The ABM Thread XLVI

[MKSheppard]

NOTE: There is no thread II, III, IV, etc. We started at XLVI as sort of a nod as to how numerous this has been over the years.

But if you're interested in other threads on this subject, try this on for size: LINK

LINK 2

Try tossing these into your preferred online LLM (Grok, ChatGPT, etc) of choice:

Starting from first principles; and without consulting the internet:

Estimate for me the capabilities of a netted system of radars against a system of targets and complex decoys mimicking the target.

TARGETS:

The primary targets we are concerned with have a conical shape with a RCS of 0.005 to 0.8 m2 at 4300 to 153 MHz and 0-30° aspect from the radar.

Complex decoys consist of objects with the same shape as well as chaff clouds covering approximately 65 kilometers in diameter spaced about 95 kilometers apart. There may be up to nine chaff clouds; making the chaff train be over 740 kilometers in length.

RADARS:

Typical Radar Horizon Distances for very large Radars are:

1640~ km target altitude = 5297~ km radar horizon (Minuteman I Apogee)
275~ km target altitude = 2180~ km radar horizon (Minuteman I Stage 3 Burnout)
130~ km target altitude = 1500~ km radar horizon (Minuteman I Stage 2/3 Staging)
38~ km target altitude = 800~ km radar horizon (Minuteman I Stage 1/2 Staging)

The Very Early Warning Radars (VEWRs) in this network specification are:

440 MHz (UHF)
10 MW Peak Power
27 Hz PRF
2 ms pulsewidth
5000 km instrumented range
Fixed Antenna Fence (Torus Antenna), 240 degree coverage
Upper Radar Fan - 7 Degree Elevation
Lower Radar Fan - 3.5 Degree Elevation

Locations are:

Clear, AK [64.2891, -149.1904]
Thule, Greenland [76.56892314671748, -68.30815221422696]

Several Precise Early Warning Radars (PEWRs) exist, with the following specifications:

425 MHz (UHF)
5 MW Peak Power
270 kW Average Power
27 Hz PRF
2000 µs pulsewidth
5000 km instrumented range
Mechanical Parabolic Antenna

Three Radars in England; 130~ m from each other, forming a 300+m baseline triangulation system [54.361171818100445, -0.6698289704347343]
One Radar at Greenland [76.56892314671748, -68.30815221422696]
One Radar at Clear, AK [64.2891, -149.1904]

The concept is that the Very Early Warning Radars (VEWRs) provide an initial "trigger" of enemy ballistic missiles when missiles "break" the radar fences which are continually monitored. Following detection; the Precise Early Warning Radars take over and provide more accurate range, elevation and azimuth information allowing missile impact zones to be calculated.

The Surveillance Radars (SR) in this network's specifications are:

420-450 MHz Frequency (UHF)
14.3 MW Peak Power
715 kW Average Power
140 degree azimuth detection volume and elevation volume scan of 93~ degrees.
Frequency Modulated and provides pulse compression with a compression ratio of 1:100.
30 meter diameter (700~ m2) Planar Phased Array with high track rate
22.86 meter centerline height of array (for radar horizon calculations)

Sensitivity - Can detect 24 cm (Basketball-sized) targets @ 2000 miles [3200 km] range

The sites for the SRs are:

1. Fairbanks AK [64.2891, -149.1904] -
2. Port Gamble (Seattle, WA) [47.853, -122.5909]
3. Malmstrom AFB [48.2878, -111.3423]
4. Grand Forks AFB [48.7264, -97.8991]
5. Detroit, MI [42.778, -83.339]
6. Sharpner's Pond (Boston, MA) [42.6467, -71.0376]

Additionally, there is a second type of radar within the network - the Engagement Radar (ER) - which handles actual intercepts.

It's specifications are:

S Band Frequency (centre about 2950 MHz) with 250 MHz Bandwidth
4.5 MW Peak Power
225 kW Average Power
3.96 meter diameter (12.3~ m2) Planar Phased Array with high track rate, giving about 100 degree coverage; up to four of these antennas can be installed, giving complete 360 degree coverage if desired at certain sites.
15-16 meter centerline height of array (for radar horizon calculations)

Sensitivity - Unknown sensitivity or performance. If we assume that the system is as sensitive as the Surveillance Radar and scaling off that radar and adjusting for RCS/frequency differences, the Engagement Radar can detect 24 cm (Basketball-sized) targets @ 1,100 km.

