Chapter 34: Case Study — A Mission to Mars

"Mars has become a kind of mythic arena onto which we have projected our earthly hopes and fears." — Carl Sagan, Cosmos

Overview

This is the chapter the whole book has been walking toward. For thirty-three chapters we have built tools — the rocket equation, orbital energy, the interplanetary Hohmann transfer, re-entry heating, life support, the discipline of mission design — each one sharpened on a smaller problem. Now we point all of them at the hardest destination a human being has any near-term hope of reaching, and we watch them converge. A mission to Mars is not a new subject. It is every subject in this book at once, forced to agree with itself.

Here is the shape of the trip, and every piece of it is something you already know how to compute. You wait for a launch window that opens only every twenty-six months (Chapter 11). You spend about $3.6\ \text{km/s}$ of delta-v to leave low Earth orbit on a trans-Mars injection, riding the same Hohmann ellipse Walter Hohmann drew on paper in 1925. You coast for the better part of a year through a radiation bath that will spend a career's dose limit (Chapter 28), correcting your aim with burns of a few meters per second. Then, in about seven minutes, you convert $5.6\ \text{km/s}$ of arrival speed into a soft touchdown on a world whose atmosphere is thick enough to incinerate you but too thin to stop you. And then — the part that makes Mars categorically harder than the Moon — you face the return, which means you must launch a rocket off another planet, which means, if you are sensible, you make its propellant out of the Martian air before you ever leave Earth.

Two long threads of this book meet here. The Hohmann-to-Mars trajectory that has run since Chapter 6 is the mission's skeleton; Starship, the fully reusable methane vehicle we have followed as a systems-engineering case study, is the most concrete attempt to fly that skeleton with human beings aboard. And all six of the book's themes appear at once, because a Mars mission is exactly the place they collide: the tyranny of the rocket equation sets the vehicle's size, mass is the enemy at every step, the environment is unforgiving in ways that kill in minutes, reusability may be the only thing that makes the economics close, orbital mechanics is the beautiful clockwork that sets the whole schedule — and history, from von Braun's paper study to a rover breathing oxygen from Martian air, tells us why each choice is the way it is.

In this chapter, you will learn to:

• Read the 26-month synodic drumbeat as the master clock of Mars exploration, and lay out the ~900-day conjunction-class mission it forces.
• Size the trans-Mars injection, and see why a crewed vehicle must be refueled in orbit before it can leave.
• Follow the cruise — its navigation, its radiation, its 3-to-22-minute conversations with Earth.
• Dissect the "seven minutes of terror" and see, in one calculation, why Mars is the graveyard of landers.
• Understand in-situ resource utilization and the Mars ascent vehicle — the answer to "you need a rocket on Mars."
• Compare the two serious human Mars architectures and assemble the round-trip delta-v budget yourself.

Learning Paths

🚀 Space Enthusiast: Read 34.1 (why you can only go every two years) and 34.4 (the seven minutes of terror — the most dramatic seven minutes in engineering), then enjoy 34.5 (making rocket fuel from thin air) and 34.6 (the two great plans for getting people there). You can take the delta-v numbers in 34.2 as given.

📐 Engineering Student: Read all of it; this is the capstone application of Parts I–V. Reproduce the trans-Mars injection of 34.2 and the Mars terminal-velocity calculation of 34.4 by hand, and do the full round-trip budget in the Mission Design Checkpoint. The ⭐⭐/⭐⭐⭐ exercises assemble a complete Mars MDR.

🎮 KSP Player: You have flown this — to Duna, KSP's Mars. Section 34.1 is the transfer window you wait for; 34.4 is why your Duna landers need so much more than a parachute (the mod-authors got the thin air right); 34.5 is the ISRU converter you bolt on to refuel for home. The real solar system is your save file with harsher numbers.

🛰️ Industry Prep: This chapter is a mission concept study in miniature. The conjunction-vs-opposition trade (34.1), the refuel-in-LEO architecture (34.2), the EDL mass wall (34.4), and the DRA-5.0-vs-Starship comparison (34.6) are exactly the trades a real Mars study argues about. The Checkpoint is your Track-C Mission Design Review, worked end to end.

34.1 The launch window and the 26-month synodic cycle

Everything about a Mars mission begins with a calendar you do not control. In Chapter 11 we found the number that rules that calendar: the synodic period of Earth and Mars, the time between successive identical alignments of the two planets, and therefore between successive launch opportunities. We compute it from the two orbital periods — Earth's $T_1 = 365.25\ \text{days}$, Mars's $T_2 = 686.98\ \text{days}$ — through the difference of their angular rates:

$$ \frac{1}{T_{\text{syn}}} = \left|\frac{1}{T_1} - \frac{1}{T_2}\right| = \left|\frac{1}{365.25} - \frac{1}{686.98}\right| = 1.282\times10^{-3}\ \text{day}^{-1}, \qquad T_{\text{syn}} = 780\ \text{days} \approx 26\ \text{months}. $$

That $780$-day beat is the master clock of Mars exploration. It is why the historical record of Mars launches clusters in tight windows roughly two years apart — 2018, 2020, 2022, 2024, 2026, 2028, and on — and why every serious plan is, at bottom, a fight to have hardware ready when a window opens. Miss one and you do not lose weeks; you lose a cycle.

🔗 Connection: the schedule tyranny is orbital mechanics, not bureaucracy. The 26-month drumbeat is not a policy or a budget cycle — it is a consequence of $1/T_{\text{syn}} = |1/T_1 - 1/T_2|$, as unnegotiable as gravity. A spacecraft that is a month late for its window cannot simply leave a month late; the cheap Hohmann geometry is gone, and, as Chapter 11 showed, buying your way to Mars off-window can cost more delta-v than the entire nominal mission. Patience is a propellant. The single most expensive words in Mars program management are "we'll catch the next launch period," because the next one is more than two years away.

The trip is a round trip, and the return has its own window

For a robotic orbiter the window is a one-way concern: launch, cruise, arrive, done. A crewed mission must come home, and here the synodic cycle turns cruel a second time. You arrive at Mars, and the geometry that would let you fly a cheap minimum-energy transfer back to Earth is not there yet — Earth and Mars must re-phase, and that takes time. You are, in the most literal sense, stranded until the sky lines up.

This forces a fundamental architectural choice, and it has a name.

• A conjunction-class (long-stay) mission takes the cheap minimum-energy Hohmann transfer in both directions and simply waits out the long surface stay in between. The bill is roughly: an $\approx 259$-day outbound cruise (the number we derived in Chapter 11), a surface stay of about $500$ days while Earth and Mars re-phase, and an $\approx 259$-day return — a total mission of roughly $900$ days, about two and a half years, most of it on the Martian surface. Every leg is minimum-energy, so this is the cheapest way to send people to Mars.
• An opposition-class (short-stay) mission refuses to wait. It cuts the surface stay to a month or two, but to do so it must fly at least one leg on a fast, steeply non-Hohmann trajectory — usually swinging past Venus for a gravity assist — at a large penalty in delta-v and, worse, a longer total time exposed to deep-space radiation, because the crew spends far more of the mission in transit. It trades surface time for a great deal of energy, and gets a more dangerous mission in the bargain.

