Diversion of Lava Flows by Aerial Bombing - Lessons from Mauna Loa Volcano, Hawaii J.P. LOCKWOOD F.A. TORGERSON U.S. Geological Survey, Hawaiian Volcano Observatory, Hawaii 96718 U.S. Air Force, Pacific Air Forces (OA), Hickam AFB, Honolulu, Hawaii 96853 ABSTRACT Lava flows from Mauna Loa volcano can travel the long distances from source vents to populated areas of east Hawaii only if heat~ insulating supply conduits (lava channels and/or lava tubes) are constructed and main- tained, so as to channelize the flow and prevent heat loss during transport. Lava is commonly directed into such conduits by horseshoe-or lyre-shaped spatter cones-loose accumulations of partially welded scoria formed around principal vents during periods of high fountaining. These conduit systems commonly develop fragile areas amenable to artificial disruption by explosives during typical eruptions. If these conduits can be broken or blocked, lava supply to the threatemng flow fronts will be cut off or reduced. Explosives were first suggested as a means to divert lava flows threatening Hilo, Hawaii during the eruption of 1881. They were first used in 1935, without significant success, when the Army Air Force bombed an active pahoehoe channel and tube system on Mauna Loa's north flank. Channel walls of a Mauna Loa flow were also bombed in 1942, but again there were no significant effects. The locations of the 1935 and 1942 bomb impact areas were determined and are shown for the first time, and the bombing effects are documented. Three days after the 1942 bombing the spatter cone surrounding the principal vent partially collapsed by natural processes, and caused the main flow advancing on Hilo to cease move- ment. This suggested that spatter cones might be a suitable target for future lava diversion attempts. Because ordnance, tactics, and aircraft deliv- ery systems have changed dramatically since 1942, the U.S. Air Force conducted extensive testing of large aerial bombs (to 900 kg) on prehistoric Mauna Loa lavas in 1975 and 1976, to evaluate applicability of the new systems to lava diversion. Thirty-six bombs were dropped on lava tubes, channels, and a spatter cone in Bull. Volcanol., Vol. 43-4, 1980 the tests, and it was verified that spatter cones are especially fragile. Bomb crater size (to 30 m diameter) was found to be inversely related to target rock density, with the largest craters produced in the least dense, weakest rock. Bomb fuze time delays of 0.05 sec caused maximum disruption effects for the high impact velocities employed (250 to 275 m/sec). Modern aerial bombing has a substantial probability of success for diversion of lava from most expected types of eruptions on Mauna Loa's Northeast Rift Zone, if Hilo is threatened and if Air Force assistance is requested. The techniques discussed in this paper may be applicable to other areas of the world threat- ened by fluid lava flows in the future. INTRODUCTION M a u n a Loa, the l a r g e s t shield volcano on e a r t h (Fig. 1), rises to over 4000 m elevation b e h i n d t h e city of Hilo, Hawaii (1980 population: 33,217). M a u n a L o a ' s N o r t h e a s t Rift Zone extends from the s u m m i t toward H i l o a n d has b e e n t h e source of lava flows t h a t have m o v e d toward Hilo six t i m e s in r e c o r d e d history (Table 1, Fig. 2A). T h e historic eruptive p a t t e r n , as well as seismic and geodetic evidence, suggests t h e possibility of a p o t e n t i a l l y d a n g e r o u s flank eruption in the future (LOCKWOOD et al., 1976). In response to public concern over this threat, several G o v e r n m e n t agencies p r e p a r e d contingency p l a n s for lava diversion should this be necessary. T h e U. S. Air Force a n d U. S. A r m y conducted extensive field testing of explo- sives on the Pohakuloa T r a i n i n g A r e a b o m b i n g range (Fig. 1). T h e A r m y also p r e p a r e d theoretical a n d field studies on the effects of conventional and hydraulic 728 J.P. L O O C K W O O D - F.A. T O R G E R S O N ••g•re 2A USA Pohakulod'~ I \ Trainin~Ar.ea ~ \ ~ "~ ] @ / t l \ ~ f o ZOk~, FiG. I - Location of M a u n a Loa and features discussed in text. (<<water b o m b . ) explosives on lava ( C U M M I N G S , 1975 a n d B A K E R , et al., 1978) a n d the U. S. A r m y Corps of Engineers prepared preliminary plans for lava diver- sion barriers ( U S A C O R P S OF ENGINEERS, 1975, 1977). T h e artificial diversion of lava flows has doubtless always been discussed w h e n e v e r populated areas have b e e n threatened b y slow-moving lava. T h e earliest recorded attempt to divert lava occured in 1669 when a flow from Etna volcano threatened Catania, Sicily (RITT- MANN, 1929, see account by FINCH and MACDONALD, 1951, p. 128). The use of explosives to divert lava was apparently first suggested in 1881, but the first actual use was during the 1935 Mauna Loa eruption. Gordon A. Macdo- nald, to whose memory this volume is dedicated, was intensely