Answers are grounded in this paper and cite their sources.
Hay baled too wet loses feed value to mould, exposes animals to mouldy feed, and can heat itself to the point of combustion. Advice to farmers is often given as a single moisture number for all hay. This paper tests that practice with a coupled model of microbial heat generation, conductive heat loss, post-baling drying, and mould growth, applied to small square, large round, and large square bales. Under the stated assumptions (ventilated storage, ambient temperature 25 °C, no preservative), the maximum safe baling moisture is 23.6% wet basis for small square bales, 17.2% for large round bales, and 16.5% for large square bales — in every case set by mould growth during drying, not by self-heating. The commonly quoted single figure of 20% passes the small square bale (mould index 0.56, peak centre temperature 25.1 °C) and fails both large classes (mould index 2.37 for large round and 3.32 for large square, against a spoilage criterion of 1). The thermal bound separates the classes even more sharply when drying stalls: at a constant 25% moisture and 25 °C ambient, a small square bale has a stable centre temperature of 25.8 °C, while both large classes have no stable equilibrium at all and run away. Two honest negatives are printed at full size. First, at the centre of a large stack the ceiling falls to 15.4%, so the single-bale ceilings are unsafe there. Second, if baled moisture does not fall, any hay above 14.0% moisture (the water activity 0.70 point in this model) moulds eventually, whatever the bale type. The ceilings are also sensitive to the drying rate, the least certain input: drying twice as slow lowers them to 19.4%, 16.0%, and 15.6%. A field rule is given: bale at or below the ceiling for the bale class, treat stack interiors by the stack ceiling, and probe the centre of any large bale baled above its ceiling, or stored in hot weather, every 1 day, because in the modelled runaway case a large bale goes from ambient to 50 °C in under 5 days. AI disclosure: This paper was prepared with the assistance of Dave, an artificial intelligence system, which wrote the simulation code, ran the computations, and drafted the text under the author's direction. All results are SIMULATED, NOT MEASURED.
No takes yet. Share an insight, caveat, or question.
William Alexander Aldred (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: