2. We find desert

moved from that ring. We then "instanta- pavement to be surprisingly more stabi-

neously" reset U. to a lower value to lizing than expected. Only when the

th pavement overlies loose sand and issimulate the exposure of the underlying sparsely distributed (Ap = 0.3) do we findfine particles, and proceed with the cal- an enhancement in sweepup mass in ex-culation until sweepup ends. cess of 10 percent. This is in spite o.' the

fact that we compute complete removalWe have computed the net sweepup of the pebble veneer (for A - 0 3) out tomass for a 50 ft SHOB burst at two approximately 425 ft/KT1IP range ( - 1.4yields (500 and 1000 KT) over three dif- km for a 1-MT burst). This is equivalentferent desert pavement configurations: to nearly one-quarter the entire sweepup(1) A single pebble veneer over highly area. However, the most significanterodible sand (U.th = 0.28 m/s), (2) a scouring of the underlying soil is re-

single pebble veneer over an Intermedi- stricted to ranges much closer to groundzero; thus the net sweepup mass Is pro-ately erodible soil ( Uth = 0.45 mis), and potoaeyrdc.

th portionately reduced.

Table 2. Sensitivity of sweepup mass to desert pavement cover for 50 ft SHOB burst.

Sweepup Mass, M(kt)

Threshold ArealShear Velocity Pebble Pebblesof Underlying Yield, W Cover, Pebbles Overlying Difference

Soil (m/s) (KT) Ap Only Fine Soil (%)

0.28 500 0.3 60.09 65.77 +9.420.5 60.09 63.67 +5.921.0 60.09 62.31 +3.66

1000 0.3 117.20 130.30 +11.180.5 117.20 125.70 +7.251.0 117.20 122.40 +4.44

0.45 500 0.3 60.09 63.90 +6.340.5 60.09 62.36 +3.781.0 60.09 60.10 +0.02

1000 0.3 117.20 126.20 +7.680.5 117.20 122.70 +4.691.0 117.20 120.40 +2.73

20

These results would be drastically differ- derlying soil; the results are indicated inent for non-ideal surfaces when thermal the figure. (Our choice U.th = 0.45 m/s islayers and precursors are accounted for.

The resultant elevated dynamic pressur- somewhat arbitrary; it probably underes-

es would likely remove the pebble veneer timates the dust pickup at the NTS since

much sooner and to a greater distance some of the desert pavement was broken

from ground zero. The net sweepup by pre-shot construction and traffic.)

mass would therefore be higher. This is These results emphasize that, comparedthe observation of to most dry barren or semi-barren dryLamar (1962), who found near complete surfaces, Yucca Valley soil is not as

removal of the desert pavement out to at readily eroded. This is due to the stabi-

least 600 ft/KTI / 3 from ground zero fol- lizing effect of the desert pavement

lowing shot SHASTA (W = 16.5 KT, pebble cover. Most other bare soils-for

SHOB = 196 ft/KTI / 3) in Yucca Valley. example, agricultural soils during a dry

Separate evidence, however, indicates winter or immediately after plowing orthat this shot did not produce a strong harvest-actually produces more sweep-precursor sLiner et al., 1975. up than the Yucca Valley surface. Forprecursor NORMAInE et aASS. 1example, for detonations at 50 to 500 ft

4,5 NORMALIZED SWEEPUP MASS. SHOB over bare loam surfaces (Ut h =

Figure 10 shows the variation in sweep- 0.78 m/s), between 1.3 and 2.2 timesup mass as a function of threshold ve- more dust would be raised. On the otherlocity normalized by that calculated for hand, if the surface were heavily vege-Yucca Valley, NTS. To compute the nor- tated (U > 3.5 m/s), the sweepup ismalization factor MNTS, we adopted the th"two-layer" model described above with reduced to anywhere from zero to aboutAp = 0.3 and U.t h = 0.45 m/s for the un- 50 percent of its NTS mass.

