Module 2

Mesopelagic (200–1000 m) — The Twilight Zone

The ocean's largest habitat: a realm of bioluminescent hunters, diel vertical migration, and the oxygen minimum zone

2.1 Diel Vertical Migration (DVM)

Diel vertical migration is the largest coordinated biomass movement on Earth. Every night, an estimated 1–10 billion tonnes of organisms — including copepods, krill, myctophid fish, squid, and jellyfish — ascend 400–800 m from the mesopelagic to the epipelagic to feed under cover of darkness, then descend at dawn to avoid visual predators.

DVM has profound biogeochemical consequences: it actively transports carbon from the surface to depth (~2–3 Gt C/yr globally), supplementing the passive sinking of marine snow. This active carbon pump is not captured in sediment trap measurements and was historically underestimated.

Energetic Cost-Benefit Model

The optimal migration behaviour can be modelled as a trade-off between feeding gain at the surface, metabolic cost of swimming, and predation risk. An organism should migrate if the net energy balance is positive:

\( E_{\text{net}} = G_{\text{surface}} - C_{\text{migration}} - M_{\text{predation}} \)

Net energy = feeding gain - swimming cost - predation mortality cost

The feeding gain at the surface depends on food concentration and feeding time (limited to dark hours, ~10 hr):

\( G_{\text{surface}} = I_{\max} \cdot \frac{C_s}{C_s + K_{1/2}} \cdot t_{\text{night}} \)

The migration cost depends on swimming speed, distance, and metabolic rate. For a copepod swimming at velocity \(v\) over distance \(2\Delta z\) (round trip):

\( C_{\text{migration}} = \left(R_{\text{active}} - R_{\text{basal}}\right) \cdot \frac{2\Delta z}{v} + R_{\text{basal}} \cdot 24\text{h} \)

\(R_{\text{active}} \approx 2\text{--}3 \times R_{\text{basal}}\) during swimming

Optimal Migration Depth

The daytime residence depth represents a predation-starvation trade-off. Deeper residence means lower predation risk (less light for visual predators) but higher metabolic cost (longer migration, colder temperature). The predation mortality rate decreases exponentially with depth:

\( M_{\text{pred}}(z) = M_0 \cdot \exp\!\left(-\frac{z}{z_{\text{vis}}}\right) \)

\(z_{\text{vis}} \approx\) 50–100 m (visual predation scale depth)

The optimal depth \(z^*\) maximises fitness (net energy minus predation cost). Taking\(\partial E_{\text{net}} / \partial z = 0\):

\( \frac{\partial C_{\text{mig}}}{\partial z} = \frac{\partial M_{\text{pred}}}{\partial z} = -\frac{M_0}{z_{\text{vis}}} \exp\!\left(-\frac{z^*}{z_{\text{vis}}}\right) \)

This yields deeper residence depths for larger organisms (lower mass-specific metabolic cost), in waters with more visual predators, and during full moon (increased light penetration). Acoustic surveys confirm these predictions: the deep scattering layer (DSL) deepens on moonlit nights.

2.2 Bioluminescence: Chemistry & Ecology

Approximately 76% of mesopelagic organisms produce light through bioluminescence. This makes the twilight zone the most bioluminescent environment on Earth. Light production has evolved independently at least 40 times in marine organisms.

Biochemistry of Light Production

All bioluminescence involves oxidation of a luciferin substrate catalysed by a luciferase enzyme (or a photoprotein):

\( \text{Luciferin} + \text{O}_2 \xrightarrow{\text{luciferase}} \text{Oxyluciferin}^* \to \text{Oxyluciferin} + h\nu \)

The asterisk denotes the electronically excited state that emits a photon upon relaxation

The quantum yield measures the efficiency of photon production:

\( \Phi = \frac{\text{photons emitted}}{\text{luciferin molecules oxidised}} \)

Firefly (terrestrial): \(\Phi \approx 0.88\); marine systems: \(\Phi \approx 0.10\text{--}0.30\)

Coelenterazine: The Universal Marine Luciferin

Coelenterazine (an imidazopyrazinone) is the most widespread marine luciferin, used by cnidarians, ctenophores, copepods, ostracods, chaetognaths, squid, and fish. Its ubiquity suggests it may function as an antioxidant in non-luminous organisms, with bioluminescence being a secondary adaptation. Peak emission: ~470 nm (blue), matching the wavelength of maximum transmission in seawater.

