Module 11: Deserts & Arid Lands

Deserts occupy roughly one-third of Earth's land surface and yet host astonishing physiological and biochemical ingenuity. This module traces the atmospheric physics that creates arid belts, the biochemistry of desiccation tolerance, the mammalian and avian heat-balance strategies that permit life at the edge of metabolic water deficit, and the coupled desert–tropical teleconnections (Sahara dust fertilising the Amazon) that knit the planet’s dry and wet realms into a single climate system.

1. Global Arid Belts & Their Drivers

Aridity is not a single phenomenon but a family of dynamical regimes. The world’s major deserts—the Sahara (9.2×10⁶ km²), Arabian, Kalahari, Australian Outback, Atacama, Gobi, and Mojave—each arise from a distinct combination of subsidence, rain-shadow, continentality, and cold-current influence.

Subtropical subsidence (Hadley descent)

The Sahara and Arabian deserts sit under the descending branch of the Hadley cell, where air warmed adiabatically at the lapse rate \(\Gamma_d \approx 9.8\,\mathrm{K\,km^{-1}}\) reaches the surface with very low relative humidity. The vertically-integrated moisture convergence

\[ P - E = -\frac{1}{g}\int_0^{p_s} \nabla\!\cdot\!(q\,\mathbf{v})\,dp \]

is strongly negative between 20° and 30° latitude, making precipitation P less than evaporation E.

Rain-shadow deserts

The Atacama (western South America) and Patagonian deserts form on the lee side of the Andes. Ascending moist air condenses on the windward slope, releasing latent heat and depleting the vapour burden. By the time the air descends the leeward slope its saturation vapour pressure (Clausius–Clapeyron) has risen sharply:

\[ \frac{d\ln e_s}{dT} = \frac{L_v}{R_v T^2} \Longrightarrow e_s \approx 6.11\,\exp\!\left(\frac{17.67\,T}{T+243.5}\right)\,\mathrm{hPa} \]

The Atacama is further desiccated by the cold Humboldt current, which suppresses convection, yielding recorded rainfall below 1 mm yr⁻¹ at Arica.

Cold, continental deserts

The Gobi (1.3×10⁶ km²) is a mid-latitude cold desert: distance from oceans combined with the rain-shadow of the Tibetan Plateau and Himalayas produces winter temperatures near −40 °C and summer highs above 40 °C. Continentality index \(C = A_T\,\sin\phi / \phi\) (Conrad 1946) exceeds 0.9 here.

Global distribution of major deserts

2. Biochemistry of Desiccation Tolerance

At cellular level, aridity is a solvent problem: proteins unfold, membranes leak, nucleic acids stack mis-registered when hydration falls below the first hydration shell. Desert organisms share a toolkit of compatible osmolytes, most notably the non-reducing disaccharide trehalose, the amino acid derivative glycine betaine, and the imino acid proline.

Trehalose and the water-replacement hypothesis

Trehalose (\(\alpha,\alpha\text{-}\mathrm{D}\text{-}\mathrm{glucopyranosyl}\text{-}\alpha,\alpha\text{-}\mathrm{D}\text{-}\mathrm{glucopyranoside}\)) hydrogen-bonds to polar protein groups and phospholipid head groups, substituting for water during drying (Crowe & Crowe 1992). Upon rapid drying the sugar forms an amorphous glass whose glass-transition temperature \(T_g\) rises as the water fraction drops:

\[ \frac{1}{T_g} = \frac{w_1}{T_{g,1}} + \frac{w_2}{T_{g,2}} \quad\text{(Gordon-Taylor)} \]

For anhydrobiotic tissue in the larva Polypedilum vanderplanki, \(T_g\)rises above ambient, immobilising macromolecules for decades.

Anhydrobiosis in Polypedilum vanderplanki

This West African chironomid midge accumulates trehalose to ~20% of dry weight during drying, entering reversible anhydrobiosis. Up-regulation of late embryogenesis abundant (LEA) proteins adds a second line of stabilisation by forming disordered protein coats around vulnerable surfaces (Watanabe 2006). Rehydrated larvae resume metabolism within hours, even after 17 years dry.

