Three numbers appear on almost every dried, dehydrated or low-moisture food specification, and they are routinely treated as if they measured the same thing. They do not.
Moisture content is how much water is in the product, expressed as a percentage of mass. Water activity is how much of that water is available to participate in reactions and to support microbial growth, expressed as a ratio between zero and one. Shelf life is a claim about how long the product remains acceptable, and it is a conclusion drawn from evidence rather than a measurement at all.
Getting the relationship right matters because a specification that controls the wrong one controls nothing.
Why moisture alone does not predict stability
Water in a food is not all in the same state. Some of it is free, behaving as ordinary liquid water. Some is bound – held by sugars, proteins, salts and polysaccharides, chemically unavailable even though it is physically present and counted by the oven.
A high-sugar fruit and a low-sugar vegetable at the same moisture percentage therefore behave completely differently. The sugar binds water, so the fruit has a lower water activity and is more stable, while the vegetable at the same moisture has more free water and supports growth. This is the reason a jam at a high moisture content is shelf stable and a cooked vegetable at the same moisture is not.
Water activity is the measurement that captures the difference. It is defined as the ratio of the vapour pressure of water in the food to the vapour pressure of pure water at the same temperature, which is the same thing as the equilibrium relative humidity of the air in a sealed headspace above the product, divided by 100.
What the water activity scale means in practice
Microorganisms have minimum water activity requirements below which they do not grow. Those thresholds are organism-specific, and they are the reason water activity is a control measure rather than a quality parameter.
- Most spoilage bacteria stop growing well before the scale approaches the values found in dried products.
- Staphylococcus aureus is the pathogen that tolerates the lowest water activity of practical concern, which is why it sets the boundary in several product categories.
- Yeasts and moulds continue to grow at lower water activities than bacteria, and osmophilic yeasts and xerophilic moulds are the organisms that actually spoil dried fruit.
- Below the growth thresholds, microbiological failure stops being the limiting mechanism and chemical change takes over: lipid oxidation, non-enzymatic browning, vitamin loss and texture change.
The practical consequence is that a low water activity product does not have an indefinite shelf life. It has a shelf life limited by chemistry and by packaging rather than by microbiology, and the failure mode is rancidity, colour change, hardening or loss of crispness rather than spoilage.
Because these thresholds are organism-specific and product-specific, and because the applicable microbiological criteria and the food safety management obligations sit in EU law – the microbiological criteria regulation and the hygiene regulation, applied through a HACCP-based system – a specification should reference the criteria that apply to the product category rather than reproduce a general table.
How each is measured
| Measurement | What it determines | Typical method | What can go wrong |
|---|---|---|---|
| Moisture content | Total water, as a percentage of mass | Oven drying to constant mass, Karl Fischer titration for low-moisture or high-sugar products, near infrared for rapid in-line checks | Oven methods drive off volatiles other than water; NIR needs calibration against a reference method for that product |
| Water activity | Available water, as a ratio | Equilibrium measurement in a sealed chamber using a dew point or capacitance sensor | Temperature sensitivity: the reading is only valid at the stated temperature, and a sample that has not equilibrated reads wrong |
| Shelf life | Time to unacceptability | Real-time storage study, supported by accelerated studies and challenge testing where relevant | An accelerated study extrapolated without validation is a prediction, not a result |
Two practical points. First, always report the temperature with a water activity result; the value moves with temperature, and a result quoted without one is incomplete. Second, moisture and water activity are not interchangeable even within one product category, because the relationship between them is a curve rather than a constant.
The sorption isotherm
That curve is the moisture sorption isotherm: a plot of moisture content against water activity at a fixed temperature. It is specific to the product and it is the tool that connects the two measurements.
The isotherm tells you three useful things.
- How much moisture corresponds to your target water activity. This is how a drying specification is set: pick the water activity that gives the stability you need, read off the moisture, and control moisture on the line because it is the faster measurement.
- How steep the curve is in your working range. Where the curve is steep, a small moisture gain produces a large water activity rise, and the product is unforgiving of packaging failure. Where it is flat, there is buffer.
