Pokhran II not fully successful: Scientist - Part-3

[samuel]

Amber,
When solving for various equations, such as fluid equations, we typically non-dimensionalize variables so that they are unit less. Sometimes this involves dividing by typical length scales, say domain size, other times it involves dividing mean densities, mean state etc. depending on the context. These approaches make the state variables non-dimensional. But this practice is only useful within an order of magnitude of the scales of interest. Dividing by a length scale about a 1000KM typically produces huge errors for things happening in the 10KM scale. Solutions will be reasonable somewhere within an order of 1000. For things very small or very large, the results will make some sense (in an average or bulk sort of way), but you can't use it to predict anything realistic. For example, if you were to calculate the peclet number, you might say it is LV/D, where L is typical length, V is typical Velocity, and D some diffusivity. You may stick this peclet number in your equations and derive it with L = 1000KM. People won't typically bother to change its value within an order of magnitude of changes to L and to understand the physical phenomena it is used for, it is also quite un-necesary. But doing this is useless when you want to use the equations to predict something, rather than just understand phenomenology. You may pick some other such number or non-dimensionalization scheme of you are aware of. The idea is the same.


If you say 50KT+/- 10KT that's one thing. But if you say, yeah by this way, I can give you an estimate of 100KT +/-100KT, ok. pretty pictures, practically useless. Yeah, telling us that the future ocean temperature will be anywhere between -45 and +45 degrees warmer is information alright, just what kind no one knows. At the end of the day you got a number and an error around it. That is what you'll use for decision making. If you tell me that number is 1000%, what kind of meaningful decisions do you think one can make assuming these erorrs are distributed uniformly?

S

[shiv]

I think it was Oppenheimer who estimated atmospheric test yields by throwing pieces of paper in the air and measuring how far they were moved by the blast. And I think it was Feynman who estimated yield by fireball size.

But once testing went underground, yield estimation climbed the gum tree.

It was Fermi who threw bits of paper (and his estimate came out to one of the best)..
Feynman was the one who (prob. the only one) saw the fireball by directly looking at it (standing behind a windshield of jeep - and confident that UV rays would be absorbed by the glass - He did not look in other direction with dark glasses on as they were ordered to do - At least that is what RPF used to tell)

Also, BTW one of the most accurate estimations of yield of Hiroshima and Nagasaki bombs came from post event measurements (courtesy of famous Japanese nuclear physicist Nishijima) at the site(s).

OK - correction noted thanks

[shiv]

From what little I have read, it seems to me, that the political and deterrence "groups" has adapted to the inability of the scicom to come up with precise means/methods of predicting a yield. So, right now, it really does not matter. Those depending on the scicom are more than happy with a +/-10% and I would not be surprised if they will tag along even with +/-25% or even more.

Yes and if one decides to use both eyes and keep them both looking for information in all directions, the stuff that comes up is fascinating.

The NoKo yield paper linked below speaks of all these previously discussed uncertainties about yield estimation. You can read the paper yourself but I will only quote one para from that paper which fits in with what AmberG said about Gen Groves (or Grover?)

http://belfercenter.ksg.harvard.edu/fil ... 07_Hui.pdf

One week after the first fission explosion on 16 July
1945, Robert Oppenheimer wrote to General Leslie R. Groves' deputy and described the
expected performance of the Trinity device in combat, as Mark rephrased, “With the
Trinity Implosion assembly system and the grade of plutonium employed, the probability
was 88 percent that the device would survive long enough without a chain being initiated
that it would provide the nominal yield of 20 kilotons; about 94 percent that it would
survive long enough that the yield would be greater than 5 kilotons (one quarter of the
nominal yield); about 98 percent that it would provide a yield in excess of one kiloton.
Only in two percent of all firings would the chain be initiated so early that the energy
release would be between the fizzle yield and one kiloton.”