α-crystallin-like small heat-shock proteins

Desert lens proteins and cytosolic chaperones of the sHsp family (Hsp20, Hsp27) bind partially unfolded clients via their conserved α-crystallin domain, preventing irreversible aggregation when cytosolic water activity drops. Their holdase activity is ATP-independent, making them metabolically cheap during anhydrobiosis.

Osmotic pressure and the Van’t Hoff law

\[ \Pi = i\,cRT,\qquad \mathrm{d}G = V\,\mathrm{d}\Pi \]

Compatible osmolytes raise \(\Pi\) without denaturing enzymes—a feat not shared by inorganic salts, which titrate surface charges. Proline accumulates to 1 M in drought-stressed cacti without inhibiting glycolysis.

3. Mammalian Heat-Stress Physiology

For a homeotherm of mass \(M\) the body’s heat-storage equation reads

\[ Mc_p\,\frac{dT_b}{dt} = \dot{M}_{\text{met}} + R_{\text{abs}} - \epsilon\sigma T_s^4 - h(T_s-T_a) - \lambda \dot{E} \]

Desert mammals exploit every term. The camel (Camelus dromedarius) allows body-temperature oscillation 34–41 °C over 24 h (Schmidt-Nielsen 1957), storing roughly 5 MJ of heat in its ~400 kg tissue and radiating it at night instead of wasting water by sweating:

\[ Q_{\text{stored}} = Mc_p\,\Delta T_b \approx 400 \cdot 3480 \cdot 7 \approx 9.7\,\text{MJ/d} \]

Equivalent to saving ~4 L of sweat per day.

Nasal counter-current condenser

On exhalation, warm moist air from the lungs passes over cool nasal turbinates and re-deposits water. Schmid-Hempel (1984) measured up to 60% humidity recovery in kangaroo rats. The device is a mass-transfer counter-current exchanger with effectiveness

\[ \eta_{cc} = \frac{T_{\text{exh}}-T_{\text{amb}}}{T_{\text{core}}-T_{\text{amb}}} \]

Carotid rete mirabile

Antelopes, oryx, and gazelles possess a carotid rete mirabile—a plexus where arterial blood bound for the brain is cooled by venous blood returning from the evaporating nasal mucosa. Taylor & Lyman (1972) showed Oryx leucoryx maintains brain temperature 2–3 °C below a body core of 41 °C, protecting central nervous tissue while the core stores heat.

Kidney urea recycling and concentration

The kangaroo rat Dipodomys lives without drinking liquid water, deriving metabolic water from seed oxidation. Its inner medullary nephrons concentrate urine to 14× plasma osmolarity (5.5 osm/kg) by extraordinary long loops of Henle (Beuchat 1996). The concentration ratio follows the counter-current multiplier

\[ \frac{C_{\text{urine}}}{C_{\text{plasma}}} \propto \exp(L/\ell_m) \]

where \(L\) is loop length and \(\ell_m\) is the mixing length of the vasa recta.

Fennec fox and ear radiators

The fennec (Vulpes zerda) has the largest ear-to-body ratio of any canid, an effective radiator of sensible heat. For two-sided surface area \(A_{ear}\) at surface temperature \(T_s\), ear sensible loss is \(Q_{ear} = h A_{ear}(T_s-T_a)\), which at \(A_{ear}\approx 0.04\,\mathrm{m^2}\) and \(\Delta T=10\,\text{K}\) gives \(\approx 6\,\text{W}\) — significant for a 1.5 kg animal with a basal metabolic rate of ~3 W.

4. Desert Birds: Flight, Feather, and Furnace

Birds cannot sweat. Their feathers are excellent insulators, and above the lower critical temperature \(T_{lc}\) evaporative water loss (EWL) rises exponentially with air temperature. Albright et al. (2017) projected that many Sonoran passerines face lethal dehydration during 6 h exposure to anticipated +4 °C summer afternoons.

