=military =nuclear =rant
People don't think about risks of nuclear war as much as they used to, but there are some new weapon systems relevant to that.
conventional ballistic missiles
If a country attacks with ballistic missiles that carry conventional
explosives but could carry nukes, that could be mistaken
for a nuclear attack and lead to nuclear retaliation. Until recently,
countries have avoided deploying similar missiles for conventional and
nuclear attacks, but that seems to be changing.
Previously, the INF
Treaty
banned all ground-launched missiles with ranges between 500 to 5500 km.
Missiles with <500 km range aren't as effective for cross-border attacks.
Missiles with >5500 km range are expensive and vulnerable enough that it's
not worth putting them close to targets or using them for conventional
weapons. Long-range ballistic missile attacks are easier to detect & give
more warning time than shorter-range missiles.
China never signed
that treaty, and now the US and Russia have withdrawn from it. In 2017 the
US accused Russia of violating it by deploying the Novator
9M729. Russia denies that, see
eg this
blog
for some details. As people have
noted the
supposed weapon isn't strategically important, and the larger missile hasn't
been seen on deployed vehicles since. My guess is, US intelligence mistook a
test of a naval weapon for a new weapon for a land-based launcher, Trump
jumped on it, and backing down would've looked "soft on Russia".
For
similar reasons, I didn't buy the US stories about Russia being responsible
for attacking its own Nordstream pipelines. And now the story
is,
that was some Ukrainian guys in a yacht, which isn't consistent with the
seismometer data. Maybe that's why US papers don't talk about the
seismometer data so much. Anyway, that was probably US torpedoes. But I'm
sure the investigation data showing otherwise will be released any day
now.
And if we go back 20 years to the Iraq War, it was obvious at the time
to anyone familiar with the science & engineering involved who looked up
some technical details that the evidence of an Iraqi nuke program was bad.
Sure, places like the NYT have terrible science reporting, but even then, it
should've been obvious that Cheney was lying because he started out by
arguing Iraq was supporting Al Qaeda and then dropped that argument when it
became apparent people weren't buying it. Yet today, we have media democrats
celebrating Dick Cheney supporting Kamala. Disgusting. But I'm getting
distracted here.
Anyway, in 2018 the US suspended the INF Treaty, and
it doesn't matter who was right, what matters now is, the treaty is quite
dead. The US was already interested in conventional ballistic
missiles but it's
been more interested since that treaty was suspended.
ballistic antiship missiles
Ships are particularly suitable targets for conventional ballistic
missiles. So, China has been developing ballistic antiship missiles, such as
the DF-21D and YJ-21.
There's a list of copes that people in the US military have been going
down regarding ballistic antiship missiles:
"You have to know where the ships
are to target them."
Yes, that's one reason China has been making
surveillance satellites. Large ships are relatively easy to find.
"Ships move so the missiles will miss."
That's why you add guidance
to missiles.
"The missiles won't be able to detect the ships at long
range, especially with a plasma sheath from reentry."
That's why you
put an antenna on the rear of the missile, and provide targeting updates to
it.
"Satellites won't have continuous real-time targeting info."
That's one reason China has been making stealthy high-altitude
surveillance UAVs.
"But you have to network and integrate all that
stuff, and China just doesn't have the experience or competence for that."
Wrong.
"You can just shoot down the missile with a railgun."
How have those railgun programs been working out?
Railguns are
not good weapons. They're sort of the Peltier
coolers of guns, in that some
people like them because they replace hot gas with electricity but they just
don't work very well.
Some people think railguns or coilguns or
whatever would be much lighter than regular guns because electromagnetism is
magic and can do whatever, but they're not. They're heavier, inevitably.
They're also expensive. (Yes, electric motors are cheap, but that's because
they often operate for billions of cycles.)
Yes, in theory, the
maximum potential speed of EM guns is higher. That's not useful, because
aerodynamic drag exists and most of the energy of cannon-sized projectiles
will be lost at long range. There are also issues with the horizon, what
with Earth being curved.
But perhaps the biggest problem with
railguns is, at long range you want guidance, and it's much harder to add
guidance to something that's fired out of a gun with high acceleration than
a missile. Unguided projectiles are only useful at shorter ranges, and a
regular gun is cheaper and lighter than a railgun.
The enthusiasm the
US Navy leadership had for railguns is something that lowered my estimation
of its competence, along with eg the Zumwalt and LCS program. But naval
aviation and submarines are basically separate departments of the USN from
the surface ships, and they seem more competent.
"OK, maybe not a
railgun, but you can hit it with a counter-missile, like any other ASM."