21

0

Jo

0 T-- I9 LO 0U') C~ LO 4

e rl

0 I0

-0-00

I o LO CJLI) ISi "/I-wdd em ~ILwO

22,

SECTION 5

IMPLICATIONS FOR TARGET AREAS

Surfaces in target areas vary widely re- - A crop or grass cover may be suffi-flecting differences in soil type, topogra- cient to completely suppress sweepupphy, moisture levels, and vegetation from a 500 KT airburst if the SHOBcover. Most real target surfaces do not is greater than 500 ft/KT" /3 (Fig. 4).resemble the Nevada desert, nor do they Thus late summer is the least vulner-remain invariant throughout the year. able time of year for sweepup; this isSmall differences in surface properties when crop and grass densities arecan lead to large changes in dust sweep- greatest.up.

e There exists an optimum burst alti-The real soil sweepup model we devel- tude between SHOB = 80 to 100 ft/oped used a single parameter, the KT / 3 at which the sweepup mass isthreshold velocity, to account for soil minimized over bare soils. Burststype, moisture, and vegetation cover, detonated below this altitude or be-Clearly this is a simplification, but one tween 300 and 400 ft/KTI / 3 tend tothat nonetheless accounts for exper- maximize dust production (Fig. 5).imental data on low-speed scouring ofreal soils. In extending this analysis to For a fixed SHOB, sweepup massnuclear sweepup, we recognize that sev- Ten to ith Isgeral potentially important physical pro- yield. The sensitivity to yield is great.cesses are treated inadequately or not at est for the most erodible soils (Fig. 6).all. Inclusion of (a) non-ideal shock ef-fects, (b) blast and thermal modification The effective sweepiup radius in-of surface properties, (c) uneven terrain creases with increasing burst altitudeeffects, and (d) post-positive phase blast and yield, and is most sensitive towinds would no doubt alter the results surface conditions over highly erod-somewhat. We anticipate that each of ible soil (Figs. 7-9).these processes modifies the real soil * Bare, dry agricultural soils (com-corrections we calculate, nevertheless, posed primarily of loam) produce upthe conclusions indicatc: to 220 percent more sweepup mass

than a desert pavement surface typi-*Targets located on cropland have the cal of the NTS (Fig. 10).

greatest dust producing potential,since the surface tends to be bare at Our results also imply that sweepupleast part of the year-especially dur- may not be uniform even within a singleing the spring plowing and immedi- target area. Intercontinental Ballisticately after harvest in the autumn Missile silo fields are quite large, typical-(Figs. 1 and 2). Moreover, some por- ly covering an area on the order and 104

tion of the land remains bare (i.e., km 2 in the U.S. and somewhat less infallow) all year around. For example, the Soviet Union. Surface conditionsin U.S. target regions, typically one- within an area so large are not necessar-fourth the area devoted to crops is ily uniform. A burst over one portion of afallow at any given time [North Dako- silo field could yield a much differentta Agricultural Statistics Service, sweepup amount and scour an area con-19881. siderably larger or smaller than a burst

23

in another portion of the field. Fratricide 0.78 m/s, our results indicate a maxi-probabilities could thus be highly vari- mum sweepup efficiency of about 0.16able. Tg/MT for a 500-KT low altitude burst,

Figure I I depicts a U.S. target area. ranging up to = 0.30 Tg/MT for a I-MT

Aside from isolated urban areas and bo- detonations. In addition, we calculate

dies of water, most of the land shown maximum sweepup radii between 3 and

here is either planted (primarily with 4 km from the burst point-roughly one-

wheat, oats, and barley) or is tall-grass third to one-half the mean spacing be-

wildland. Yet the soil upon which this tween silos. Smaller sweepup radii and

vegetation grows is not the same across lower efficiencies are to be expected from

the entire region; roughly half the silos strikes against those silos located on

are located on loamy soil which is well- moist or poorly drained soils, or follow-

drained and hence dry (except for peri- ing attacks during less vulnerable times

ods immediately after a rainfall or in of year such as summer or when the

irrigated zones). Thus, assuming U~th surface is snow-covered or frozen.

