EFFECTS AND WEIGHT OF FIRE

The Weapon of Artillery

Last updated 31 May 2002

CONTENTS

WHAT IS WEIGHT OF FIRE?

Weight of fire concerns the quantity, density and intensity of artillery fire used to attack a target.  In essence it is about the effectiveness ('doing the right thing') and efficiency ('doing the thing right') of artillery fire at the target end.  The British undertook lots of research into this during World War 2 and in 1943 the War Office established the Fire Effect Committee.

BRITISH RESEARCH IN WORLD WAR 2

Broadly, research proceeded in two phases, initially theoretical investigations, experiments and trials in UK.  Then field studies by 1 Operations Research Section (ORS) in Italy and 2 ORS with 21 Army Group in NW Europe.  There are indications of some work in the Burma theatre but this seems to have been bequeathed to the Indian Government.  Studies examined hostile and fratricidal artillery fire, and friendly fire on subsequently captured positions.  Both sections had several artillery officers.  The morale effects were a particular concern and there was a lot analysis trying to separate this effect from the others.  

Initial work concentrated on HE fragmentation, the theoretical effectiveness of shells and the vulnerability of particular targets.  This being combined into usable data as ‘area of effect’ (AoE), originally called 'vulnerable areas', for each combination of shell and target type.  This led to such things as the definition of a casualty and, in contrast to the US, UK adopted a probability approach.  For example, “a 50% military loss occurs when a wound causes a disability lasting approximately 6 days” – the 100% loss was a 45-day disability.  This is a very good analytical approach but perhaps less useful as a guide to units in the field.

As an example of target vulnerability research, the mean presentation area of a human body across a variety of postures was calculated as 4.2 ft sq with vulnerable organs occupying 43% of the total at the front and 36% at the back.  Similar approaches can be used for equipment targets.

The Effects of HE Shells

The physical effects from High Explosive (HE) shells are caused by three things, in descending order of importance:

• Fragments or 'splinters' from the shell casing when it fractures under pressure from the expanding gases of the exploding filling, often erroneously called 'shrapnel'.  These fragments rely on kinetic energy for their effect.  A key point is that a high proportion of these fragments are wasted (they go up in the air or straight into the ground).

• Blast can be significant with very large calibre shells and all shells if they detonate in a confined space or very close (inches not yards) to a vulnerable hard target.  Blast is the pressure wave created by the expanding gases released from the fractured shell casing, it causes overpressure, which in turn causes damage.  When a shell penetrates the ground the gas pressure created by the detonation displaces material to form a crater.

• Flash from the heated gases, generally insignificant for artillery HE weapons, the blast and splinters will usually do more damage but flammable materials may catch fire.

Shrapnel was widely used in World War1, but not in World War 2, it was invented at the end of the 18th Century.  In its 20th Century form a time fuze detonated a propelling charge in a carrier shell with a low angle of descent to fire a few hundred balls forwards and downwards in a narrow cone, like a shotgun.

Kinetic energy is the product of a fragment's mass and velocity (½(mass × velocity²)).  The fragmentation of shells and fragment velocity varies depending on the amount and type of explosive and the design of the shell body and type of steel  Initially velocity is close to the detonation speed of the explosive, several thousand feet per second.  By 1941 British research determined that the best size for an anti-personnel splinter was under 1/25 oz, significantly less than existing designs.  The 1907 criteria, apparently developed by France, was 58 ft-lbs to create an incapacitating wound, British research with small fragments suggested closer to 5 ft-lbs was all that was needed.

Most British shells used standard engineering steel, '19-ton' strength; in contrast the US used '23-ton'.  Using normal as opposed to high strength steel made it easier to produce shells.  However, it also meant that shell walls had to be thicker to survive firing stresses, which left less volume for explosive filling and required less energetic explosive to fracture the weaker steel into useful sized fragments.  Table 1 shows the HE percentage in some World War 2 shells.

Table 1 - Shell and Filling Weights

This table shows the 4.5-in shell was seriously deficient in explosive power having only about 3.8-lbs of HE, less than 105-mm!

When an HE shell explodes most of the fragments are projected sideways and forward from the shell's axis, almost nothing goes backwards.  Sideways includes up and down, the latter going into the ground and the former into the air then falling to the ground with very little energy.  Larger fragments tend to go forwards because they have greater mass and less velocity from the explosion so are more affected by the shell's terminal velocity.  Air-burst shells make use of the fragments projected downwards.  The spread of fragments from an HE shell depends on the angle of descent, Figure 1 shows this in plan view.