- Whether the product will pick up or give off water in a given environment. A product below its equilibrium with the ambient air will absorb; above it, it will lose. That is what drives moisture migration between components in a mixed product, and it is why a crisp freeze dried piece in a cereal goes soft.
Hysteresis is worth knowing about: the adsorption and desorption curves are not identical, so a product that dried down to a moisture is not in the same state as the same product that absorbed up to it.
Establishing a defensible shelf life
A shelf life is a conclusion, and the file behind it needs four elements.
- A definition of the end point. Unacceptable on what basis – microbiological, sensory, chemical, physical, or a labelling parameter such as a nutrient content claim? Different end points give different answers on the same product.
- Storage conditions that reflect reality, including the packaging that will actually be used. A shelf life determined on foil laminate does not transfer to a polyethylene bag.
- A study. Real-time storage with defined sampling points is the reference. Accelerated testing at elevated temperature is a legitimate screening tool where the degradation mechanism is temperature-dependent and the acceleration factor has been validated for that product, and it is not a substitute for real-time data on a new product.
- A safety margin, and a statement of the assumptions. Shelf life determination is the food business operator’s responsibility under EU food law, and the operator has to be able to justify the date it applies.
For frozen product the logic is different again. Microbial growth is arrested by temperature rather than by water activity, so shelf life is set by quality deterioration – dehydration, freezer burn, oxidation, texture loss – and is governed by the storage temperature and its stability. Shelf life of frozen fruit covers what that means in practice, and shelf life against best before in food trade covers the date-marking side.
Where the numbers go wrong
A single result on a composite sample. Moisture and water activity vary within a lot, and in a dried product they vary within a bag. One measurement on a well-mixed composite describes the average and says nothing about the wettest piece, which is where mould starts. For a product near its limit, the sampling plan matters as much as the limit.
Equilibration not allowed for. A water activity reading taken before the sample has equilibrated with the sensor chamber reads low or high depending on which direction it is moving. A result that arrives suspiciously fast is usually a result taken too early.
Temperature drift. Both the measurement and the product are temperature sensitive. A sample measured warm in a warm laboratory and a sample measured cool are not comparable, and neither is comparable to the product sitting in a container in summer.
The wrong method for the matrix. Oven drying a high-sugar fruit can caramelise the sample and drive off volatiles that are not water, giving a moisture result that is too high. Karl Fischer titration is the reference for those matrices, and a specification that names “moisture” without naming the method has left the door open.
Packaging assumed rather than specified. Water activity is a property of the product in equilibrium with its environment. Change the barrier and you change the trajectory. If the shelf life was established on one packaging format, the specification has to name it.
Mixed products treated as one product. In a mix – fruit and cereal, nut and dried fruit, a seasoning blend – the components equilibrate with each other, and the resulting water activity sits somewhere between the starting values. The stable component drags the unstable one down and the wet one drags the dry one up. Formulating a mix means matching water activities, not moisture contents.
What belongs in the specification
- Moisture, with the method named and the tolerance stated.
- Water activity, with the maximum, the method and the temperature at which it is measured.
- Where relevant, the sorption behaviour or at least a statement of the packaging required to hold the water activity.
- The shelf life, the storage conditions it assumes and the date-marking format.
- The end point on which the shelf life was determined, so that a dispute about acceptability has a reference.
A specification that carries a moisture figure alone is controlling the easy measurement and leaving the meaningful one open. A specification that carries water activity without the temperature, or a shelf life without the storage conditions, is doing the same thing in a different place.
Water activity measurement sets out the reference method behind the figure. In the dried fruit category the same relationship has to account for two further variables: sulphur dioxide, which suppresses spoilage independently of water activity, and the packaging barrier, which decides how long the equilibrium the number describes actually holds.
Vorezan does not perform analysis and does not set shelf lives. Where a shelf life, a microbiological criterion or a date-marking decision has regulatory consequences, it should be established with the operator’s own laboratory and, where appropriate, verified against the criteria in force at the time.