[samuel]

There are three distinct things one must typically consider:

1. What the theory says. In many problems this typically is useful to understand the phenomena, not predict it. The error bars here are order of magnitude type. It is amazing to hear these physicists talk about things and say oh to an order of magnitude. Most engineers say, balls man, what good is that?

2. What the method implementing theory does when suitable system identification is carried out on the system. In many problems this is typically useful to predict the "mean field" or "average" response or effect. The error bars here are varied but are typically good in the vicinty of the parameters where system was identified.

3. Using both method and observation to infer in an ongoing way (many hypotheses and tests). This allows us to predict the effect or response over a broad range of conditions and is best useful for decision making. Will my bomb work, how many are needed etc. It is still limited in that it is a useful method for interpolating, not extrapolating.

1 is typical physics, 2 is basic engineering, 3 is where the world is going from economics to biology.

4. Finally, the thing we care most about, extrapolating. Nature is not scale invariant, fractals not withstanding. Things that work on one scale don't at other. Climate changes, 50KT bomb does its dynamics differently than a 500KT bomb. When a scientist extrapolates, raise your eyebrows. When a politician does it, raise your eyebrows. When you want to do it, consult the horoscope section

JMT
S

[shiv]

The other question that should be asked in view of what has happened on this forum is that if yield estimation is so inaccurate, how does that make Chidambaram's estimate of yield right and Santhanam's wrong?

Read the Outlook article:
http://www.outlookindia.com/article.aspx?262027

Regarding the fission device

With respect to the fission bomb, which gave more than 20 KT for sure,

About the TN devise Santhanam says:

About 20 to 25 KT.

From another source Santhanam says about the TN device:
http://beta.thehindu.com/opinion/op-ed/article21311.ece

The fission bomb yield from the DRDO’s seismic instrumentation was 25 +2 kiloton and left a crater 25 metres in diameter. If the TN device had really worked with a yield of 50 +2 kt, it should have left a crater almost 70 metres in diameter.

If you add up the figures of Santhanam's words he is splitting up the S1("Thermonuclear) and S2 (fission) devices differently from BARC

Santhanam says that the fission device yielded 25+2 (Take that as 25+/-2) so 23 to 27 kt (never mind how he arrived at that accuracy). He says that the TN device yielded 20-25 kt

What Santhanam is saying is that the total yield of the May 11 tests was a minimum of 43 kt and a maximum of 52 kt. In other words Santhanam is not significantly disputing the total yield of the May 11 tests as quoted by BARC. His figures add up to match what BARC said.

But what Santhanam is saying is that the fission device yielded 25 kt and TN device yielded 20-25 kt. What BARC is saying is that the fission device of 1974 vintage gave 15 kt and the rest came from the TN device.

[samuel]

Right, so is the fundamental disagreement about

50 = S1 + S2 + S3

where there are infinitely many solutions without further evidence?
(I know I heard something about wave interference etc. here, but is the basic disagreement of the type don't know how to or how much to ascribe to each one of these?)

S

[Amber G.]

Samuel - In Physics it quite common to "simplify" a formula and make constant like e=1 (which in MKS unit is 1.6*10^(-19) coulomb) - without any unit and thus dimensionless .. or h=1 (plank's constant) or velocity of light c = 1.. I have even seen articles where e=h=c=1.

But that has *nothing* to do with uncertainty or 'order of magnitude' calculation. To be honest, I have never seen ( this is 3 decades of teaching physics) the technique you presented associated with that (to be sure, it could be my fault, I simply could not follow what you are trying to say)

Sure when one estimates order of magnitude, one estimates order of magnitude. If one wants to makes some variables/parameters dimensionless one does that. There is nothing appox or mathematically wrong with that, (if done correctly From what I understand it has nothing to do with increasing or decreeing preciseness.

For yield "centered around 100KT necessarily (or quite often) does not mean, say 100KT+- 20KT..( If some relationship in the formula is logarithmic - just to give an example the value could be , say (just an example) 50KT to 200KT with " most prob" value at 100KT ...