Sandgrouse belly-feather water transport

Male Pteroclidae (sandgrouse) have specialised ventral feathers whose barbules curl on wetting to hold up to 25 mL of water, which they fly tens of kilometres to carry to thirsty chicks (Cade & Maclean 1967). The barbule geometry exploits capillary pressure \(P_c = 2\gamma\cos\theta / r\), with feather pore radius \(r\sim 10\,\mu\mathrm{m}\)yielding retention pressures of several kPa.

Cactus wren, roadrunner, and Houbara bustard

The cactus wren (Campylorhynchus brunneicapillus) nests inside cholla cactus (Cylindropuntia) whose spines provide both mechanical protection and a thermal buffer. The greater roadrunner (Geococcyx californianus) employs gular flutter (rapid pulsation of the mouth floor) rather than whole-body panting, raising respiratory ventilation without the mechanical work of breathing. The Houbara bustard (Chlamydotis undulata) lowers daytime activity to crepuscular windows and uses sand-bathing to abrade ectoparasites and cool by conduction.

Desert lark metabolic scaling

Tieleman & Williams (2000) showed arid-zone larks have basal metabolic rates 15–40% below mesic relatives of the same mass, an allometric shift \(\mathrm{BMR} = a M^{0.67}\) with depressed prefactor \(a\). Lower BMR reduces endogenous heat production and cumulative pulmonary water loss.

Evaporative water loss equation

\[ \mathrm{EWL}(T_a) = \mathrm{EWL}_0\,\exp\!\bigl[b\,(T_a - T_{lc})\bigr] \]

For small passerines, \(b \approx 0.15\text{-}0.20\,\mathrm{K^{-1}}\); dehydration kills at ~15% body-mass loss (Wolf 2000).

5. Desert Reptiles: Ectotherm Engineering

Reptiles are ectotherms: their body temperature \(T_b\) closely follows operative temperature \(T_e\), the equilibrium temperature of a non-metabolising object in the same thermal environment. The energy budget reduces to

\[ T_e = T_a + \frac{R_{\text{abs}} - \epsilon\sigma T_a^4}{\rho c_p g_{Ha}} \]

Fringe-toed lizards and substrate dynamics

Uma scoparia and Uma inornata possess elongated, keel-scaled toe fringes that act as sand-skis. They minimise ventral contact with scorching sand (daytime surface temperatures of 70 °C) by stilting and rapid locomotion, and dive into loose aeolian sand for thermal refugia.

Galápagos land iguana thermoregulation

Conolophus subcristatus on Fernandina Island basks on black volcanic rock at dawn to attain activity temperatures of ~37 °C, then seeks shade when \(T_e>42\,^\circ\mathrm{C}\). Its thermal inertia \(\tau_{therm}\propto M^{1/3}\) means a 6 kg individual responds to environmental change on the scale of tens of minutes rather than seconds.

Desert tortoise water economy

Gopherus agassizii stores water in the urinary bladder (up to 40% of body mass) as a reserve, tolerating plasma osmolarities above 400 mOsm/kg. It excavates burrows 1–2 m deep, where soil temperatures are buffered to near annual mean and humidity is high.

6. CAM Photosynthesis & Desert Botany

To photosynthesise by day without bleeding water, cacti and succulents evolved Crassulacean acid metabolism (CAM). At night, stomata open and CO₂ is fixed by PEP carboxylase into oxaloacetate, stored as malate in vacuoles. By day, stomata close; malate is decarboxylated, releasing CO₂to RuBisCO for normal Calvin-cycle carbon fixation.

\[ \text{Night: } \mathrm{PEP} + \mathrm{HCO_3^-} \xrightarrow{\text{PEPC}} \mathrm{OAA} \to \mathrm{malate} \]

\[ \text{Day: } \mathrm{malate} \to \mathrm{CO_2} \xrightarrow{\text{RuBisCO}} \text{Calvin cycle} \]

Water-use efficiency (mol CO₂ fixed per mol H₂O lost) of CAM plants reaches 0.02—roughly 5–10× that of C₃ plants. The saguaro (Carnegiea gigantea) stores up to 80% water by mass, pleating accordions that expand after monsoon rains. Welwitschia mirabilis of the Namib harvests fog via leaf stomata that open when relative humidity exceeds ~75%, reaching ages over 1500 yr.

