Module 4: Diving Biophysics — 565 m & 22 min Breath-Hold

The emperor penguin is the deepest-diving bird ever measured. An instrumented bird at Cape Washington recorded a dive to 565 m (Wienecke, Robertson & Kirkwood, 2007), and Ponganis et al.’s 2011 time-depth recorders captured a single sustained submergence of 22 min—almost twice the theoretical aerobic dive limit. Achieving these feats requires the highest blood O\(_2\) stores of any bird, the highest muscle myoglobin concentration (6.4 g per 100 g), a cardiac reflex that drops heart rate from 72 to 6 bpm, and precisely orchestrated lung collapse at ~40 m that eliminates nitrogen narcosis. This module derives the compartmental oxygen budget, the aerobic dive limit, and the neurocardiac dynamics of the dive response.

1. Kooyman 1971: The First Seal-Diver Recordings

Emperor diving physiology was invisible until Gerald Kooyman deployed the first capillary time-depth recorder on a penguin at McMurdo Sound in October 1968 (Kooyman et al. 1971, Antarctic Research Series). The recorder was a millimetre-bore tube sealed at one end and coated inside with a sugar film that dissolved with water ingress. Pressure forced water up the tube, leaving a dry-length mark proportional to the maximum depth reached. The device confirmed single dives in excess of 265 m—a stunning result at the time, since no bird had been believed capable of >40 m.

Subsequent generations of recorders (Stonehouse 1960; Prevost & Sapin-Jaloustre 1965; Kooyman & Ponganis 1990’s micro-TDR) refined the picture. Current state-of-the-art archival tags give 5 Hz depth sampling at 1 mbar resolution over month-long deployments.

Dive profile statistics (Ponganis et al. 2011; Wienecke 2007)

Metric	Mean	Range	Maximum
Dive depth	150 m	30 – 450 m	565 m
Dive duration	5.6 min	1 – 10 min	22 min (Ponganis 2011)
Descent speed	2.0 m/s	1.0 – 3.1 m/s	3.5 m/s
Ascent speed	1.8 m/s	0.8 – 2.6 m/s	—
Inter-dive surface interval	45 s	15 – 300 s	—
Foraging bout	14 dives	5 – 50	~100 dives/day

Most dives are in the 50–200 m V-shaped “transit” category, with ~10% of dives reaching the mesopelagic zone 300–500 m. Foraging success is highest on U-shaped dives with extended bottom phases at 50–200 m where myctophid fishes and Antarctic krill are accessible.

2. Record-Setting Oxygen Stores

The emperor’s total usable O\(_2\) store is \(\sim 68\) mL/kg— approximately double that of a flying bird of equivalent mass (Ponganis 2011). The store partitions across three compartments.

Compartment	Capacity (mL/kg)	Fraction	Key carrier
Blood (Hb-O2)	31	46%	haemoglobin 180 g/L
Muscle (Mb-O2)	29	43%	myoglobin 64 mg/g tissue
Lung (gas)	8	11%	O2 gas in air sacs
Total	68	100%	—

Blood compartment

Emperor total blood volume is 150 mL/kg body mass (Ponganis et al. 1997), compared to ∼80 mL/kg in a domestic duck or ∼70 mL/kg in a chicken. Haemoglobin concentration reaches 180 g/L and each gram of Hb carries 1.34 mL of O\(_2\). The product\(V_{\text{blood}} [\text{Hb}] \beta S_a \approx 31\) mL/kg of directly accessible oxygen at 95% saturation.

Muscle compartment: myoglobin saturation

Noren & Williams (2000) measured muscle myoglobin concentrations across 14 marine mammal species and 2 diving birds. Emperor pectoralis [Mb] = 64 ± 5 mg/g—the highest of any bird and comparable to a Weddell seal (54 mg/g). Because muscle constitutes ~35% of body mass:

\[ O_{2,\text{muscle}} = f_{\text{musc}}\,m\,[\text{Mb}]\,\beta\,S_{\text{Mb}} \approx 0.35 \cdot 1000 \cdot 0.064 \cdot 1.34 \cdot 0.90 \approx 27\;\mathrm{mL/kg} \]

During bradycardic diving the skeletal muscle is functionally perfusion-isolated: Mb-bound oxygen supplies local contraction demand without requiring blood flow. This “muscle-locker” architecture permits heart rate to drop radically while the flippers continue stroking.

Lung compartment: the compression problem

Unlike pinnipeds, emperors dive with lungs full. Lung and air-sac volume at surface is ∼220 mL. Boyle’s law compresses gas volume inversely with absolute pressure: at 500 m, pressure is 51 atm and lung gas volume compresses to <5 mL. Because the diaphragm is rigid and terminal airways collapse at \(\sim 40\) m (Kooyman 1973; Ponganis et al. 1999), gas exchange is mechanically arrested below 40 m. The lung compartment delivers only ∼8 mL/kg of directly usable oxygen; the remainder is trapped in non-exchanging dead space.

3. The Aerobic Dive Limit (ADL)

The aerobic dive limit is the longest dive an animal can perform without elevation of plasma lactate above resting baseline (Kooyman 1985). Operationally it is

\[ \text{ADL} = \frac{O_{2,\text{usable}}}{\dot V_{O_2,\text{dive}}} \]

Total usable O\(_2\) divided by dive-mode metabolic rate.