(Or more direct example... 10^3=1000 but 10^(3+-.1) is between (10^2.9=794 while 10^3.1=1259 and it would be wrong to put that kind of range as 1000+- a constant)

Anyway this may be getting tomuch OT so, unless some one asks, I will shutup.

[shiv]

Right, so is the fundamental disagreement about

50 = S1 + S2 + S3

where there are infinitely many solutions without further evidence?
(I know I heard something about wave interference etc. here, but is the basic disagreement of the type don't know how to or how much to ascribe to each one of these?)

S

OK let me add further thoughts. I would ignore S3 as sub kiloton and less than 5% of the total yield.

If you were testing 2 nuclear devices of low yield placed in shafts A and B oriented East-West about 1000 meters apart where would you place your accelerometers and other close-in measuring equipment, knowing that they could be destroyed by the shock of either of the explosions?

Would you have two separate rings of measuring devices - in a ring around each shaft? How many?

Would you place any measuring device in the region that is roughly between the two shafts? Or would you simple place a ring of measuring devices around the twin shafts in a rough oval shape where no measuring device could be crushed from both sided by shock waves arriving from both sides?

Or would you simply place a line of accelerometers (and whatever) East of shaft A to pick up shaft A signals firs and West of shaft B to pick up shaft B signals first?

[samuel]

Ah but nondimensional numbers are really common in many studies of fluids -- peclet, prandtl, froude, reynolds, raleigh, ekman...the list of numbers is very long and you might want to, for example, look at Batchelor's book on fluids. These are NOT universal constants (or some scaling of them). They are ways to characterize something about the dynamics, thermodynamics etc. so that things can appropriately be understood in a scale invariant way.

it would be somewhat silly thing to assume universal constants as contributing to uncertainty I think, but parameterization of equations has everything (a lot) to do with uncertainty. I am pretty sure this is OT now, but I am happy to point in directions I am aware of...

S

[samuel]

We should just consider what kind of interference pattern is expected to decide how to place the sensors. But what do we know of what sensors they used?

OK, let's drop S3.
Can we assume they were exactly synchronized and what was the exact geometry of these blasts (sorry, I'll look up). Arun said something about an ARC sensor at 30o phase.


PS: Have a look here:
http://id.mind.net/~zona/mstm/physics/w ... ence1.html

[Amber G.]

Shiv - Thanks for the summary.
To add: "yield estimation" accuracy (obviously) depends on data. KS and RC obviously have access to more fine (and reliable) data then crude data available in open source.

But what Santhanam is saying is that the fission device yielded 25 kt and TN device yielded 20-25 kt. What BARC is saying is that the fission device of 1974 vintage gave 15 kt and the rest came from the TN device

.

The split between fusion/fisson, IMO (my guess only) could not come from seismic readings, crater size or sacrificial xray sensors etc...If one knows the design of the device and the isotope contents measured in cavity (which is obviously classified information) one would obviously know..but without that I don't know how one can estimate the ratio of fusion/fission yield.

Of course, above para should be taken FWIW - I am not (by any means) an expert in this field and have very rudimentary knowledge of how one does measure such things...

[NRao]

Amber/Samuel,

This has been posted before, so apologies if you have already checked it out, but the following article has some numbers that might quench you thirst and reduce the fire here:

Cratering Phenomenology and Yield Estimation

It has a graph at the very end that SEEMS to be a standard (from my jaunts across the web based silk route).

Two items of interest:
1) This author SEEMS to have approximated some, so, obviously the result set also will be approx, and
2) It does have examples that are directly applied to the S1 (TN) at POK-II. This is of great importance since it does contradict Sathanam WRT crater

#1 really should not impact the debate WRT Santhanam, but does (obviously) provide a slightly different picture and therefore, one may argue, a computed yield.

Amber,

In R=K*Y^(1/n), n varies from 3.0 to 3.4, Y = yield, K is based on the soil composition and the graph in the above article will throw some "light" on that (beyond that it is guesstimate and approximation), R is computed.