7. Climate Change & Desert Futures

The greening–browning paradox

Satellite NDVI shows that since 1982 large parts of the Sahel have greened under elevated CO₂ and recovered monsoon rains (Liu 2020, Remote Sensing of Environment). Conversely, the American southwest has browned under deeper drought. Rising CO₂ improves water-use efficiency (\(\partial A/\partial c_a > 0\) with stomatal closure), offsetting some drying—but only up to a ceiling beyond which heat stress dominates.

Wet-bulb temperature and the survivability limit

Raymond et al. (2020, Science Advances) showed that coastal Persian Gulf and Pakistan stations have already exceeded a wet-bulb temperature of 35 °C—the theoretical survivability limit for humans and many mammals at rest in the shade. Wet-bulb obeys implicitly

\[ e_s(T_w) - e_a = \gamma (T - T_w),\quad \gamma = \frac{c_p p}{0.622\,L_v} \]

The Charney biogeophysical feedback

Charney (1975) argued that overgrazing-induced increases in albedo cool the boundary layer, inducing subsidence and reducing rainfall—a positive feedback that can push a system across a dry tipping point. Modern coupled GCMs confirm multiple stable equilibria of the Sahara for insolation windows near the mid-Holocene (Claussen 1999). The feedback can be summarised as

\[ \frac{dP}{dA} < 0,\quad \frac{dA}{dV} < 0 \;\Rightarrow\; \frac{dP}{dV} > 0 \]

where A = albedo, V = vegetation cover, P = precipitation.

Desertification feedbacks on land–atmosphere coupling

Loss of vegetation reduces evapotranspiration, shifting the surface energy partition from latent to sensible heat (Bowen ratio \(\beta = H/\lambda E\) rising from <1 to >5). Higher sensible flux deepens the planetary boundary layer and suppresses cloud formation, further reducing precipitation.

8. The Sahara–Amazon Dust Bridge

NASA’s CALIPSO lidar (Yu et al. 2015) quantified the trans-Atlantic dust plume from the Saharan air layer: approximately 182 Tg yr^-1 of mineral dust leaves North Africa, of which ~27 Tg are deposited on the Amazon basin. Crucial is its phosphorus content—~0.08% by mass—delivering 22,000 t P yr^-1that compensates the ~23,000 t P yr^-1 leached by Amazonian rainfall. The Bodélé Depression (a dry palaeolake in Chad) alone supplies the bulk of this flux, making a dust-lofting cyclone in the Sahel a direct fertiliser of the world’s largest rainforest.

\[ F_P^{\text{dust}} \approx \rho_{\text{dust}}\, f_P\, u_*^3 / g \]

where \(u_*\) is friction velocity and \(f_P\) the phosphorus mass fraction.

The dust also shadows Atlantic sea-surface temperature (lowering SST by ~0.2 K in the main development region) and thereby weakens hurricanes (Evan 2011). Sahel greening, by suppressing dust emission, may therefore reduce Amazon P input even as it improves West African agriculture—a textbook trade-off between local and remote ecosystems.

Simulation: Oryx Heat Balance & Rete Brain Cooling

Steady-state energy budget for an Arabian oryx under varying solar load and wind speed, including evaporative cooling and the carotid rete's selective brain-cooling response.

Python

script.py130 lines

import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

# Desert Heat Balance: Arabian oryx (Oryx leucoryx)
# Energy budget at the animal surface:
#   M + R_abs = R_emit + H_conv + lambda * E_evap + H_cond
# Where:
#   M        = metabolic heat production (W)
#   R_abs    = absorbed shortwave + longwave radiation (W/m^2)
#   R_emit   = emitted longwave radiation ~ eps*sigma*T_s^4
#   H_conv   = convective loss = h*(T_s - T_air)
#   E_evap   = evaporative water loss (kg/s), lambda = 2.45e6 J/kg
# The oryx uses a carotid rete mirabile to permit selective brain
# cooling while body core oscillates 34-41 C (Schmidt-Nielsen 1957).