Bacterial Bioluminescence: Vibrio fischeri

Many deep-sea fish and squid produce light via symbiotic bacteria housed in specialised light organs (photophores). Vibrio fischeri uses a quorum-sensing mechanism: light production activates only above a threshold cell density, regulated by the autoinducer N-acyl homoserine lactone (AHL). The luciferase reaction uses a long-chain aldehyde:

\( \text{FMNH}_2 + \text{O}_2 + \text{RCHO} \xrightarrow{\text{lux}} \text{FMN} + \text{RCOOH} + \text{H}_2\text{O} + h\nu_{490} \)

Ecological Functions of Bioluminescence

Counter-illumination: Ventral photophores match downwelling light, eliminating the organism's silhouette when viewed from below. Used by hatchetfish, squid, krill. Requires precise matching of intensity and angular distribution.
Predator avoidance (burglar alarm): Bright flash startles predator or attracts a larger predator to attack the attacker. Used by dinoflagellates, some jellyfish.
Prey attraction: Anglerfish lure (esca) contains luminous bacteria. Dragonfish (Malacosteus) uniquely produces far-red light (~700 nm) invisible to most deep-sea organisms, acting as a covert searchlight.
Communication: Species-specific flash patterns for mate recognition, analogous to firefly signalling. Used by ostracods and some squid.

2.3 The Oxygen Minimum Zone (OMZ)

The OMZ is a layer of severely depleted dissolved oxygen (\(<0.5\) mL/L, or \(<20\;\mu\)mol/kg) found at 200–1000 m depth. It arises from the imbalance between biological oxygen consumption (decomposition of sinking organic matter) and oxygen supply (lateral advection and vertical mixing):

\( \frac{\partial [\text{O}_2]}{\partial t} = \underbrace{\nabla \cdot (K \nabla [\text{O}_2])}_{\text{diffusion}} + \underbrace{\mathbf{u} \cdot \nabla [\text{O}_2]}_{\text{advection}} - \underbrace{R_{\text{bio}}(z)}_{\text{consumption}} \)

Steady-State O\(_2\) Balance

In a simplified 1D vertical model at steady state, vertical advection (upwelling \(w\)) balances biological consumption:

\( w \frac{\partial [\text{O}_2]}{\partial z} + \kappa \frac{\partial^2 [\text{O}_2]}{\partial z^2} = R_{\text{bio}}(z) \)

The consumption rate \(R_{\text{bio}}(z)\) follows the sinking organic matter flux, which attenuates with depth according to the Martin curve (see Section 2.5). The strongest consumption occurs just below the euphotic zone where fresh organic matter first encounters heterotrophic bacteria.

OMZ Expansion Under Climate Change

Global OMZs have expanded by 3–8% since 1960 (Schmidtko et al., 2017). Two mechanisms drive this:

Reduced O\(_2\) solubility: Warmer water holds less dissolved gas. Henry's law: \([\text{O}_2]_{\text{sat}} = K_H \cdot p_{\text{O}_2}\), where \(K_H\)decreases ~2% per °C of warming.
Increased stratification: Surface warming strengthens the pycnocline, reducing vertical mixing and O\(_2\) ventilation of subsurface waters. This effect dominates at mid-latitudes.

OMZ expansion compresses the habitable depth range for aerobic organisms, with cascading effects on fisheries and biogeochemistry. It also enhances denitrification, producing N\(_2\)O (a potent greenhouse gas) — a positive feedback discussed inClimate Science Module 6.

2.4 Pressure-Adapted Enzymes & TMAO

Organisms in the mesopelagic experience pressures of 20–100 atm (2–10 MPa). Their proteins must function under conditions that would denature mesophilic enzymes.

Le Chatelier's Principle and Protein Stability

Pressure shifts equilibria toward states with smaller molar volume. Protein folding involves burying hydrophobic residues in a core with imperfect packing, creating small voids. The volume change upon unfolding:

\( \Delta V_{\text{unfold}} = V_{\text{unfolded}} - V_{\text{folded}} < 0 \)

Typically \(\Delta V \approx -50\;\text{to}\;-100\;\text{mL/mol}\) for globular proteins

Since \(\Delta V < 0\), pressure favours the unfolded (denatured) state. The equilibrium constant for unfolding increases with pressure according to:

\( K(P) = K(1\;\text{atm}) \cdot \exp\!\left(-\frac{\Delta V \cdot (P - 1)}{RT}\right) \)

For \(\Delta V = -80\;\text{mL/mol}\) at 500 atm:\(\ln(K(500)/K(1)) = 80 \times 10^{-6} \times 500 \times 101325 / (8.314 \times 283) \approx 1.7\), so \(K\) increases by a factor of ~5.5 — significant destabilisation.

TMAO as a Piezolyte

Trimethylamine N-oxide (TMAO) is the primary osmolyte used by deep-sea fish to counteract pressure effects on proteins. TMAO concentration increases linearly with depth:

\( [\text{TMAO}] \approx 40 + 0.39 \times z\;\;\text{(mmol/kg, z in m)} \)

Yancey et al. (2014): TMAO increases ~40 mmol/kg per 1000 m

TMAO stabilises proteins by strengthening the hydration shell around hydrophobic groups. It increases the thermodynamic cost of exposing hydrophobic residues to water, making the folded state more favourable even under high pressure. The free energy of stabilisation:

\( \Delta G_{\text{stab}} \propto [\text{TMAO}] \cdot \Delta A_{\text{SASA}} \)

\(\Delta A_{\text{SASA}}\): change in solvent-accessible surface area upon unfolding

Yancey et al. (2014) proposed that the maximum depth at which fish can survive (~8200 m) is set by the TMAO concentration approaching levels that would interfere with other cellular processes. Below this depth, only invertebrates (amphipods, with different osmolyte strategies) are found.

