Comparing Drone Stockpile Surveys with Manual Volume Estimates

How stockpile volumes are measured today

Every operation that moves, stores or sells bulk material needs to know how much it has on hand. Gravel, coal, compost, grain, ore, or construction aggregates, the numbers drive inventory value, reorder timing, and shipping plans. For decades the standard approach was a manual measurement: someone walks the stockpile with a rover and a prism, takes a grid of points, and later computes a volume from those surveyed elevations.

In the past few years an alternative has become practical. A small multi-rotor drone flies over the same stockpile, captures several hundred overlapping images, and photogrammetry software turns those images into a dense point cloud or a digital surface model. The volume is then calculated from that model. The method is often called aerial volume measurement, and it is being adopted by quarries, recycling yards, ports, and farms across many countries.

Both approaches have their advocates. The right choice depends on the size of the site, the frequency of surveys, the required accuracy, and the cost of downtime while measuring takes place. This article compares the two methods so that a site manager or procurement officer can weigh the trade-offs for their own operation.

What manual volume surveys involve

A manual stockpile survey typically uses a total station or a GNSS rover. The surveyor walks around the pile and takes elevation readings at points that capture the shape. On a well-defined pile with gentle slopes a few dozen points might be enough. On a complex pile with multiple ridges, gullies, and an irregular footprint the surveyor may take several hundred points to get a reliable surface model.

The work is physically demanding. The surveyor has to climb the pile, which can be unstable, and must keep the rover rod vertical. Loose material shifts underfoot. Rain, heat, dust, and poor light all slow the process. A large stockpile can take half a day or more to measure, and the site may need to pause operations in that area while the survey is done. For a busy quarry or a port that loads ships continuously, that downtime has a real cost.

Once the field data is collected, the surveyor downloads the points into office software, cleans any outliers, and builds a triangulated surface. The volume above a base plane is then computed. The whole workflow, field work plus office processing, can take a full day for a single large pile.

What drone stockpile surveys involve

A drone survey replaces the ground walk with an aerial flight. The pilot plans a flight path that covers the stockpile with sufficient overlap, typically 75% forward overlap and 65% side overlap. The drone flies at a consistent altitude, often 60 to 100 metres above the pile, and captures images at a set interval. The entire flight for a stockpile of a few hectares takes between 10 and 30 minutes.

Back in the office, photogrammetry software aligns the images using common points and generates a dense point cloud. From that cloud a digital surface model is created. The volume calculation is essentially the same as the manual method, the difference is the density of the input data. Where a manual survey might have 200 points, a drone survey can produce tens of thousands or even millions of points, each representing a measured elevation.

The extra density captures every bump and hollow. That can improve accuracy, but it also means the processing takes more computer time. Most users today upload the images to a cloud platform such as nationaldroneservices.net, where the processing, point cloud generation, and volume calculation happen automatically. The turnaround from flight to final number can be under an hour.

Accuracy and error sources

Accuracy is the first question most operators ask. Both methods can achieve similar results when conditions are ideal, but the error sources are different.

Manual surveys suffer from point spacing. If the surveyor misses a ridge or a depression, the surface model will be wrong. The error is systematic, it depends on the skill of the surveyor and the shape of the pile. On a flat, regular pile the error is small. On a pile with complex geometry the error can be several percent of the total volume. Repeat surveys by the same person also vary, because the points chosen each time are not identical.

Drone surveys have their own error sources. The main ones are ground control, image quality, and vegetation. Without enough ground control points, physical targets with known coordinates, the model can drift or warp. If the stockpile is covered with grass or weeds, the photogrammetry sees the top of the vegetation, not the material surface. Dust, fog, or low sun can also degrade the images. In good conditions, with proper ground control and clean images, drone surveys routinely match or exceed the accuracy of a careful manual survey.

Cost and time considerations

Manual surveys have a low equipment cost. A good GNSS rover costs a few thousand dollars, and most sites already own one. The labour cost is in the surveyor's time and the lost productivity while the pile is being measured. For a site that needs monthly or weekly volumes, that labour cost adds up.

