Greenhouse moisture removal is a mass-balance question before it is an equipment question. Calculate the moisture source and outdoor-air exchange for each relevant period, keep signs visible, then state what remains unknown.
Why calculate the moisture load by period?
A greenhouse can have the same floor area while its moisture requirement changes with crop stage, irrigation, screen position, ventilation permission, outdoor humidity and event duration. A 24-hour average can hide a night transition or a short post-irrigation period. This method produces a bounded water-vapour removal requirement for a stated control volume and interval. It does not produce a real device capacity, model selection or field performance result.
Start with a boundary and a period
Define whether the control volume is a whole greenhouse, compartment, crop zone or test chamber. Identify the crop and growing-medium boundary; specify the period, temperature, humidity target, pressure basis, ventilation state and storage terms. Keep separate what enters as water, what leaves as drainage or condensate, what is stored, what exchanges with outdoor air and what is removed by a process. Count a measured term once only when its boundary and period are clear.
ET is not automatically pure transpiration
Evapotranspiration (ET) is evaporation plus plant transpiration from the stated crop-and-growing-medium boundary. If media evaporation, wet surfaces or irrigation is not independently resolved, ET is a combined source for that boundary. Do not label it pure transpiration and then add the same evaporation again.
For a measured period with all other water flows zero or separately accounted for, use ET_mass = irrigation − drainage − Δstorage. Express every term as kg of water. Positive Δstorage retains water in the boundary; a negative value releases stored water. Storage is actual water mass, not an uncorrected change in total plant or pot weight: harvest, additions and other non-water changes can affect weighing. Do not count internal returns twice. FAO defines evaporation and transpiration; applying a water balance requires the matching measured boundary and period.
| Storage change | ET mass | 24-hour average |
|---|---|---|
| +2 kg | 12 − 3 − 2 = 7 kg | 7/24 ≈ 0.292 kg/h |
| −1 kg | 12 − 3 − (−1) = 10 kg | 10/24 ≈ 0.417 kg/h |
These are hypothetical checks: 12 kg irrigation, 3 kg drainage, the stated storage change and all other flows zero. A period average does not establish a night peak. Use representative records when the question concerns a short event, and keep ET separate from the whole-greenhouse balance until other sources and sinks are accounted for.
Convert outdoor air on a dry-air basis
Outdoor exchange contributes moisture according to the difference in humidity ratio, not relative humidity alone. If n_ref is an air-change rate for the matching period and V_zone is the effective controlled volume, Q_ref = n_ref × V_zone. State whether Q_ref is actual, standard or another reference volume flow. Internal circulation is not outdoor exchange.
Using the specific-volume and humidity-ratio conventions in ASHRAE Psychrometrics, for a consistent reference state, m_da = Q_ref / v_ref, where v_ref is mixture volume per kg of dry air. W is kg water per kg dry air; convert g/kg to kg/kg before using it. Equal dry-air flows need not mean equal actual volume flows at different states.
For a well-mixed zone with balanced dry-air flow, L_exchange = m_da × (W_outdoor − W_zone). Positive adds moisture to the zone; negative removes moisture. Outdoor RH or cold weather alone does not establish the sign because temperature and pressure determine humidity ratio. The greenhouse psychrometric calculation guide explains the vapour-pressure, humidity-ratio and dew-point conversions.
Hypothetical signed exchange example
Assume Q = 800 m³/h, v = 0.8 m³/kg dry air, outdoor W = 0.006 and zone W = 0.010 kg/kg. Then m_da = 800/0.8 = 1,000 kg dry air/h and L_exchange = 1,000 × (0.006 − 0.010) = −4 kg water/h. Over three constant hours, the contribution is −12 kg water. These values show the sign and duration; they are not a greenhouse measurement or design condition.
Do not turn an opening percentage into airflow. Leakage depends on wind, temperature difference, fans, openings and airtightness. A tracer-decay result applies to its tested weather, mixing and configuration; ACH50 at 50 Pa is an airtightness result, not operating ACH. If the exchange basis is unsupported for the project period, leave it unknown rather than choosing a coefficient. NIST infiltration guidance explains driving forces, while the NBS tracer-decay reference describes method conditions. Changes in background concentration, inter-zone exchange or crop CO2 sources and sinks need to be accounted for before interpreting a decay as outdoor exchange.
Combine internal sources and outdoor exchange
The EnergyPlus moisture-balance reference describes sources, exchange and storage terms. Let G be the total internal moisture source counted once for the same period, in kg water/h. Under steady humidity ratio, a well-mixed zone with balanced outdoor dry-air flow and zero net storage has D = G + m_da × (W_outdoor − W_target). D is the bounded vapour removal requirement in kg water/h. The equation assumes the target zone is the mixed exhaust state and that other surface or adjacent-zone exchanges are zero or already included.