Worked Example: the conjunction-class timeline (Tier 2).

Strategy first. Add the three legs the synodic geometry forces: outbound cruise, surface wait, return cruise. We take the outbound and return as the Chapter 11 Hohmann cruise, and the surface stay as the standard re-phasing wait.

$$ > t_{\text{mission}} \approx \underbrace{259\ \text{days}}_{\text{outbound}} + > \underbrace{\sim 500\ \text{days}}_{\text{surface (re-phasing wait)}} + > \underbrace{259\ \text{days}}_{\text{return}} \approx 1{,}018\ \text{days} \approx 2.8\ \text{years}. > $$

Sanity check. NASA's reference human Mars studies converge on almost exactly this: outbound and return transits of roughly six to nine months each, bracketing a surface stay of about a year and a half, for a total near $900$ days. The $\sim 500$-day surface stay is not a choice — it is how long the crew must wait for the return window to open. This single fact reshapes everything downstream: it sets the consumables the mission must stockpile or manufacture, the radiation dose the crew accumulates, the power and habitat the surface base must provide, and the reliability every system must hold for nearly three years with no repair depot. A weekend trip this is not.

📜 From History: the plan is older than the rocket. In 1948 Wernher von Braun wrote Das Marsprojekt, a startlingly detailed engineering study of a crewed Mars expedition — a flotilla of ships assembled in Earth orbit, riding minimum-energy transfers to Mars and back. He got the architecture right decades before any of it was buildable: assemble in orbit because the rocket equation forbids launching the whole thing at once, and fly the cheap Hohmann because there is no affordable alternative. Like Tsiolkovsky's equation and Hohmann's transfer before it, the Mars mission existed as correct physics long before the hardware — the recurring pattern of this whole book. What has changed since von Braun is not the trajectory (it is the same ellipse) but the vehicle, and that is the story of the rest of this chapter.

🔄 Check Your Understanding 1. Why can't a crewed Mars mission simply turn around and come home a week after arriving? 2. A conjunction-class mission is described as "cheaper but longer" than an opposition-class one. Cheaper and longer in what, exactly?

Answers

1. Because the cheap minimum-energy return transfer requires Earth and Mars to be correctly re-phased, and that alignment does not recur until roughly $500$ days after arrival. Leaving earlier would demand a fast, high-energy trajectory the mission cannot afford — you are held on the surface by orbital mechanics, not by choice. 2. Cheaper in delta-v (both legs are minimum-energy Hohmann transfers) but longer in total mission time (~900 days, most of it on the surface). The opposition-class mission is the reverse: a short surface stay bought with much more delta-v and more time in deep-space radiation.

34.2 Trans-Mars injection

The mission's first burn is the one that commits it to another planet. The spacecraft is in a low Earth parking orbit, moving at about $7.7\ \text{km/s}$; it must be flung onto the heliocentric Hohmann ellipse that carries it to Mars. That flinging burn has a name we now make formal — a name Chapter 11 used when it computed the number, and one this chapter owns.

Definition (trans-Mars injection, TMI). The trans-Mars injection is the propulsive maneuver that raises a spacecraft from a low Earth parking orbit onto an Earth-departure hyperbola whose hyperbolic excess velocity $v_\infty$ places it on the heliocentric transfer to Mars. It is the first leg of the patched-conic Mars trajectory of Chapter 11 — the burn that turns an orbit around Earth into a trajectory toward Mars. Its size is set by the departure characteristic energy $C_3 = v_\infty^2$ the transfer demands.

We do not need to re-derive the number, because Chapter 11 already did, from first principles. For the minimum-energy Earth-to-Mars Hohmann, the departure requires a hyperbolic excess of $v_{\infty,\oplus} = 2.95\ \text{km/s}$, a characteristic energy of

$$ C_3 = v_{\infty,\oplus}^2 = (2.95)^2 \approx 8.7\ \text{km}^2/\text{s}^2, $$

and, burning from a $300\ \text{km}$ parking orbit where the circular speed is $7.73\ \text{km/s}$, a trans-Mars injection burn of

$$ \Delta v_{\text{TMI}} = \sqrt{v_{\infty,\oplus}^2 + \frac{2\mu_\oplus}{r_{\text{park}}}} - \sqrt{\frac{\mu_\oplus}{r_{\text{park}}}} = 11.32 - 7.73 = 3.59\ \text{km/s} \approx 3.6\ \text{km/s}. $$

That $\approx 3.6\ \text{km/s}$ is the same figure the delta-v map of Chapter 3 has quoted since the start of the book. It is a small burn by the standard of reaching orbit in the first place — recall that getting from Earth's surface to LEO cost about $9.4\ \text{km/s}$. Once you are in low Earth orbit, you are, in Heinlein's phrase, halfway to anywhere; Mars is only another $3.6\ \text{km/s}$ down the road. The reason interplanetary flight is hard is not this burn. It is everything the burn has to carry.

Why a crewed vehicle must refuel before it can leave

Here the tyranny of the rocket equation reasserts itself, and it does so in a way that shapes the entire Starship architecture. A $3.6\ \text{km/s}$ delta-v is cheap in velocity but not in mass, because the rocket equation converts velocity into a mass ratio, and a mass ratio multiplies whatever you are pushing. The heavier the crewed stack, the more propellant that modest burn devours.

Worked Example: the trans-Mars injection propellant for a Starship-class vehicle.

Strategy first. The stack post-burn (its "dry" mass for this maneuver) is the empty ship plus its payload. Use the rocket equation to find the mass ratio for $\Delta v_{\text{TMI}} = 3.6\ \text{km/s}$ with a methalox engine, then get the propellant as $m_p = m_f(m_0/m_f - 1)$.

Take a Starship stack in LEO with a dry mass of about $120\ \text{t}$ and $100\ \text{t}$ of payload, so the post-injection mass is $m_f = 220\ \text{t}$ (aspirational figures — Tier 3). Its Raptor engines burn liquid methane and oxygen at a vacuum specific impulse of about $380\ \text{s}$ (Chapter 17; Appendix H), so $$ > v_e = I_{sp}\,g_0 = 380 \times 9.80665 = 3{,}727\ \text{m/s} = 3.73\ \text{km/s}. > $$ The mass ratio for the injection is $$ > \frac{m_0}{m_f} = e^{\Delta v_{\text{TMI}}/v_e} = e^{3.6/3.73} = e^{0.966} = 2.63, > $$ so the propellant needed in low Earth orbit is $$ > m_p = m_f\left(\frac{m_0}{m_f} - 1\right) = 220\ \text{t} \times (2.63 - 1) = 220 \times 1.63 \approx 358\ \text{t}. > $$ Sanity check and the punchline. A Starship's tanks hold on the order of $1{,}200\ \text{t}$ of propellant, but a ship that has just fought its way to orbit arrives there nearly empty — it spent that propellant getting off Earth. It cannot carry both a full LEO-injection load and a heavy payload up in one go, because the rocket equation already ate the propellant on the way up. So the $\approx 358\ \text{t}$ it needs for trans-Mars injection has to be delivered to it in orbit by a series of tanker flights — perhaps three to five refuelings, each ferrying up $\sim 100\ \text{t}$ of propellant. Orbital refueling is not a convenience in this architecture; the rocket equation requires it.