interested in methods of lava diversion in his beloved Hawaii and published extensively on the feasibility and theory of bombing. His suggestions on bombing strategy form much of the foundation of the work reported in this paper. HISTORICAL BACKGROUND Lava from Mauna Loa's Northeast Rift Zone came within less than two kilome- ters of Hilo harbor in the s . m m e r of 1881, and the then-small village of Hilo appeared doomed. Small rock walls were built to divert the flow, but were overrun. Princess Ruth, who came from Hawaii Kingdom offices in Honolulu to pray, is widely credited with stopping the flow but~ nevertheless, taking no chances, the Hawaii Public Works Department sent down ~<a considerable quantity of gunpowder>> to attempt disruption of the flow front (ANoNY- MOUS, 1940). This is the earliest recorded suggestion for the use of explosives to divert lava, although the eruption stopped before the gunpowder was used. In the 1920's, Lorrin Thurston, influen- TABLE 1 - Historic M a u n a Loa flank eruptions that have threatened Hilo, Distance of vent Distance Principal Yea~ Duration (days) from Hllo Harbor (km) travelled (km) conduit type 1852 20 40 24 lava channel 1855-56 450 53 43 lava tubes. c h a n n e l s 1880-81 280 45 43 l a v a t u b e 1899 19 50 ~7 lava channel. ~ubes 1935 42 46 27 l a v a tube. channels 1942 13 43 28 l a v a c h a n n e l s DIVERSION OF LAVA FLOWS BY AERIAL BOMBING 7 2 9 tial Hawaii publisher and President of the Hawaii Volcano Research Association, advocated the use of explosives to divert lava, and outlined the basic theory of lava tube flow disruption in a newspaper article that appeared at the time of a predicted eruption of Hualalai volcano. THURSTON wrote (1929): <<The plan which I suggest is to locate, at a favorable place, a flow-hole opening into a lava tube through which molten lava is running to the front. Suspend over this blow-hole a charge of T N T suspended by a block and tackle attached to a tripod. Arrange a device to be tripped by a rope several hundred feet long, which will drop the charge of high explosive, so that within a few seconds after it is dropped it will explode with violence, such as to disrupt and destroy the tube, forcing the flow of lava to the surface to again cover the territory covered by the Krst lava>~. W. A. JAGGAR, Director of the Hawaiian Volcano Observatory, initially considered lava tubes the only target, felt disruption would be simple, and wrote (1931): <<There would be no necessity of employing an army or TNT. Any contractor with a couple of men and some blasting powder, carried by pack mules from the ranch, and guided by volcanolo- gists, can do the job>~. Four years later Jaggar was to have his chance to attempt lava tube disruption, but at this time he decided to use both the Army and TNT! THE BOMBING OPERATION OF 1935 Mauna Loa erupted high on the Northeast Rift Zone on November 21, 1935. On November 27, a new vent opened on Mauna Loa's north flank, at 8,800 feet. Lava from this vent flowed northward and initially ponded in the saddle between Mauna Loa and Mauna Kea; Hilo was not threatened. Abruptly, however, on December 22, the ponded lava broke its natural levees and began to flow rapidly in a narrow lobe directly toward Hilo. Within four days the new flow had covered 8 of the remaining 32 km to Hilo. Hilo and its vital watershed were now definitely threatened by the flow, and T. A. Jaggar, who by then saw the need for better delivery systems than mules, had ample reason to seek the help of the Army to bomb the flow. The.use of aircraft to bomb the flow was suggested to Jaggar by Guido Giacometti of Ola'a (JAGGAR, 1956, p. 153). The following account is taken from Jaggar's reports (1936, 1945) and from interviews with several pilots who participated in the bombing mission. After a brieffmg on the situation, the U. S. Army Air Corps sent ten Keystone B-3 and B 4 bombers (1920's-vintage, fabric- covered biplanes from the 23rd and 72nd Bombardment Squadrons) to Hilo on December 26th to prepare for the bombing operation. The Army's Hawaiian Department G-2 (Intelligence Officer) who helped develop the bombing plans and who handled press announcements was the then-Lt. Col. George S. Patton, Jr. (BLUMENSON, 1972; CRAVEN, 1959). Jaggar briefed the pilots on the theory of lava tube disruption and flew over Mauna Loa that afternoon to select targets. December 27th dawned clear, and in two separate flights that day, the aircraft dropped 36 to 40 bombs on fluid pahoehoe issuing from the lower vent. Half of these bombs were for aiming purposes and contained only small black powder charges. Half were 275 kg MK I demolition bombs loaded with 161 kg of T N T and armed with 0.1 sec time delay fuzes. Five of these twenty bombs struck molten lava directly; most of the others impacted solidified lava along the flow channel margins (Fig. 2B). The bombs that struck molten lava left no clear craters; craters apparently were filled by lava