2.5 Marine Snow & the Martin Curve

Marine snow consists of sinking aggregates of dead phytoplankton, fecal pellets, mucus sheets, and associated bacteria. These particles are the primary vehicle for transferring carbon from the surface to the deep ocean (the biological carbon pump).

Stokes Settling Velocity

The sinking speed of marine snow particles is governed by Stokes' law for small Reynolds number:

\( v_s = \frac{2 r^2 \Delta\rho\,g}{9 \mu} \)

r: particle radius, \(\Delta\rho = \rho_p - \rho_w\): excess density, \(\mu\): dynamic viscosity

Typical values: r = 0.5–5 mm, \(\Delta\rho \approx 10\text{--}100\;\text{kg/m}^3\), giving \(v_s \approx 10\text{--}200\;\text{m/day}\). Larger and denser particles (fecal pellets, diatom aggregates) sink faster, spending less time in the mesopelagic and losing less carbon to bacterial decomposition.

Martin Curve: Flux Attenuation

Martin et al. (1987) empirically described the decay of sinking organic carbon flux with depth using a power law:

\( F(z) = F_0 \left(\frac{z}{z_0}\right)^{-b} \)

\(F_0\): flux at reference depth \(z_0\) (usually 100 m), b \(\approx\) 0.86 (Martin exponent)

Derivation sketch: If the remineralisation rate at depth \(z\) is proportional to \(F(z)\), then \(dF/dz = -\lambda(z) F\) with\(\lambda(z) = b/z\) (increasing slowly with depth). Integrating:

\( \int_{F_0}^{F} \frac{dF'}{F'} = -\int_{z_0}^{z} \frac{b}{z'}\,dz' = -b \ln\!\left(\frac{z}{z_0}\right) \)

\( F(z) = F_0 \exp\!\left(-b \ln\frac{z}{z_0}\right) = F_0 \left(\frac{z}{z_0}\right)^{-b} \)

Consequences for carbon sequestration: at 1000 m, only \((1000/100)^{-0.86} \approx 14\%\) of the export flux remains. At 4000 m, only \(\approx 4\%\). The exponent \(b\) varies regionally: higher \(b\) (more attenuation) in productive waters with labile organic matter; lower \(b\) in oligotrophic regions with refractory particles.

The efficiency of the biological pump in sequestering CO\(_2\) depends critically on \(b\): a 25% decrease in \(b\) would increase deep-ocean carbon storage by ~0.5 Gt C/yr, with significant implications for climate projections.

2.6 Mesopelagic Fish & Global Biomass Estimates

The mesopelagic zone harbours an enormous but poorly quantified biomass of fish, dominated bymyctophids (lanternfish, family Myctophidae, ~250 species) andbristlemouths (Cyclothone spp., possibly the most abundant vertebrate genus on Earth). Traditional trawl surveys estimated mesopelagic fish biomass at ~1 Gt, but acoustic surveys suggest the true figure may be 10–15 Gt(Irigoien et al., 2014), making it the largest fish biomass on the planet.

The discrepancy arises because mesopelagic fish are highly effective at net avoidance. Acoustic backscattering measurements detect the deep scattering layer (DSL), first discovered during WWII when sonar operators noticed a “false bottom” that rose at night. The DSL corresponds to the daytime aggregation of vertically migrating organisms.

Acoustic Biomass Estimation

Fish biomass is estimated from volume backscattering strength \(S_v\) (dB re 1 m\(^{-1}\)):

\( S_v = 10 \log_{10}\!\left(\sum_{i} n_i \sigma_{bs,i}\right) \)

\(n_i\): number density of scatterers, \(\sigma_{bs}\): backscattering cross-section (depends on swim bladder)

Converting acoustic backscatter to biomass requires knowing the target strength of individual organisms, which depends on body size, swim bladder type (gas-filled vs. fat-filled vs. absent), and tilt angle. Myctophids with gas-filled swim bladders have target strengths ~10 dB higher than those without, potentially biasing biomass estimates by an order of magnitude.

Carbon Cycling Role

Mesopelagic fish play a dual role in carbon cycling:

Active carbon transport: Feeding at the surface and respiring/defecating at depth, DVM fish transport an estimated 0.5–1.5 Gt C/yr below the permanent thermocline.
Fragmentation of marine snow: Mesopelagic organisms feed on sinking particles, breaking large aggregates into smaller, slower-sinking fragments. This increases the Martin curve exponent \(b\), reducing passive export efficiency.

The net effect on the biological carbon pump depends on the balance between active transport (enhancing deep carbon) and fragmentation (reducing passive flux). Current estimates suggest the active transport dominates, making mesopelagic fish net contributors to carbon sequestration with direct implications for climate-biodiversity coupling.