It's not clear that USN ships can currently intercept an incoming
ballistic missile, but it's at least possible with something like their
current equipment. The problem is, what if it's not a unitary warhead?
If a ballistic missile carries a bundle of metal rods, it can release
them a ways away from the target, and intercepting those is impractical. So,
the missile would need to be intercepted before releasing its payload.
Because interceptors take some time to accelerate and reach the intercept
point, the incoming missiles would need to be detected at more than 2x the
payload release range. The required detection range is not feasible.
The effectiveness of long-rod penetrators is one reason ships are a suitable
target for ballistic missiles: with most targets, they'd overpenetrate and
waste most of their energy, but ships are thick.
No, I'm not writing original military advice for China. This is all trivially obvious stuff that was discussed on public forums 20 years ago.
chinese strategy
So,
conventional ballistic missiles seem useful and unlikely to just go away. As
for how they might be used, my impression of Chinese strategy wrt Taiwan is:
- Make nukes
and hypersonic glide vehicles to deter a nuclear response.
- Use the
threat of ballistic antiship missiles to keep carriers away.
- Make
anti-satellite weapons to respond to US attacks on satellites used for
targeting ships.
- Use long-range AA missiles to target tanker aircraft
and AWACS to keep away refueled fighter aircraft. Give the missiles AESA
radar to handle jamming.
- With US fighter aircraft kept out of range,
use many air-launched antiship missiles to overwhelm the defenses of
destroyers not with carriers.
- Use land-based missiles and UAVs to
attack targets in Taiwan.
- Intercept long-range cruise missiles with
ships using the Chinese version of Aegis.
- Lay naval mines on the east
side of Taiwan to support a blockade.
- Set up passive bistatic radar to
detect stealth bombers over land.
- various anti-submarine things
That's not getting into the
economic stuff like stockpiling food & ore, replacing coal imports, building
certain chemical plants, etc. Which, in terms of timing, is what really
worries me: the amount of stockpiling doesn't make sense to me as
very-long-term preparation. (That's the only reason I'm even thinking about
this stupid military stuff instead of something normal like metabolic
engineering or AI self-improvement.)
Another issue here is that China
seems to have very different views of the game theory of escalation than the
US does.
My impression of the planned US response to all that is:
- Build B-21
bombers. (At high costs.)
- Start
multiple
different
programs for
long-range air-to-air missiles. (None use ramjets. Obviously ramjets are
correct for long-range AA missiles, but the US military seems to like them
less than China & Russia.)
- Build more F-35s. (But production rate is
currently maxed, which is part of why they're making the F-15EX too.)
-
Abandon AWACS and big tankers, and use more F-35s for those roles instead
with buddy refueling. (But this is a big sacrifice in terms of radar and
fuel ranges, and it's more expensive.)
- Use air-launched cruise missiles
with autonomous targeting, eg the LRASM.
That's right, autonomous
targeting. Despite some popular misconceptions, current US military policy
does allow for use of fully-autonomous weapons. What the actual policy
documents say is, systems with autonomous targeting must have a "senior
review" before being fielded, and their actions must reflect the commander's
intent.
In theory, it should be possible to build B-21s for 3x the
cost per payload as the discounted purchase price of a 787, which would be
1/3 their actual cost. And South Korea can build equivalent destroyers to US
ones for 1/3 the cost. China also seems to get around 3x the military stuff
per nominal dollar as the US.
Yes, the F-35 is better than current
Chinese aircraft. But as an engineer, it doesn't feel great to see your
country's leadership depending on better technology to make up for worse
strategy and planning and manufacturing. Especially when the companies
responsible for that tech advantage are relying on a university system
that's been taken over by publication metrics, full of visiting chinese grad
students, and (based on the papers I see) no longer better at engineering
than the high-tier chinese universities. What America does have is
people like me...who don't get used effectively. I can't even say the work
I've been paid for has clearly been of net benefit to society.
nuclear cruise missiles
Launches of large ballistic missiles make lots of heat and are easy to
detect with satellites. Ballistic missiles fly high and can be detected at
long ranges with line-of-sight radars. Cruise missiles launches are harder
to detect, and they fly lower, so the horizon limits radar detection ranges
more.
For decades, there was an international consensus about
nuclear-armed cruise missiles: be ready to make them, but don't actually
make them, because they're unnecessary and destabilizing. Now, countries are
making them. For example,
Zircon is nuclear-capable. As
for the US, SLCM-N had been cancelled, but Congress recently
mandated
development of it.