immediately, but may be marked by discontinuous rims of spatter. Col. William C. Capp (USAF, ret.), a pilot who bombed the lower target, reported direct hits on the channel, observing that a sheet of red, molten rock was thrown up to about 200' elevation and that flying debris made small holes in his lower wing. Bombs that 730 J.P. L O O C K W O O D - F ~ . T O R G E R S O N 155 e 3 0 ' W [at~'-.v q 19* 3 0 ' N FIG. 2A - Generalized map of the Northeast 155" 0 0 ' W I 19" 4 8 ' N MAUNA KEA ~ , ~ , . _ _ , . . . . ~ A ~ N / " Area of fiqure 2(; ~ ~ ~ - I - ~ J KILAUEA Northeast Rift Zone / ... o l ~km, / Rift Zone, showing historic lava flows above Hilo. impacted on solidified, vesicular pahoehoe along the flow margin produced craters averaging 6.7 m diameter and 2.0 m depth (1) (Fig. 3). The craters formed in large part from compaction of the vesic- ular lava, as the volume of crater ejecta is much less than crater volume. Several of the black-powder <<pointer>> bombs failed to explode. One apparently impacted a thin pahoehoe overflow at the channel edge; it penetrated the crust and underlying molten (?) lava, but not the underlying solid flow (Fig. 4). Pilots observed that several bombs collapsed thin lava tube roofs, although in no case was sufficient roof material imploded into the tube to cause blockage. The extrusion of lava ceased within a week, however, and JAGGAR wrote (1936) that the bombing caused the fluid pahoehoe to thicken and block the vent by the process of gas release and destruc- tion of the ((equilibrium of self-heating>>. He further believed that the bombing (i) All crater dimensions referred to through- out the text are those of the apparent craters. Dimensions of the t r u e craters (without ejecta fallback) are greater. stopped the flow of lava, and wrote, (1939, p. 2) that: ~(The smashing of the tunnel had cooled the oncoming liquid so that it L 0 Ikm I I 1935 lava Lava tube channel : : : : : Eruptive vent • Bomb crater FIG. 2 B - L o w e r 1935 vent, showing b o m b e d areas. DIVERSION OF LAVA FLOWS BY AERIAL BOMBING ~ - - ~ 1942. Lava ~***4~ Eruptive Vent ~.~":~- Spatter . . . . . . Flow Channel .'~.J'~bx",?" = Bomb Crater 28 APRIL ~'~'~ ,,.~ : . . . . . . . . . . . . . . . . . x - - 4 - 1 0 MAY 731 Fro. 2C - Lower 1942 vents, showing bombed areas. d a m m e d itself. T h i s confirmed t h e t h e o r y t h a t t h e b o m b i n g solidified t h e t u n n e l lava back into t h e h e a r t of t h e m o u n t a i n . W i t h twelve river hits o u t of sixteen, a n d liquid thrown up h u n d r e d s of feet, t h e r e c a n be no question w h a t e v e r t h a t the b o m b i n g s t o p p e d the flow~. G r o u n d e x a m i n a t i o n of t h e b o m b i n g sate showed no evidence t h a t the b o m b i n g h a d i n c r e a s e d viscosity, and we agree with the conclusion of STEARNS a n d MACDONALD (1946, p. 94) t h a t ~the cessa- tion of t h e 1935 flow soon after t h e b o m b i n g m u s t be c o n s i d e r e d a coinci- dence >~. L e s s t h a n s e v e n y e a r s were to elapse, however, b e f o r e b o m b s were to be again d r o p p e d on m o l t e n lava. THE BOMBING OPERATION OF 1942 A b o u t 4:40 AM, on April 28th, after a two d a y s u m m i t eruption, lava broke out on the N o r t h e a s t Rift Z o n e at 9,400 feet, 43 k m above Hilo. T h e eruptions was d e s c r i b e d b y MACDONALD (1943) a n d in a n u n p u b l i s h e d m a n u s c r i p t by JAGGAR (1942). T h e following account is t a k e n in m o s t p a r t from t h e s e sources, since t h e e r u p t i o n itself was classified owing to war- t i m e security; extensive m i l i t a r y r e c o r d s of t h e e r u p t i o n a n d b o m b i n g o p e r a t i o n were m a d e b u t have since b e e n lost. M a c d o n a l d a n d J a g g a r write t h a t a two- k m long ~curtain of fire>~ fissure fed a flow t h a t m o v e d v e r y quickly downslope a n d t r a v e l l e d 10 k m in t h e first six hours. I t e n t e r e d d e n s e rain forest, a n d flow p r o g r e s s was difficult to m o n i t o r b e c a u s e of thick s m o k e a n d clouds. R e p o r t s of t h e flow p r o g r e s s a r e few a n d in p a r t c o n t r a d i c t o r y after t h a t first day. On April 30 its progress was m e a s u r e d at 3 k m in 24 hours. A t t h a t time t h e flow front was a t a b o u t 3,000 f e e t elevation, less t h a n 6 k m a b o v e t h e vital O l a ' a F l u m e . MACDONALD a n d ABBOTT (1970, p. 64) s t a t e d that: ~On M a y 1 t h e r e a p p e a r e d to be i m m i - n e n t d a n g e r of t h e flow successively cutting t h e flume t h a t s u p p l i e d w a t e r to the town of M o u n t a i n View, blocking the highway a r o u n d t h e island - which was vital to m i l i t a r y operations, a n d d e s t r o y i n g p a r t of Hilo a n d p o s s i b l y p a r t of its harbor. I t was d e c i d e d to t r y to d i v e r t t h e flow b y aerial bombing>>. 