The much smaller antenna area (12.3 m2 vs. 700 m2) is the dominant factor reducing range vs the Surveillance Radar, despite the higher gain per area at S-band.

It is for this reason that the Engagement Radar buildings are designed with the Antenna Faces "plugged", to enable the current 3.96m diameter antennas to be replaced with much larger 9m diameter [63.6 m2] antennas. If this were to happen, the detection range on 24 cm spheres would rise to about 2,400~ km.

The sites for the ERs (and their associated faces)

1. Fairbanks AK [64.2891, -149.1904] (North + West Faces)
2. Port Gamble (Seattle, WA) [47.853, -122.5909] (North + West Faces)
3. Malmstrom AFB [48.2878, -111.3423] (360 degree Face Coverage)
4.Grand Forks AFB [48.7264, -97.8991] (360 degree Face Coverage)
5. Chicago, IL (North Face)
6. Detroit, MI [42.778, -83.339] (North + East Faces)
7. New York City, NY (North Face)
8. Sharpner's Pond (Boston, MA) [42.6467, -71.0376] (North + East Faces)
9. Hawaii, HI (360 degree Face Coverage)
10. San Francisco, CA (North Face)
11. Los Angeles, CA (North Face)
12. Salt Lake City, UT (North Face)
13. F.E. Warren AFB (360 degree Face Coverage)
14. Whiteman AFB (360 degree Face Coverage)
15. Washington DC (360 degree Face Coverage)
16. Dallas, TX (North Face)
17. Albany, GA (360 degree Face Coverage)

The FAIRBANKS, SEATTLE, MALMSTROM, GRAND FORKS, DETROIT and BOSTON ERs are located within about 40 kilometers of the SRs also located there, enabling the Engagement Radars to act as back up Surveillance Radars if the Surveillance Radars there are damaged or destroyed. Presumably in an "improved" system, these co-located Engagement Radars would have their antennas enlarged to 9.1m diameter.

There is one last radar we need to cover, a specialized Instrumentation Radar (IR) that will become available about five years after the system is functional, with the following specification:

L-Band (1250 MHz Centrepoint)
15.4 MW Peak Power
920 kW Average Power
0.6 Degree Beamwidth
136 degree azimuth detection volume
29 meter diameter (660~ m2) Planar Phased Array with high track rate

Two different operating modes:

In wideband mode, it uses a 1 ms pulse with a 200 MHz bandwidth (obtained using linear frequency modulation pulse compression) at frequencies between 1,175 and 1,375 MHz. This mode limits it to within 22.5 degrees of its boresite, but it provides a range resolution of about 1.14 meters.

In narrowband operation, it uses frequencies between 1,215 to 1,250 MHz. In search, it uses 1 MHz pulses with lengths of 1.5 or 2.0 ms, and in track it uses six different 5 MHZ bandwidth pulses with lengths between 0.15 and 1.5 ms.

Overall, typical accuracies are 3 m in range and 0.02 degrees in angle. Sensitivities are similar to the earlier Surveillance Radars and Engagement Radars; providing theoretical ranges in excess of 7,000 km against 24 cm diameter (basketball-sized) targets at extremely high altitudes. In reality, the Instrumentation Radar is radar horizon limited to about 5,000 km ranges against ICBMs at their apogee.

The Instrumentation Radar is located at [52.737219496720655, 174.09111311290823] and faces [52.7380112881138, 174.08986414891154].

===

There is a bonus radar which may come online about 20 years after system IOC:

Sea Based Radar (SBR); with the following specifications:

X Band (9500 MHz Centrepoint)
1000 MHz Bandwidth
Range Resolution 25~ cm
452.6 kW Peak power
90 kW Average Power
17.85 meter diameter (384 m2) Planar Phased Array
Can track an object the size of a baseball (7.3~ cm diameter) from 4,700 km distance.

Location is generally near Hawaii.

============================

Questions to ponder based on the data above:

1.) What are the implications of the Very Early Warning Radars (VEWRs) and Precise Early Warning Radars (PEWRs) in relation to the system and early target cuing for the Surveillance Radars in the system?