sigma = 5.670374419e-8   # Stefan-Boltzmann
eps   = 0.96             # skin emissivity
alpha_sw = 0.30          # shortwave absorptivity (pale oryx pelage)
rho_air  = 1.12          # kg/m^3 at ~35 C
c_p      = 1005          # J/kg/K
lam      = 2.45e6        # latent heat of vaporisation, J/kg
A_body   = 1.8           # m^2 surface area (70 kg oryx)
M_rest   = 70.0          # W basal metabolic heat
T_air = np.linspace(20, 50, 200) + 273.15  # K
S_sun = np.array([200, 600, 1000])         # W/m^2 solar
u_wind = np.linspace(0.2, 6.0, 200)        # m/s

# Convective coefficient (Mitchell 1976): h ~ 10.5 * u^0.6 for ungulates
def h_conv(u):
    return 10.45 * (u ** 0.6)

# Solve for steady surface temperature at each T_air, for fixed wind
def solve_Ts(Ta, S, u, E_evap=0.0):
    # Newton iteration on energy balance
    Ts = Ta + 2.0
    for _ in range(60):
        R_abs  = alpha_sw * S + eps * sigma * Ta**4
        R_emit = eps * sigma * Ts**4
        H_cv   = h_conv(u) * (Ts - Ta)
        Q_ev   = lam * E_evap / A_body
        F = M_rest/A_body + R_abs - R_emit - H_cv - Q_ev
        dF = -4*eps*sigma*Ts**3 - h_conv(u)
        Ts -= F/dF
    return Ts

fig, ((ax1, ax2), (ax3, ax4)) = plt.subplots(2, 2, figsize=(14, 11))
fig.patch.set_facecolor('#0a0a1a')
for ax in (ax1, ax2, ax3, ax4):
    ax.set_facecolor('#0a0a1a')
    ax.tick_params(colors='#94a3b8')
    for sp in ax.spines.values():
        sp.set_color('#334155')
    ax.grid(True, color='#334155', alpha=0.4)

colors = ['#22d3ee', '#facc15', '#f43f5e']
for S, c in zip(S_sun, colors):
    Ts = np.array([solve_Ts(Ta, S, u=2.0) - 273.15 for Ta in T_air])
    ax1.plot(T_air - 273.15, Ts, '-', color=c, linewidth=2.5,
             label=f'S = {S} W/m^2')
ax1.plot(T_air-273.15, T_air-273.15, ':', color='#94a3b8', label='T_s = T_air')
ax1.axhline(41, color='#f97316', linestyle='--', alpha=0.7,
            label='Lethal T_body ~ 41 C')
ax1.set_xlabel('Air temperature (C)', color='#94a3b8')
ax1.set_ylabel('Skin / core temperature (C)', color='#94a3b8')
ax1.set_title('Oryx body temperature vs ambient (wind = 2 m/s)',
              color='#e2e8f0', fontweight='bold')
ax1.legend(facecolor='#1e293b', edgecolor='#334155',
           labelcolor='#cbd5e1', fontsize=9)

# Wind sweep at T_air = 45 C
Ts_wind = np.array([solve_Ts(45+273.15, 900, u) - 273.15 for u in u_wind])
ax2.plot(u_wind, Ts_wind, '-', color='#f472b6', linewidth=2.5,
        label='Dry skin (no sweating)')
Ts_wind_ev = np.array([solve_Ts(45+273.15, 900, u, E_evap=5e-5)-273.15
                      for u in u_wind])
ax2.plot(u_wind, Ts_wind_ev, '-', color='#22d3ee', linewidth=2.5,
        label='Panting 5e-5 kg/s')
ax2.axhline(41, color='#f43f5e', linestyle='--', alpha=0.7,
            label='Lethal threshold')
ax2.set_xlabel('Wind speed (m/s)', color='#94a3b8')
ax2.set_ylabel('Skin temperature (C)', color='#94a3b8')
ax2.set_title('Convective cooling benefit of wind (T_air = 45 C)',
              color='#e2e8f0', fontweight='bold')
ax2.legend(facecolor='#1e293b', edgecolor='#334155',
           labelcolor='#cbd5e1', fontsize=9)