732 J.P. LOOCKWOOD - F . A . TORGERSON Fla. 3 - Aerial view of 1935 bomb craters, looking north. Ruy Finch, Volcanologist at the Hawaiian Volcano Observatory, flew over the flow on the morning of May 1 to select targets for bombing later that day. Finch selected as the primary target a long pahoehoe channel at 7,000 feet about 7 km below the principal eruptive vent (then stabilized at 9,200 feet, see Fig. 2C). Below that point four individual channels coalesced into one. Finch hoped to break down the levee walls of the channel and thus rob the main flow of its lava supply, at least temporarily. When the bombers (possibly twin-engined B-18's) reached the area that afternoon, the primary target was obscured by clouds, so that upper, secondary targets, also selected by Finch, were bombed. Reports on the duration and extent of the bombing vary. All bombing was apparently done on May 1, although in one report FINCH (1942) also refers to bombing on the morning of May 2. The number of bombs dropped is not known. Finch says ~a few~ or ~<severab~ in different reports; MACDONALD(1943, p. 254) says ~about 16 demolition bombs, some weighing 300 and others 600 lbs;~. The locations of the bombing had never been described, but in a 1977 overflight G. A. Macdonald pointed out the general FIG. 4 - ~Pointer bomb~ that penetrated 1935 pahoehoe flow alongside channel. DIVERSION OF LAVA FLOWS BY AERIAL BOMBING 733 area. B y e x a m i n a t i o n of a p u b l i s h e d photo of one explosion (STEARNS a n d MACDON- ALD, 1946, Pl. 35) a n d of a n o t h e r b o m b b u r s t shown in a c l a n d e s t i n e movie film t a k e n during t h e bombing, t h e b o m b e d a r e a s were l o c a t e d a n d are shown in Fig. 2C. D u r i n g g r o u n d inspection of the two c r a t e r e d areas, only two entire craters were found; o t h e r s h a d a p p a r e n t l y b e e n covered b y l a v a s of the continuing erup- tion. T h e two craters m e a s u r e d 6.4 a n d 9.1 m in d i a m e t e r , a n d 2.2 and 3.0 m in d e p t h . A b u n d a n t s h r a p n e l a n d c r a t e r ejecta were found s c a t t e r e d over pre-1942 kipukas at the u p p e r b o m b site, showing t h a t several o t h e r b o m b s h a d i m p a c t e d a n d d e t o n a t e d in this a r e a b u t l a t e r h a d b e e n covered b y a ' a lava of t h e continuing eruption. O n e u n e x p l o d e d d u d b o m b , found in a k i p u k a less t h a n 5 m from a l a t e r 1942 flow, h a d n o t d e t o n a t e d from t h e h e a t (Fig. 5). I t was identified b y e x p e r t s from t h e 6th Explosive O r d n a n c e D i s p o s a l ( E O D ) D e t a c h m e n t of Ft. S h a f t e r as a pre-WWII vintage M K I, 275 kg (600 ib) d e m o l i t i o n bomb, t h e s a m e t y p e u s e d in t h e 1935 bombing. A n a l y s i s of the craters i n d i c a t e d the b o m b s were also fuzed in t h e 0.1 sec t i m e - d e l a y mode. T h e bomb, containing 161 kg of T N T , was d e s t r o y e d by an A r m y E O D t e a m on J u n e Fro. 5 - Unexploded 275 kg MK I bomb, dropped in 1942. 5, 1980, 38 y e a r s after it was d r o p p e d . A l t h o u g h t h e d e t o n a t i o n c a m e too l a t e to p l a y a role in l a v a diversion, it d i d gener- a t e a strong seismic signal t h a t was r e c o r d e d on t h e H a w a i i a n Volcano O b s e r - v a t o r y network. A t t h e lower b o m b i n g site (7,800 feet), r i b b o n s of b a s a l t glass are widely scat- t e r e d on t h e surface. T h i s s u g g e s t e d a b o m b h a d i m p a c t e d m o l t e n lava, since t h e r e was no s p a t t e r i n g at t h e v e n t a n d b e c a u s e t h e o n l y i d e n t i f i a b l e c r a t e r was in prehistoric lava. On t h e b a n k of a p a h o e h o e c h a n n e l 25 m d o w n s l o p e from the eruptive vent, t h e aft third of a M K I b o m b was found. T h i s b o m b casing h a d b a s a l t glass i n j e c t e d into t h e tail cone s p a n n e r (Fig. 6) a n d it was obvious t h a t it h a d b e e n i m m e r s e d in m o l t e n lava. T h e b o m b h a d a p p a r e n t l y i m p a c t e d in fluid p a h o e h o e (causing t h e ~(splash;) of b a s a l t glass), b u t h a d failed to d e t o n a t e . On heating, t h e T N T riffling h a d ignited a n d b u r n e d , r u p t u r i n g t h e b o m b case. T h e e x p a n s i o n of b u r n i n g gases e j e c t e d t h e tail section from t h e p a h o e h o e channel. T h e effects of the b o m b i n g are difficult to ascertain, as t h e c r a t e r e d a r e a s were largely covered b y l a v a after M a y 1. MACDONALD (1943, p. 254) o b s e r v e r e d t h a t b o m b s d r o p p e d on t h e lower v e n t h a d ((no a p p r e c i a b l e effect)). H e r e p o r t e d t h a t a t t h e u p p e r site ~(a [narrow a'a] levee was b r o k e n a n d a s m a l l lava s t r e a m e s c a p e d to one side of t h e m a i n flows, b u t t h e t o p o g r a p h i c d e p r e s s i o n t h a t guided the original flow also guided t h