import math
G0 = 9.80665  # standard gravity, m/s^2

def injection_prop(dv_kms, isp_s, m_post_t):
    """Propellant (t) for a burn of dv (km/s) with an engine of specific impulse isp (s),
    where the post-burn mass (empty vehicle + payload) is m_post_t (tonnes)."""
    ve = isp_s * G0                             # effective exhaust velocity, m/s
    mass_ratio = math.exp(dv_kms * 1000 / ve)   # rocket equation (Chapter 3)
    return m_post_t * (mass_ratio - 1)

# A Starship-class stack in LEO: ~120 t dry + ~100 t payload = 220 t; methalox Raptor, Isp ~380 s (vac).
tmi = injection_prop(3.6, 380, 220)
print(f"TMI propellant needed in LEO: {tmi:.0f} t")
# Expected output:
# TMI propellant needed in LEO: 358 t

🔗 Connection: reusability makes the arithmetic close. Launching $\sim 358\ \text{t}$ of propellant to orbit in tanker-loads sounds absurd until you remember reusability is changing everything (theme 5). If each tanker is thrown away, refueling a single Mars ship costs a fleet of expendable rockets and the mission is a non-starter. If each tanker flies home, refuels, and launches again within days — the Falcon-9-and-beyond model of Chapter 22, climaxed for Starship in Chapter 38 — then those five flights are five turnarounds of one vehicle, not five rockets built and discarded. The refuel-in-orbit architecture is only sane because the launch vehicle is reusable. This is the clearest single place where the reusability theme is not a nicety but a precondition for going to Mars at all.

⚠️ Common Misconception: "Just build one giant rocket that flies straight to Mars from the ground." The rocket equation forbids it, for the same reason it forbids single-stage-to-orbit with useful payload (Chapter 3). To reach Mars in one launch you would have to stack the Earth-to-orbit delta-v ($\sim 9.4$) and the trans-Mars injection ($\sim 3.6$) into a single mass ratio — and $e^{13/v_e}$ with a chemical $v_e$ is a mass ratio in the dozens, leaving essentially nothing for payload. Splitting the job — reach orbit, then refuel, then inject — is staging by another name: you refill the tank at the top of the hill instead of dragging all the fuel up from the bottom. Mass is the enemy, and the enemy compounds exponentially; the only way to beat it is to not carry what you do not yet need.

🔄 Check Your Understanding 1. The trans-Mars injection is only $\sim 3.6\ \text{km/s}$, far less than the $\sim 9.4\ \text{km/s}$ to reach orbit. Why, then, does it demand hundreds of tonnes of propellant for a large crewed ship? 2. In one sentence, why is orbital refueling required rather than merely convenient for a Starship Mars mission?

Answers

1. Because the propellant a burn needs is not set by the delta-v alone but by the delta-v times the mass being pushed, through the rocket equation's mass ratio ($e^{\Delta v/v_e}$). A modest delta-v applied to a $220\ \text{t}$ stack still means $\sim 358\ \text{t}$ of propellant — the mass ratio of $2.63$ multiplies everything. 2. Because a ship that reaches LEO has spent nearly all its propellant getting there, so it cannot also carry a full injection load and payload in a single launch — the injection propellant must be delivered to it in orbit.

34.3 The cruise

The injection burn ends, the engines fall silent, and the spacecraft begins the longest, quietest, and in some ways most dangerous phase of the mission: about eight and a half months of coasting around the Sun, engine off, on the transfer ellipse we built in Chapter 11. Nothing dramatic happens. That is precisely the problem — for the better part of a year, a crew and its machines must simply keep working, alone, in an environment engineered to wear them both down.

Navigation: designed with conics, flown with integration

Patched conics designed this trajectory, but patched conics will not fly it. Recall the caution of Chapter 11: the neat two-body ellipse is a design approximation, good to a percent or two, and a percent of a hundred-million- kilometer trip is a miss of a million kilometers. To actually arrive, navigators throw the conics away and integrate the real equations of motion — every planet's pull, the Sun's radiation pressure, the small errors in the injection burn — and then trim the aim with a handful of small trajectory-correction maneuvers (TCMs), each a few meters per second, spread across the cruise. A total TCM budget of about $0.1\ \text{km/s}$ is typical. The navigation itself is done from Earth by the tools of Chapter 26: ranging and Doppler tracking through the Deep Space Network, refined by delta-DOR, locate the spacecraft to kilometers across hundreds of millions of them.

📜 From History: the mission lost to a mismatched unit. The single most famous navigation failure in planetary spaceflight happened on exactly this cruise. In 1999 the Mars Climate Orbiter was lost at arrival because one team's software produced a thruster impulse in pound-force-seconds while the navigation software expected newton-seconds; the tiny, unmodeled bias accumulated over the months of cruise until the orbiter arrived too low and broke apart in the atmosphere. No subsystem failed — the interface between two correct pieces failed, the exact failure mode systems engineering exists to prevent (Chapter 29). It is why "units on every number" is a rule of this whole book, and why a Mars navigator trusts nothing that is not sanity-checked twice. On a hundred-million-kilometer coast, a small persistent error is not small.

Radiation: spending a career in eight months

The cruise is where the crew accumulates most of its radiation dose, and the numbers are sobering. From Chapter 28, the deep-space galactic-cosmic-ray dose rate — measured for real by the detector aboard the Curiosity rover on its own cruise to Mars — is about $1.8\ \text{mSv}$ per day. Over the outbound leg:

$$ D_{\text{cruise}} \approx 1.8\ \frac{\text{mSv}}{\text{day}} \times 259\ \text{days} \approx 466\ \text{mSv} \approx 0.47\ \text{Sv}. $$

That is the one-way transit dose, before a single day on the surface. A round trip's two cruises alone deliver nearly $0.9\ \text{Sv}$, against a NASA career limit on the order of $0.6\ \text{Sv}$ (set to hold the lifetime fatal-cancer risk near a few percent). The cruise, in other words, can spend a whole career's radiation allowance — which is why the mission carries a storm shelter (Chapter 28) of water, food, and waste to ride out the acute dose of a solar particle event, and why flying faster — cutting the exposure time — is one of the strongest arguments for the high-thrust nuclear propulsion we meet in the architectures of 34.6.

🔧 Engineering Reality: the cruise is a spacecraft on life support, in every sense. "Coasting" hides how much must keep running. For eight and a half months the vehicle must generate power far from a weaker Sun (Chapter 25), reject the crew's and electronics' heat to space (Chapter 24), recycle air and water through a closed-loop life-support system (Chapter 28) — because at $\sim 5\ \text{kg}$ of consumables per person-day, an open-loop crew of four would need over eighteen tonnes for the round trip — and hold attitude and comm lock the whole way. A dead subsystem in deep space cannot be swapped from a cargo flight; there is no cargo flight. Space is an unforgiving environment, and the cruise is that theme stretched across three-quarters of a year.