2.7 Sensory Adaptations to Permanent Twilight

The mesopelagic is a world of extreme low light: at 500 m in clear ocean water, daylight intensity is ~10\(^{-8}\) of surface values. Organisms have evolved remarkable sensory adaptations:

Tubular Eyes

Many mesopelagic fish (e.g., Opisthoproctus, the barreleye) have evolved tubular eyes — cylindrical rather than spherical — that maximise light collection from a narrow field of view directly above. The retinal sensitivity is enhanced by a tapetum lucidum (reflective layer) and densely packed rods with visual pigment\(\lambda_{\max} \approx 480\) nm, matching the blue bioluminescence spectrum.

The photon catch rate for a tubular eye of aperture area \(A\) viewing a bioluminescent flash at distance \(d\):

\( \dot{N} = \frac{F_{\text{flash}} \cdot A}{4\pi d^2} \cdot \eta_{\text{retina}} \cdot e^{-K_d \cdot d} \)

\(F_{\text{flash}}\): flash photon output, \(\eta_{\text{retina}}\): quantum efficiency of rhodopsin (~0.7)

Lateral Line and Mechanoreception

In near-complete darkness, many mesopelagic predators rely on mechanoreception via an enhanced lateral line system. The canal neuromasts detect pressure gradients from prey swimming movements. Detection range scales with prey size and swimming speed:

\( r_{\text{detect}} \propto \left(\frac{v_{\text{prey}} \cdot L_{\text{prey}}^2}{\nu}\right)^{1/3} \)

\(\nu\): kinematic viscosity; detection range is typically 1–3 body lengths

Electroreception

Some deep-sea sharks and rays retain ancestral electroreceptors (ampullae of Lorenzini) capable of detecting electric fields as weak as 5 nV/cm. The bioelectric field generated by a swimming fish at distance \(r\) attenuates as \(\sim r^{-3}\) (electric dipole). In the mesopelagic, where visual detection is unreliable, electroreception provides an additional sensory modality for locating prey at distances of 0.5–1 m.

Diel Vertical Migration Cycle

Day versus night depth distributions of migrating organisms, with the OMZ band and bioluminescence zone indicated:

2.8 The Mesopelagic Carbon Budget

The mesopelagic zone is the critical gateway for carbon sequestration. Carbon enters via three pathways: passive sinking (marine snow, fecal pellets), active transport (DVM), and physical transport (subduction of dissolved organic carbon). The total flux into the mesopelagic is approximately 10–15 Gt C/yr, of which only ~0.5–2 Gt C/yr reaches the deep ocean below 1000 m (the “sequestration depth” where carbon is isolated from the atmosphere for >100 years).

The transfer efficiency \(T_{\text{eff}}\) quantifies what fraction of the export flux at the base of the euphotic zone survives to the base of the mesopelagic:

\( T_{\text{eff}} = \frac{F(1000)}{F(z_{eu})} = \left(\frac{1000}{z_{eu}}\right)^{-b} \)

For \(z_{eu} = 100\) m and b = 0.86: \(T_{\text{eff}} \approx 14\%\)

The remaining 86% is remineralised in the mesopelagic, fuelling bacterial metabolism and sustaining the mesopelagic food web. This remineralisation is the primary source of the OMZ (see Section 2.3) and contributes to the nutricline that supplies nutrients to the surface upon upwelling.

The partitioning of remineralised carbon between dissolved inorganic carbon (DIC) and dissolved organic carbon (DOC) in the mesopelagic affects long-term sequestration. Refractory DOC (\(\tau > 1000\) yr) constitutes ~630 Gt C in the global ocean — one of the largest organic carbon pools on Earth, comparable to atmospheric CO\(_2\).

Bacterial Respiration Stoichiometry

Mesopelagic bacteria remineralise sinking organic matter following the Redfield stoichiometry, consuming O\(_2\) and releasing CO\(_2\), NO\(_3^-\), and PO\(_4^{3-}\):

\( (\text{CH}_2\text{O})_{106}(\text{NH}_3)_{16}(\text{H}_3\text{PO}_4) + 138\text{O}_2 \)

\( \to 106\text{CO}_2 + 16\text{HNO}_3 + \text{H}_3\text{PO}_4 + 122\text{H}_2\text{O} \)

The apparent oxygen utilisation (AOU) measures how much O\(_2\)has been consumed since a water parcel was last at the surface:\(\text{AOU} = [\text{O}_2]_{\text{sat}}(T,S) - [\text{O}_2]_{\text{observed}}\). High AOU indicates extensive remineralisation and old water age. In the most intense OMZs, AOU exceeds 250 \(\mu\)mol/kg, corresponding to water ages of centuries.