Scale: J Moist loam1 c 14 km R 4I VIIE2 Fine i< 100 urn)

Deeo (- 20 cm)E _'Well-drained loam

SDI * eepf20cmt

I=Poorly.drained loamRockyShallow K< 20 cm)

ete sand

Source: Aandahl [1982]; Omodt et al. [1968].

Figure 1 1. Soils in the Minot AFB, North Dakota ICBM silo area(dots indicate silo locations).

24

SECTION 6

CONCLUSIONS

We find that the threshold shear velocity completely suppress sweepup. By ex-provides a useful quantitative measure trapolating our results to real soil condi-of the susceptibility of a surface to nu- tions in target regions, we expect evenclear sweepup. Wide variations in this larger variations in sweepup mass couldparameter occur naturally with seasonal result under attack conditions. We findand regional differences in vegetation that sweepup is minimal as long as thecover, soil moisture, soil type, and land vegetation cover remains intact; at highuse. We find that such variations can overpressures or high thermal fluxes,lead to large changes in the initial the stabilizing influence is diminished.sweepup mass raised by a low altitude The threshold velocities at which plantairburst-particularly over surfaces lodging, breakage, and uprooting occurwhich are dry and minimally vegetated, are not yet established, but it is never-When vegetation is present, the surface theless clear that sweepup is greatly re-is strongly stabilized; in fact, a dense duced.agricultural cover can in some cases

25

SECTION 7

LIST OF REFERENCES

Aandahl, A. M., Soils of the Great Plains, and Natural, Sources and Transport, NewUniversity of Nebraska Press, Omaha, York Academy of Sciences, New YorkNebraska, 1982. City, 1980.

Alderfer, R. B., "Soil," McGraw-Hill Ency- Gillette, D. A., "Threshold Friction Velo-clopedia of Science and Technology, Vol. cities for Dust Production for Agricultur-12, McGraw-Hill, Inc., New York City, al Soils," J. Geopyhs. Res., Vol. 93, 1988,1977. pp. 12645-12662.

Bacon, D.. P., T. P. Dunn, and R. A. Sar- Gillette, D. A., et al., "Threshold Veloci-ma, Late-Time Cloud Modeling, Vol. 1: ties for Input of Soil Particles into the AirCases N1-N6, Science Applications Inter- by Desert Soils," J. Geophys. Res., Vol.national Corporation, McLean, Virginia, 85, 1980, pp. 5621-5630.SAIC-88/1856, 1988. Gillette, D. A., et al., "Threshold Friction

Bagnold, R. A., The Physics of Blown Velocities and Rupture Moduli for

Sand and Sand Dunes, Methuen and Crusted Desert Soils for the Input of Soil

Co., Ltd., London, 1941. Particles Into the Air," J. Geophys. Res,,Vol. 87, 1982, pp. 9003-9015.

Brode, H. L., Airblastfrom Nuclear Gillette, D. A. and K. J. Hanson, "SpatialBursts-Analytic Approximations, Pacific- and Temporal Variability of Dust Produc-Sierra Research Corporation, Los An- tion Caused by Wind Erosion in thegeles, California, PSR 1419-3, 1987. United States," J. Geophys. Res., Vol. 94,

Businger. J. A., "Turbulent Transfer in 1989, pp. 2197-2206.

the Atmospheric Surface Layer," in D. A. Gillette, D. A. and R. Passi, "ModelingHaugen (ed.), Workshop on Micrometeorol- Dust Emission Caused by Wind Ero-ogy, American Meteorological Society, sion," J. Geophys. Res., Vol. 93, 1988,Boston, Massachusetts, 1973. pp. 14233-14242.

