After landing: slope, soil and penetration in dropped object studies

A dropped object study counts a hit where the object lands. The landing point is where the method stops: the object falls, it arrives on the seabed somewhere in the drop zone, and whether it hits a pipeline, a manifold or an umbilical is decided at that point. For most seabeds and most objects that is a sound simplification. The object arrives, spends nearly all of its energy in the first contact, and stays more or less where it is.

Not always. Three things happen after landing that the standard hit model does not follow: the object can move, it can go into the seabed, and the seabed under the target can give way. Each one shifts the answer, and they do not all shift it in the same direction. This article sets out what each one does, when it matters, and how to account for it in a study without inventing a model you cannot defend.

What the method assumes at the seabed

In a DNV-RP-F107 style analysis the hit probability comes from two parts: the probability that an object lands in a given area, and the share of that area the target occupies. The target's footprint is widened by the breadth of the object, so a 10 m tubular landing beside a pipeline can still count as a hit even though its centre missed the line. That breadth allowance is the method's only concession to the size of the object, and it is applied at the landing point.

DORAS follows the same approach on a 1 m grid: a landing probability for every cell, and a breadth-adjusted target area within each cell, as set out in sections 2.3 and 2.4 of the methodology. Neither the RP's hit model nor DORAS follows an object after it lands. There is no slope, run-out or soil model in either. What happens next is an engineering judgement, and it belongs in the study's assumptions where the next reader can see it.

Slopes: when the landing point is not the stopping point

On a sloping seabed some objects keep going. Which ones depends far more on shape than on mass:

The seabed matters as much as the object. Firm sand, gravel, rock and hard clay let objects run; soft clay tends to catch them. Local features matter more than the average gradient: trench walls, rock berms, scour pits, the mound around a protection structure and the shoulders of a free span all steer objects towards the lowest point, and the lowest point is often exactly where the pipeline is.

The direction of the error depends on where the target sits. A target downslope of the drop zone collects objects that landed upslope of it and rolled, so the landing count under-states its exposure. A target upslope loses some objects that land on it and then move off, but in practice an object that lands on a line and then rolls has usually already struck it. The net effect is non-conservative for downslope targets and roughly neutral elsewhere.

The energy of a run-out hit is a different matter. An object that has already struck the seabed and rolled has lost most of the energy it arrived with. Run-out hits are low-energy hits. They matter most where the damage threshold is low: flexible flowlines, umbilicals, coatings and small-bore lines, and wherever the hit frequency itself is the reported figure, for example for snagging or displacement.

Accounting for run-out without a run-out model

The aim is not to build a soil mechanics model for every study. It is to decide whether the effect is material, and if it is, to bound it with a method the reviewer can follow.

1. Screen the slope. Take the gradient across the drop zone from the bathymetry, with its source and date. On gentle slopes over soft clay, say so in the study and move on: that is a legitimate conclusion, provided the basis is written down.
2. Identify the movers in the lift manifest. Round and slender objects that can land side-on are the candidates. If the lift manifest already groups objects by shape, this is a filter, not new work.
3. Widen the target's catchment upslope, for the movers only. An object that lands within its run-out distance upslope of a target and moves towards it will reach the target. The simplest defensible allowance is to extend the target's exposed footprint upslope by an assumed run-out distance, and to apply that extended footprint only to the movers' share of the lifts. It is an approximation, it is conservative if the run-out distance is chosen generously, and it can be done with the target geometry the analyst already enters.
4. Run it as a sensitivity and report the difference. If the downslope targets' hit frequencies barely move, the effect is not material and the study says so. If they move a lot, that is a finding in its own right, and usually an argument for moving the lift path rather than for refining the run-out estimate.
5. Keep the energy honest. Do not carry the full arrival energy into the run-out hits. Either treat them in the lowest energy band, or report them as hit frequency only and state that they are not expected to cause damage above a stated level.

Penetration: how far the object goes into the seabed

Penetration of the object into the soil matters mainly for buried lines. Burial depth is real protection: the cover absorbs energy before anything reaches the pipe. But the credit depends on the object as well as the cover. A compact or flat object arriving on soft clay spreads its load and stops near the surface. A slender object arriving end-on concentrates its load on a small area and can go much deeper for the same energy.

So a burial credit that is set purely as an energy allowance can be generous for slender objects arriving end-on. The practical check is to ask which objects in the manifest could arrive end-on, and whether the cover depth is credible against them. Where it is not, give those objects a reduced or zero burial credit and record why.

Soil under the target: where some of the energy goes

The third effect works the other way. A pipeline resting on soft soil is pushed into the seabed when it is struck, and part of the impact energy goes into deforming the soil rather than denting the pipe. On rock or hard ground the pipe takes all of it. Treating the line as rigidly supported is the conservative choice, and it is what most studies do without saying so.

That conservatism can be large on very soft clay, and it is tempting to claim it. If you do, the basis has to be shown: the soil description and its source, how the energy reduction was estimated, and which targets it was applied to. A credit with no basis is a number nobody can check.

What to write down

Most of the value here is in the assumptions register, not the calculation. For any study over a seabed that is not flat and uniform, the next analyst should be able to find:

• the bathymetry source, its date, and the gradient range across each drop zone;
• the soil description and where it came from;
• which objects were treated as movers, and why;
• any run-out allowance, its distance, and the energy assigned to run-out hits;
• any burial credit, and whether it was reduced for slender objects arriving end-on;
• whether any soil-absorption credit was taken under the target, and its basis.

Each of those can legitimately be "not considered, because...". What matters is that the reason is written down and the reader can disagree with it.

The landing point is where the counting stops

The standard method counts hits where objects land, and for most studies that is the right level of detail. Slopes, soil and penetration are second-order for most targets and first-order for a few: the downslope flexible, the buried line under a pipe-handling route, the manifold in a hollow. The job is to find those few, bound the effect with a method that can be checked, and say plainly what was and was not modelled. DORAS gives you the landing-point picture at 1 m resolution, so you can see which targets sit close enough to the edge of a drop zone, or low enough on a slope, to deserve the second look. See the features page for what the hit-frequency maps show.