e n e w branch. I t flowed p a r a l l e l to t h e m a i n river for a s h o r t distance, a n d t h e n r e j o i n e d it)~. FL'~CH s p e c u l a t e d (1942, p. 6) t h a t t h e collapse of t h e source cone a t 9,200 f e e t t h r e e d a y s l a t e r (discussed below) m a y have b e e n a , b a c k i n g ups) effect of t h e M a y 1 b o m b i n g 2 k m downslope, b u t this a p p e a r s impossible. A n u n p u b l i s h e d H a w a i i a n Volcano O b s e r v a t o r y r e p o r t for April, 1942 s t a t e s that, after t h e lack of success on M a y 1 , p r e p a r a t i o n s were being m a d e to use 2000 lb b o m b s , b u t on a r e c o n n a i s s a n c e flight on M a y 2 t h e flow front was o b s e r v e d to be widening a n d thickening. D o w n s l o p e p r o g r e s s was slower, a p p r e - 7 3 4 J.P, LOOCKWOOD -F.A. TORGERSON FIG. 6 - Detailed view of taft cone area of bomb fragment ejected from molten pahoehoe after burning (without detonation). Note basalt glass under fuze spanner (arrows). Threaded tail fuze receptacle is 5 cm in diameter. h e n s i o n was s o m e w h a t l e s s e n e d , a n d t h e r e s e e m e d no n e e d for i m m e d i a t e f u r t h e r bombing~>. D u r i n g the e v e n i n g of M a y 4, p a r t of t h e s o u t h e r n wall of t h e 9,200 foot s p a t t e r cone s u d d e n l y c o l l a p s e d (Fig. 7). A s m a l l e r portion of t h e n o r t h e r n walt also c o l l a p s e d b u t t h e flow from t h a t side was small, a n d t h e b r e a c h was soon b l o c k e d b y new s p a t t e r . On t h e south side, however, t h e c o l l a p s e d s p a t t e r wall was quickly carried away by a m a j o r new flow, which f o r m e d a large c h a n n e l south of the e s t a b l i s h e d conduit for t h e principal flow, now a d v a n c i n g a t t h e 3,000 foot elevation. W i t h its s u p p l y of l a v a m u c h r e d u c e d (less t h a n h a l f t h e v o l u m e was now b e i n g supplied) t h e m a i n flow s t o p p e d advanc- ing on M a y 5th or 6th, b e c a u s e of t h e effect of t h e s p a t t e r cone collapse (MACDONALD, 1943, p. 250). A f t e r M a y 4, no n e w flows e x t e n d e d over 10 k m from t h e u p p e r vent. N o t i n g t h e effects of this n a t u r a l collapse, FINCH (1942, p. 6) s u g g e s t e d t h a t source v e n t s p a t t e r cones s h o u l d be a favorable t a r g e t for future l a v a diversion a t t e m p t s . t h e i r l a t e r a l m a r g i n s quickly solidify a n d m o v e m e n t is soon r e s t r i c t e d to n a r r o w conduits within t h e largely s t a t i o n a r y flow. T h e s e conduits s u p p l y m o l t e n rock from t h e source v e n t s to t h e advancing f r o n t of t h e flow a n d insulate t h e m o l t e n m a t e r i a l from excessive h e a t loss during t r a n s p o r t . T H E O R Y A N D S T R A T E G Y OF B O M B I N G During an eruption, as fluid lava flows of the Hawaiian type m o v e downslope, FIO. 7 - Aerial view of the 1942 spatter cone at 9,200 feet~ looking west. The partial cone collapse of May 4, 1942 caused the new channel on the south (left) side of the cone. DIVERSION OF LAVA FLOWS BY AERIAL BOMBING 7 3 5 The conduits consist of natural-leveed channels in a'a flows, and of channels or lava tubes (~pyroducts>)(2)) or both in pahoehoe flows. T h e insulation efficiency of these conduits, especially lava tubes, is remarkable. In 1973, lavas from the Kilauea Mauna Ulu eruption dropped in temperature less than 15°C after 12 km of transport in a shallow lava tube system ( S W A N S O N and FABBI, 1973, p. 654). Such supply conduits must be estab- lished and maintained if lava is to be transported long distances from the vents. These supply conduits, and the spatter cones which commonly channelize lava flow into them are fragile, and frequently break down during eruptions due to natural causes. The purpose of aerial bombing is to deliberately disrupt lava supply conduit systems at the time and place desired, in order to deprive the active flow front of the constant lava supply required for forward movement. Three types of targets susceptible to artificial disruption have been identified (FINCH and MACDONALD, 1951, p. 129) and discussed as bombing targets. These are (1) lava tubes, proposed as targets by Thurston and Jaggar, and bombed in 1935; (2) channel levees, bombed in 1942; and (3) source vent spatter cones, proposed as potential targets by Finch. For lava tubes, the objective is to implode a thick section of tube roof over flowing lava, causing the flow to become choked with solid debris. If sufficient solid debris is suddenly mixed with flowing lava, the tube may be blocked, causing lava from upstream to overflow through the broken roof and spread onto lower- standing terrane to the side of the tube system. For lava channels, a place must be found where the bordering levee is narrow, unstable, and where the flow is moving at a higher elevation than the