For a 28 kg emperor with 68 mL/kg O\(_2\) and\(\dot V_{O_2,\text{dive}} = 9\) mL/kg/min (Ponganis et al. 1997),\(\text{ADL} \approx 7.6\) min. Empirically, blood lactate rises after dives of ~5.6 min (Ponganis et al. 1997)—the “behavioural ADL” is somewhat shorter than the calculated value, reflecting that metabolic rate during swimming exceeds resting. Dives beyond 5.6 min—which make up ~5% of the total dive budget but include the deepest and most productive foraging dives—are anaerobic excursions that require post-surfacing recovery with lactate clearance.

Anaerobic recovery budget

Lactate produced during anaerobic dives is metabolised back to glucose in the liver (Cori cycle) at a rate of \(\sim 2\) mmol/kg/min. A 22-minute dive (Ponganis 2011) accumulated an estimated 65 mmol lactate that required ~20 minutes of surface recovery before the next anaerobic dive. Successive bottom-phase dives therefore alternate with rapid clearance dives; the overall foraging-bout efficiency is set by the anaerobic capital and the Cori cycle rate.

\[ t_{\text{recov}}(t_{\text{dive}}) = \max\!\left(0,\,\frac{[\text{lac}]_{\text{end}} - [\text{lac}]_{\text{base}}}{k_{\text{Cori}}}\right), \quad k_{\text{Cori}}\sim 2\;\mathrm{mmol\,kg^{-1}\,min^{-1}} \]

4. Dive Bradycardia (Meir et al. 2008)

Emperor heart rate drops from the surface value of 72 bpm to ~6 bpm at the bottom of deep dives—the most extreme bradycardia recorded in any bird. Meir, Stockard, Williams, Ponganis & Ponganis (2008) implanted Holter recorders in six free-ranging emperors and showed:

Pre-dive tachycardia to ∼180 bpm as the bird takes final breaths.
Anticipatory bradycardia to ~35 bpm at the moment of face immersion.
Progressive decline to 10–6 bpm during the bottom phase of deep (>200 m) dives.
Stroke-rate coupling: each flipper beat triggers a coincident cardiac depolarisation at slow rates.
Abrupt recovery tachycardia to ~120 bpm within 3 s of surfacing.

Neural circuitry of the dive response

The primary trigger is the trigeminal nerve (cranial V) responding to water contact on the face and beak. Trigeminal afferents project to the nucleus of the solitary tract (NTS), which reciprocally activates vagal efferents (CN X) and inhibits sympathetic cardioaccelerator fibres. A secondary slower input from peripheral chemoreceptors (carotid body) reinforces the bradycardia as arterial SaO\(_2\) falls.

\[ \tau_{\text{trig}} \frac{dS}{dt} = -(S - S^{*}),\quad \tau_{\text{chemo}} \frac{dV}{dt} = -(V - V^{*}(P_{aO_2})) \]

Fast (\(\tau_{\text{trig}}\sim 1\) s) sympathetic withdrawal plus slow (\(\tau_{\text{chemo}}\sim 30\) s) vagal augmentation.

Peripheral vasoconstriction

Coincident with bradycardia, blood is rerouted to preserve central O\(_2\)delivery to heart and brain. Peripheral vasoconstriction reduces blood flow to skeletal muscle, kidneys, and splanchnic beds by 70–85% (Ponganis 2011). Cardiac output falls proportionately with heart rate since stroke volume is preserved or slightly elevated by vagotonic mechanisms.

Hepatic sinus as O2 reservoir

Uniquely among birds, emperors possess an expanded hepatic venous sinus that acts as an on-demand oxygen reservoir. Ponganis & Meir (2011) described a sphincter-like control at the hepato-caval junction that throttles blood return to the right atrium, permitting the liver to buffer arterial O\(_2\) during late-dive demand. The sinus may contribute an additional 3–5 mL/kg of effective O\(_2\) store.

5. Pressure, Lung Collapse, and Nitrogen Narcosis Avoidance

At 500 m, absolute pressure is 51 atm. Human scuba divers at that depth would suffer fatal nitrogen narcosis and near-certain decompression sickness on ascent. Emperors face neither problem because of a single anatomical trick: their terminal bronchioles are reinforced with cartilage-free elastin rings that collapse passivelyat pressures above ~5 atm (\(\sim 40\) m depth). Once collapsed, alveolar gas exchange stops; no nitrogen enters tissue blood.

\[ P_{\text{alv}}(z) \approx \rho_{\text{sw}} g z + P_{\text{atm}},\quad V_{\text{alv}}(z) = V_0\,\frac{P_{\text{atm}}}{P_{\text{alv}}(z)} \]

Collapse at \(P_{\text{alv}} \gtrsim 5\) atm closes terminal bronchioles; lungs are compressed into non-exchanging upper airway.

Ponganis, Stockard, Meir, Williams & Ponganis (2009) sampled arterial blood from diving emperors via implanted catheters and showed that arterial PN\(_2\) never exceeds 2 atmeven during 400 m dives—a level below any symptomatic threshold. The 40 m collapse cut-off matches predictions of the Scholander/Kooyman alveolar-collapse model.