Hypothetical total-balance example
Assume G = 3 kg/h, m_da = 200 kg dry air/h, W_outdoor = 0.012 and W_target = 0.010. The outdoor contribution is 200 × (0.012 − 0.010) = 0.4 kg/h, so D = 3 + 0.4 = 3.4 kg/h. This is a mass-balance output for these assumptions, not verified equipment capacity. If the outdoor term is negative, it reduces removal in that period; it does not erase the internal source.
A negative D means these assumptions imply a net moisture deficit at the target, not negative dehumidifier capacity. With ventilation fixed, dehumidification alone cannot maintain that target. Reconsider permitted airflow, the target and other processes; do not subtract a negative-demand period from a later positive removal requirement.
For intervals that can each be reasonably treated as steady with negligible net storage, use co-occurring inputs and calculate the theoretical positive removal requirement as Σ(max(D_i, 0) × Δt_i). This is a conditional demand total, not actual collected condensate or proof of equipment performance. Where humidity or material storage changes significantly, use a transient balance instead of applying this steady expression unchanged. Do not combine the maximum source, maximum humidity difference and longest duration from different periods. Do not add an unspecified generic peak factor. A short event needs a short-period source, exchange and storage balance.
| Period | G (kg/h) | m_da (kg dry air/h) | W_out | W_target | Exchange | D (kg/h) |
|---|---|---|---|---|---|---|
| A | 4.0 | 200 | 0.006 | 0.010 | −0.8 | 3.2 |
| B | 3.0 | 200 | 0.012 | 0.010 | +0.4 | 3.4 |
The fictional rows show why the largest internal source is not automatically the largest removal period: the exchange term changes sign. They are not weather records, seasonal recommendations or a capacity guarantee.
Convert mass rate to water volume
Keep kg of water as the primary quantity. The USGS water-density reference supports using approximately 1 kg/L for liquid water near ordinary conditions, so a hypothetical 3.4 kg/h sustained for 24 h gives 81.6 kg/day, approximately 81.6 L/day. Six hours at 3.4 kg/h followed by 18 h at 1 kg/h gives 3.4 × 6 + 1 × 18 = 38.4 kg/day, approximately 38.4 L/day. Density varies with temperature, so these are communication conversions, not drainage, purity or rating claims.
Never extend a single reading or nominal daily figure across a full day without an operating schedule. A daily total is an integral of period rates; it is not evidence of a peak rate.
Latent heat is a thermal conversion, not input power
For a stated phase-change state, Q_latent (kW) = m_dot (kg/h) × h_fg (kJ/kg) / 3,600. Hypothetically, 3 kg/h × 2,400 kJ/kg ÷ 3,600 = 2 kW. This is one latent heat component, not electrical input power, total greenhouse cooling load or equipment energy use. Use property values appropriate to the adopted state and do not add the same latent term twice.
The NIST water and steam overview explains latent heat and state-dependent water properties. Use a property value matching the adopted state; the hypothetical 2,400 kJ/kg above is an illustration.
Process airflow is a different quantity
For one steady dehumidifying airstream with known effective inlet and outlet humidity ratios, m_da = D/(W_in − W_out) and Q_in = m_da × v_in. Assume D = 3.4 kg/h, W_in = 0.012, W_out = 0.008 and v_in = 0.85 m³/kg dry air. Then m_da = 850 kg dry air/h and Q_in = 722.5 m³/h. These are assumed inputs, not promised outlet conditions or equipment performance.
Process airflow is not outdoor-ventilation flow, HAF-fan sizing, duct pressure or canopy coverage. If W_in ≤ W_out, the assumed process is not viable. Coil limits, controls, distribution, drains and access require separate validation; see the uneven crop humidity zones guide before treating process airflow as room distribution.
What this method can and cannot conclude
| Can calculate | Must remain open until evidence exists |
|---|---|
| Period-based internal source and outdoor exchange | Actual equipment capacity at project inlet condition |
| Signed kg/h and integrated kg/day or approximate L/day | Real peak performance, model quantity or redundancy choice |
| Latent heat for a stated vapour rate | Electrical input power or total energy balance |
| A hypothetical process-airflow requirement | Whether real dry air reaches the crop zone |
Use the existing greenhouse design-data page for project-input collection, the greenhouse dehumidifier sizing case for a separate condition-dependent case, and the agriculture application page for wider context.
A practical calculation record
Record the boundary, period, source terms, storage convention, outdoor-air path, pressure and volume-flow reference, humidity-ratio inputs, sign convention, equations, units and uncertainty. Mark each input measured, justified or hypothetical. Leave a term open when its period, boundary or state basis is unsupported. This keeps the method useful for engineering review without turning an unresolved load into a product claim.