immediately adjacent terrain. In this case the objective is to breach the levee and allow lava to pour onto the lower terrain. Ideally, debris from the breached levee should be thrown into the downstream (2) This more descriptive term was proposed by COAN (1844. p. 46). part of the old channel to impede flow, but in practice this would be difficult. The spatter cones that form around the principal vents of most Hawaiian erup- tions are typically horseshoe- or lyre- shaped, and force lava being erupted at the vent to enter the channel or tube system at its open, downstream end. T h e spatter cone typically supports molten lava at an elevation above the surround- ing terrain, and the objective is to breach the cone wall so as to provide a new flow path that would divert newly-erupted lava from the established conduit system. For each of these three types of targets, the strategy remains the same - to deprive the primary flow channel of some or all of its fresh supply of lava, thus quickly stopping the advance of the main flow. This was demonstrated by the natural diversion of lava from the prin- cipal flow coduit on May 4, 1942 (Fig. 7). Tactics of this disruption must, however, vary for each potential target. Extensive testing was carried out in 1975 and 1976 to learn optimum methods of disrupting each of these target types with aerial bombs (Lockw~OOD, 1979). TESTING PROGRAM OF 1975 AND 1976 In response to the threat of a poten- tially disastrous eruption of Mauna Loa following the July 1975 eruption (LocK- WOOD et al., 1976), local authorities asked Federal military agencies to prepare contingency plans for lava diversion. Effects of explosives on disruption of lava supply conduits were not known, so the U. S. Army and U. S. Air Force conducted field tests in prehistoric lavas of the Pahakuloa Training Area (PTA) on the north slope of Mauna Loa (Fig. 1). We here report on a program of testing conducted under an informal cooperative study between the USAF Pacific Air Forces and the U.S. Geological Survey's Hawaiian Volcano Observatory. T h e testing involved aircraft that were in Hawaii for training exercises and had been already scheduled for bombing practice. Only the target and ordnance selection were altered. The tests were conducted on 736 J . P . L O O C K W O O D - F . A . T O R G E R S O N 5 FIG. 8 - Aerial view of Pu'u 6,995' Pohakuloa Training Area, looking south. a n d n e a r P u ' u (hill) 6,955 on t h e w e s t e r n edge of P T A (Fig. 8). Pu'u 6,955 is a typical large (35 m high) s p a t t e r cone f o r m e d b e t w e e n 1,000-2,000 y e a r s ago. T a r g e t s were p l a c e d on this s p a t t e r cone a n d on its l a v a c h a n n e l s downslope, as well as on a l a v a t u b e s y s t e m of a n adjoin- ing y o u n g e r p a h o e h o e flow 1 k m to t h e n o r t h e a s t . T h e t e s t i n g p r o g r a m was d e s i g n e d to evaluate t h e effects on each of t h e t h r e e t a r g e t t y p e s by d i f f e r e n t b o m b s , explosive fillings, fuzes, a n d d e l i v e r y aircraft. D u r i n g t e s t s on four different occasion in 1975 a n d 1976 a t o t a l of 36 b o m b s were d r o p p e d . T h e s e consisted of 22 M K 8 2 500 lb. (225 kg) a n d 14 MK84, 2000 lb (900 kg) G e n e r a l B o m b s . T h e b o m b s were filled with e i t h e r H-6 or T r i t o n a l explo- sive, a n d were fitted with e i t h e r nose or tail fuzes, or both. B o t h F-4 a n d A-7D aircraft of t h e 23 rd a n d 33 rd T a c t i c a l F i g h t e r Wings were e m p l o y e d for b o m b delivery. Of t h e initial t e n MK82 b o m b s d r o p p e d , e i g h t a p p a r e n t l y failed to d e t o n a t e , or d e t o n a t e d a t too g r e a t a d e p t h to form surface craters. A p o r t a b l e s e i s m o g r a p h was d e p l o y e d to m o n i t o r s u b s e q u e n t tests. Fuze configuration was changed for t h e n e x t test, a n d t h e remRinlng twelve M K 82 b o m b s d e t o n a t e d as expected. T h e s e b o m b s p r o d u c e d c r a t e r s averaging 10.6 m in d i a m e t e r a n d 1.7 m in d e p t h in t h e s p a t t e r cone a n d 6.8 m in d i a m a t e r a n d 1.8 m in d e p t h on <<tube-fed pahoehoe>> (SWANSON, D. A., 1973) l e v e e s of t h e c h a n n e l downslope from P u ' u 6,995. T h e sizes of t h e b o m b craters v a r i e d directly with fuze d e l a y - t i m e settings. F o r t h e vesic- FIG. 9 - Aerial view of Pu'u 6,995', looking north. A M K 84 b o m b has just exploded in the cone's northwest flank. Crater in foreground is 30 m in diameter. D I V E R S I O N O F LAVA F L O W S B Y A E R I A L B O M B I N G 737 ular, tube-fed p a h o e h o e the n o n d e l a y - fuzed b o m b p r o d u c e d a 4.6 m d i a m e t e r crater; t h e 0.025 sec d e l a y a 6.4 m d i a m e t e r crater; a n d t h e 0.05 sec d e l a y a 9.4 m d i a m e t e r crater. D u e to the rela- tively small effects of t h e s e bombs, t h e r e m a i n d e r of t h e testing