The conversation delay

And the crew is, in a way the ISS crew never is, alone. As Chapter 28 noted, the one-way light time from Earth to Mars runs from about $3\ \text{minutes}$ near closest approach to about $22\ \text{minutes}$ when the planets are on opposite sides of the Sun. A question and its answer can be three-quarters of an hour apart. Real-time help from Earth is physically impossible, which is why deep-space missions are built for autonomy (Chapter 27): the crew and the flight software must be able to diagnose and act on their own. That principle is about to be tested to its absolute limit, because the next seven minutes happen faster than a signal can even reach the people who built the spacecraft.

🔄 Check Your Understanding 1. Why does the mission plan for trajectory-correction maneuvers at all, if Chapter 11 already computed the transfer exactly? 2. A round-trip Mars mission's two cruises deliver roughly $0.9\ \text{Sv}$. Why is that number alarming, and name one architectural lever that reduces it.

Answers

1. Because Chapter 11's transfer is a patched-conic design approximation and the real trajectory, integrated with all perturbations plus small injection errors, drifts from it; TCMs of a few m/s trim the aim so the spacecraft actually hits Mars's capture corridor. 2. It is comparable to a whole career radiation limit ($\sim 0.6\ \text{Sv}$), spent before any surface time — so a single mission can max out a crew member's allowed lifetime dose. The strongest lever is a faster transit (less time exposed), which favors high-thrust propulsion such as nuclear thermal; a storm shelter additionally guards against the acute dose of a solar particle event.

34.4 Mars arrival and EDL — the "seven minutes of terror"

The spacecraft crosses into Mars's sphere of influence with the hyperbolic excess velocity we computed in Chapter 11, $v_{\infty,\text{Mars}} = 2.65\ \text{km/s}$, and falls inward. By the top of the atmosphere, about $125\ \text{km}$ up, it has traded that excess for the depth of Mars's gravity well and is moving at

$$ v_{\text{entry}} = \sqrt{v_{\infty,\text{Mars}}^2 + \frac{2\mu_{\text{Mars}}}{r_{\text{atm}}}} = \sqrt{(2.65)^2 + \frac{2(4.283\times10^4)}{3{,}490}} \approx \sqrt{7.0 + 24.5} = 5.6\ \text{km/s} $$

(the same $\approx 5.6\ \text{km/s}$ Chapter 11 warned you to compute relative to Mars, not the Sun). Now the spacecraft must shed all of it — five and a half kilometers per second of speed — and set down gently, in about seven minutes, on its own, with Earth watching a recording of events that have already finished. NASA's engineers named this the "seven minutes of terror," and the name is not marketing.

Definition (entry, descent, and landing, EDL). Entry, descent, and landing (EDL) is the sequence of events that takes a spacecraft from the top of a planet's atmosphere to a stationary, intact landing on the surface: entry (hypersonic deceleration behind a heat shield, dumping most of the kinetic energy as heat), descent (supersonic and subsonic slowing, by parachute and/or propulsion, with the heat shield jettisoned), and landing (the final touchdown, by legs, airbags, a sky crane, or a landing burn). At Mars it is compressed into roughly seven minutes and, because of the communication delay, must be fully autonomous.

Why is EDL the part of a Mars mission that engineers lose sleep over? Because Mars's atmosphere is a trap built to be the worst of both worlds.

🚪 Threshold Concept: Mars's atmosphere is thick enough to kill you and too thin to save you. At the surface, Mars's air is about $0.6\%$ as dense as Earth's — a near-vacuum by our standards. Yet a spacecraft arrives moving so fast that even this whisper of gas piles up into a hypersonic shock that heats the vehicle like a re-entering Earth capsule (Chapter 7) — so you must carry a heat shield, paying its full mass. But that same thin air cannot stop you: there is not enough of it to slow a heavy vehicle to a soft landing with parachutes, the way Earth's dense atmosphere does for a returning capsule. Earth's atmosphere is thick enough to both burn and brake; the vacuum of the Moon does neither, so you just fire an engine all the way down; but Mars sits in the cruel middle — it gives you all the heating problem and only part of the braking solution. Once you truly grasp that Mars is "too much atmosphere to ignore, too little to rely on," every strange thing about Mars landers — the parachutes that deploy at supersonic speed, the sky cranes, the retropropulsion — stops being exotic and becomes the only thing that could possibly work.

Let us prove the cruel middle with a calculation — the one number that explains why no spacecraft has ever simply parachuted onto Mars.

Worked Example: why a parachute cannot land you on Mars.

Strategy first. A parachute's steady-fall (terminal) speed is where drag balances weight: $v_{\text{term}} = \sqrt{2mg / (\rho\, C_d A)}$. Compute it for the same capsule under the same parachute in Earth's air and in Mars's, and compare. The only things that change are the surface gravity $g$ and the atmospheric density $\rho$.

Take a $3{,}000\ \text{kg}$ capsule under a large $20\ \text{m}$-diameter parachute (area $A = \pi(10)^2 = 314\ \text{m}^2$), drag coefficient $C_d \approx 0.5$. Earth: $g = 9.81\ \text{m/s}^2$, $\rho = 1.225\ \text{kg/m}^3$. Mars: $g = 3.71\ \text{m/s}^2$, surface $\rho \approx 0.020\ \text{kg/m}^3$ (Tier 2). Then $$ > v_{\text{Earth}} = \sqrt{\frac{2(3000)(9.81)}{(1.225)(0.5)(314)}} = \sqrt{\frac{58{,}860}{192.4}} > = \sqrt{306} = 17.5\ \text{m/s}, > $$ $$ > v_{\text{Mars}} = \sqrt{\frac{2(3000)(3.71)}{(0.020)(0.5)(314)}} = \sqrt{\frac{22{,}260}{3.14}} > = \sqrt{7{,}089} = 84\ \text{m/s}. > $$ Sanity check and meaning. On Earth the parachute settles the capsule to about $17\ \text{m/s}$ — fast, but survivable with crushable legs or airbags. On Mars the identical parachute leaves it screaming in at about $84\ \text{m/s}$ — nearly $190\ \text{mph}$, straight into the ground. And this is the best case, computed at the densest air right at the surface; higher up, where the chute actually deploys, the density is lower still and the terminal speed even worse. The ratio is roughly $\sqrt{\rho_{\text{Earth}}/ \rho_{\text{Mars}}} \approx \sqrt{60} \approx 8$ times faster, softened a little by Mars's weaker gravity. A parachute alone can never land a spacecraft on Mars. Something else must finish the job — and that "something else" is the whole drama of Mars EDL.

import math

def terminal_velocity(m, g, rho, Cd, A):
    """Steady-fall speed where drag balances weight: v = sqrt(2 m g / (rho Cd A))."""
    return math.sqrt(2 * m * g / (rho * Cd * A))