Figure 1 - Fragmentation Patterns from the Point of Burst at Different Angles of Descent

The Effects of Terrain

The effects of terrain are many and varied and can markedly reduce the direct effects of bursting shells.  In soft ground a shell will penetrate slightly deeper than on hard ground, this means that the crater will absorb more of the fragments.  Of course the efficiency of the fuze's instantaneous mechanism is a factor; inefficient fuzes penetrate deeper giving fewer useful fragments (this was a significant problem with the fuzes used in the early part of World War 1).  When shells were required to penetrate the ground before their fuzes detonated ('delay') the British practice was to fire with fuze caps on.

Fragments fly in straight-lines out from the shell burst with gravity pulling them down as they lose velocity.  Very few targets are on football fields.  'Normal' open ground is 'rough', it has natural folds, small dips and hollows, furrows, ditches, bunds, etc.  These all provide troops with protection from ground bursting weapons, not to mention direct fire projectiles.  'Natural' or 'average’ ground offers about 5 times as much protection to a prone soldier as an 'unnatural' level surface like a foorball field.  Then there are the more obvious results of human activity such as buildings and walls, and military activity, notably trenches.  However, air-burst shells direct their fragments into and behind this natural or artificial protection.

Buildings are a further complication, and their protective properties are dependent on the amount of artillery fire directed at them and the material used to build them.  The blast effect of shells will damage buildings, particularly if there are direct hits, and if there are enough hits the building will be reduced to rubble.  However, most masonry or concrete buildings will stop fragments.  The flash of detonation can ignite flammable materials.

Then there is vegetation.  Fragments and blast will strip foliage and eventually reduce large trees to shattered trunks.  Initially thick vegetation may cause some shells to 'air-burst' but the branches and so on will absorb many splinters, one test for the 58 ft-lbs criterion was that a fragment penetrated about 1 inch into wood.  In heavy bombardments the blast will move the loose and shattered vegetation on the forest floor to the edge of the impact area or pile it up against obstacles.

Flying debris can be a hazard, particularly rubble in built up areas when large shells are used.  In either soft or hard ground artillery shells do not cause a noticeable hazard from flying spoil and forest debris usually offers little danger.  

British research also investigated using artillery fire to cut wire entanglements and clear mines.  This work showed that there were some useful effects, blast being quite effective against some types of Teller mine and shell fragments easily cut wire.  However, they were not reliable or efficient methods due to the natural dispersion of the shells.

The Effects of Target Posture

It's also useful to note how vulnerability changes with target posture because it suggests the relative amounts of fire needed in different circumstances.  The following estimates the relative risks of becoming a casualty to ground-burst shells on ‘average’ ground:-

BATTLEFIELD EFFECTS

The direct effects of an HE shell are one way of looking at the effects of artillery fire.  However, they have to be related to the battlefield.  

British operational research scientists defined artillery effects on the battlefield stating them “in order of their ease of achievement”, although they are all happening in some degree simultaneously.  They were:-

The last two, sometimes called “physical effects,” are much easier to analyse and in the final year of the war most of the research focussed on the first two (“psychological effects”) through operations analysis.  The definition of demoralisation is particularly important; it is specific and is different from a more general deterioration in morale.  There was no information about recovery rates or how long the effect lasted.  There were additional caveats to demoralisation including that fire must be continuous over the period and that the target troops must not be fully protected against the bombardment – in deep or concrete bunkers, etc.  Note to that “a lack of will to resist continuing for some time after the end of the bombardment” is different from the time taken for neutralised troops to decide that fire has ended and move to their fighting positions.

Of course there are also tertiary effects, notably disruption and delay caused by taking evasive or protective action, evacuating casualties and repairing damage, and so on.  And casualties will usually reduce, even if only temporarily, the efficiency of the unit suffering them and may reduce capability if casualties or damaged equipment are not promptly replaced.

Quantitative Effects

British researchers put much effort into investigating and quantifying the effects of artillery fire.  The early focus was on physical effects.  Later work focussed on the much more difficult psychological ones.

Early work established splinter patterns and their relationship with the angle of descent and the all-important AoEs for different types of target.  Figure 2 shows an example of AoE contours.

Figure 2 – 25-pr HE ground-burst, angle of descent 20º - Percentage Casualties to men standing in the open around the point of burst of the shell.

However, the shape of the fragment pattern depends on the shell's angle of descent, while the size of the AoE depends on the type of target and the lethality of the shell.  Nevertheless, data for shells, mortar bombs and aircraft bombs against various targets was provided and explained in Army Operational Research Group Report No 179 'Lethal and Material Effects of Gunfire and Bombing on Land Targets - A Record of the Present State of Knowledge' 20 March 1944.  It was updated in Report No 234, although the changes mainly concerned aircraft bombs, including relative areas for the blast effects of aircraft bombs.  It also considered the effectiveness of phosphorus munitions, best summarised as 'not very'.