# Energy budget pie at desert noon, S=1000, T_air=43 C
Ta = 43 + 273.15; u = 2.0; S = 1000
Ts = solve_Ts(Ta, S, u)
R_abs  = (alpha_sw * S + eps * sigma * Ta**4) * A_body
R_emit = eps * sigma * Ts**4 * A_body
H_cv   = h_conv(u)*(Ts - Ta) * A_body
# Assume remaining closes via evaporation
E_need = (M_rest + R_abs - R_emit - H_cv) / lam
labels_pie = ['Longwave out', 'Convection', 'Evaporation', 'Metabolic waste']
vals = [R_emit, max(H_cv, 0), max(E_need*lam, 0), M_rest]
ax3.pie(vals, labels=labels_pie,
        colors=['#22d3ee', '#facc15', '#f472b6', '#4ade80'],
        textprops={'color': '#e2e8f0'},
        autopct='%1.0f%%')
ax3.set_title(f'Heat budget at T_air=43C, S=1000 W/m^2 (E={E_need*3600*1000:.0f} g/h)',
              color='#e2e8f0', fontweight='bold')

# Carotid rete brain cooling
T_body = np.linspace(36, 42, 200)
T_brain_no_rete = T_body
# Rete allows cooling of arterial blood by 2-3 C (Taylor 1972)
T_brain_rete = T_body - 2.5*np.tanh(0.7*(T_body - 38))
ax4.plot(T_body, T_brain_no_rete, '--', color='#f43f5e', linewidth=2,
         label='No carotid rete')
ax4.plot(T_body, T_brain_rete, '-', color='#22d3ee', linewidth=2.5,
         label='Selective brain cooling')
ax4.fill_between(T_body, T_brain_rete, T_brain_no_rete,
                 color='#22d3ee', alpha=0.15,
                 label='Brain thermal window')
ax4.axhline(41, color='#f43f5e', linestyle=':', alpha=0.7)
ax4.set_xlabel('Core body temperature (C)', color='#94a3b8')
ax4.set_ylabel('Hypothalamic temperature (C)', color='#94a3b8')
ax4.set_title('Carotid rete mirabile: selective brain cooling',
              color='#e2e8f0', fontweight='bold')
ax4.legend(facecolor='#1e293b', edgecolor='#334155',
           labelcolor='#cbd5e1', fontsize=9)

plt.tight_layout()
plt.savefig('output.png', dpi=120, bbox_inches='tight', facecolor='#0a0a1a')
print('Oryx heat balance simulation complete.')
print(f'At T_air=43 C, required evaporative loss: {E_need*3600*1000:.1f} g/h')
print(f'Wind from 0.5 -> 5 m/s reduces T_s by {Ts_wind[0]-Ts_wind[-1]:.1f} C')

Click Run to execute the Python code

Code will be executed with Python 3 on the server

Simulation: Desert Bird Evaporative Water Loss & Survival

Exponential EWL curves above thermoneutrality, allometric scaling of mass-specific water loss, and Albright 2017-style lethal-temperature shifts for five desert species under future warming.

Python

script.py121 lines

import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

# Desert bird evaporative water loss and survival time
# Dawson (1982), McKechnie & Wolf (2010), Albright et al. (2017 PNAS)
# Above the lower critical temperature T_lc, EWL rises non-linearly:
#     EWL(T_a) = EWL_0 * exp(b * (T_a - T_lc))    (g H2O / h)
# Allometric scaling of body water store (Calder 1984):
#     W_total = 0.65 * M_body   (kg water / kg body)
# Lethal dehydration at ~15 % body mass loss (Wolf 2000)

species = {
    'Houbara bustard (1.8 kg)':  dict(M=1.800, EWL0=1.8,  T_lc=36, b=0.13, c='#22d3ee'),
    'Sandgrouse (0.25 kg)':      dict(M=0.250, EWL0=0.40, T_lc=34, b=0.14, c='#facc15'),
    'Cactus wren (0.04 kg)':     dict(M=0.040, EWL0=0.06, T_lc=35, b=0.18, c='#f472b6'),
    'Roadrunner (0.30 kg)':      dict(M=0.300, EWL0=0.55, T_lc=36, b=0.15, c='#4ade80'),
    'Desert lark (0.022 kg)':    dict(M=0.022, EWL0=0.035,T_lc=34, b=0.20, c='#f97316'),
}

