A word borrowed from wine

Winemakers use the French word terroir for the way a place shows in what grows there: its soil, its slope, its climate and the hands that work it. In viticulture the idea has been studied for decades, and the evidence is that the main effects run through a few measurable things, chiefly how much water the vine gets, how warm it is and how much light reaches the fruit (Leeuwen and Seguin 2006). The Chinese tradition has its own word for the same intuition, daodi (道地), “of the proper region”, applied to medicinal materials whose best sources were recorded over centuries. Chapter 7 returns to that tradition. This chapter asks what terroir is in chemical terms.

The answer begins with a distinction. A plant’s primary metabolism (photosynthesis, respiration, the making of sugars, amino acids and cell walls) keeps it alive. Its secondary metabolism makes molecules it does not need in order to live from day to day but that help it survive particular threats: phenolic compounds that absorb ultraviolet light, pungent compounds that deter insects, tannins that make tissue unpalatable. Many of the molecules people value in herbs and fruits are secondary metabolites, and their amounts respond to the environment much more than the amounts of primary metabolites do.

Two sites can therefore produce different raw material from the same plant for two separate reasons. The first is how much carbon the plant has to spend, which is set by photosynthesis minus respiration. The second is how it spends that carbon, which is set by signals such as light, cold and drought. This chapter models the first, because it can be calculated; the second needs measurement crop by crop.

The Arrhenius equation

Almost every chemical reaction in a living cell runs faster when it is warmer. In 1889 Svante Arrhenius described how the rate constant $k$ of a reaction depends on absolute temperature $T$ (Arrhenius 1889):

$$k = A\,e^{-E_a/RT}$$

where $E_a$ is the activation energy, $R$ the gas constant (8.314 J mol$^{-1}$ K$^{-1}$) and $A$ a constant. Plant physiologists more often use $Q_{10}$, the factor by which a rate rises for a 10 °C increase. Plant respiration typically has a $Q_{10}$ near 2 over the ordinary range of temperatures, although plants partly acclimate when temperatures stay high or low for days (Atkin and Tjoelker 2003). The two descriptions are linked:

import math

R = 8.314                      # gas constant, J/(mol·K)

def arrhenius_ratio(Ea, T1, T2):
    """How many times faster a reaction runs at T2 than at T1 (°C)."""
    return math.exp(Ea / R * (1 / (T1 + 273.15) - 1 / (T2 + 273.15)))

# A Q10 of 2 near 20 °C corresponds to this activation energy:
Ea = R * math.log(2) / (1 / 288.15 - 1 / 298.15)
print(f"Ea for Q10 = 2 between 15 and 25 °C: {Ea / 1000:.1f} kJ/mol")
for lo, hi in ((5, 15), (15, 25), (25, 35)):
    print(f"  Q10 from {lo} to {hi} °C: {arrhenius_ratio(Ea, lo, hi):.2f}")
# Ea for Q10 = 2 between 15 and 25 °C: 49.5 kJ/mol
#   Q10 from 5 to 15 °C: 2.10
#   Q10 from 15 to 25 °C: 2.00
#   Q10 from 25 to 35 °C: 1.91

Day and night

Photosynthesis happens only in the light. Respiration happens all the time, by day and by night. A plant’s net gain in a day is roughly the carbon fixed in daylight minus the carbon respired over 24 hours. Respiration depends on temperature through the Arrhenius relation, so warm nights cost a plant carbon that cool nights do not.

That is the basis of a common explanation of why highland crops are often richer in sugars and other stored compounds: warm days for photosynthesis, cool nights that slow respiration. The idea is right in direction but incomplete, because photosynthesis has its own temperature optimum. When days are too hot, photosynthesis falls and daytime respiration rises. A model makes the trade-off visible.

import math

def day_balance(t_mean, swing, t_opt=25.0, width=10.0, r20=0.12, q10=2.0):
    """One day, hour by hour. Temperature is a sine wave peaking at 14:00;
    light is a sine arch from 06:00 to 18:00. Units are arbitrary:
    1.0 = gross photosynthesis per hour at full light and t_opt."""
    gross = resp_day = resp_night = 0.0
    for i in range(24 * 60):                       # one-minute steps
        h = i / 60
        temp = t_mean + swing * math.cos(2 * math.pi * (h - 14) / 24)
        light = max(0.0, math.sin(math.pi * (h - 6) / 12))
        gross += light * math.exp(-((temp - t_opt) / width) ** 2) / 60
        resp = r20 * q10 ** ((temp - 20) / 10) / 60
        if light > 0:
            resp_day += resp
        else:
            resp_night += resp
    net = gross - resp_day - resp_night
    return gross, resp_day, resp_night, net

print("site                      gross  resp.day  resp.night   net  lost")
sites = (("cool highland, 21 ± 6 °C", 21, 6), ("warm lowland, 27 ± 4 °C", 27, 4))
for name, tm, sw in sites:
    g, rd, rn, n = day_balance(tm, sw)
    print(f"{name:25s} {g:5.2f}  {rd:8.2f}  {rn:10.2f}  {n:5.2f}  {100 * (rd + rn) / g:3.0f}%")
print()
print("same mean (22 °C), wider swing:")
for sw in (2, 6, 10):
    g, rd, rn, n = day_balance(22, sw)
    print(f"  ± {sw:2d} °C: night respiration {rn:.2f}, net {n:.2f}")
# site                      gross  resp.day  resp.night   net  lost
# cool highland, 21 ± 6 °C   7.33      1.97        1.25   4.11   44%
# warm lowland, 27 ± 4 °C    6.03      2.75        2.02   1.26   79%
#
# same mean (22 °C), wider swing:
#   ±  2 °C: night respiration 1.54, net 4.08
#   ±  6 °C: night respiration 1.34, net 3.78
#   ± 10 °C: night respiration 1.19, net 2.23

The cool highland site loses less than half its gross photosynthesis to respiration; the warm lowland site loses almost four-fifths. The second part of the output holds the surprise. At the same mean temperature of 22 °C, a wider day–night swing does lower night-time respiration, but beyond a few degrees the net gain falls, because the hot afternoons push photosynthesis past its optimum and raise daytime respiration. The exponential shape of the Arrhenius curve means that an average temperature does not tell you the average rate: a day that swings between 12 °C and 32 °C respires more, in total, than a day held at 22 °C.

The figure shows the general result. At a mean of 18 °C, a wider swing helps, because it lifts the afternoons towards the optimum while the nights stay cool. At 26 °C, any swing hurts. “A large day–night difference” is good for a crop only when the days it produces are near what the crop prefers.

Two places, two plants

Dendrobium officinale grows naturally as an epiphyte or lithophyte, on tree bark or on damp, shaded rock, in subtropical eastern China. The Yandang Mountains in Yueqing, Zhejiang, are a Cretaceous caldera made largely of rhyolite, a fine-grained volcanic rock (UNESCO World Heritage Centre 2001). A plant rooted in a thin layer of moss on rhyolite has little soil to draw on, gets water mostly from rain and mist, and is exposed to the light and temperature regime of a particular cliff face. Ginger grown in the uplands of Bentong, in Pahang, Malaysia, grows in soil, in the tropics, where the seasons change little and the main variable is the difference between hill and lowland temperatures; Bentong ginger is protected as a geographical indication (Khalid et al. 2026).

The model above can say only the most general thing about these two places: a site with cooler nights loses less carbon to respiration, and so has more to allocate, as long as its days are not too far from the optimum. It cannot say how much of that carbon becomes polysaccharide in a Dendrobium stem, or gingerol in a rhizome. That depends on the plant’s signalling, on its age at harvest and on its genetics, and has to be measured.