2.) What is the implication of the Fairbanks radar deployment (SR and ER) for early tracking of inbound targets?

3.) What are the implications of the extremely long baselines for the entire radar network system in terms of accuracy and resolution on incoming targets; if all six of the Surveillance Radars can see the same target complex?

4.) Related to #2, what are the implications of fact that all six of the Surveillance Radars are semi co-located with Engagement Radars with much higher frequencies than the SRs; meaning that we also have six Engagement Radars on a very long baseline?

5.) What are the implications of the frequency diversity between the Surveillance Radars and the Engagement Radars?

6.) What would the effect be of other radars in the network having side angle aspect views on target complexes? For example, what if the Grand Forks site at 48.7264, -97.8991 is the target? How would the rest of the integrated network see the inbound target complexes?

8.) What would be the result of reducing the network of radars to just an engagement+surveillance radar pair at Malmstrom and an engagement radar at Grand Forks?

9.) What would be the results of reducing the network to just Malmstrom's engagement+surveillance radar pair?

10.) What are the implications of the Instrumentation Radar joining the network five years after IOC?

11.) What are the implications of the Sea Based Radar joining the network twenty years after IOC?

Thank you for your time; and show your work.

Once it answers, try tossing this in:

Starting from first principles; and without consulting the internet:

Estimate for me the capabilities of the earlier networked radar system with the following interceptor configurations against incoming target cloud complexes with target complex velocities of 6.7 to 7.5 km/sec:

********

MID COURSE NUCLEAR; using the same radar network described earlier, with the following potential interceptors:

Basic Nuclear Mid-Course Interceptor

Command Guided, simple one way datalink telling missile where to go.
5 megaton warhead tailored for X-Ray Fluence.
4.1 km/sec Delta-V, three-stages, final stage with warhead has simple vents to terminate thrust early if commanded. Cannot restart.

Improved Nuclear Mid-Course Interceptor

Command Guided, simple one way datalink telling missile where to go.
1 megaton warhead tailored for X-Ray Fluence.
6.1 km/sec Delta-V, three stages, final stage has a "loiter" mode, enabling it to coast after third stage burnout and then restart upon ground command via igniting a "terminal engagement" solid propellant motor.

Mid-Course Nuclear Missile Launcher Sites:

1. Fairbanks AK [64.2891, -149.1904] (40~ missiles)
2. Port Gamble (Seattle, WA) [47.853, -122.5909] (40~ missiles)
3. Malmstrom AFB [48.2878, -111.3423] (40~ missiles)
4. Grand Forks AFB [48.7264, -97.8991] (40~ missiles)
5. Chicago (40~ missiles)
6. Detroit, MI [42.778, -83.339] (40~ missiles)
7. New York City (40~ missiles)
8. Boston, MA [42.6467, -71.0376] (40~ missiles)
9. San Francisco, CA (40~ missiles)
10. Los Angeles, CA (40~ missiles)
11. Salt Lake City, UT (40~ missiles)
12. Whiteman AFB, Missouri (40~ missiles)
13. Washington DC (40~ missiles)
14. Dallas, TX (40~ missiles)
15. Albany, GA (40~ missiles)

===

NUCLEAR EFFECTS:

DTRA-IR-10-22: Collateral Damage to Satellites from an EMP Attack (Aug 2010) DTIC ADA531197 gives the following equations for an exoatmospheric nuclear detonation:

X-Rays (cal/cm2): (6.4 x 10^3) * (Y/R^2)
Neutrons (n/cm2): (1.6 x 10^15) * (Y/R^2)
Gamma (rads [SI]): (2.5 x 10^5) * (Y/R^2)

Where:

R: distance in kilometers
Y: yield in megatons

On 5 March 1969 during a NSC meeting, SecDef Packard stated that SPARTAN's warhead (5MT X-Ray) had a kill radius of 12 miles on soft targets and 4 miles on hard targets.