T_air = np.linspace(25, 55, 300)   # C

# Panel 1: EWL curves
for name, p in species.items():
    ewl = p['EWL0'] * np.exp(p['b'] * np.maximum(T_air - p['T_lc'], 0))
    ax1.plot(T_air, ewl, '-', color=p['c'], linewidth=2.2, label=name)
ax1.axvline(50, color='#f43f5e', linestyle='--', alpha=0.5, label='Extreme heat (50 C)')
ax1.set_xlabel('Air temperature (C)', color='#94a3b8')
ax1.set_ylabel('Evaporative water loss (g H2O / h)', color='#94a3b8')
ax1.set_yscale('log')
ax1.set_title('Desert bird EWL above thermoneutral zone',
              color='#e2e8f0', fontweight='bold')
ax1.legend(facecolor='#1e293b', edgecolor='#334155',
           labelcolor='#cbd5e1', fontsize=8)

# Panel 2: survival time (hours until 15 % water loss) no re-drinking
for name, p in species.items():
    W_tot = 0.65 * p['M'] * 1000   # g water
    W_lethal = 0.15 * p['M'] * 1000
    ewl = p['EWL0'] * np.exp(p['b'] * np.maximum(T_air - p['T_lc'], 0))
    t_surv = W_lethal / ewl
    ax2.plot(T_air, t_surv, '-', color=p['c'], linewidth=2.2, label=name)
ax2.axhline(6, color='#f43f5e', linestyle='--', alpha=0.7,
            label='Active period ~ 6 h')
ax2.set_xlabel('Air temperature (C)', color='#94a3b8')
ax2.set_ylabel('Dehydration survival time (h)', color='#94a3b8')
ax2.set_yscale('log')
ax2.set_title('Time to 15 % body-water loss (no drinking)',
              color='#e2e8f0', fontweight='bold')
ax2.legend(facecolor='#1e293b', edgecolor='#334155',
           labelcolor='#cbd5e1', fontsize=8)

# Panel 3: allometric scaling of mass-specific EWL at 45 C
M_range = np.logspace(-2, 1, 200)   # 10 g to 10 kg
# Mass-specific EWL at hyperthermia ~ M^-0.3 (Tieleman & Williams 2000)
ewl_spec = 5.0 * M_range**(-0.30)
ax3.plot(M_range*1000, ewl_spec, '-', color='#22d3ee', linewidth=2.5,
         label='Mass-specific EWL ~ M^-0.30')
# Observed species points
for name, p in species.items():
    ewl_spec_point = (p['EWL0']*np.exp(p['b']*(45-p['T_lc']))) / (p['M']*1000)
    ax3.plot(p['M']*1000, ewl_spec_point, 'o', color=p['c'],
             markersize=10, markeredgecolor='white', label=name)
ax3.set_xlabel('Body mass (g)', color='#94a3b8')
ax3.set_ylabel('Mass-specific EWL at 45 C (g H2O / g body / h)', color='#94a3b8')
ax3.set_xscale('log')
ax3.set_yscale('log')
ax3.set_title('Allometry of hyperthermic water loss',
              color='#e2e8f0', fontweight='bold')
ax3.legend(facecolor='#1e293b', edgecolor='#334155',
           labelcolor='#cbd5e1', fontsize=8)

