nx_water_activity.nx

nx_water_activity.nx -- FOOD-SCIENCE SUITE / WATER ACTIVITY + HURDLE rung. Water activity is the second great preservation variable after pH, and it falls straight out of arithmetic this ecosystem is already doing: the same dissolved-solute count that sets the freezing point sets aw, because both are colligative. nx_icecream already computes those moles, so aw costs one Raoult ratio rather than a new model. aw = n_water / (n_water + n_solute) THE POINT IS THE HURDLE MODEL, NOT THE NUMBER. Food is not made safe by one barrier but by whichever barrier actually holds, and until now the ecosystem could reason about two of them in isolation: nx_chem_solution owns the pH 4.6 line, nx_thermal_process owns lethality and the cold chain. Neither knew about water. wa_hurdle_verdict composes all three and NAMES the barrier doing the work: aw < 0.85 -> safe by WATER ACTIVITY (FDA non-PHF); honey and milk powder live here, and nothing else matters pH < 4.6 -> safe by ACID low-acid, ambient -> only a 12D thermal process will do low-acid, chilled -> pasteurisation plus the COLD CHAIN Naming the hurdle matters: "safe" without knowing which barrier holds is how a reformulation quietly removes the only thing that was working. AND IT IS WILLING TO SAY THE UNCOMFORTABLE THING. An ice cream mix comes out at aw ~0.977, which is well above the 0.93 that C. botulinum needs. The mix is NOT protected by water activity, and this rung says so rather than reporting a reassuring number -- that is precisely why the cold chain is load-bearing for frozen dessert. Units: aw is a 0..1 ratio carried as _q4 (x10000), because the decisions sit at the third decimal (0.850 vs 0.858) and _q3 would round them away. Grounding (cited; researcher-groundable): raoult_law_water_activity_ideal_solution fda_21cfr113_114_potentially_hazardous_food_aw_0_85 icmsf_minimum_water_activity_for_growth_by_organism beuchat_xerophilic_mould_growth_limit_0_61 genealogy_id: hurdle_technology + nishi_food_science_suite