2.9 Ecological Significance & Conservation Challenges

The mesopelagic is increasingly targeted for industrial exploitation. Proposed mesopelagic fisheries could harvest millions of tonnes of lanternfish and krill for aquaculture feed and fish oil. However, the ecological consequences are poorly understood:

Carbon pump disruption: Removing mesopelagic fish would reduce active carbon transport, potentially weakening the biological pump by 0.5–1 Gt C/yr — equivalent to ~5% of current anthropogenic CO\(_2\) emissions.
Food web cascades: Mesopelagic organisms are critical prey for cetaceans, seabirds, tuna, and deep-diving predators. Overfishing could cascade through multiple trophic levels.
Slow recovery: Mesopelagic fish have lower growth rates and longer generation times than epipelagic species, making populations vulnerable to overexploitation.

The mesopelagic is sometimes called the “ocean's twilight zone” not only for its darkness but for our ignorance of its biology. Major research programs (NASA EXPORTS, JETZON) are working to quantify mesopelagic processes and their role in the global carbon cycle, directly connecting toclimate science andclimate-biodiversity research.

Simulation: DVM Energy Budget & OMZ Dynamics

Quantitative models of the diel vertical migration energy trade-off, OMZ formation and expansion under warming, bioluminescence spectra, and the Martin curve for particle flux:

Python

script.py70 lines

import numpy as np
import matplotlib.pyplot as plt

fig, axes = plt.subplots(2, 2, figsize=(16, 12))
fig.patch.set_facecolor('#0a0a1a')

# --- Panel 1: DVM Energy Budget ---
ax = axes[0, 0]
z_day = np.linspace(200, 800, 300)  # daytime depth (m)

# Parameters
I_max = 50  # max ingestion (ug C/ind/day)
C_s = 200   # surface food conc (ug C/L)
K_half = 100
t_night = 10  # hours of feeding
R_basal = 5  # basal respiration (ug C/ind/day)
v_swim = 50  # swimming speed (m/hr)
z_vis = 80   # visual predation scale depth
M0 = 20      # max predation mortality cost (ug C equiv/day)

# Feeding gain (constant, independent of day depth)
G = I_max * C_s / (C_s + K_half) * (t_night / 24)
G_arr = np.ones_like(z_day) * G

# Migration cost (increases with distance)
travel_time = 2 * z_day / v_swim  # hours round trip
R_active = 2.5 * R_basal
C_mig = R_active * travel_time / 24 + R_basal * (24 - travel_time) / 24

# Predation cost (decreases with depth)
M_pred = M0 * np.exp(-z_day / z_vis)

# Net energy
E_net = G_arr - C_mig - M_pred

ax.plot(z_day, G_arr, color='#34d399', linewidth=2, label='Feeding gain')
ax.plot(z_day, C_mig, color='#fbbf24', linewidth=2, label='Migration cost')
ax.plot(z_day, M_pred, color='#f87171', linewidth=2, label='Predation cost')
ax.plot(z_day, E_net, color='#60a5fa', linewidth=3, label='Net energy')
ax.axhline(0, color='white', linestyle=':', alpha=0.3)

# Optimal depth
z_opt = z_day[np.argmax(E_net)]
E_opt = np.max(E_net)
ax.axvline(z_opt, color='#818cf8', linestyle='--', alpha=0.5)
ax.plot(z_opt, E_opt, '*', color='#818cf8', markersize=15)
ax.annotate(f'Optimal depth\n{z_opt:.0f} m', xy=(z_opt, E_opt),
            xytext=(z_opt + 80, E_opt + 5), color='#c4b5fd', fontsize=10,
            arrowprops=dict(arrowstyle='->', color='#c4b5fd'))

ax.set_xlabel('Daytime residence depth (m)', color='white', fontsize=11)
ax.set_ylabel('Energy (\u03bcg C/ind/day)', color='white', fontsize=11)
ax.set_title('DVM Energy Budget Model', color='#60a5fa', fontsize=13, fontweight='bold')
ax.legend(fontsize=9, facecolor='#1a1a2e', edgecolor='#1e40af', labelcolor='white')
ax.set_facecolor('#0f0f2a')
ax.tick_params(colors='white')
ax.grid(True, alpha=0.2, color='#1e40af')