# Panel 4: heatwave warming shifts lethal T_air (Albright 2017)
dT_future = np.arange(0, 5.1, 0.5)
lethal_T = []
for dT in dT_future:
    # Lethal T_air: where 6 h exposure exceeds dehydration threshold
    row = []
    for name, p in species.items():
        W_lethal = 0.15 * p['M'] * 1000
        def f(Ta):
            return p['EWL0']*np.exp(p['b']*np.maximum(Ta - p['T_lc'], 0))*6 - W_lethal
        lo, hi = 25.0, 60.0
        for _ in range(80):
            m = 0.5*(lo+hi)
            if f(m) > 0:
                hi = m
            else:
                lo = m
        row.append(m - dT)   # future lethal T given climate dT shift
    lethal_T.append(row)
lethal_T = np.array(lethal_T)
for i, (name, p) in enumerate(species.items()):
    ax4.plot(dT_future, lethal_T[:, i], '-', color=p['c'],
             linewidth=2.2, marker='o', label=name)
ax4.axhline(45, color='#f43f5e', linestyle='--', alpha=0.5,
            label='Already common summer high')
ax4.set_xlabel('Regional warming (C)', color='#94a3b8')
ax4.set_ylabel('Effective lethal T_air (C)', color='#94a3b8')
ax4.set_title('Climate-shifted lethal temperatures (Albright 2017 style)',
              color='#e2e8f0', fontweight='bold')
ax4.legend(facecolor='#1e293b', edgecolor='#334155',
           labelcolor='#cbd5e1', fontsize=8)

plt.tight_layout()
plt.savefig('output.png', dpi=120, bbox_inches='tight', facecolor='#0a0a1a')
print('Desert bird EWL simulation complete.')
for name, p in species.items():
    W_lethal = 0.15*p['M']*1000
    ewl45 = p['EWL0']*np.exp(p['b']*(45-p['T_lc']))
    print(f'  {name}: survival at 45 C = {W_lethal/ewl45:.1f} h')

Click Run to execute the Python code

Code will be executed with Python 3 on the server

Key References

• Schmidt-Nielsen, K. (1957). “Body temperature of the camel and its relation to water economy.” American Journal of Physiology, 188, 103–112.

• Cade, T. J. & Maclean, G. L. (1967). “Transport of water by adult sandgrouse to their young.” Condor, 69, 323–343.

• Taylor, C. R. & Lyman, C. P. (1972). “Heat storage in running antelopes: independence of brain and body temperatures.” American Journal of Physiology, 222, 114–117.

• Charney, J. G. (1975). “Dynamics of deserts and drought in the Sahel.” Quarterly Journal of the Royal Meteorological Society, 101, 193–202.

• Mitchell, J. W. (1976). “Heat transfer from spheres and other animal forms.” Biophysical Journal, 16, 561–569.

• Crowe, J. H. & Crowe, L. M. (1992). “Anhydrobiosis: a strategy for survival.” Advances in Space Research, 12, 239–247.

• Beuchat, C. A. (1996). “Structure and concentrating ability of the mammalian kidney: correlations with habitat.” American Journal of Physiology, 271, R157–R179.

• Claussen, M. et al. (1999). “Simulation of an abrupt change in Saharan vegetation in the mid-Holocene.” Geophysical Research Letters, 26, 2037–2040.

• Tieleman, B. I. & Williams, J. B. (2000). “The adjustment of avian metabolic rates and water fluxes to desert environments.” Physiological and Biochemical Zoology, 73, 461–479.

• Wolf, B. (2000). “Global warming and avian occupancy of hot deserts; a physiological and behavioral perspective.” Revista Chilena de Historia Natural, 73, 395–400.

• Watanabe, M. (2006). “Anhydrobiosis in invertebrates.” Applied Entomology and Zoology, 41, 15–31.

• Evan, A. T. et al. (2011). “Influence of African dust on Atlantic tropical cyclone activity.” Journal of Climate, 24, 2688–2706.

• Yu, H. et al. (2015). “The fertilizing role of African dust in the Amazon rainforest: A first multiyear assessment based on data from Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations.” Geophysical Research Letters, 42, 1984–1991.

• Albright, T. P. et al. (2017). “Mapping evaporative water loss in desert passerines reveals an expanding threat of lethal dehydration.” PNAS, 114, 2283–2288.

• Raymond, C., Matthews, T. & Horton, R. M. (2020). “The emergence of heat and humidity too severe for human tolerance.” Science Advances, 6, eaaw1838.

• Liu, Y. Y. et al. (2020). “Recent Sahel greening reversed by climate change.” Remote Sensing of Environment, 251, 112108.

← Module 10: Temperate & Boreal Biomes Module 12: Amazon & Tropical Forests →