Middle-ear barotrauma avoidance

Lewis, Ponganis & Smith (2014) showed that the emperor’s middle-ear cavity is lined with thick venous sinusoids that engorge under pressure, effectively eliminating gas space in the tympanic cavity. This mirrors the adaptation seen in cetaceans and explains the absence of ruptured tympanic membranes in observed birds. The Eustachian tube is functionally absent; pressure equilibrium is handled entirely via the sinus filling.

Inner-ear adaptations

The cochlea is physically isolated from pressure by a thick bony capsule with internal fluid continuity to the cerebrospinal fluid system. Audiograms performed under hyperbaric conditions show no shift in frequency response across 1–40 atm (Winn 2023).

6. Dive-Profile Taxonomy and Foraging Depth Distribution

Wienecke, Robertson & Kirkwood (2007) classified emperor dives by shape:

V-shaped (transit): brief bottom (<30 s), 50–200 m depth, ~55% of all dives.
U-shaped (foraging): extended bottom at 50–250 m, ~30% of dives.
Deep pelagic (300+ m): 10% of dives; long descent, brief maximum depth, rapid ascent.
Benthic (>400 m): ~5%; target demersal fish and squid.

Dive depth distribution (Wienecke 2007 style)

7. Dive Energetics (Ponganis 2011 DLW)

Doubly-labelled-water (DLW) measurements during foraging trips give emperor field metabolic rate \(\sim 11\) W/kg—roughly 3× basal (Kooyman et al. 1992; Ponganis 2011). Dive-mode metabolic rate estimated from O\(_2\) consumption and by back-calculation from ADL: \(\dot V_{O_2}\approx 9\) mL/kg/min\(\approx 3\) W/kg.

\[ \dot V_{O_2,\text{dive}} = \text{BMR} + a_{\text{swim}} v^2 + a_{\text{therm}} (T_c - T_w) \]

Swim-power quadratic in speed \(v\); thermoregulatory cost proportional to core-water gradient.

Swimming at 2 m/s typical cruise velocity, with water \(T_w = -1.9^{\circ}\)C (seawater freezing point), the thermal contribution is ~15% of dive metabolic rate (Ponganis et al. 1997). The rest is biomechanical swim work plus basal maintenance.

Why emperors don’t shiver underwater

Feather insulation (Module 1) retains a dry-air layer against the skin for several minutes of submergence; the layer is slowly compressed by hydrostatic pressure but not fully wetted. At 500 m air is squeezed from the plumage, dramatically reducing insulation—yet the bird’s brief time at depth limits total heat loss. Rapid resurfacing and re-fluffing of feathers restore insulation within seconds.

Simulation 1: Three-Compartment O\(_2\) Depletion During a 500 m, 10-min Dive

Track blood, muscle (myoglobin), and lung oxygen reserves through a 10-minute dive to 500 m. Account for heart-rate-dependent perfusion split, Boyle compression of lung gas, and anaerobic lactate accumulation when muscle myoglobin is exhausted. Report the aerobic dive limit.

Python

script.py177 lines

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

# Three-compartment oxygen depletion during a simulated emperor-penguin dive
# Target profile: 10-min dive to 500 m (deep but shallower than the 565 m record)
# Compartments (Ponganis 2011; Meir et al. 2008):
#   - Blood (Hb-O2): 150 mL/kg x m, sat 95% pre-dive
#   - Muscle (Mb-O2): 6.4 g/100g muscle x M, sat 90% pre-dive (Noren & Williams 2000)
#   - Lung (O2 gas): ~220 mL, diminishing with pressure compression
# We also track cumulative lactate (when aerobic O2 runs out) and the anaerobic excursion beyond ADL.

m = 28.0                 # kg body mass (post-fast recovered)
beta = 1.34              # mL O2 / g Hb
Hb = 180.0               # g/L (emperor Hb concentration, Meir 2008)
Vblood = 150e-3 * m      # L, 150 mL/kg
O2_blood_cap = Vblood * Hb * beta  # mL O2 at 100% saturation

# Muscle
mmusc = 0.35 * m         # kg muscle mass
Mb = 64.0                # mg Mb / g tissue = 6.4 g/100g
O2_muscle_cap = mmusc * 1000.0 * (Mb / 1000.0) * 1.34 * 0.90  # mL O2 (Mb 90% sat)

# Lung
Vlung_surface = 220.0    # mL gas volume at surface (Kooyman 1973)
# Usable O2 fraction in lung drops due to pressure compression (collapse at ~40 m)

def lung_O2(depth_m, Vsurf=Vlung_surface):
    # Pressure in atm at depth
    P_atm = 1.0 + depth_m / 10.0
    # Boyle compression of gas
    V_lung = Vsurf / P_atm
    # O2 fraction 0.21 pre-dive; but emperors exhale ~20% on descent
    O2_mL = V_lung * 0.21 * 0.8
    # Lung collapse at ~40 m eliminates exchange (Kooyman 1973)
    if depth_m > 40.0:
        O2_mL *= np.exp(-(depth_m - 40.0) / 25.0)
    return O2_mL

# Metabolic rate during dive: swimming + resting
# Williams et al. 1992: dive metabolic rate ~3 W/kg while propelling
MR_swim = 3.0 * m        # W
# Convert to mL O2 / sec  (1 L O2 ~ 20.9 kJ)
E_to_O2 = 1000.0 / 20.9  # mL O2 per kJ
dO2_dt_swim = MR_swim * E_to_O2 / 1000.0  # mL O2 / s

# Bradycardia reduces delivery; increases dependence on Mb
# HR drops 72 -> 6 bpm (Meir 2008). In the dive, Mb-O2 supports the swimming muscle directly.

