Where the plant lives

Wild Dendrobium officinale is an epiphyte or lithophyte: it grows on tree bark and on rock, not in soil. Its records come from humid mountains across southern China, from the Dabie Mountains and eastern Zhejiang to Guangxi and south-eastern Yunnan, where it roots in thin pads of moss and humus on rock faces and trunks. Such places share three physical features. Rain runs off within minutes, so the roots are never waterlogged. The air moves, so leaves and stems dry quickly after rain or mist. And the nights are markedly cooler than the days. Each feature has a counterpart in the greenhouse, and each can be measured.

Carbon fixed at night

Most plants take in carbon dioxide by day, when light drives photosynthesis. Plants with crassulacean acid metabolism (CAM) open their stomata at night instead, store the carbon as malic acid, and close their stomata in the heat of the day, which saves a great deal of water. Many epiphytic orchids use CAM, and the genus Dendrobium contains species that are strictly C3, species that are strictly CAM, and species that switch between the two.

Night-time uptake matters for a grower because it happens when the air is cool and humid, so each unit of carbon costs less water. Chapter 2 returns to this when it compares warm and cool nights.

Saturation, deficit and dew point

Air can hold only so much water vapour, and the limit rises steeply with temperature. The saturation vapour pressure $e_s$ is well described by the Tetens formula (Tetens 1930; Murray 1967):

$$e_s(T) = 0.6108 \exp\!\left(\frac{17.27\,T}{T + 237.3}\right)\ \text{kPa},$$

with $T$ in °C. The actual vapour pressure is $e_a = e_s \cdot \text{RH}/100$. Two derived quantities matter. The vapour-pressure deficit $\text{VPD} = e_s – e_a$ measures how strongly the air pulls water from wet surfaces, including the inside of a leaf. The dew point is the temperature to which a surface must cool before water condenses on it from that air.

import math

def es(t):
    """Saturation vapour pressure over water, kPa (Tetens 1930, in Murray's 1967 form)."""
    return 0.6108 * math.exp(17.27 * t / (t + 237.3))

def dew_point(t, rh):
    """Temperature at which the air's actual vapour pressure becomes saturating, °C."""
    g = math.log(rh / 100) + 17.27 * t / (t + 237.3)
    return 237.3 * g / (17.27 - g)

# Greenhouse states: (label, air temperature °C, relative humidity %)
states = [("day, open vents", 26, 65), ("day, closed and misted", 26, 92),
          ("night, well ventilated", 14, 80), ("night, closed", 14, 97)]
for label, t, rh in states:
    ea = es(t) * rh / 100
    vpd = es(t) - ea
    td = dew_point(t, rh)
    print(f"{label:24s} {t:2d} °C {rh:2d} % RH: VPD {vpd:4.2f} kPa, "
          f"dew point {td:4.1f} °C ({t - td:3.1f} K below the air)")
# day, open vents          26 °C 65 % RH: VPD 1.18 kPa, dew point 18.9 °C (7.1 K below the air)
# day, closed and misted   26 °C 92 % RH: VPD 0.27 kPa, dew point 24.6 °C (1.4 K below the air)
# night, well ventilated   14 °C 80 % RH: VPD 0.32 kPa, dew point 10.6 °C (3.4 K below the air)
# night, closed            14 °C 97 % RH: VPD 0.05 kPa, dew point 13.5 °C (0.5 K below the air)

Relative humidity on its own is a poor guide. Air at 97 % RH and 14 °C has a VPD of only 0.05 kPa and a dew point just half a degree below the air. Any surface that falls half a degree below the air will become wet.

Why a leaf gets wet at night

On a clear night a leaf radiates heat to the sky or to a cold roof, and so it becomes colder than the surrounding air. The air warms it back through the thin layer of still air around the leaf, the boundary layer. The faster the air moves, the thinner that layer and the closer the leaf stays to air temperature (Campbell and Norman 1998).

import math

# A leaf at night loses long-wave radiation to a cold roof or sky and is warmed by the air
# through its boundary layer. Steady state: h · (T_air − T_leaf) = R_loss.
R_LOSS = 30.0      # W m⁻², net long-wave loss of the upper leaf face under a clear night sky
D = 0.03           # m, characteristic leaf width (a Dendrobium leaf is about 2–3 cm wide)
CP = 29.3          # J mol⁻¹ K⁻¹, molar heat capacity of air

def h_leaf(u):
    """Heat transfer coefficient of both leaf faces, W m⁻² K⁻¹. Boundary-layer conductance
    0.135·√(u/d) mol m⁻² s⁻¹ per face, × 1.4 for turbulent field air (Campbell & Norman 1998)."""
    return 2 * CP * 1.4 * 0.135 * math.sqrt(u / D)

def dew_point(t, rh):
    g = math.log(rh / 100) + 17.27 * t / (t + 237.3)
    return 237.3 * g / (17.27 - g)

T_AIR = 14.0
for rh in (85, 95):
    td = dew_point(T_AIR, rh)
    print(f"night air {T_AIR:.0f} °C, {rh} % RH: dew point {td:.1f} °C")
    for u in (0.05, 0.2, 0.5, 1.0):
        dT = R_LOSS / h_leaf(u)
        leaf = T_AIR - dT
        state = "WET: dew forms" if leaf < td else "dry"
        print(f"   air speed {u:4.2f} m/s: leaf {dT:3.1f} K below the air, {leaf:4.1f} °C -> {state}")
# night air 14 °C, 85 % RH: dew point 11.5 °C
#    air speed 0.05 m/s: leaf 2.1 K below the air, 11.9 °C -> dry
#    air speed 0.20 m/s: leaf 1.0 K below the air, 13.0 °C -> dry
#    air speed 0.50 m/s: leaf 0.7 K below the air, 13.3 °C -> dry
#    air speed 1.00 m/s: leaf 0.5 K below the air, 13.5 °C -> dry
# night air 14 °C, 95 % RH: dew point 13.2 °C
#    air speed 0.05 m/s: leaf 2.1 K below the air, 11.9 °C -> WET: dew forms
#    air speed 0.20 m/s: leaf 1.0 K below the air, 13.0 °C -> WET: dew forms
#    air speed 0.50 m/s: leaf 0.7 K below the air, 13.3 °C -> dry
#    air speed 1.00 m/s: leaf 0.5 K below the air, 13.5 °C -> dry

Two ways of keeping leaves dry follow directly: lower the humidity, so that the dew point is further below the air, or move the air, so that the leaf stays closer to air temperature. Small fans that keep air moving at 0.3–0.5 m/s across the plants do the second at modest cost. Under a roof that blocks the clear sky, or with a thermal screen drawn at night, the radiative loss is smaller and so is the risk.

Wetness and disease

Keeping leaves and stems dry therefore lowers the risk of infection for many pathogens without any spray. It does not remove the risk: pathogens also spread through infected planting material, contaminated substrate and wounds, and some, such as S. rolfsii, live in the substrate itself. Clean planting stock, clean bark and good drainage are part of the same strategy.