27 6 
PACIFIC SCIENCE, Vol. XII, July, 1958 
500 feet wide) along the barrier to provide a 
path of easy movement for the lava. 
No better design for stream crossings has 
been found than that suggested by the U. S. 
Engineer Department in the 1940 report. A 
bundle of concrete pipes 24 to 48 inches in 
diameter should be laid parallel to the stream 
course, and anchored in place with concrete. 
Above these, the barrier may consist of the 
same loose rubble as elsewhere. 
Concrete underpasses, with concrete stop- 
logs, have been suggested by the department 
for highway crossings. An alternative, and 
much less costly method, would be to leave a 
gap in the barrier for the highway to pass 
through, and provide a pile of loose rubble 
near one side of the gap that can quickly be 
pushed into place by bulldozers, thus closing 
the gap when the flow approaches it. One ad- 
vantage of the barrier proposed in 1950 (Fig. 
7) is that no special crossing structure is re- 
quired at the highway between Hilo and Olaa. 
One segment of the barrier ends just up slope 
from the highway, natural topography then 
guiding the flow across the highway to a point 
where it will be controlled by the next segment. 
A flat top on the barrier is unnecessary, 
though it would do no harm. It has been sug- 
gested that the barrier might be built with a 
flat top broad enough to accommodate either 
a one- or two-lane highway. However, this 
would add greatly to the cost, both because of 
the much greater bulk of material that would 
have to be obtained and put in place, and be- 
cause of the higher standards that would have 
to be set for the material and the greater 
care that would have to be used in construc- 
tion. To successfully divert lava flows, the bar- 
rier need not even approach the standards 
necessary for a highway fill. 
The precise height of barrier needed can be 
determined only by detailed surveys. The bar- 
rier must be higher than average where it 
crosses depressions, but can be lower where it 
is superimposed on natural ridges. The height 
of the barrier designed by the U. S. Engineer 
Department in 1940 averaged about 40 feet 
for an available channel width throughout 
most of its course of approximately 3,000 feet. 
The height was determined by the cross- 
sectional area of the channel behind the bar- 
rier that was considered necessary to contain 
a lava flow of the dimensions that might rea- 
sonably be expected to enter the area. Logi- 
cally enough, the problem was approached on 
the basis of hydrodynamics, assuming that the 
lava would behave much like a stream of water 
under the same circumstances. As noted ear- 
lier, however, we now realize more clearly that 
lava does not behave wholly like water. The 
sides of the flow rise steeply to heights of 
many feet above the surrounding terrain or 
above a restraining barrier. It is therefore not 
necessary to build a barrier to a height equal 
to the full depth of the lava flow it is intended 
to divert. I believe that a barrier with an aver- 
age height of 25 to 30 feet following the 1940 
alignment would be adequate. 
CONCLUSIONS 
As a result of the foregoing considerations, 
I believe (1) that lava flows are certain to enter 
the city and harbor of Hilo eventually unless 
something is done to prevent their entry; (2) 
that they can be successfully diverted from the 
city and harbor by properly located and con- 
structed barriers; (3) that no other method can 
be relied upon to divert the flows; (4) that 
construction of the barriers in advance of the 
eruption is preferable, but that barriers prob- 
ably can be constructed in time even after the 
flow has started to advance toward Hilo; (5) 
that the barriers need consist only of loose 
rubble obtained locally and pushed into place 
by bulldozers; and (6) that the barrier align- 
ment proposed by the U. S. Engineer Depart- 
ment in 1940 is adequate to protect the center 
of the city and the harbor, but an alignment 
farther southwest is necessary if it is desired to 
protect all of the city. 
REFERENCES 
Baldwin, E. D. 1953- Notes on the 1880-81 
lava flow from Mauna Loa. Volcano Letter 
520: 1-3. 