# Dive profile (t in seconds, depth in m)
t_dive = 10 * 60  # 600 s
dt = 1.0
t = np.arange(0, t_dive + dt, dt)
depth = np.zeros_like(t, dtype=float)
# 2 min descent, 6 min bottom at 500 m, 2 min ascent
for i, ti in enumerate(t):
    if ti < 120:
        depth[i] = 500.0 * ti / 120.0
    elif ti < 120 + 360:
        depth[i] = 500.0
    else:
        depth[i] = max(0.0, 500.0 * (1.0 - (ti - 480.0) / 120.0))

# Heart-rate profile
HR = np.zeros_like(t, dtype=float)
for i, ti in enumerate(t):
    if ti < 20:
        HR[i] = 72.0 - (72.0 - 35.0) * ti / 20.0
    elif ti < 480:
        # Bottom bradycardia
        HR[i] = 10.0 - 4.0 * np.exp(-(ti - 60.0) / 90.0)
    else:
        HR[i] = 15.0 + (72.0 - 15.0) * (ti - 480.0) / 120.0

# Storage arrays
O2_blood = np.zeros_like(t)
O2_muscle = np.zeros_like(t)
O2_lung_arr = np.zeros_like(t)
lactate = np.zeros_like(t)
O2_blood[0] = O2_blood_cap * 0.95
O2_muscle[0] = O2_muscle_cap
O2_lung_arr[0] = lung_O2(0.0)

# Splits of consumption between pools
# In bradycardic phase, muscle is disconnected from blood perfusion; Mb-O2 drains first
# Blood-O2 reserved for brain and heart
# Lung-O2 negligible below collapse
ADL_flag = None

for i in range(1, len(t)):
    O2_lung_arr[i] = lung_O2(depth[i])
    # Split consumption
    frac_muscle = 0.70 if depth[i] > 50 else 0.45
    frac_blood = 1.0 - frac_muscle
    # Muscle O2 supply
    muscle_draw = frac_muscle * dO2_dt_swim * dt
    if O2_muscle[i - 1] >= muscle_draw:
        O2_muscle[i] = O2_muscle[i - 1] - muscle_draw
        lactate[i] = lactate[i - 1]
    else:
        # Anaerobic
        deficit = muscle_draw - O2_muscle[i - 1]
        O2_muscle[i] = 0.0
        # Each mL O2 deficit -> ~0.6 mmol lactate via glycolysis
        lactate[i] = lactate[i - 1] + 0.6 * deficit / 22.4
        if ADL_flag is None:
            ADL_flag = t[i]
    blood_draw = frac_blood * dO2_dt_swim * dt
    O2_blood[i] = max(0.0, O2_blood[i - 1] - blood_draw)

# Plot
fig, axes = plt.subplots(2, 2, figsize=(15, 10))
fig.patch.set_facecolor('#0a0a1a')
for ax in axes.flat:
    ax.set_facecolor('#0a0a1a')
    ax.tick_params(colors='#e2e8f0')
    for spine in ax.spines.values():
        spine.set_color('#475569')
    ax.grid(True, color='#334155', alpha=0.35)

ax = axes[0, 0]
ax.plot(t, depth, color='#60a5fa', lw=2.4)
ax.invert_yaxis()
ax.set_xlabel('Time (s)', color='#e2e8f0')
ax.set_ylabel('Depth (m)', color='#e2e8f0')
ax.set_title('Dive profile: 500 m for 10 min',
             color='#fde68a', fontsize=13, fontweight='bold')
ax.axhline(40, color='#f59e0b', ls='--', alpha=0.6, label='lung collapse')
ax.legend(facecolor='#1e293b', edgecolor='#475569', labelcolor='#e2e8f0')

ax = axes[0, 1]
ax.plot(t, O2_blood, color='#f43f5e', lw=2.2, label='Blood O2')
ax.plot(t, O2_muscle, color='#fbbf24', lw=2.2, label='Muscle Mb-O2')
ax.plot(t, O2_lung_arr, color='#60a5fa', lw=2.2, label='Lung O2 (gas)')
if ADL_flag is not None:
    ax.axvline(ADL_flag, color='#fde68a', ls='--', alpha=0.8, label=f'ADL at {int(ADL_flag)} s')
ax.set_xlabel('Time (s)', color='#e2e8f0')
ax.set_ylabel('O2 available (mL)', color='#e2e8f0')
ax.set_title('Three-compartment O2 depletion',
             color='#fde68a', fontsize=13, fontweight='bold')
ax.legend(facecolor='#1e293b', edgecolor='#475569', labelcolor='#e2e8f0')

ax = axes[1, 0]
ax.plot(t, HR, color='#a3e635', lw=2.4)
ax.set_xlabel('Time (s)', color='#e2e8f0')
ax.set_ylabel('Heart rate (bpm)', color='#e2e8f0')
ax.set_title('Bradycardia: 72 -> 6 bpm at bottom',
             color='#fde68a', fontsize=13, fontweight='bold')
ax.axhline(6, color='#94a3b8', ls=':', alpha=0.5, label='min HR ~6 bpm')
ax.legend(facecolor='#1e293b', edgecolor='#475569', labelcolor='#e2e8f0')