Chepil, W. S., "Influence of Moisture on Glasstone, S. and P. J. Dolan, The Ef-Erodibility of Soil by Wind," Soil Sci. Soc. fects of Nuclear Weapons, Third Edition,Amer. Proc., Vol. 20, 1956, pp. 288-292. United States Department of Defense

and Energy Research and DevelopmentDorrance, W. H., Viscous Hypersonic Administration, Washington, D.C., 1977.Flow, McGraw-Hill, Inc., New York City,1962., Hartenbaum, B., Lofting of Particulates

by a High Speed Wind, Defense NuclearGillette, D. A., "Tests with a Portable Agency, Washington, DC, DNA 2737F,Wind Tunnel for Determining Wind Ero- 1971.sion Threshold Velocities," Atmos. Envi-ron., Vol. 12, 1978, pp. 2309-2313. Lamar, D. L., Preliminary Study of the

Distribution of Desert Pavement in Rela-Gillette, D. A., "Major Contributions of tion to Blast Areas at the Nevada ProvingNatural Primary Continental Aerosols: Grounds, RAND Corporation, Santa Mo-Source Mechanisms," in T. J. Kneip and nica, California, Report D-10270-PR,P. J. Lioy (eds.), Aerosols: Anthropogenic 1962.

26

Liner, R. T., ct al., Nuclear Precursor Phe- Rosenblatt, M., Introduction to Nuclearnomenology and Sweep-Up Dust Cloud Dust/Debris Cloud Formation, DefenseModel Development, Defense Nuclear Nuclear Agency, Washington, DC, DNAAgency, Washington, DC, DNA 3781F, 5832T, 1981.1975.

Schlamp, R. J., K. L. Schuckman, andMarshall, J. K., "Drag Measurements in M. Rosenblatt, Sweep-Up Layer Charac-Roughness Arrays of Varying Density terstics After a I Megaton Burst at HOBand Distribution," Agric. Meteor., Vol. 8, = 1000 Feet, CRT 5360-IT, California1971, pp. 269-292. Research and Technology, Woodland

Hills, California, 1982.Mirels, H., "Blowing Model for TurbulentBoundary-Layer Ingestion," A/AA J., Vol. Schlichting, H., Boundary Layer Theory,22, 1984, pp. 1582-1589. McGraw-Hill Book Co., New York, New

York, 1968.Morris, P. J., Forest Blowdown from Nu-clear Airblast, Defense Nuclear Agency, Shinn, J. H., et al., "Observations ofWashington, DC, DNA 3054F, 1973. Dust Flux in the Surface Boundary Lay-

er for Steady and Nonsteady Cases," inNorth Dakota Agricultural Statistics Ser- R. J. Englemann and G. A. Sehmelvice, North Dakota Agricultural Statistics, (eds.), Atmosphere-Surface Exchange of1988, Report No. 57, North Dakota Agri- Particulate and Gaseous Pollutantscultural Statistics Service, Fargo, North (1974), Technical Information Center,Dakota, 1988. U.S.Dcpartment of Energy, 1976.

Oda, K., Suzuki, M., and T. Udagawa, Versteegen, P., D. Rault, and R. Hillen-"Varietal Analysis of Physical Characters dahl, Effects of ICBM Basing, Vol. 5: Ther-in Wheat and Barley Plants Relating to mal Layer Experiments and ModelLodging and Lodging Index," Bull. Nat. Development, Defense Nuclear Agency,Inst. Agri. Sc., Series D, No. 15, 1966, Washington, D.C., DNA-TR-87-171-V5,pp. 55-91, (in Japanese with English December 1989.summary).

Versteegen, P., private communication,Omodt, H. W., et al., The Major Soils of Science Applications International Cor-North Dakota, Bulletin No. 472, Depart- poration, McLean, Virginia, 1989.ment of Soils, Agricultural ExperimentStation. North Dakota State University, Westphal, D. L., 0., B. Toon, and T. N.Fargo, January 1968. Carlson, "A Two-Dimensional Numerical

Investigation of the Dynamics and Micro-Owen, P. R., "Saltation of Uniform physics of Saharan Dust Storms," J.Grains in Air," J. Fluid Mech., Vol. 20, Geophys. Res., Vol. 92, 1987, pp.1964, pp. 225-242. 3027-3049.