# --- Panel 2: OMZ profile + warming expansion ---
ax = axes[0, 1]
z = np.linspace(0, 2000, 500)

def o2_profile(z, T_surf=25, strat_factor=1.0):
    # O2 saturation decreases with temperature
    O2_sat = 8.0 - 0.08 * T_surf  # simplified Henry's law\n    # OMZ from biological consumption (Martin-like)\n    z0 = 100\n    consumption = np.where(z > z0,\n        6.0 * strat_factor * (z / z0)**(-0.86) * np.exp(-((z - 500)**2) / (2 * 250**2)) * 5,\n        0)\n    # Recovery at depth (thermohaline)\n    recovery = np.where(z > 600, (O2_sat * 0.7) * (1 - np.exp(-(z - 600) / 1200)), 0)\n    O2 = O2_sat - consumption + recovery\n    return np.clip(O2, 0.1, 10)\n\nscenarios = [\n    (20, 0.8, '#34d399', 'Pre-industrial (20\u00b0C)'),\n    (25, 1.0, '#60a5fa', 'Present (25\u00b0C)'),\n    (27, 1.3, '#fbbf24', 'RCP4.5 (+2\u00b0C)'),\n    (29, 1.6, '#f87171', 'RCP8.5 (+4\u00b0C)'),\n]\nfor T, sf, color, label in scenarios:\n    O2 = o2_profile(z, T, sf)\n    ax.plot(O2, z, color=color, linewidth=2, label=label)\n\nax.axvline(0.5, color='#ef4444', linestyle='--', alpha=0.5, linewidth=1.5)\nax.text(0.7, 50, 'Hypoxic
threshold', color='#fca5a5', fontsize=9)\nax.fill_betweenx(z, 0, 0.5, alpha=0.05, color='#ef4444')\n\nax.set_xlabel('Dissolved O\u2082 (mL/L)', color='white', fontsize=11)\nax.set_ylabel('Depth (m)', color='white', fontsize=11)\nax.set_title('OMZ Expansion Under Warming', color='#60a5fa', fontsize=13, fontweight='bold')\nax.legend(fontsize=9, facecolor='#1a1a2e', edgecolor='#1e40af', labelcolor='white')\nax.invert_yaxis()\nax.set_facecolor('#0f0f2a')\nax.tick_params(colors='white')\nax.grid(True, alpha=0.2, color='#1e40af')\n\n# --- Panel 3: Bioluminescence emission spectra ---\nax = axes[1, 0]\nlam = np.linspace(350, 700, 500)  # nm\n\ndef emission_spectrum(lam, peak, width):\n    return np.exp(-((lam - peak)**2) / (2 * width**2))\n\nspectra = [\n    (470, 25, '#60a5fa', 'Coelenterazine (470 nm)
Copepods, jellyfish'),\n    (490, 30, '#34d399', 'Bacterial (490 nm)
Vibrio fischeri'),\n    (440, 20, '#818cf8', 'Dinoflagellate (440 nm)'),\n    (510, 35, '#fbbf24', 'Firefly squid (510 nm)'),\n    (700, 40, '#f87171', 'Dragonfish far-red (700 nm)
Malacosteus niger'),\n]\nfor peak, width, color, label in spectra:\n    I_em = emission_spectrum(lam, peak, width)\n    ax.plot(lam, I_em, color=color, linewidth=2, label=label)\n    ax.fill_between(lam, 0, I_em, alpha=0.1, color=color)\n\n# Water transmission window\nax.axvspan(450, 510, alpha=0.08, color='#38bdf8')\nax.text(480, 0.95, 'Max water
transmission', color='#7dd3fc', fontsize=8, ha='center')\n\nax.set_xlabel('Wavelength (nm)', color='white', fontsize=11)\nax.set_ylabel('Relative emission', color='white', fontsize=11)\nax.set_title('Bioluminescence Emission Spectra', color='#60a5fa', fontsize=13, fontweight='bold')\nax.legend(fontsize=7, facecolor='#1a1a2e', edgecolor='#1e40af', labelcolor='white', loc='upper right')\nax.set_facecolor('#0f0f2a')\nax.tick_params(colors='white')\nax.grid(True, alpha=0.2, color='#1e40af')\n\n# --- Panel 4: Martin curve ---\nax = axes[1, 1]\nz = np.linspace(100, 6000, 500)\nz0 = 100  # reference depth\n\nfor b, color, label in [(0.60, '#34d399', 'b = 0.60 (oligotrophic, refractory)'),\n                         (0.86, '#60a5fa', 'b = 0.86 (Martin global mean)'),\n                         (1.20, '#fbbf24', 'b = 1.20 (productive, labile)'),\n                         (1.60, '#f87171', 'b = 1.60 (highly labile)')]:\n    F = (z / z0)**(-b)\n    ax.plot(F * 100, z, color=color, linewidth=2.5, label=label)\n\n# Key depths\nfor depth, label in [(200, 'Base epipelagic'), (1000, 'Base mesopelagic'),\n                      (4000, 'Abyssal')]:\n    ax.axhline(depth, color='#94a3b8', linestyle=':', alpha=0.3)\n    F_martin = (depth / z0)**(-0.86) * 100\n    ax.annotate(f'{label}: {F_martin:.1f}%', xy=(F_martin, depth),\n                xytext=(F_martin + 15, depth - 100), color='#94a3b8', fontsize=8,\n                arrowprops=dict(arrowstyle='->', color='#94a3b8', lw=0.8))\n\nax.set_xlabel('Fraction of export flux remaining (%)', color='white', fontsize=11)\nax.set_ylabel('Depth (m)', color='white', fontsize=11)\nax.set_title('Martin Curve: F(z) = F\u2080(z/z\u2080)\u207b\u1d47', color='#60a5fa', fontsize=13, fontweight='bold')\nax.legend(fontsize=8, facecolor='#1a1a2e', edgecolor='#1e40af', labelcolor='white')\nax.invert_yaxis()\nax.set_facecolor('#0f0f2a')\nax.tick_params(colors='white')\nax.grid(True, alpha=0.2, color='#1e40af')\nax.set_xlim(0, 105)\n\nplt.tight_layout()\nplt.savefig('output.png', dpi=150, bbox_inches='tight', facecolor='#0a0a1a')\nprint('DVM and OMZ dynamics simulation complete')