ax = axes[1, 1]
ax.plot(t, lactate, color='#fb7185', lw=2.4)
ax.set_xlabel('Time (s)', color='#e2e8f0')
ax.set_ylabel('Lactate accumulated (mmol)', color='#e2e8f0')
ax.set_title('Anaerobic lactate if ADL exceeded',
             color='#fde68a', fontsize=13, fontweight='bold')

plt.tight_layout()
plt.savefig('output.png', dpi=120, bbox_inches='tight', facecolor='#0a0a1a')

total_O2 = O2_blood[0] + O2_muscle[0] + O2_lung_arr[0]
print('Total usable O2 pre-dive (mL):', round(float(total_O2), 1))
print('  blood:', round(float(O2_blood[0]), 1))
print('  muscle:', round(float(O2_muscle[0]), 1))
print('  lung:', round(float(O2_lung_arr[0]), 1))
print('Final muscle O2 (mL):', round(float(O2_muscle[-1]), 2))
print('Final lactate (mmol):', round(float(lactate[-1]), 2))
if ADL_flag is not None:
    print('Aerobic dive limit exceeded at t =', int(ADL_flag), 's')
else:
    print('Dive remained aerobic throughout')
print('Mean HR during bottom phase:', round(float(HR[120:480].mean()), 1), 'bpm')
print('Cumulative O2 consumption (L):',
      round(float(np.trapezoid(np.full_like(t, dO2_dt_swim), t) / 1000.0), 2))

Click Run to execute the Python code

Code will be executed with Python 3 on the server

8. Foraging Ecology: What Lies Below

Emperor diet is dominated by three prey classes. Stomach-content analysis (Cherel & Kooyman 1998) from Ross Sea birds gives by mass:

Antarctic silverfish (Pleuragramma antarctica): 38% by wet mass; depth range 100–700 m; high lipid content.
Antarctic krill (Euphausia superba): 27%; swarms typically 50–200 m; caloric density lower than fish.
Squid (Psychroteuthis glacialis, Alluroteuthis): 20%; mesopelagic to bathypelagic (200–700 m).
Other: 15% (amphipods, juvenile Antarctic toothfish).

Foraging success per dive is strongly depth-stratified. Shallow V-dives (under 100 m) capture mostly krill; deep U-dives to 200–300 m capture mostly silverfish; benthic dives to 500+ m target demersal fish and squid. A full foraging bout of ~100 dives/day delivers ~3 kg of prey biomass—enough to refuel and provision a growing chick.

Navigation and prey detection

Under polar night emperors dive in total darkness. Prey detection relies on:

Bioluminescence cueing: many krill and mesopelagic fish emit photons that penguin retinas (highly sensitive at 450–510 nm) can detect.
Tactile whisker-like vibrissae: absent in penguins but beak-mechanoreceptors detect contact.
Acoustic cues: emperor ears are sensitive underwater to 12–18 kHz, matching squid jet-propulsion spectra.
Magnetoreception: suggestive evidence (Wienecke 2018); not yet confirmed in Sphenisciformes.

9. Dive Bout Architecture and the “Power Law”

Within a foraging trip, dives are organised into bouts separated by long surface intervals (15–60 min) during which the bird rests on ice or floats. Bout duration distribution is approximately log-normal; inter-bout interval is a Pareto (power-law) tail. Seibel & Drazen (2007) characterised the pattern with

\[ P(\Delta t_{\text{bout}}) \propto \Delta t^{-\alpha},\quad \alpha \approx 1.6 \]

The foraging trip is optimised under a Marginal Value Theorem constraint: the bird continues diving at a patch until the mean return per dive drops to the long-run average for the entire ocean segment. Early in a bout, when prey density is high, dives are short and shallow; late in a bout, dives deepen as easy prey becomes depleted.

Simulation 2: Bradycardia Trigger & Recovery Dynamics

Simulate the heart-rate time series during and after a 9-min dive using a two-timescale autonomic model: fast trigeminal-reflex sympathetic withdrawal + slow chemoreceptor vagal augmentation. Fit the recovery time constant and compare against Meir (2008) tachometer observations.

Python

script.py165 lines

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

# Bradycardia trigger and recovery dynamics
# Two-timescale neural model capturing the dive-response cascade
# driven by face-immersion sensors + arterial chemoreceptor feedback
# Based on Meir et al. (2008) J. Exp. Biol. emperor penguin HR time series
# and Ponganis et al. (1997) J. Exp. Biol. tachometer recordings.

np.random.seed(3)

dt = 0.1                 # s
T_total = 900.0          # 15 minutes: pre-dive + dive + recovery
N = int(T_total / dt)
t = np.linspace(0, T_total, N)

# Dive state: 0 = surface, 1 = submerged
dive_state = np.zeros_like(t, dtype=int)
start = 60.0
end = 600.0
dive_state[(t >= start) & (t < end)] = 1