27

DISTRIBUTION LISTDNA-TR-90-71

DEPARTMENT OF DEFENSE DEPARTMENT OF THE NAVY

ASSISi ANT TO THE SECRETARY OF DEFENSE NAVAL RESEARCH LABORATORYATTN. EXECUTIVE ASSISTAN, ATTN: CODE 2627

ATTN- CODE 4770 G COOPERSTEINDEFENSE INTELLIGENCE AGENCY ATTN: CODE 7920 A WILLIAMS

ATTN. D SUNSHINEATTN: DT-2A G WEBER NAVAL SURFACE WARFARE CENTERATTN- RTS-2B ATTN K42 LVALGE

DEFENSE NUCLEAR AGENCY NAVAL WEAPONS EVALUATION FACILITYATTN SPSD ATTN: CLASSIFIED LIBRARYATTN: SPWE C GALLOWAYATTN. TDTR OFFICE OF CHIEF OF NAVAL OPERATIONS

2 CYS ATTN: TITL ATTN.' OP 654

DEFENSE NUCLEAR AGENCY DEPARTMENT OF THE AIR FORCEATTN: TDNM

2 CYS ATTN- TDTT W SUMMA AERONAUTICAL SYSTEMS DIVISIONATTN. ASD/ENSSS H GRIFFIS

DEFENSE TECHNICAL INFORMATION CENTER2 CYS ATTN: DTIC/FDAB AIR UNIVERSITY LIBRARY

ATTN: AUL-LSESTRATEGIC AND THEATER NUCLEAR FORCES

ATTN: DR E SEVIN BALLISTICS SYSTEMS DIVISION/MYATTN: ASMS

THE JOINT STAFF ATTN- MYEATTN- JAD/SFDATTN. J8 NUC FORCE ANAL DIV FOREIGN TECHNOLOGY DIVISION

ATTN. SDMDA S SPRINGDEPARTMENT OF THE ARMY

HQ USAF/XOXFSDEP CH OF STAFF FOR OPS & PLANS ATTN' AFXOOTS

ATTN DAMO-NCZPHILLIPS LABORATORY

HARRY DIAMOND LABORATORIES ATTN: HO R DUFFNERATTN. SLCHD-NW-P-CORRICAN ATTN: NTA A SHARP

ATTN- NTCAU S ARMY ATMOSPHERIC SCIENCES LAB ATTN: NTED J RENICK

ATIN. SLCAS-AR-M R SUTHERLAND ATTN: NTES

U S ARMY BALLISTIC RESEARCH LAB WRIGHT RESEARCH & DEVELOPMENT CENTERATTN SLCBR-SS-T ATTN: WRDC/MLBT W F ANSFACHATTN- SLCBR-TB-B G BULMASH

DEPARTMENT OF ENERGYU S ARMY ENGR WArERWAYS EXPER STATION

ATN' C WELCH CEWES-SE-R DEPARTMENT OF ENERGYATTN D RICKMAN CEWES-SE-R OFFICE OF MILITARY APPLICATIONSATTN- E JACKSON CEWES-SD-R ATTN. OMA/DP-225ATTN. F DALLRIVA CEWES.SS.RATTN J BALSARA 'EWES-SS-R LAWRENCE LIVERMORE NATIONAL LAB

ATTN. L-262 PAUL GUDIKSENU S ARMY NUCLEAR & CHEMICAL AGENCY ATTN' L-439 J KELLER

ATTN MONA-NU D BASH ATTN: L-81 R PERRETTATTN L-85 P CHRZANOWSKI

U S ARMY NUCLEAR EFFECTS LABORATORYATTN ATAA.TDC R BENSON

Dist-1

DNA.TR-90-71 (DL CONTINUED)