Drone surveys shift the cost from labour to equipment and software. A capable survey-grade drone costs more than a rover, and the photogrammetry software often requires a subscription. However, the flight time is short, and one person can fly the drone from a safe distance while the site continues working. For frequent surveys the per-survey cost drops, and the reduced downtime can offset the initial investment. Many sites now contract the flight to a specialist service, paying per survey and avoiding the capital outlay entirely. National Drone Services is one example of a nationwide provider that handles the capture and processing for clients who prefer not to operate their own drone.

When each method suits best

There is no universal winner. The best method depends on the specific needs of the site.

Manual surveys are still a good fit when:

• The stockpile is small and simple in shape. A single conical pile of sand, for example, can be measured quickly with a rover and a few points.
• The site has no drone operator, no budget for a drone, and only needs an occasional volume check.
• Weather or airspace restrictions make drone flights impossible on a regular basis.
• The surveyor is already on site for other work, so the marginal cost of measuring the pile is low.

Drone surveys become attractive when:

• The stockpile is large, irregular, or has steep faces that are dangerous to walk.
• Volumes are needed frequently, weekly or even daily, so the labour cost of manual surveys is high.
• The site cannot afford downtime. A drone can fly while loaders and trucks continue moving material below.
• A permanent record of the pile shape is wanted for reconciliation or dispute resolution. The dense point cloud from a drone flight serves as a timestamped 3D snapshot.

Combining the two approaches

Some sites use both methods. They run a manual survey once to establish a baseline and then use drone flights for subsequent checks. The manual survey provides a set of ground control points that help anchor the drone models, improving their accuracy. Over time the drone data can be compared with the earlier manual numbers to validate the consistency of the aerial volume measurement process.

Another hybrid approach uses a drone for the top surface of the pile and a rover for the base. If the ground around the stockpile is uneven or obscured by vegetation, the drone may not capture a clean base plane. A few rover shots at the edges fix that, and the two data sets are merged in the processing software. Platforms like nationaldroneservices.net support this kind of mixed workflow, allowing users to upload both image sets and ground survey points into the same project.

Practical steps for getting started

A site manager who wants to compare the two methods for their own operation can run a simple test. Pick one stockpile that is typical of the site. Have a manual survey done in the usual way. On the same day, have a drone survey flown. Compare the two volume numbers. The difference will show the practical accuracy gap for that particular pile and that particular flight setup. Repeat the test a few times, and the pattern becomes clear.

If the drone numbers consistently match the manual numbers within an acceptable tolerance, often 1% to 3% for most bulk materials, then the site can switch to drone surveys with confidence. If the difference is large, the cause is usually worth investigating. It may be a ground control issue, a flight altitude that is too high, or a pile surface that is too reflective or too dark for good image matching.

National Drone Services and other providers often offer a trial flight where they handle the whole process and compare the result with the site's existing manual data. That removes the risk of buying equipment before knowing whether the method works for the specific material and site layout.

Regulatory and safety factors

Manual surveys involve real physical risk. Climbing a stockpile of loose material is one of the more dangerous tasks a surveyor can do. Slips, falls, and being struck by moving equipment are all possible. Drone surveys eliminate that risk entirely, because the operator stays on level ground.

Drone flights come with their own regulatory requirements. In most countries the operator needs a remote pilot certificate, and the flight must stay within visual line of sight unless a waiver is obtained. The aircraft must be registered and insured. These are not major hurdles, but they do mean that a site cannot simply buy a drone and start flying the next day. The training and paperwork take a few weeks to complete.

Making the decision

The choice between manual and drone stockpile surveys comes down to a balance of accuracy, cost, safety, and convenience. For a one-off measurement of a simple pile, a manual survey is hard to beat. For ongoing volume tracking at a busy site, the drone method usually wins on speed and safety, and it can match or exceed the accuracy of a manual survey when done properly.

Many sites today use both, picking the method that fits each survey's purpose. A quarterly stocktake might be done by drone for speed and a full 3D record, while a quick weekly check on a single bin might stay manual. The important thing is to understand the strengths and weaknesses of each approach so that the numbers you rely on for inventory, payment, or planning are as trustworthy as they need to be.

National Drone Services on social media Facebook