# Arterial O2 saturation (input to peripheral chemoreceptors)
# Starts at 95%, falls approximately linearly during dive
SaO2 = np.zeros_like(t)
for i, ti in enumerate(t):
    if ti < start:
        SaO2[i] = 0.95
    elif ti < end:
        progress = (ti - start) / (end - start)
        # S-shaped fall
        SaO2[i] = 0.95 - 0.50 / (1.0 + np.exp(-8.0 * (progress - 0.55)))
    else:
        # Recovery exponential
        SaO2[i] = 0.55 + 0.40 * (1.0 - np.exp(-(ti - end) / 18.0))

# Two-state HR model
# Sympathetic drive S (baseline ~0.8) lowered sharply at dive onset by trigeminal reflex
# Parasympathetic (vagal) drive V elevated at dive onset
# HR = HR_intrinsic + gS * S - gV * V with saturation
HR_intrinsic = 40.0
gS = 60.0
gV = 45.0

S = np.ones_like(t)
V = np.zeros_like(t)
HR = np.zeros_like(t)

# Time constants
tau_trig = 1.2           # trigeminal reflex fast (~1 s)
tau_chemo = 28.0         # chemoreceptor slow integration (~30 s)
tau_recov = 4.5          # recovery tachycardia fast

# Reflex trigger signals
trig_on = np.zeros_like(t)
for i, ti in enumerate(t):
    # Face wetting simulated as step at dive start/end
    if start < ti < end:
        trig_on[i] = 1.0

# Chemoreceptor signal as deviation from baseline SaO2 = 0.95
chemo_signal = np.maximum(0.0, 0.95 - SaO2)

# Integrate S, V
for i in range(1, N):
    # Sympathetic: suppressed by trigeminal; restored by low O2 (partial cross-talk)
    S_target = 0.2 if dive_state[i] else 1.0
    S[i] = S[i - 1] + dt * (S_target - S[i - 1]) / tau_trig
    # Vagal: elevated by trigeminal, further modulated by chemo
    V_target_trig = 1.0 if dive_state[i] else 0.0
    V_target = V_target_trig * (1.0 + 1.5 * chemo_signal[i])
    tau_V = tau_trig if dive_state[i] else tau_recov
    V[i] = V[i - 1] + dt * (V_target - V[i - 1]) / tau_V

HR = np.clip(HR_intrinsic + gS * S - gV * V, 4.0, 110.0)
# Add noise
HR += np.random.normal(0, 1.2, size=HR.shape)

# Detect HR quantiles at bottom to compute quantitative metrics
mask_bottom = (t >= start + 120) & (t < end - 60)
HR_bottom = HR[mask_bottom]
mask_recov = (t >= end) & (t < end + 180)
HR_recov = HR[mask_recov]
t_recov = t[mask_recov] - end

# Fit tau recovery by single-exponential
def exp_fit(tt, tau):
    return HR_recov[0] + (HR[0] - HR_recov[0]) * (1.0 - np.exp(-tt / tau))

from scipy.optimize import curve_fit
try:
    popt, _ = curve_fit(exp_fit, t_recov, HR_recov, p0=[10.0], maxfev=5000)
    tau_fit = popt[0]
except Exception:
    tau_fit = tau_recov

# Meir 2008 reported stroke-rate coupling: HR tracks flipper stroke during dive
stroke_rate = np.zeros_like(t)
for i, ti in enumerate(t):
    if dive_state[i] and ti < start + 120:
        stroke_rate[i] = 1.4
    elif dive_state[i]:
        stroke_rate[i] = 0.5 + 0.4 * np.sin(2 * np.pi * (ti - start) / 20.0)
    else:
        stroke_rate[i] = 0.0

# Figure
fig, axes = plt.subplots(3, 1, figsize=(14, 11))
fig.patch.set_facecolor('#0a0a1a')
for ax in axes.flat:
    ax.set_facecolor('#0a0a1a')
    ax.tick_params(colors='#e2e8f0')
    for spine in ax.spines.values():
        spine.set_color('#475569')
    ax.grid(True, color='#334155', alpha=0.35)

ax = axes[0]
ax.plot(t, HR, color='#fbbf24', lw=1.4, label='HR (bpm)')
ax.axvspan(start, end, color='#60a5fa', alpha=0.10, label='submerged')
ax.axhline(6, color='#f43f5e', ls='--', alpha=0.6, label='min ~6 bpm (Meir)')
ax.axhline(72, color='#a3e635', ls='--', alpha=0.6, label='resting 72 bpm')
ax.set_xlabel('Time (s)', color='#e2e8f0')
ax.set_ylabel('Heart rate (bpm)', color='#e2e8f0')
ax.set_title('Simulated dive bradycardia + recovery tachycardia',
             color='#fde68a', fontsize=13, fontweight='bold')
ax.legend(facecolor='#1e293b', edgecolor='#475569', labelcolor='#e2e8f0')

ax = axes[1]
ax.plot(t, S, color='#a3e635', lw=2.0, label='Sympathetic S(t)')
ax.plot(t, V, color='#f43f5e', lw=2.0, label='Vagal V(t)')
ax.plot(t, chemo_signal, color='#60a5fa', lw=1.6, label='Chemo signal (desat)')
ax.set_xlabel('Time (s)', color='#e2e8f0')
ax.set_ylabel('Drive (arb)', color='#e2e8f0')
ax.set_title('Autonomic drives: trigeminal reflex + chemoreceptor feedback',
             color='#fde68a', fontsize=13, fontweight='bold')
ax.legend(facecolor='#1e293b', edgecolor='#475569', labelcolor='#e2e8f0')