Next they produced estimated weights of fire to achieve these effects.  These are in terms of 25-pr equivalence and shown in the next table.

Table 2 – Intensity and Density

*Note - The morale and neutralisation data need to be treated with caution; the evidence for achieving the defined demoralisation in 15 minutes was based on a single operation, at Wesel, during the Rhine crossings.  Before this it was thought that at least 4 hours were needed.  There is also doubt about 25-pr effect equivalence for neutralisation because there were indications that neutralisation correlated with the number of rounds fired rather than their lethality.  

Equivalence Between Calibres

During World War 2 there were far more calibres than now so one need was relating the effects of one calibre of shell to another.  Using a standard target of men in slit trenches, a good approximation of relative effect was the square root of the weight of explosive filling.  Of course this ignores the different power of different explosives.

The following table shows some World War 2 equivalence values where 25-pr equals 1.

Table 3 – 25-pr Equivalence

A notable point is that smaller shells are proportionally more effective than larger ones, remembering that the fragmentation effects are being compared.  Of course larger shells can be fired further, their greater explosive content makes them more effective against more solid targets, if they hit them, and their blast effects more significant.

What it Means on the Ground

Putting Tables 2 and 3 together reveals how many shells of different calibres are needed to achieve the different effects per 10,000 yds², ie 100 × 100 yds.

Table 4 - Density and Intensity per 100 × 100 yds

One often asked question is about the effect of indirect artillery fire on tanks.  One example helps, in 1944 the German IX Corps in Italy reported that artillery fire was the largest cause of its tanks losses, it seems that this was usually from medium and heavy guns controlled by air OPs.  The second largest source was German destruction of damaged or broken-down tanks to prevent their capture (mechanical reliability was not a feature of German tanks).  Other tanks, anti-tank, air attack and mines well below the first two as the causes of tank losses.

CALCULATING WEIGHT OF FIRE AND FINDING THE RESULTS OF FIRE

Calculating density (rounds per 100 yd square) or intensity (rounds per 100 yd square per minute), involves two main steps:-

• Finding the number of rounds to achieve the effect.
• Finding how many rounds to fire so that the necessary number hit the target.

Alternatively the same approach can be used to work out what results would have been for a given amount of fire.  The first step being to find how many shells probably hit the target and then what their result was.

Additional considerations include duration of intensity, the distribution of aim-points and the number of guns required.  The calculations are not absolute, like all gunnery they depend on statistics and probability theory.  However, before doing any calculations the data, AoE, has to be available for the expected target types and their postures.  Further data is also required as will be seen below.

Area of Effect

Figure 1 showed how AoE varies with the angle of descent and a low angle pattern with its conspicuous ‘butterfly’ wings becoming more rounded.

• The AoE becomes more circular as the angle of descent increases and the number of effective fragments increases because fewer are going into the ground or upwards before falling ineffectually.  
• Next, the AoE reflects a probability of effect in the area, not that any target element in the area will suffer the effect and that none outside it will.  
• Finally, the AoE definition may assume that a target in the open is on flat ground without any undulations, etc, providing cover against ground burst shells.

Unless direct hits are sought, HE air-burst is more effective than ground-burst for two reasons: for a given angle of descent there are more useful splinters, particularly at lower angles, and because the splinters strike downwards, instead of horizontal, they reach into holes and hollows.  Clearly, the extent of increased effectiveness will depend on the target.  Post World War 2 trials found that against dug-in targets proximity (VT) fuzed shells varied from about 1.2 to 2.5 times as effective as ground-burst.  However, some data from the war indicated that air-burst could be as much as 10 times as effective as ground-burst.  

One important aspect of air-burst is the height of burst (HOB), 10 yds is about optimal and if the bursts are too high then the effectiveness of the splinters is significantly reduced.  Getting this HOB by predicting the fuze length of 'time' fuzes, whether clockwork or powder burning, was virtually impossible so HOB had to be ranged.  The benefit of VT was its correct and consistent HOB without ranging it.

How many shells must hit the target to get the results?

The first step in weight of fire planning is calculating the number of shells to achieve the required effect on the target. For example:-

• A target has 10 elements in an area 150 yds × 150 yds (22,500 yds²), the aim is to inflict a particular level of damage on any 4 of them (40%) and the AoE for this damage is 10 yds radius (314 yds₂).
• Then N = P/100 × S/A, where N is the number of shells to be fired, P is the required percentage of damaged target elements, S is the target’s area and A is the AoE.  Here N = 29 (density 13 rounds per 100 yd square).