A = math.pi * 10**2                         # 20 m parachute -> ~314 m^2
earth = terminal_velocity(3000, 9.81, 1.225, 0.5, A)   # same capsule, same chute
mars  = terminal_velocity(3000, 3.71, 0.020, 0.5, A)   # ...on two different worlds
print(f"Earth terminal speed under chute: {earth:.0f} m/s")
print(f"Mars  terminal speed under chute: {mars:.0f} m/s")
# Expected output:
# Earth terminal speed under chute: 17 m/s
# Mars  terminal speed under chute: 84 m/s

The EDL toolbox, and the mass wall

Because no single method suffices, every Mars lander chains several, in a frantic seven-minute relay:

1. Hypersonic entry behind a heat shield. The vehicle enters at $\sim 5.6\ \text{km/s}$ blunt-side forward, and an ablative or rigid heat shield turns the bulk of that kinetic energy into heat and a glowing wake — the physics of Chapter 7, applied to a thinner atmosphere. This single phase removes most of the speed, bleeding the vehicle down from Mach 25 or so to a couple of times the speed of sound.
2. A supersonic parachute. Deployed at roughly Mach 2 into air that will not tolerate a subsonic deployment, a disk-gap-band parachute drags the vehicle down further — but, as we just proved, only to a few hundred kilometers per hour. It is a step, not a solution.
3. The terminal method — and here the mission's mass decides everything. For a light payload you can finish with airbags (Mars Pathfinder in 1997, and the twin rovers Spirit and Opportunity in 2004, bounced to a stop) or with a rocket-powered sky crane — the astonishing maneuver by which Curiosity (2012) and Perseverance (2021) fired retro-rockets to a hover and then lowered the rover on cables to the surface before flying the descent stage away to crash at a safe distance. For a heavy payload, parachutes become useless dead weight and you must slow the whole way down on rocket thrust — supersonic retropropulsion, firing engines into your own hypersonic flow.

🔧 Engineering Reality: there is a Mars-landing mass wall, and humans are on the wrong side of it. The heritage EDL chain — heat shield, supersonic parachute, sky crane — has been pushed to roughly one tonne of landed mass, and Perseverance (about a one-tonne rover) sits near that ceiling. The reason is the parachute: to land more mass you need a bigger chute, but a bigger chute cannot be deployed fast enough or hold together in Mars's thin, hypersonic-then-supersonic flow. A crewed lander needs to put tens of tonnes on the surface — twenty to forty times what any parachute-based system has ever delivered. There is no parachute that scales that far. The only physics that does is propulsive descent: fire large engines against the incoming flow and ride them down, exactly the propulsive-landing skill developed for Earth reuse (Chapter 22) now turned outward. This is why Starship plans no parachutes at all — it enters belly-first to bleed speed aerodynamically, then flips and lands on its Raptors. Crossing the Mars mass wall is not an incremental improvement on Curiosity; it is a different way of landing entirely, and it is the single hardest piece of hardware between us and a human footprint on Mars.

🔧 Engineering Reality: the terror is real because the delay is real. From atmospheric entry to touchdown is about seven minutes. The one-way light time to Mars during a landing is on the order of ten to twenty minutes. So by the time the "we have touched the atmosphere" signal reaches Earth, the spacecraft has already been alive or dead on the surface for many minutes — the room in mission control is watching history, unable to help, unable even to know. Every decision of EDL — when to deploy the chute, when to cut it, when to ignite the descent engines — must be made by the vehicle itself, in real time, correctly, the first time, with no human in the loop. It is the purest test of the autonomy of Chapter 27 that the space program has: the machine flies the most dangerous seven minutes of the mission entirely alone.

Whether the spacecraft dives straight to the surface (direct entry, as the landers do) or first brakes into orbit is itself a choice. It can capture propulsively (the Mars orbit insertion of Chapter 11), or — to save that propellant — it can use the atmosphere to capture, which brings us to the term Chapter 11 promised this chapter would define.

Definition (aerocapture). Aerocapture is an arrival maneuver that uses a single, deep pass through a planet's atmosphere to shed enough energy — as heat, against a heat shield — to convert the arrival hyperbola directly into a bound orbit, replacing all or most of the propulsive orbit-insertion burn. It is the aggressive cousin of aerobraking (Chapter 11), which nibbles energy over hundreds of grazing passes; aerocapture does it in one plunge and demands pinpoint guidance through a razor-thin corridor — too shallow and you skip back out into space, too steep and you burn up or crash. It can save on the order of $2\ \text{km/s}$ of insertion delta-v at Mars, has been demonstrated at Earth, and is studied intensively for heavy Mars missions, but has not yet been flown at Mars.

🔄 Check Your Understanding 1. Explain, in terms of atmospheric density, why Mars EDL needs both a heat shield (like Earth entry) and rocket engines or airbags (unlike Earth entry). 2. What is the difference between aerocapture and aerobraking, and what does aerocapture buy you?

Answers

1. Mars's air, though only $\sim 0.6\%$ of Earth's density, is enough at hypersonic speed to build a searing shock, so a heat shield is mandatory — but it is far too thin to brake a heavy vehicle to a soft landing with parachutes (terminal speed $\sim 84\ \text{m/s}$), so a final active method (retropropulsion, sky crane, or airbags) is also mandatory. Earth's dense air does both jobs; Mars does only the first.
2. Aerobraking uses many shallow grazing passes to slowly shrink an already-captured orbit over weeks; aerocapture uses one deep pass to capture directly from the arrival hyperbola into a bound orbit. Aerocapture buys you the propulsive orbit-insertion burn (up to $\sim 2\ \text{km/s}$ at Mars) — at the price of a heat shield and a razor-thin guidance corridor.

34.5 Surface operations, ISRU, and the return problem

Suppose the seven minutes end well and the crew stands on Mars. They now face the problem that makes Mars categorically harder than any place humans have been: they have to get back off it. The Moon was forgiving here — its escape velocity is only $2.4\ \text{km/s}$, and Apollo simply carried the small ascent stage and its propellant down with it. Mars is not forgiving. Its escape velocity is $5.03\ \text{km/s}$, ascent to a low Mars orbit costs on the order of $4\ \text{km/s}$, and — through the tyranny of the rocket equation — the return propellant is the most expensive mass in the entire mission.

Here is why. Every kilogram of return propellant you bring from Earth must itself be launched to LEO, then injected toward Mars (paying the $3.6\ \text{km/s}$ trans-Mars injection to accelerate it), then carried through the searing, mass-limited EDL of 34.4 and set gently on the surface — and only then is it available to burn for the trip home. Each of those legs multiplies the mass through a rocket equation of its own. A kilogram of methane on the launch pad at Mars, brought from Earth, costs you many kilograms in low Earth orbit. This is the rocket equation stacked, and it is brutal.

The answer is one of the most elegant ideas in all of mission design.