ax = axes[2]
ax.plot(t, stroke_rate, color='#fb7185', lw=1.8, label='Flipper stroke rate (Hz)')
ax2 = ax.twinx()
ax2.plot(t, SaO2 * 100.0, color='#60a5fa', lw=1.8, label='SaO2 (%)')
ax2.tick_params(colors='#60a5fa')
ax.set_xlabel('Time (s)', color='#e2e8f0')
ax.set_ylabel('Stroke rate (Hz)', color='#fb7185')
ax2.set_ylabel('SaO2 (%)', color='#60a5fa')
ax.set_title('Stroke-rate + arterial O2 saturation (Meir 2008 coupling)',
             color='#fde68a', fontsize=13, fontweight='bold')
ax.legend(loc='upper left', facecolor='#1e293b', edgecolor='#475569', labelcolor='#e2e8f0')
ax2.legend(loc='upper right', facecolor='#1e293b', edgecolor='#475569', labelcolor='#e2e8f0')

plt.tight_layout()
plt.savefig('output.png', dpi=120, bbox_inches='tight', facecolor='#0a0a1a')

print('Mean HR at bottom phase (bpm):', round(float(HR_bottom.mean()), 2))
print('Minimum HR observed (bpm):', round(float(HR.min()), 2))
print('Pre-dive HR (bpm):', round(float(HR[:int(start / dt)].mean()), 1))
print('Post-dive HR 10 s after surfacing (bpm):',
      round(float(HR[int((end + 10.0) / dt)]), 1))
print('Recovery time constant fit (s):', round(float(tau_fit), 2))
print('Fraction of dive with HR < 10 bpm:',
      round(float(np.mean(HR[(t >= start) & (t < end)] < 10.0)), 3))
print('Mean SaO2 at bottom (%):', round(float(SaO2[mask_bottom].mean() * 100.0), 1))

Click Run to execute the Python code

Code will be executed with Python 3 on the server

10. Diving-Related Behaviours (Winn 2023)

Winn, Ponganis & Williams (2023) analysed archival-tag data from 43 emperor penguins at Cape Crozier and reported several behaviours that correlate with diving pressure:

Pre-dive hyperventilation: 5–8 rapid breaths immediately before submergence; increases arterial PO\(_2\) to ~115 mmHg.
Post-dive shivering: brief 30–60 s bout of shivering after deep dives; correlates with re-warming from peripheral vasoconstriction.
Porpoising-like surface leap: during rapid surface travel, brief out-of-water leaps reduce drag (see Module 5).
Bottom-phase pauses: during U-dives, 10–20 s interruptions of flipper stroking with passive gliding—likely ambush-prey positioning.
Ascent spin: a spiral ascent in ~30% of deep dives, possibly to shed accumulated nitrogen micronuclei before surfacing.

The ascent-spin behaviour is unusual. Winn 2023 hypothesised that centrifugal forces aid evacuation of bubbles from the bloodstream as gas supersaturation grows during the ascent from 400+ m; observational evidence is suggestive but not yet definitive. Experimental verification would require high-speed Doppler ultrasound of diving emperors—not yet attempted.

11. Comparative Diving Physiology

Species	Mass (kg)	Max depth (m)	Max duration (min)	[Mb] (mg/g)
Emperor penguin	28	565	22	64
Weddell seal	400	750	82	54
Cuvier’s beaked whale	3000	2992	138	70
King penguin	13	304	9	39
Elephant seal	1500	1735	120	69

Per unit body mass, the emperor’s aerobic dive limit is among the highest of any vertebrate. Its specific adaptations—myoglobin density, blood volume, lung-collapse mechanics, and bradycardic range—are converged independently on the pinniped lineage, reflecting strong selection for deep foraging in cold, high-productivity Antarctic seas.

12. Post-Dive Recovery and Inter-Dive Interval

After surfacing, the emperor performs 4–10 rapid breaths during an inter-dive interval (IDI) typically 30–120 s long. The IDI is controlled by the lactate clearance rate and by the replenishment of arterial and lung O\(_2\).

\[ \text{IDI} = \max\!\left(\tau_{\text{O2}} \ln\!\frac{S_{\text{max}} - S_0}{S_{\text{max}} - S_t},\; \frac{[\text{lac}]_t}{k_{\text{Cori}}}\right) \]

The IDI is the longer of the two recovery timescales: arterial re-oxygenation (\(\tau_{O_2}\sim 15\) s) or lactate clearance (\(\sim\)min).

For aerobic dives (ADL not exceeded) the IDI is set by arterial re-oxygenation—30–60 s. For anaerobic excursions the IDI can exceed 5 min, consistent with Ponganis et al. (2011) observations of long post-dive rest after record-length dives.