With the DTRA-IR-10-22 equations that works out to:

85~ cal/cm2 for Soft RVs
700~ cal/cm2 for Hard RVs

********

MID-COURSE NON-NUCLEAR; using the same radar network described earlier, with the following potential interceptor:

Basic HTK Mid-Course Interceptor

10.6 km/sec Delta-V, three-stage launcher that lofts a single 64 kg HTK interceptor onto the desired trajectory. The HTK interceptor contains an IR sensor with a 2 degree FOV and a 300 km detection distance, plus a Two-Way Data-Link to ground stations.

Improved HTK Mid-Course Interceptor

10.9 km/sec Delta-V, three-stage launcher that lofts seven (7) HTK interceptors onto the desired trajectory. The HTK interceptors mass approximately 5 kg and contain an IR sensor with a 2 degree FOV and a 150 km detection distance, plus a Two-Way Data-Link to ground stations.

Launcher Site:

1. Fort Greely, AK [63.95862794730161, -145.7095501743137] (40~ missiles)

Also consider cases if a second launcher site is built here:

2. Fort Drum, NY [44.06949663368277, -75.74441960176631] (40~ missiles)

Consider a third point:

3. Ship [37.091230566442576, -157.07159941182468] armed with either:

Basic HTK Naval Mid-Course Interceptor
3 km/sec Delta-V, three-stage launcher that lofts a small 16.75 kg HTK interceptor with a 2 degree FOV and a 200 km detection distance.

Improved HTK Naval Mid-Course Interceptor
4.5 km/sec Delta-V, three-stage launcher that lofts a small 16.75 kg HTK interceptor with a 2 degree FOV and a 300 km detection distance.

Consider a fourth point:

4. An advanced joint service HTK Boost/Mid-Course Phase Interceptor that can be ground launched or sea launched.
6.86 km/sec Delta-V, three stage vehicle that lofts a 130 kg HTK interceptor and reaches 6 km/sec velocity in 60 seconds.

Consider a fifth point:

5. There is a network of space based infrared sensors in GEO staring at all potential enemy missile fields. The sensitivity of these IR sensors is such that air-to-air missile motor (200~ kg total launch mass) plumes can be seen in air-to-air engagements between aircraft. What would be the effect of including this early warning system into our system architecture?

Consider a sixth point:

6. A relocatable networked X-Band (8.55 – 10 GHz) radar that can be transported on C-17s exists. The antenna is an AESA planar array with an effective area of 9.2 m2 with a peak power of 405.5 kW and average power of 81.1 kW. Current estimates are that it can detect (S/N 20) a 0.01 m2 target at 870+ km and discriminate (S/N 100) that same target at 580 km.

Currently, these radars are deployed at the following locations:

Kyogamisaki, Japan [35.76630251582859, 135.19511732389296]
Shariki, Japan [40.96723200356672, 140.32087708495268]
Soseong-ri, South Korea [36.04693467197404, 128.22479354770897]
Guam [13.521328876781203, 144.81461900172286]

Assume that they are pointed at the threat launch azimuth as necessary.

********************************

Show your work for all cases and points.

[MKSheppard]

Random things coughed out by the current (Nov 2025) free version of Grok:

2. Implications of the Fairbanks site (SR + ER co-location in Alaska)

Fairbanks is the only continental site that can see ICBMs launched from Siberia or Kamchatka while they are still in powered flight or very early midcourse (horizon ~2200 km at 275 km altitude, ~5300 km at apogee). It therefore gets the longest possible coherent observation time (~12–15 minutes on some trajectories) and the highest elevation angles (objects appear almost overhead). High elevation = smallest range-rate spread across a large penetration-aid cloud, making it easier to keep the entire complex inside one or a few resolution cells early on and watch how the cloud evolves. Co-location of a large UHF SR and an S-band ER means the same object set can be handed over instantly to a higher-frequency, narrower-beam radar for fine discrimination without any cueing delay or geometry change.

3. Accuracy/resolution when all six SRs can illuminate the same threat complex

The six SR sites span a baseline of roughly 4200 km east-west (Boston to Fairbanks) and 2200 km north-south. When all six have line-of-sight to the same object at, say, 800 km altitude, the geometric dilution of precision (GDOP) becomes extraordinarily small.