LOS ALAMOS NATIONAL LABORATORYATTN A S MASON HERCULES, INCATTN' J NORMAN ATTN. P MCALLISTERATTN. R W SELDENATTN. R W WHITAKER HORIZONS TECHNOLOGY, INCATTN: R S DINGUS ATTN: W T KREISS

SANDIA NATIONAL LABORATORIES INSTITUTE FOR DEFENSE ANALYSESATTN' A CHABAI DIV 9311 ATTN. CLASSIFIED LIBRARY

ATTN: E BAUEROTHER GOVERNMENT

JAYCORCENTRAL INTELLIGENCE AGENCY ATTN. W FLATHAU

ATTN. OSWR/NEDATTN OSWR S WALLENHORST KAMAN SCIENCES CORP

ATTN R RUETENIKDEPARTMENT OF DEFENSE CONTRACTORS ATTN' T STAGLIANO

ACUREX CORP KAMAN SCIENCES CORPATTN C NARDO ATTN: F SHELTON

ATTN B KINSLOWAEROJET PROPULSION DIVISION

ATTN- A COLLINS KAMAN SCIENCES CORPATTN D MOFFETT

AEROSPACE CORP ATTN DASIACATTN D RAUSCHATTN H BLAES M2 250 KAMAN SCIENCES CORPORATIONATTN- L SELZER M4/925 ATTN: DASIAC

APPLIED RESEARCH ASSOCIAl ES, INC LACHEL PIEPENBURG AND ASSOCIATES, INCATTN. J KEEFER ATTN' D PIEPENBURG

APTEK, INC LOCKHEED MISSILES & SPACE CO, INCATTN T MEAGHER ATTN. P J SCHNEIDER

ATTN' R K JAMiSONBOEING TECHNICAL & MANAGEMENT SVCS, INC ATTN. T J KELIHER

ATTN- R BURWELLMCDONNELL DOUGLAS CORPORATION

CALIFORNIA RESEARCH & TECHNOLOGY, INC ATTN. L COHENATTN. K KREYENHAGEN

PACIFIC-SIERRA RESEARCH CORPCALIFORNIA RESEARCH & TECHNOLOGY, INC 2 CYS ATTN: R A GAJ

ATTN J THOMSEN 2 CYS ATTN R SMALL

CALSPAN CORP R & D ASSOCIATESATTN M DUNN ATTN- R ROSSATTN- M HOLDEN ATTN. T A MAZZOLA

CARPENTER RESEARCH CORP R & D ASSOCIATESATTN H J CARPENTER ATTN GGANONG

G B LABORATORY, INC R & D ASSOCIATESATTN G BURGHART ATTN. J WEBSTER

GENERAL ATOMICS, INC RAND CORPAT-N CHARLES CHARMAN ATTN' LIBRARY

GENERAL ELECTRIC CO S-CUBEDATTN B MAGUIRE ATTN C NEEDHAM

GENERAL RESEARCH CORP SCIENCE APPLICATIONS INTL CORPATTN W ADLER ATTN J STODDARD

H & H CONSULTANTS, INCATTN W HALL

Dist-2

DNA-TR-90-71 (DL CONTINUED)

SCIENCE APPLICATIONS INTL CORP TOYON RESEARCH CORP

ATTN. D BACON ATTN- J CUNNINGHAM

ATTN. J COCKAYNEATTN' PVERSTEEGEN TRW INCATTN. W LAYSON ATTN: A ZIMMERMAN

ATTN M SEIZEW

SCIENCE APPLICATIONS INTL CORP ATTN- P BRANDT

ATTN J MANSHIP ATTN. R BACHARACH

SOUTHERN RESEARCH INSTITUTE WEIDLINGER ASSOC, INC

ATTN. C PEARS ATTN. DARREN TENNANTATTN: H LEVINE

SRI INTERNATIONALATTN J COLTON WEIDLINGER ASSOCIATES, INC

ATTN' M SANAI ATTN- M BARON

TECH REPS, INCATTN- F MCMULLAN

Dist-3