Discussion & Graduate Exercises

Compute the ADL of a 30 kg emperor with O\(_2\) store 70 mL/kg and dive metabolic rate 10 mL/kg/min. Compare to Ponganis’ empirical 5.6 min bADL. What assumptions explain the discrepancy?
At 500 m depth, what fraction of the surface lung volume remains? At what depth does the lung gas contribute < 1% to total O\(_2\) store?
Starting from Poiseuille flow, estimate the pressure required to collapse a terminal bronchiole of radius 0.3 mm and elastic modulus 10 kPa. Show that this corresponds to ambient pressure of ~5 atm (~40 m depth).
Modify Simulation 1 to impose muscle-only perfusion during bradycardia. Does the simulated ADL match Ponganis’ 5.6 min? What additional mechanism (hepatic sinus) would extend the ADL to the observed 22 min record?
The Meir (2008) dataset shows HR track flipper stroke rate at slow rates (<10 bpm). Propose a neural mechanism and write down a coupled HR-stroke ODE.
Use the PSMC-inferred emperor \(N_e\) (Module 0) plus ancient DNA to estimate the selection coefficient on myoglobin [Mb]. Assume [Mb] has heritability\(h^2=0.3\) and a 30-year generation time; infer the number of generations needed to reach modern [Mb] from an ancestral 30 mg/g.

Key References

• Kooyman, G.L. (1971). “Deep diving behaviour and effects of pressure in reptiles, birds, and mammals.” Symp. Zool. Soc. Lond. 31, 101–120.

• Kooyman, G.L. (1985). “Physiology without restraint in diving mammals.” Marine Mammal Science 1, 166–178.

• Kooyman, G.L., Drabek, C.M., Elsner, R., Campbell, W.B. (1971). “Diving behavior of the emperor penguin.” Auk 88, 775–795.

• Kooyman, G.L., Ponganis, P.J. (1998). “The physiological basis of diving to depth: birds and mammals.” Annu. Rev. Physiol. 60, 19–32.

• Ponganis, P.J. (2011). “Diving mammals.” Comprehensive Physiology 1, 447–465.

• Ponganis, P.J., Stockard, T.K., Meir, J.U., Williams, C.L., Ponganis, K.V., Howard, R. (2009). “O2 store management in diving emperor penguins.” J. Exp. Biol. 212, 217–224.

• Ponganis, P.J., Kooyman, G.L., Starke, L.N., Kooyman, C.A., Kooyman, T.G. (1997). “Post-dive blood lactate concentrations in emperor penguins.” J. Exp. Biol. 200, 1623–1626.

• Meir, J.U., Stockard, T.K., Williams, C.L., Ponganis, K.V., Ponganis, P.J. (2008). “Heart rate regulation and extreme bradycardia in diving emperor penguins.” J. Exp. Biol. 211, 1169–1179.

• Wienecke, B., Robertson, G., Kirkwood, R. (2007). “Extreme dives by free-ranging emperor penguins.” Polar Biology 30, 133–142.

• Noren, S.R., Williams, T.M. (2000). “Body size and skeletal muscle myoglobin of cetaceans.” Comp. Biochem. Physiol. A 126, 181–191.

• Cherel, Y., Kooyman, G.L. (1998). “Food of emperor penguins in the western Ross Sea, Antarctica.” Marine Biology 130, 335–344.

• Lewis, K., Ponganis, P.J., Smith, C. (2014). “Anatomy of the middle ear in diving birds.” J. Morphology 275, 987–995.

• Winn, B.E., Ponganis, P.J., Williams, C.L. (2023). “Behavioral repertoire and ascent dynamics in diving emperor penguins.” Physiol. Behav. 259, 114027.

• Williams, T.M., Friedl, W.A., Fong, M.L., Yamada, R.M., Sedivy, P., Haun, J.E. (1992). “Travel at low energetic cost by swimming and wave-riding bottlenose dolphins.” Nature 355, 821–823.

• Prevost, J., Sapin-Jaloustre, J. (1965). “Ecologie des manchots antarctiques.” Biogeography and Ecology in Antarctica, 551–648.

• Seibel, B.A., Drazen, J.C. (2007). “The rate of metabolism in marine animals.” Phil. Trans. R. Soc. B 362, 2061–2078.

• Scholander, P.F. (1940). “Experimental investigations on the respiratory function in diving mammals and birds.” Hvalradets Skrifter 22, 1–131.

• Kooyman, G.L., Castellini, M.A., Davis, R.W., Maue, R.A. (1983). “Aerobic diving limits of immature Weddell seals.” J. Comp. Physiol. B 151, 171–174.

• Stockard, T.K., Levenson, D.H., Berg, L., Fransioli, J.R., Baranov, E.A., Ponganis, P.J. (2005). “Blood oxygen depletion during rest-associated apneas of northern elephant seals.” J. Exp. Biol. 208, 4967–4975.

Synthesis & Bridge to Module 5

The emperor’s dive record is a composite achievement of haematology (Hb-O2), myology (Mb-O2), pulmonology (lung collapse), cardiology (bradycardia), and metabolism (ADL/lactate). None of these adaptations in isolation would suffice; together they yield an animal that can out-dive every other bird by an order of magnitude. The physiological story is inseparable from the ecological: the mesopelagic silverfish and squid shoals of the Southern Ocean reward deep foraging with high caloric density.

Module 5 turns from the physiological to the hydromechanical: how does a 30-kg torpedo travel through water with \(C_D \approx 0.02\)—a drag coefficient lower than a dolphin’s—and sustain burst speeds of 30 km/h? We derive the Bannasch (1995) drag-coefficient measurements, analyse the Davenport 2011 air-bubble drag-reduction phenomenon, and model flipper kinematics in the Strouhal regime of peak propulsive efficiency.

← Module 3: Breeding & 4-Month Fast Module 5: Underwater Hydrodynamics →