Click Run to execute the Python code

Code will be executed with Python 3 on the server

Simulation: TMAO Depth Profile & Particle Sinking

TMAO concentration as a function of depth, protein stability under pressure, and Stokes settling velocities for different particle types:

Python

script.py127 lines

import numpy as np
import matplotlib.pyplot as plt

fig, axes = plt.subplots(2, 2, figsize=(16, 12))
fig.patch.set_facecolor('#0a0a1a')

# --- Panel 1: TMAO concentration vs depth ---
ax = axes[0, 0]
z = np.linspace(0, 10000, 500)
TMAO = 40 + 0.39 * z / 10  # mmol/kg (Yancey fit: ~40 mmol/kg per 1000m)
# Better fit: TMAO ~ 40 + 39 * (z/1000) mmol/kg
TMAO = 40 + 39 * z / 1000

ax.plot(TMAO, z, color='#818cf8', linewidth=2.5)
ax.fill_betweenx(z, 0, TMAO, alpha=0.1, color='#818cf8')

# Fish depth limit
ax.axhline(8200, color='#f87171', linestyle='--', linewidth=1.5)
ax.text(200, 8400, 'Maximum fish depth (~8200 m)', color='#fca5a5', fontsize=10)
ax.text(200, 8700, 'TMAO toxicity limit', color='#fca5a5', fontsize=9)

# Zone boundaries
for d, lab in [(200, 'Epipelagic'), (1000, 'Mesopelagic'),
               (4000, 'Bathypelagic'), (6000, 'Abyssal')]:
    ax.axhline(d, color='#38bdf8', linestyle=':', alpha=0.2)
    ax.text(10, d + 100, lab, color='#7dd3fc', fontsize=8)

ax.set_xlabel('[TMAO] (mmol/kg)', color='white', fontsize=11)
ax.set_ylabel('Depth (m)', color='white', fontsize=11)
ax.set_title('TMAO Concentration vs Depth', color='#60a5fa', fontsize=13, fontweight='bold')
ax.invert_yaxis()
ax.set_facecolor('#0f0f2a')
ax.tick_params(colors='white')
ax.grid(True, alpha=0.2, color='#1e40af')

# --- Panel 2: Protein stability under pressure ---
ax = axes[0, 1]
P = np.linspace(1, 1100, 500)  # atm
R_gas = 8.314  # J/mol/K
T = 283  # K (10 degC, typical mesopelagic)

# K(P)/K(1) = exp(-dV * (P-1) * 101325 / (R*T))
# dV in m^3/mol (negative for unfolding)
for dV, tmao, color, label in [
    (-80e-6, 0, '#f87171', 'No TMAO (\u0394V=-80 mL/mol)'),
    (-80e-6, 100, '#fbbf24', 'TMAO 100 mmol/kg'),
    (-80e-6, 250, '#34d399', 'TMAO 250 mmol/kg'),
    (-50e-6, 0, '#818cf8', '\u0394V=-50 mL/mol (piezophile)')]:
    # Unfold equilibrium constant ratio
    K_ratio = np.exp(-dV * (P - 1) * 101325 / (R_gas * T))
    # TMAO stabilisation: reduces effective dV
    tmao_effect = np.exp(-tmao * 0.0015 * np.ones_like(P))  # simplified
    K_eff = K_ratio * tmao_effect
    # Fraction folded = 1/(1 + K)
    f_folded = 1.0 / (1.0 + K_eff * 0.001)  # K at 1 atm ~ 0.001 (mostly folded)
    ax.plot(P, f_folded * 100, color=color, linewidth=2, label=label)

ax.axhline(50, color='white', linestyle=':', alpha=0.2)
ax.text(50, 48, '50% folded', color='#94a3b8', fontsize=9)
ax.set_xlabel('Pressure (atm)', color='white', fontsize=11)
ax.set_ylabel('Fraction folded (%)', color='white', fontsize=11)
ax.set_title('Protein Stability vs Pressure', color='#60a5fa', fontsize=13, fontweight='bold')
ax.legend(fontsize=8, facecolor='#1a1a2e', edgecolor='#1e40af', labelcolor='white')
ax.set_facecolor('#0f0f2a')
ax.tick_params(colors='white')
ax.grid(True, alpha=0.2, color='#1e40af')