The SLCM-N concept is basically "put a nuke in a
Tomahawk". Why did people think that's a good idea? Well, here's an
article
supporting it. The argument is:
"Fielding SLCM-Ns would demonstrate that the United States has optimal capabilities to respond to small-scale nuclear use by an adversary. The current relative lack of flexible options for a nuclear response may leave adversaries with the mistaken notion that they could “get away with” small-scale nuclear use without facing unacceptable consequences."
Maybe you see the problem here.
stealthy cruise missiles
Stealthy cruise missiles are a relatively recent development. See
LRASM and Storm
Shadow for some examples.
The "Storm Shadow" is a stealthy missile, which is why it went with a
turbojet to maximize its IR signature, despite that giving half the range of
a turbofan. As a bonus, a turbojet costs maybe $100k less, which is why it's
a bargain $2.5 million.
Consider the following situation:
A stealthy cruise missile is launched from an unknown location, perhaps a submarine, by an unknown country. It flies to a military base of a nuclear power, and is detected only briefly before the nuke it carries goes off.
Now what?
The more
countries develop stealthy cruise missiles capable of carrying nukes, the
more of a problem this becomes.
hypersonic glide vehicles
Any ballistic missile moves at "hypersonic" speeds, but the term "hypersonic
weapons" has recently been used specifically for fast-maneuvering
waverider hypersonic weapons. And then some people deliberately
conflate those with anything fast.
The main purpose of hypersonic
glide vehicles is to maneuver nukes during reentry to avoid potential
missile defenses. The combination of high speed and high accelerations makes
hitting them impractical. Some examples include Russia's
Avangard
and China's DF-ZF.
Missile
defense not being reliable against incoming ballistic missiles is just a
continuation of the past situation, but there are a couple differences with
hypersonic glide vehicles:
- Because
they have some lift, these glide vehicles have a trajectory that's lower and
more flat. That reduces radar detection ranges somewhat.
- Existing
missile defenses could plausibly stop 1 or 2 nukes. If there's some smaller
country that only has a few nukes, them having hypersonic glide vehicles for
those would put them in a stronger position relative to the US.
hypersonic tech
Why do we
have these hypersonic glide vehicles now, but not 50 years ago?
They're basically waveriders that
operate at a range of speeds. Building hypersonic wind tunnels is expensive,
but China has been making some recently. And good CFD modeling only became
practical relatively recently, because of the computational requirements.
As for hypersonic cruise missiles, you obviously need propulsion, but
once upon a time, people tried to make scramjets and they just didn't work.
What changed?
With scramjets, air goes through the combustion chamber
quickly, and swirling the air enough for good mixing would take too much
energy. (Even ramjets have some combustion inefficiency compared to
turbojets because of the high flow speeds.) Also, a very long combustion
chamber would have lots of friction with the supersonic air.
My
understanding is, the key to working scramjets was hydrogen. It's a small
molecule that diffuses quickly. It can also burn faster than hydrocarbons:
oil burning is actually a rather complex reaction.
Back in 1991,
weeks before the USSR dissolved, it tested "Kholod", the first successful
scramjet. That demonstrated the hydrogen-fueled scramjet engine concept
working. In 2004, NASA successfully combined that with a waverider design
with control surfaces in the
X-43A. The X-43C was
supposed to demonstrate that with hydrocarbon fuel, but they couldn't get it to
work and cancelled the project.
Obviously hydrogen has low density
and is hard to store, which is bad for fast missiles, so you probably want
some chemical hydrogen storage system. Well, I could speculate on what
they're doing for eg Zircon, but that seems beyond the scope of this post.
Of course, using stored hydrogen for fuel reduces range, and supersonic
speeds reduce range, so these hypersonic cruise missiles will always have
shorter ranges than turbofan cruise missiles and large ballistic missiles.
That being the case, I suppose the main strategic role Zircon fills is
shorter-range nuclear strike that's hard to intercept and hits faster than a
ballistic missile, probably in under 10 minutes. (Some Zircon missiles were
used in Ukraine, but that was probably testing instead of a cost-effective
attack. And Ukraine claimed to intercept one, but I think that was a
Kinzhal.)
anti-satellite weapons
If a
conflict between China and the US over Taiwan happens, I think there's a
fairly high chance of both sides attacking satellites.
Some
satellites are used to monitor missile launches and buildup of military
forces. If surveillance satellites get destroyed, it's harder to do that. On
the other hand, there are a lot more satellites than there used to be, and
their capability is somewhat redundant.
When I said "China seems to
have different views of the game theory of escalation than the US does" this
is one of the things I was talking about. China seems to consider attacks on
satellites a much smaller escalation than US leadership does.