However, this simple model assumes that no target element will be hit effectively more than once.  This is statistically reasonable up to about 10% casualties, higher than this more shells need to be fired to compensate for ‘overhitting’, if 100% casualties are sought then the calculations must be for about 400%.  In the above example N should be 36 for 40% casualties.  A second assumption is that fire will be evenly distributed across the target, and in reality this was almost never the case in World War 2 because the guns fired parallel so the troop layout dictated the spread of individual gun's aim-points.

For example, modern thinking suggests that 30% casualties results in a target being militarily 'destroyed'.  The fine print of definitions and whether or not this is true are not considered further!  However, using the figures in Table 4, and the target of 'casualties to troops in weapon pits', suggest that for each 100 yds × 100 yds then 40 × 15 = 600 25-pr shells are required.  Compensating this for overhitting raises the number to 690. 

When this is compared to the 8 - 32 shells per hour for neutralisation, then the implications of neutralisation versus destruction becomes clear, even without making allowance for shells that miss the target or are ineffective (see below).  Conversion factors in Table 3 can be used to convert to other types of shell.

AoE Shape

Another necessary adjustment is for the shape of the AoE.  This varies with angle of descent, which in turn varies with range and charge.  This problem of AoE shape is one of the keys to accurately estimating the extent of the effect on the target, and is one that was barely touched in the World War 2 work.

How many shells must be fired?

The second calculation step is to increase the theoretical number of shells to compensate for some of them missing the target or being ineffective.  There are several causes of this.  The first three concern 'accuracy' in a general sense.

Target Location Error – there is always some inaccuracy, although it's small for targets that have been effectively ranged (adjusted).

Accuracy of Fire – there are many possible causes of predicted fire inaccuracy - the distance between where the shells were aimed and the mean point of their impact.  However, their magnitude at the target is closely related to range.  The work of the ORS identified lack of accuracy (errors) in predicted fire as a major problem.  

Dispersion – the PEs in range and line, their size varies with range and charge.  Round to round variations in MV are the primary source of the first.  However, the importance of range dispersion depends on the size of the target and the relationship between the line of fire and the target axis.  Against larger target areas whose long axis is parallel to the line of fire dispersion may be a good thing, particularly at shorter ranges.

Slope – depending on the relationship between the direction of the slope of the ground and the line of fire, shells may fall outside the target area or not fully cover it.  1 ORS in Italy seem to have been the only group to incorporate this, no doubt because of its importance in the Italian terrain.

Angle of Descent - the shape of the AoE varies with the angle of descent and is relative to the line of fire.  So the line of fire and the shape of the effect may have more or less effect depending on the layout of the target, the relationship between the line of fire and the target's axis and the amount of dispersion.

Blinds – there will always be some shells that don't explode.

The approach to the first three is to enlarge the target area to hold the PEs, typically combining them using root mean squares.  For adjusted fire the first two become a small PE depending on the adjustment precision.  Slope corrections make the target area asymmetric in relation to its ‘centre’.  Blinds are a percentage matter.  

Finally, considerations of accuracy and particularly dispersion, together with AoE shape come together with the relationship between the long axis of the target and the line of fire.  If the line of fire is across the axis of a long narrow targets far more rounds will be wasted than if the line of fire is along the target's axis.  This is because most dispersion is along the line of fire.

It will immediately be apparent that predicting compensation for accuracy, dispersion, AoE shape and slope again depends on where the guns are, and poses a real ‘chicken and egg’ problem if there is a choice of firing units.

Accuracy and dispersion, including dispersion compensating for inaccuracy, are reviewed in more detail in "Errors and Mistakes".

How good are the calculations?

A key question is the goodness of the models.  British researchers summarised by an example giving the expected number of casualties as 9%.  They then said that it might be as low a 5% or as high as 15% but not as low as 2 or 3% or as high as 30 or 40%.  However, it could be argued that there is not, even today, a good model capable of handling all the variables and being used to either estimate the number of rounds required to produce the desired effects or to estimate the casualties for a given number of rounds.

Data quality is also an issue.  Physical effects are relatively easy to model and validate, although the latter may be expensive and apparently similar targets can vary widely in terms of their vulnerability.  Psychological effects are a different matter, realistic experiments and trials are out of the question on ethical grounds (at least in Western countries, although there are some tantalising hints that the Soviets may have experimented).  Therefore only war provides the data, but is not a good environment for well managed trials and experiments!

Copyright © 2001, 2002 Nigel F Evans. All Rights Reserved.