u~lized the larger M K 84 b o m b s (Fig. 9). T h e results of this testing are s u m m a - rized in Fig. 10. E l e v e n craters were f o r m e d from the fourteen M K 84 b o m b s t h a t were dropped, since two b o m b s exploded on the s a m e spot, one collapsed a lava tube roof a n d did n o t form a crater, a n d one was a dud ( s e i s m o m e t e r s recorded impact, b u t showed no explo- sion). Of the eleven craters, four were f o r m e d on dense, tube-fed p a h o e h o e ( d = 2.41 g]cc) 1 k m n o r t h e a s t of P u ' u 6,995, one f o r m e d in vesicular (<fountain- fed>~ p a h o e h o e ( d = 1.885 g]cc) along a lava channel 150 m d o w n s t r e a m from the s p a t t e r cone, a n d six f o r m e d in the loose s p a t t e r of Pu'u 6,995 itself (d = 0.53 g/cc). T h e m e a s u r e d densities are averages of several fist-s/re samples. Because of large cavities in each of the t a r g e t rock types, bulk densities of the overall areas in which the craters f o r m e d are s o m e w h a t less. As is seen in Fig. 10, crater d i m e n s i o n s are greatly influenced by rock type, a n d t h e l a r g e s t craters are formed in the l e a s t dense, w e a k e s t rock (spatter). F u r t h e r , fuze t i m e - d e l a y is an i m p o r t a n t factor in the craters f o r m e d in s p a t t e r because of t h e high i m p a c t velocities (250 to 275 In] sec) i m p a r t e d to t h e b o m b s b y t h e deliv- ery aircraft; fuze d e l a y s m u s t be s h o r t to avoid excessive p e n e t r a t i o n . No d e l a y (surface detonation) results in shallow craters with poor energy coupling. T h e s h o r t e s t Lime d e l a y e m p l o y e d on t h e s p a t t e r cone (0.025 sec) r e s u l t e d in a n i n t e r m e d i a t e - s i z e crater. T h e l o n g e s t t i m e d e l a y (0.25 sec) r e s u l t e d in the s m a l l e s t crater, one o b s e r v e d to h a v e f o r m e d b y collapse of a subsurface explosion c h a m b e r (camouflet). B o m b s a r m e d with a 0.05 sec d e l a y fuze produce o p t i m u m cratering effects in s p a t t e r cones. B o m b i n g effects extend far b e y o n d t h e craters. L a r g e cracks, m o s t l y concentric to the craters, are found up to 10 m from crater r i m s (Fig. 11). S p a t t e r t h a t was loosely w e l d e d a n d solid to walk on before the explosions b e c a m e loose a n d friable up to 50 m from some craters. Clearly, t h e b o m b s s u b s t a n t i a l l y w e a k e n the cones. T h e b o m b s which i m p a c t e d on dense, strong, tube-fed p a h o e h o e n e a r a lava tube s y s t e m were a r m e d with various fuze delays, but all d e t o n a t e d a t t h e surface, on impact, with no delay. T h e s e b o m b s were filled with the explosive H-6, which has b e e n shown to be p r o n e to d e t o n a t e on i m p a c t with h a r d targets (FACE, 1967). C l e a r l y a less sensitive explosive is n e e d e d if such h a r d t a r g e t s are to b e d i s r u p t e d in the future. All of t h e s e surface explosions p r o d u c e d shallow craters, a n d one t h a t struck the lip of a 7 o. o 4 - o 3 - = o 2 - < / ~ OS ~ N O NO I I 5 I0 too.ale b=~b c,0t=r) Qo~ ©o0 t I 1 I O 1 5 2 0 2 5 3 0 AppQrent crtlter d i a m e t e r ( m ) FIG. 10 - Dimensions of MK 84 bomb craters, showing effect of target lithology and fuze time- delay. 7 3 8 J.P. L O O C K W G O D - F.A. T O R G E R S O N Fro. 11 - Margin of bomb crater in Pu'u 6,995'. lava tube window enlarged the window from 4.6 × 4.6 m to 7.6 × 11.0 m. The 6 m diameter lava tube had a roof of 1.5 m average thickness. The implosion of this tube roof sent an air-shock wave through- out an extensive tube system, as evidenced by clouds o f dust that were forced out of numerous small windows and cracks up to 600 m from the bomb impact point. It does not appear, however, that sufficient material was imploded into any tube to have caused blockage of an active lava tube flow. Principal conclusion of the bombing tests at PTA were as follows: 1) Aerial bombing can disrupt vulner- able portions of typical lava supply conduit systems that could carry lava to Hilo. The large size of potential targets requires the use of large bombs. The 900 kg MK 84 General Purpose bomb would be effective. 2) Crater formation mechanisms vary with the physical properties of the target rocks. Craters in strong, dense a'a or pahoehoe form in large part by excavation of debris and have conspicuous crater rims. Craters in weak, low density spatter form largely by compaction of pre-existing rock; rims are absent or minor. Some craters in spatter f o r m entirely by collapse into subsurface explosion cham- bers (camouflets) for bombs which pene- trate to excessive depth. Bombs detonat- ing in vesicular pahoehoe of intermediate density form craters by both excavation and compaction processes. 