Definition (in-situ resource utilization, ISRU). In-situ resource utilization (ISRU) is the practice of manufacturing a mission's consumables — most importantly propellant, but also oxygen, water, and building material — from resources found at the destination, rather than launching them from Earth. For Mars, the headline application is making rocket propellant from the Martian atmosphere and, where available, subsurface water ice, so the return vehicle can be fueled on the surface instead of hauling its propellant across the solar system.

Definition (Mars ascent vehicle, MAV). A Mars ascent vehicle (MAV) is the rocket that lifts the crew (or samples) from the Martian surface to Mars orbit or onto a trans-Earth trajectory. Because its propellant is the most leverage-heavy mass in the architecture, the MAV is the natural customer for ISRU: in the leading plans it is landed empty (or nearly so) and fueled on Mars from locally made propellant before the crew ever departs Earth.

The chemistry of Mars ISRU is chemistry you already met. Mars's atmosphere is about $96\%$ carbon dioxide — an endless free feedstock of carbon and oxygen. Feed that $\text{CO}_2$ and some hydrogen into the Sabatier reaction you saw close the life-support loop in Chapter 28:

$$ \text{CO}_2 + 4\,\text{H}_2 \;\longrightarrow\; \text{CH}_4 + 2\,\text{H}_2\text{O}, $$

and you get methane — rocket fuel — plus water. Electrolyze the water (Chapter 28's $2\,\text{H}_2\text{O} \rightarrow 2\,\text{H}_2 + \text{O}_2$) and you recover the oxygen to burn the methane with, and the hydrogen to feed back into the Sabatier reactor. The product is liquid methane and liquid oxygen — methalox — which is not a coincidence: it is exactly the propellant Starship's Raptor engines burn (Chapter 17). The choice of methane, made chapters ago for its clean combustion and storability, pays off here as the one chemical rocket propellant you can manufacture on Mars from the air and a little water. The hydrogen is the only ingredient in short supply; a mission either brings it (light, but bulky and hard to store) or mines Martian water ice and electrolyzes that.

🚪 Threshold Concept: make your return ticket at the destination. The instinct is to pack everything you will need before you leave — including the propellant to come home. The rocket equation makes that instinct ruinous, because return propellant brought from Earth is multiplied by every leg it rides through. The threshold idea, due to Robert Zubrin's Mars Direct and adopted by nearly every serious plan since, is to leave the return propellant behind and manufacture it there. Land an empty ascent vehicle and a propellant plant, let it quietly make methane and oxygen from the Martian atmosphere for a year or two — and do not launch the crew from Earth until the return vehicle is confirmed full and waiting. ISRU does not merely save mass; it inverts the risk, so that no astronaut is ever sent to Mars without a fueled ride home already sitting on the surface. Once you see the mission this way, ISRU stops being an optional technology and becomes the keystone that makes a crewed Mars mission close on mass and on safety at the same time.

Let us size the payoff.

Worked Example: the Mars ascent vehicle, fueled on Mars.

Strategy first. Size the propellant to ascend from the Martian surface to a low Mars orbit with the rocket equation, using an ISRU methalox engine. Then argue the leverage: this is propellant you did not have to launch from Earth, inject to Mars, and land.

Ascent from the surface to a low, $\sim 150\ \text{km}$ Mars orbit costs about $\Delta v_{\text{asc}} = 4.1\ \text{km/s}$ (Tier 2; the circular speed there is $\sim 3.5\ \text{km/s}$ plus gravity and drag losses through the thin air). Take a methalox ascent engine at $I_{sp} = 360\ \text{s}$, so $v_e = 360 \times 9.80665 = 3{,}530\ \text{m/s} = 3.53\ \text{km/s}$, lifting a $5\ \text{t}$ crew capsule (crew, cabin, samples). The mass ratio is $$ > \frac{m_0}{m_f} = e^{4.1/3.53} = e^{1.162} = 3.20, > $$ so the ascent propellant is $$ > m_p = m_f\left(\frac{m_0}{m_f} - 1\right) = 5\ \text{t} \times (3.20 - 1) = 11\ \text{t}. > $$ Sanity check and the leverage. About $11\ \text{t}$ of methalox lifts the $5\ \text{t}$ capsule to Mars orbit — a propellant fraction near $70\%$, entirely ordinary for a small launch stage. Now the leverage: had those $11\ \text{t}$ been brought from Earth, they would first have had to be landed on Mars (through the mass-limited EDL of 34.4), which means injected toward Mars (another mass ratio), which means launched to LEO (another mass ratio) — a stack of rocket equations that turns $11\ \text{t}$ on the Martian pad into many tens of tonnes in low Earth orbit. Making it on Mars instead deletes that entire chain. This is why NASA's own studies find ISRU one of the largest single mass savings available to a human Mars mission, and it is mass is the enemy (theme 4) answered by refusing to carry the enemy at all.

📜 From History: a rover has already done it. ISRU is not a paper concept. In April 2021 a microwave-oven-sized instrument called MOXIE, riding on the Perseverance rover, pulled carbon dioxide from the Martian atmosphere and split it into carbon monoxide and breathable oxygen ($2\,\text{CO}_2 \rightarrow 2\,\text{CO} + \text{O}_2$), producing a few grams of oxygen an hour and running successfully more than a dozen times across a Martian year. A few grams an hour is tiny — a real ascent vehicle needs to make oxygen at more than a hundred times that rate for a year or more — so the challenge is scale, not principle. But for the first time in history, a machine on another planet made a consumable humans could use, out of that planet's own air. The keystone of the return trip has been demonstrated; what remains is to build it a thousand times bigger.

Around the propellant plant, the rest of surface operations is every subsystem of Part IV transplanted to a dusty, cold, low-gravity world: power for the ISRU plant and the habitat — likely a fission reactor of the Kilopower class, because a $500$-day dust-storm-prone stay strains solar arrays (Chapter 25); thermal control against nights that plunge below $-80\,^\circ\text{C}$ (Chapter 24); a pressurized habitat and its closed-loop life support (Chapter 28); and the human factors of a crew confined for a year and a half with a $22$-minute tether to home. ISRU is the star of the surface phase, but it runs on a power plant and inside a habitat that are themselves triumphs of everything earlier in the book.

🔄 Check Your Understanding 1. Why is return propellant brought from Earth so much more "expensive," per kilogram, than the same propellant made on Mars? 2. MOXIE made oxygen but no fuel. Which reaction from Chapter 28 would a full plant add to make methane, and what Martian resources does the whole chain consume?

Answers

1. Because a kilogram of return propellant carried from Earth must be launched to LEO, injected toward Mars, and landed through EDL — each leg multiplying its mass through a separate rocket equation — so it costs many kilograms in LEO. Propellant made on Mars skips that entire stacked chain; it never has to be accelerated across the solar system. 2. The Sabatier reaction, $\text{CO}_2 + 4\text{H}_2 \rightarrow \text{CH}_4 + 2\text{H}_2\text{O}$, which makes methane (and water to electrolyze for oxygen and recycled hydrogen). The chain consumes Martian atmospheric $\text{CO}_2$ (abundant, $\sim 96\%$ of the air) and hydrogen — either brought from Earth or obtained by mining and electrolyzing Martian water ice.

