The Complete Guide to Silage Inventory Measurement

Why silage inventory measurement matters

Feed is the largest single cost on a dairy operation — typically 50–60% of operating expenses. Within that cost, silage shrink represents 5–20% of inventory loss on most operations, and over 30% on some. Yet most dairies still measure silage inventory with the same tools they used twenty years ago: a tape, a probe, and a spreadsheet.

The disconnect is meaningful. Every percentage point of feed-system shrink reduction recovers exactly 1% of the annual feed bill. On a 2,500-cow dairy carrying ~$6 million of annual feed cost, the difference between 15% shrink and 10% shrink is $300,000 of recovered margin every year. On a 1,500-cow dairy at ~$3.8M of feed, the same 5-point cut is ~$190,000. The same operation cannot identify where shrink is concentrated, target interventions, or measure improvement without inventory measurement that holds up at the pile level.

Accurate silage inventory measurement is the foundation for:

• Shrink quantification — what gets measured gets managed.
• Run-out forecasting — when does each silo run dry, and when do you need to start a transition?
• Harvest planning — how many acres of corn silage do you need next year?
• Feed-purchase timing — when do you bid for outside feed and at what volume?
• Operating-partner conversations — nutritionists, agronomists, packing crews respond to numbers.

Measurement methods compared

Modern silage solutions cluster into three measurement methods, each with a different trade-off curve. Manual probing is the cheapest and least accurate. Drone photogrammetry is the most accurate and most equipment-dependent. 2.5D satellite mapping splits the difference for operations without drone access. The right choice depends on accuracy required, equipment available, and cadence.

The right method depends on the operation. A single-site dairy with two bunk silos can manage with manual measurement plus quarterly drone flights. A multi-site operation with a dozen piles benefits dramatically from automated workflows that combine drone and satellite data. For the broader question of how drones fit into a working dairy program, see the drones on dairy farms guide.

Drone photogrammetry, step by step

Drone-based 3D measurement is currently the most accurate practical method for silage inventory. The workflow has three parts: setup, flight, and processing.

1. Setup: Ground Control Points (GCPs)

Before any drone flight, place surveyed reference markers around the pile. GCPs are the single largest factor in drone measurement accuracy — they are what make drone-based volumetrics trustworthy enough for operational decisions. More on GCPs and accurate positioning →

Five to eight GCPs per pile is typical. They should be placed at the corners and midpoints of the pile perimeter, with surveyed positions accurate to centimeter level. RTK-GPS rovers handle this in 10–15 minutes per site.

2. Flight: structured grid pattern at 80% overlap

The drone flies a pre-planned grid pattern at 50–100 feet altitude, taking photos with 80% forward overlap and 70–80% side overlap. A consumer-grade quadcopter (DJI Mavic class) is sufficient for most operations; enterprise drones with built-in RTK reduce reliance on GCPs but cost considerably more.

Flight time per pile is typically 8–15 minutes. A skilled operator can fly four to six piles per hour.

3. Processing: photogrammetry to 3D model to volume

Drone images are uploaded to photogrammetry software, which:

1. Identifies common features across overlapping images.
2. Triangulates 3D positions of those features.
3. Produces a point cloud, then a textured mesh of the pile surface.
4. Anchors the mesh to known coordinates using the GCP positions.
5. Measures volume directly from the mesh, integrated against a defined base plane.

Modern pipelines do this in minutes per pile. The output is a volume number that holds up against ground-truth weights consistently within a few percent.

2.5D satellite mapping (no drone)

Not every operation has a drone or wants to fly one. 2.5D satellite mapping fills the gap for these operations: pile volumes are estimated from recent commercial satellite imagery in a browser, with no equipment beyond a laptop.

The trade-off is accuracy. Satellite-based pile measurement typically runs ±5–10% — better than manual measurement, worse than drone with GCPs. The advantage is speed. A multi-site operation can measure every pile in an afternoon without leaving the office.

Most operations use both: drones for the headline accuracy on the largest piles, satellite for routine quarterly checks across the rest of the operation. SilagePlan handles both workflows in a single dashboard.

Density sampling: turning volume into tonnage

Volume is half the equation. The other half is density — pounds of dry matter per cubic foot.

Industry references for dry matter density:

• Bunker silos: 14–16 lbs DM/ft³ when properly packed.
• Drive-over piles: 10–13 lbs DM/ft³.
• Ag bags: 15–17 lbs DM/ft³ depending on bagger.

Density isn't uniform. It is highest near the bottom of the pile where the weight of overlying silage compresses the lower layers, and lowest near the top. Sampling at multiple depths gives a more accurate average than a single face core.

Density also varies by harvest year, packing equipment, and crew. Sample your own piles rather than relying on industry averages — the difference between 12 and 15 lbs DM/ft³ on a 50,000-ft³ pile is 75 tons of dry matter, or roughly $13,500 at $180/ton.

Quantifying shrink over time

Repeated measurements at known intervals, integrated with feedout records, reveal silage shrink at the pile level over time.

The math:

Shrink (%) = 1 − (DM tons fed out + DM tons remaining) ÷ DM tons put up

Where:

• DM tons put up = harvest volume × density × DM% — measured at the post-harvest baseline flight.
• DM tons fed out = sum of mixer-load records over the period, converted to dry matter.
• DM tons remaining = current pile volume × current density × DM% — measured at the current flight.

The advantage of repeated measurement is that shrink becomes visible at the pile level. Operations typically discover that one or two piles consistently lose more than the rest, often traceable to a specific harvest day, a packing crew, or a cover quality issue. Inventory data tells you where to look.

Forecasting run-out and feed adequacy

Once inventory is measured accurately, run-out forecasting becomes mechanical.

Days of feedout = (DM tons available × 2000) ÷ (lactating cows × DMI per cow)

For a 1,000-cow herd at 55 lbs DMI per cow per day, every 100 tons of silage dry matter buys ~3.6 days of feedout. The point isn't precision — it's whether you have 60 days of feed or 120 days of feed before the next harvest. The difference is measured in trips to the feed mill, ration transitions, and cash-flow timing.

Use the silage volume calculator to estimate days-of-feedout for your operation directly.

Forecasting is most useful when integrated with harvest planning. Aurox customers commonly use SilagePlan inventory data to back into next year's corn-silage acreage requirement, factoring in expected shrink, expected DMI, and a small safety margin.

Building a silage inventory measurement workflow

A typical workflow on a multi-site dairy:

1. Post-harvest baseline. Within 30 days of harvest, fly every pile with drone and GCPs. This is the opening inventory.
2. Density sampling. Sample 3–5 cores per pile at the face. Calibrate the volume-to-tonnage assumption.
3. Quarterly drone or satellite check. Re-measure each pile at intervals. The pace should match the operation's risk tolerance for inventory drift.
4. Feedout records integration. Pull mixer-load data into the same system that holds inventory. Compute shrink as the gap.
5. Run-out forecasting. Update projected end-of-feed dates after each measurement. Trigger ration-transition planning 60–90 days before run-out.
6. Annual harvest planning. Use the prior year's measured shrink, current cow numbers, and expected DMI to size next harvest.

SilagePlan implements this workflow as a single product — drone or satellite imagery in, measured inventory + shrink + run-out forecasts out. The point isn't the software. The point is having a workflow that survives at scale.

Frequently asked questions