3) Spatter cones are the component of lava-supply conduit systems most vulner- able to disruption by aerial bombing. Effects are not limited to the .craters; spatter cone walls are fractured and weakened up to 50 m from crater rims. 4) For bombs emplaced at high aircraft release velocities (non free-fall), fuze delay times of 0.05 sec appear to cause optimum disruption of spatter cones. 5) Aerial bombs can make large craters on upper channel levees, but the width of these levees precludes effective disruption except where they are particularly narrow. 6) Lava tubes with thick, hard roofs will be difficult to collapse by aerial bombard- ment. Use of a less sensitive explosive bomb fill, such as Tritonal, would improve disruption effects. Technical details and extensive crater documentation for these tests are given by TORGERSON (1976) and TORGERSON and BEVINS •(1976). ADVANTAGES AND DISADVANTAGES OF BOMBING Bombing offers the following principal advantages: 1) Negligible environmental and legal impact. Lava will be diverted to barren Government-owned lands at higher eleva- tions. The legal problems of lava diver- sion onto private land can be avoided, and valuable biological habitats can be saved from destruction. 2) Flexibility of response. Lava diver- sion targets can be changed minutes before bombing in response to changing volcanic conditions. 3) Low costs. Bombing has the following disadvan- tages: 1) Religious beliefs. Many people in Hawaii have deeply held religious convic- tions that the use of explosives during an eruption would be an insult to Pele, the Hawaiian volcano goddess. Because of this, bombing is repugnant to many polit- DIVERSION OF LAVA FLOWS BY AERIAL BOMBING 7 3 9 ical leaders, a n d explosives m a y be e m p l o y e d only as a l a s t r e s o r t - possibly after t h e o p t i m u m t i m e for bombing. 2) Vulnerable targets may not appear. Large, n e a r - v e n t s p a t t e r cones have only d e v e l o p e d in six of the seven N o r t h e a s t Rift Zone eruptions of historic time. T h e lava t h a t flowed into Hilo in t h e spring a n d s u m m e r of 1881 was e r u p t e d directly from a crack without fountaining; a s p a t t e r cone a n d lava channels never formed. T h e t u b e s y s t e m t h a t conducted lava to Hilo h a d a very thick roof t h a t m a y have b e e n difficult or p e r h a p s impos- sible to collapse by aerial bombing. For this flow, ground e m p l a c e m e n t of large explosive charges or lava diversion b a r r i e r s would have b e e n required to effect successful diversion. I n 1935 a s p a t t e r cone f o r m e d at the upper, b u t n o t the lower vent. 3) U n k n o w n effects on molten rock. Questions r e m a i n a b o u t the effectiveness of b o m b s on m o l t e n or very high-tempera- ture rock. Will t h e y d e t o n a t e ? W h a t will the explosive effects on i n c a n d e s c e n t rock be? 4) Persistent bad weather. P e r s i s t e n t cloud cover above the lava flow could p r e v e n t the b o m b i n g accuracy required. Clearly, i m p o r t a n t questions remain, a n d contingency p l a n n i n g for the protec- tion of Hilo should include several m e t h o d s of lava diversion (lava diversion barriers, w a t e r application, ground e m p l a - c e m e n t of explosives, a n d bombing). THE FUTURE M o s t lava flows of the type likely to e n d a n g e r Hilo in the future could p r o b a b l y be d i v e r t e d from harmful p a t h s b y aerial bombing. F o r success, vulnerable a r e a s in the lava supply conduit s y s t e m m u s t develop, m u s t b e identified, a n d m u s t be i n t e r d i c t e d at the time of m a x i m u m vulnerability. Typically, this time of m a x i m u m vulnerability will occur within one to two weeks after the eruption begins. P r e s e n t contingency plans call for local stockpiling of required ordnance, a n d for delivery aircraft to be brought to Hawaii within 72 hours after initiation of a t h r e a t e n i n g eruption, t h e n s t a n d b y for use if n e e d e d . T h e following scenario, b a s e d on M a u n a L o a ' s historic behavior, d e p i c t s t h e role such aircraft could p l a y if their services were r e q u e s t e d for lava diversion: M a u n a Loa flank eruptions above Hilo will begin with an eruption in M o k u a - weoweo, M a u n a L o a ' s s u m m i t caldera. T h i s s u m m i t eruption will only l a s t one to t h r e e days a n d will be followed i m m e - diately or within a few days b y an outbreak along the N o r t h e a s t Rift Zone, b e t w e e n 8,000 and 12,000 feet elevation. T h e initial lava will flow from one or m o r e elongate fissures, t h e typical H a w a i i a n ~(curtain of f'lre~. R a t e s of lava production will initially be high, fountains will b e low, a n d lava will travel very fast - up to 15 km in the first 24 hours, because the slopes are relatively s t e e p (averaging 8 ° b e t w e e n 6,000 and 13,000 feet), a n d because the erupting lavas will b e hot, gas-rich, and highly fluid. Below 6,000 feet the rate of flow advance will begin to slow