34.6 Human Mars architectures: DRA 5.0 and Starship

Everything in this chapter — the window, the injection, the cruise, the EDL, ISRU, the return — is a piece that a mission architecture must assemble into a single coherent whole that closes on mass, cost, risk, and schedule, exactly the systems-engineering synthesis of Chapter 29. Two serious architectures dominate the conversation, and comparing them is the best possible capstone trade study, because they make opposite bets on almost every choice.

NASA's Design Reference Architecture 5.0

DRA 5.0 (2009) is NASA's canonical human Mars study — the careful, conservative, government reference against which other plans are measured. Its defining features:

• Conjunction-class, long-stay. A crew of six, minimum-energy transfers both ways, roughly a $900$-day mission with about $500$ days on the surface — the timeline of 34.1.
• Split mission and pre-deployment. Cargo — the surface habitat and, critically, the Mars ascent vehicle with its ISRU plant — is sent ahead on an earlier launch window, uncrewed, on slow minimum-energy trajectories. The ascent vehicle makes its return propellant from the Martian atmosphere before the crew leaves Earth, so the crew departs only once a fueled ride home is confirmed on the surface — the risk-inversion of 34.5.
• High-thrust propulsion for the crew. To hold the crew's transit time (and thus radiation dose) down, DRA 5.0's baseline uses nuclear thermal propulsion (Chapter 21) — hydrogen heated by a reactor to roughly double a chemical engine's specific impulse — for the trans-Mars injection of the crewed stack, with aerocapture and aerobraking to save arrival propellant.
• Heavy-lift, assembled in orbit. Several launches of a Saturn-V-class heavy lifter (the SLS lineage) loft the pieces, which are joined in LEO before departure — von Braun's 1948 insight, still binding.

DRA 5.0 is expendable, cautious, and mass-optimized to the last kilogram. It is the architecture of an agency that must not lose a crew.

SpaceX's Starship

Starship makes the opposite bet on nearly everything, wagering that reusability (theme 5) changes the arithmetic so completely that brute force beats finesse:

• Fully reusable, methalox, refueled in orbit. Each ship reaches LEO nearly empty and is topped up by tanker flights (the $\sim 358\ \text{t}$ of 34.2) before trans-Mars injection — many turnarounds of a reusable vehicle instead of a stack of expendable stages.
• The ship is the lander and the MAV. No separate descent stage, no parachutes: the same vehicle enters Mars belly-first to bleed speed aerodynamically, flips, and lands propulsively on its Raptors (Chapter 22's landing skill, turned outward), then — refueled on the surface by ISRU methalox (34.5) — launches itself back toward Earth. One vehicle plays every role.
• Mass, not thrift. Where DRA 5.0 counts kilograms, Starship's bet is to make the vehicle so large and so cheap-to-reuse that it can land $100\ \text{t}$-class payloads and simply carry margin the older architecture could never afford. It crosses the Mars-landing mass wall of 34.4 by being built, from the start, for propulsive descent at that scale.

The full Starship story — Raptor's full-flow staged combustion, the catch-and-reuse economics, the iterative development — is the climax of Chapter 38; here it is the concrete vehicle that turns this chapter's trajectory into a plan for people. Its numbers remain aspirational (Tier 2/3) — the vehicle is still in development (Chapter 30; Appendix H) — but the architecture is a clean, falsifiable answer to every problem in this chapter.

Worked Example: the two architectures as a trade study (Chapter 29 framing).

Criterion	DRA 5.0	Starship
Mission class	Conjunction, long-stay, crew ~6	Conjunction, long-stay, larger crews (goal)
Reusability	Expendable	Fully reusable (design goal)
Crew propulsion	Nuclear thermal (high $I_{sp}$, fast transit)	Chemical methalox + orbital refueling
Arrival	Aerocapture / aerobraking	Direct entry, propulsive landing
Landed mass per flight	Optimized, modest	$\sim 100\ \text{t}$-class (goal)
Return	Pre-deployed MAV, ISRU-fueled	The ship itself, ISRU-fueled on surface
Maturity / risk	High-heritage, conservative	Low-heritage, aggressive, unproven

There is no single "right" column, exactly as Chapter 29 warned: the answer depends on how you weight mass against cost against schedule against risk. DRA 5.0 minimizes the chance of losing a crew at the cost of expendable heavy lift and nuclear development; Starship bets that reuse collapses the cost so far that raw mass margin becomes affordable, at the cost of proving an entirely new way to land. What they share is more telling than what divides them: both fly the same Chapter-11 Hohmann, both wait on the same 26-month window, both close the return with ISRU, and both are, at bottom, answers to the tyranny of the rocket equation. The trajectory is settled physics; only the vehicle is still being argued.

🔧 Engineering Reality: what still stands between us and Mars. Neither architecture is blocked by a missing equation — the orbital mechanics has been solved since Hohmann. What remains is engineering at the edge of the possible: landing tens of tonnes through the EDL mass wall (34.4); scaling ISRU from MOXIE's grams to a MAV's tonnes (34.5); a radiation dose that spends a career limit (34.3); closed-loop life support reliable for three years with no repair (Chapter 28); and a cost that a government or a company can actually bear. These are the open problems the book's final chapters look toward, and they are the subject of Chapter 39's survey of where spaceflight is going. Mars is not waiting on a discovery; it is waiting on a great deal of very hard engineering — which is, after all, what this entire book has been about.

🔄 Check Your Understanding 1. DRA 5.0 and Starship make opposite choices about reusability and crew propulsion. For each, name the choice and the single biggest thing that choice is trying to buy. 2. Name two things the two architectures have in common, and explain why they must share them.

Answers

1. Reusability: DRA 5.0 is expendable (buying high heritage and conservative, proven hardware); Starship is fully reusable (buying a collapse in per-flight cost that makes mass margin affordable). Crew propulsion: DRA 5.0 uses nuclear thermal (buying a faster transit and lower radiation dose); Starship uses chemical methalox with orbital refueling (buying propellant it can also manufacture on Mars, at the cost of refueling flights). 2. Both fly the minimum-energy Chapter-11 Hohmann transfer and both wait for the 26-month synodic window — because those are set by orbital mechanics, not by engineering choice — and both close the return problem with ISRU, because the stacked rocket equation makes bringing return propellant from Earth prohibitive for either.

Mission Design Checkpoint: the Mars-track payoff — a complete mini-MDR

This is the payoff of the whole Design Your Mission project for Track C, and a worked template for every other track: a complete Mission Design Review for a crewed Mars mission, assembled from the pieces you have built across the book. If your mission is a Mars orbiter (Track C), this is the human sibling of your design; if it is Track A, B, or D, work through it as the fully worked example of how every subsystem you sized becomes one coherent architecture. It is Chapter 29's synthesis, aimed at the hardest target.