# --- Panel 3: Stokes settling velocity ---
ax = axes[1, 0]
r = np.linspace(0.01e-3, 5e-3, 300)  # radius in m (0.01 to 5 mm)
rho_w = 1025  # kg/m^3
mu = 1.08e-3  # Pa.s (seawater at 10 degC)
g = 9.81

particles = [
    (50, '#34d399', 'Phytoplankton cells (\u0394\u03c1=50)'),
    (100, '#60a5fa', 'Marine snow (\u0394\u03c1=100)'),
    (200, '#fbbf24', 'Copepod fecal pellets (\u0394\u03c1=200)'),
    (500, '#f87171', 'Diatom aggregates (\u0394\u03c1=500)'),
    (1000, '#a78bfa', 'Salp fecal pellets (\u0394\u03c1=1000)'),
]
for drho, color, label in particles:
    vs = 2 * r**2 * drho * g / (9 * mu)  # m/s
    vs_m_day = vs * 86400  # m/day
    ax.plot(r * 1e3, vs_m_day, color=color, linewidth=2, label=label)

ax.set_xlabel('Particle radius (mm)', color='white', fontsize=11)
ax.set_ylabel('Settling velocity (m/day)', color='white', fontsize=11)
ax.set_title('Stokes Settling Velocity', color='#60a5fa', fontsize=13, fontweight='bold')
ax.legend(fontsize=8, facecolor='#1a1a2e', edgecolor='#1e40af', labelcolor='white')
ax.set_facecolor('#0f0f2a')
ax.tick_params(colors='white')
ax.grid(True, alpha=0.2, color='#1e40af')
ax.set_yscale('log')
ax.set_ylim(0.1, 1e5)

# --- Panel 4: Carbon flux reaching depth (Martin curve integration) ---
ax = axes[1, 1]
z0 = 100  # reference depth
z_range = np.linspace(100, 6000, 500)
F0 = 10  # Gt C/yr export at 100m

# Martin curve for different b values
for b, color, label in [(0.6, '#34d399', 'b=0.60'), (0.86, '#60a5fa', 'b=0.86'),
                         (1.2, '#fbbf24', 'b=1.20'), (1.6, '#f87171', 'b=1.60')]:
    F = F0 * (z_range / z0)**(-b)
    # Cumulative remineralised = F0 - F(z) = total carbon consumed above z
    remineralised = F0 - F
    ax.plot(z_range, F, color=color, linewidth=2, label=f'Flux (b={b})')
    ax.fill_between(z_range, 0, F, alpha=0.05, color=color)

# Sequestration threshold (~1000 yr residence time)
ax.axhline(F0 * (1000/z0)**(-0.86), color='#818cf8', linestyle='--', alpha=0.5)
ax.text(4500, F0 * (1000/z0)**(-0.86) + 0.1,
        f'Flux at 1000 m: {F0 * (1000/z0)**(-0.86):.1f} Gt C/yr', color='#c4b5fd', fontsize=9)

ax.set_xlabel('Depth (m)', color='white', fontsize=11)
ax.set_ylabel('Organic C flux (Gt C/yr)', color='white', fontsize=11)
ax.set_title('Carbon Flux vs Depth (Biological Pump)', color='#60a5fa', fontsize=13, fontweight='bold')
ax.legend(fontsize=9, facecolor='#1a1a2e', edgecolor='#1e40af', labelcolor='white')
ax.set_facecolor('#0f0f2a')
ax.tick_params(colors='white')
ax.grid(True, alpha=0.2, color='#1e40af')

plt.tight_layout()
plt.savefig('output.png', dpi=150, bbox_inches='tight', facecolor='#0a0a1a')
print('TMAO and particle sinking simulation complete')

Click Run to execute the Python code

Code will be executed with Python 3 on the server

References

Martin, J. H., Knauer, G. A., Karl, D. M., & Broenkow, W. W. (1987). VERTEX: carbon cycling in the northeast Pacific. Deep-Sea Research, 34(2), 267–285.
Schmidtko, S., Stramma, L., & Visbeck, M. (2017). Decline in global oceanic oxygen content during the past five decades. Nature, 542, 335–339.
Yancey, P. H., Gerringer, M. E., Drazen, J. C., Rowden, A. A., & Jamieson, A. (2014). Marine fish may be biochemically constrained from inhabiting the deepest ocean depths. PNAS, 111(12), 4461–4465.
Haddock, S. H. D., Moline, M. A., & Case, J. F. (2010). Bioluminescence in the sea. Annual Review of Marine Science, 2, 443–493.
Steinberg, D. K. & Landry, M. R. (2017). Zooplankton and the ocean carbon cycle. Annual Review of Marine Science, 9, 413–444.
Breitburg, D., Levin, L. A., Oschlies, A., et al. (2018). Declining oxygen in the global ocean and coastal waters. Science, 359(6371), eaam7240.
Sutton, T. T. (2013). Vertical ecology of the pelagic ocean: classical patterns and new perspectives. Journal of Fish Biology, 83(6), 1508–1527.
Widder, E. A. (2010). Bioluminescence in the ocean: origins of biological, chemical, and ecological diversity. Science, 328(5979), 704–708.