Simple triangulation geometry: two radars 4000 km apart with 0.1° azimuth accuracy each give cross-range accuracy ≈ (range × beamwidth in radians) / √2 ≈ 800 km × 0.00175 / 1.4 ≈ 1 km at 800 km slant range. With six radars the error collapses to tens of metres in 3D position and a few m/s in velocity after only a few seconds of data. Range resolution on the UHF SRs after pulse compression is ~ (c/(2×bandwidth)) × 100 (compression ratio) — assuming even a modest 1 MHz chirp the compressed pulse is ~1.5 m range resolution. Combined with the angular accuracy from the long baseline, the net can resolve and track individual objects separated by only a few tens of metres anywhere in the midcourse volume over CONUS. A 65 km diameter chaff cloud is therefore resolvable into thousands of individual resolution cells from multiple aspects simultaneously.

4. Six co-located SR+ER pairs on the same long baseline

The ERs are S-band (~3 GHz, λ ≈ 0.1 m) with ~6–7× shorter wavelength than the UHF SRs. Angular resolution therefore improves by roughly the same factor for the same physical aperture, but the ER apertures are much smaller (12.3 m² today, 63.6 m² upgraded). Even in the baseline 3.96 m dish configuration, the ER beam is <0.5° wide versus ~1.3–1.5° on the 30 m UHF array. Once upgraded to 9 m dishes the ER beams become ~0.15–0.2°. Six such radars on the same 4000+ km baseline give sub-10 m position accuracy and sub-metre range resolution (S-band typically has wider instantaneous bandwidth). This is effectively space-object-tracking quality on a continental scale against uncooperative targets.

5. Frequency diversity UHF ↔ S-band

UHF (420–450 MHz): long wavelength, relatively insensitive to shape details, sees travelling-wave and broadside specular returns more strongly, penetrates chaff clouds better, but poor imaging.

S-band (≈2950 MHz): Rayleigh-region scattering from small features becomes significant, nose-on specular from a cone is much weaker, but edge/corner returns are stronger, and range resolution is inherently better for the same fractional bandwidth.

A real warhead and a lightweight replica of identical external shape will have almost identical UHF returns from most aspects but can differ dramatically in S-band signature because the mass distribution affects subtle travelling-wave and creeping-wave terms. Chaff dipoles cut for UHF will be poor reflectors at S-band and vice versa. Simultaneous multi-frequency illumination therefore gives an immediate classification lever that single-frequency nets lack.

6. Value of oblique/side-aspect looks (e.g., attack on Grand Forks)

If the target is Grand Forks itself, the local SR/ER sees the incoming complex almost nose-on until very late (low elevation angles, large grazing). Nose-on is the aspect where sophisticated cone-shaped decoys are designed to mimic the real RV most closely. The other five SR/ER sites, however, see the complex at 30–90° off nose-on because of the Earth’s curvature and their lateral displacement. At those aspects:

Real heavy RVs remain stable; lightweight balloon decoys or tumbling replicas precess or nutate visibly.

Conical specular glints move off the tip.

Chaff clouds appear elongated along the line of sight from the side, making Doppler spread different.

Micro-Doppler from any rotation is far higher when viewed from the side.

Result: even if the local radar is saturated or jammed, the rest of the net discriminates the threat complex perfectly.

8. Reduced network: only Malmstrom SR+ER + Grand Forks ER

Coverage collapses to the northern/central US. Anything aimed west of ~110°W or east of ~90°W arrives with little or no midcourse tracking. Only two illumination geometries (roughly 800 km baseline) → GDOP worsens by factor ~5–7, position accuracy degrades to hundreds of metres, discrimination becomes heavily dependent on frequency diversity alone. Saturation resistance drops dramatically; the threat designer only needs to defeat two radars instead of six. Still better than nothing for attacks on the missile fields themselves, but strategically trivial.

9. Single site: Malmstrom SR+ER only

Now a classic single-radar problem. One geometry, one frequency pair, no triangulation. Accuracy ~ few km at 1000 km altitude, resolution limited to the better of the two radars (~10–20 m). Decoys that mimic the real RV from that single aspect (easy to do) become credible. Chaff corridors aligned with the radar line-of-sight can blind the volume. The system is now vulnerable to 1970s-level penetration aids.

[MKSheppard]

As I mentioned before; the Anti ABM people collapse everything to the classic "single radar" problem.