The MDR, on one table. A Mission Design Review answers one question — does this mission close? — by laying every piece down and checking they agree:

The delta-v budget — and the ISRU twist. The master constraint (Chapter 29) is still the delta-v budget, but a Mars mission splits it not just by who pays (launcher vs. spacecraft) but by where the propellant comes from — and that second split is the whole architecture. Some legs must be launched from Earth; the return legs are manufactured on Mars:

The code. Roll up the budget with mission.py from Chapter 29 (roll_up_dv and size_vehicle), keeping the two propellant sources separate, and size the ISRU-fueled ascent stage:

from mission import roll_up_dv, size_vehicle    # from astrotools (Chapter 29)

# Split the budget by WHERE THE PROPELLANT COMES FROM -- the heart of a Mars architecture.
earth_launched = [                               # must be lifted from Earth
    ("TMI (LEO -> trans-Mars)",         3600, 0.05),
    ("mid-course TCMs",                  100, 0.20),
    ("Mars landing burn (post-entry)",   600, 0.10),
]
mars_made = [                                    # manufactured on Mars by ISRU
    ("MAV ascent (surface -> low Mars orbit)", 4100, 0.05),
    ("TEI (low Mars orbit -> Earth)",          2100, 0.05),
]

_, dv_earth = roll_up_dv(earth_launched)
_, dv_mars  = roll_up_dv(mars_made)
print(f"Earth-launched delta-v: {dv_earth:.0f} m/s")
print(f"Mars-made  delta-v:     {dv_mars:.0f} m/s")

# Size just the MAV ascent stage: a 5 t crew capsule to low Mars orbit on ISRU methalox (Isp 360 s).
_, mav_dv = roll_up_dv([("MAV ascent", 4100, 0.05)])
m_prop, m_wet = size_vehicle(mav_dv, isp=360, payload=5000)   # kg
print(f"MAV ascent propellant (ISRU methalox): {m_prop/1000:.1f} t")
print(f"MAV wet mass on the Mars pad:          {m_wet/1000:.1f} t")
# Expected output:
# Earth-launched delta-v: 4560 m/s
# Mars-made  delta-v:     6510 m/s
# MAV ascent propellant (ISRU methalox): 11.9 t
# MAV wet mass on the Mars pad:          16.9 t

How it closes, and how it feeds the capstone. Read the two totals: about $4.6\ \text{km/s}$ of margined delta-v the mission must launch from Earth, and about $6.5\ \text{km/s}$ it makes on Mars. That second number is delta-v the rocket equation would otherwise have punished at every leg — launched, injected, and landed — and ISRU deletes it from the Earth-launched mass entirely. The MAV that provides part of it needs only about $12\ \text{t}$ of locally made propellant, none of it hauled from Earth. When you assemble your full Mission Design Review in Chapter 40, this is the pattern: a delta-v budget as the spine, split by who pays and where the propellant comes from, each number traceable to a chapter, closing on a vehicle you could actually build. For Track C, this is the mission.

Summary

A Mars mission is the whole book, forced to agree with itself. Carry these forward:

Numbers worth memorizing: window every $\approx 26$ months; TMI $\approx 3.6\ \text{km/s}$; cruise $\approx 259$ days; Mars entry $\approx 5.6\ \text{km/s}$; parachute terminal speed on Mars $\sim 84\ \text{m/s}$ (so parachutes alone cannot land); Mars escape $5.03\ \text{km/s}$; ISRO propellant = methalox by Sabatier from $96\%$-CO$_2$ air.

Spaced Review

Retrieval strengthens memory. Answer from memory before checking, then look back at the cited chapter. This chapter revisits Chapter 11, Chapter 28, and Chapter 29.

1. (§34.2, Ch. 11) The trans-Mars injection needs a hyperbolic excess of $v_\infty = 2.95\ \text{km/s}$. What departure characteristic energy $C_3$ is that, and roughly what injection burn from a $300\ \text{km}$ parking orbit — and why is the burn so much less than $v_\infty$ plus the escape speed?
2. (§34.1, Ch. 11) Where does the 26-month launch cadence come from — what quantity is it, and what does it depend on?
3. (§34.3, Ch. 28) A crew of four flies an open-loop 900-day Mars mission. Roughly how many tonnes of consumables is that at $\sim 5\ \text{kg}$ per person-day, and what does the answer tell you about closing the loop?
4. (§34.3, Ch. 28) Against a NASA career dose limit near $0.6\ \text{Sv}$, why is the ~0.47 Sv of a single cruise leg such a serious constraint, and what is a "storm shelter" defending against?
5. (§34.6, Ch. 29) The DRA-5.0-vs-Starship comparison is a trade study. Why is there no single "correct" winner, and what is the driving requirement of a crewed Mars mission?

Answers

1. $C_3 = v_\infty^2 = (2.95)^2 \approx 8.7\ \text{km}^2/\text{s}^2$; the burn is $\Delta v_{\text{TMI}} \approx 3.6\ \text{km/s}$. It is far less than $v_\infty + v_{\text{esc}}$ because the spacecraft is already moving at $\sim 7.7\ \text{km/s}$ in the parking orbit and you pay only the difference — and the speeds add in quadrature ($v_{\text{peri}} = \sqrt{v_\infty^2 + v_{\text{esc}}^2}$) because energy goes as speed squared. Burn low, burn fast — the Oberth effect. 2. It is the synodic period, the time between identical Earth–Mars alignments: $1/T_{\text{syn}} = |1/T_1 - 1/T_2|$, so it depends on the difference of the two planets' orbital rates — $\approx 780$ days for Earth and Mars. 3. About $4 \times 900 \times 5 = 18{,}000\ \text{kg} \approx 18\ \text{t}$ — far beyond what you would want to launch and land, which is exactly why a multi-year mission must recycle air and water (closed-loop life support) rather than carry it all. 4. Because a single one-way cruise already spends most of a lifetime allowed dose, so a round trip can exceed the career limit — radiation is a first-rank Mars problem. A storm shelter defends against the acute dose of a solar particle event, using water/food/waste already aboard as shielding. 5. Because the "best" architecture depends on how you weight mass, cost, schedule, and risk — the trade study exposes the weighting rather than hiding it. The driving requirement is returning the crew alive, which is what forces ISRU (a fueled ride home), the radiation defenses, and the multi-year reliability standard.

What's Next

We have flown the whole mission — waited for the window, injected, cruised, survived the seven minutes, made fuel from the air, and come home. But look at what a Mars campaign leaves behind: spent stages abandoned in orbits around Earth and the Sun, a discarded heat shield and descent stage on the Martian surface, hardware strewn across the solar system in the name of exploration. Scale that up — thousands of launches, mega-constellations, a genuine space economy — and a new problem comes into focus, one that is not about how to get somewhere but about how to keep the places we use usable. The orbits around Earth are a finite commons, and we are filling them with debris that threatens the very access spaceflight depends on. In Chapter 35 we turn from ambition to responsibility: the Kessler syndrome, the debris environment, the law and traffic management of orbit, and the sustainability of a spacefaring civilization. Getting to Mars is the dream this book has chased; being a responsible actor in space is the condition for keeping the dream — and the planet it launches from — intact.