Samuel,

That graph (in the paper) - from a statistical PoV - is the biggest joke. So far, it looks like - to me at least - it is a non-linear regression based on some 4-6 data points.

The reason I call it a joke is that the entire world is basing their entire decision process AND deterrence to some extent on THIS graph!!!!!!!

Think about that. That is how far these yahoos are from reality!!!! Millions of lives.

Which is why i would not debate too long on "deterrence". It is so fluid and so nebulous that it can be considered to be very, very ridiculous.

[samuel]

NRao,
I studied this paper in some detail and was struck by how much things were being based on those curves in the back...I actually agree with the bottom of your message for me without any serious reservation.
I will try to find my post on it...but basically, need more tests. This is too ad-hoc methodology.

S

[NRao]

This is too ad-hoc methodology.

Absolutely ........................................................................ if you are prospecting for O&G.

But it seems to me that when it comes to testing nukes and computing their yield pretty much anything goes. Within reason i would imagine.

Take a closer look at the same graph. There are some data points that are stand-alone!!!!!!!!!!!!!!!!!!!!!

Pre-Schooner, Buckboard, Buggy. The last one is part of the Hard-rock (POK-II S1) graph!!!!!

However, to your point of "ad-hoc methodology" I am not sure that it can be corrected in the field. In the lab, lasers, etc, I do not know - perhaps they have a solution.



BUT, the bottom line is that the BARC team has computed everything based n these equations (RR article). This is standard proc for this type work.

What I am suggesting is that we stick to these equations and plug any new (RR article) data points we get and see if it resolves anything between Santhanam and GoI team.

[shiv]

In the interests of truth and reconciliation, bearing in mind the passionate and sometimes angry discussions that we have had in the last 130 plus pages of this discussion I would like make the following post;

Three data points were put forward by 3 members in the discussions

1) That the May 11 1998 TN device (S1() was meant to have a yield of 200 kt
2) That the actual design of the fizzled device was known and it was known to have a fissionable tamper (of the type that devices of yields in the hundred of kt and even megatons have)
3) That the fizzled thermonuclear device was actually expected to yield 1 megaton.

The three people who said this were, as far as my recall goes, in total agreement with the statements of Dr. K Santhanam.

In this connection I want to point out Dr. Santhanam's statement to the media in 2009:
http://www.outlookindia.com/article.aspx?262027

Between 1995 and May 1998, the device for a thermonuclear test was (made) ready. It was a device for 45 KT; anything above it would have caused a venting of radiation that would have been a violation of the Partial Test Ban Treaty to which India is a party.

Given the standing that Dr Santhanam has in India - I estimate that no BRF member has even a fraction of that standing in terms of knowledge of what happened in Pokhran on May 11 1998 compared with Dr. Santhanam. No prizes fro guessing whose statement I personally would chose to believe as credible.

[shiv]

PS: Have a look here:
http://id.mind.net/~zona/mstm/physics/w ... ence1.html

Samuel I am not looking at interference at all.

I am thinking of measuring devices that are placed so close that they are destroyed by the shock as suggested in some statement by RC - which I need to dig up.

That means that the devices are hardly going to pick up both shock waves with a suitable time lag. They will measure just the nearby one up to a cut off level and then die.

Unless I am mistaken CORRTEX is designed around actual crushing of cable - but there is some report somewhere that CORRTEX for Pokhran 1998 did not work. (Need to search for that).

[Bade]

There are three distinct things one must typically consider:

1. What the theory says. In many problems this typically is useful to understand the phenomena, not predict it. The error bars here are order of magnitude type. It is amazing to hear these physicists talk about things and say oh to an order of magnitude. Most engineers say, balls man, what good is that?

No, physics can predict as in the classical domain very accurately. What you cannot do is measure the prediction exactly. As an example the momentum and energy conservation predicts the track of a fundamental particle exactly, but the issue is with measuring the parameters exactly which introduces both statistical as well as systematic errors.

In the domain of nuclear physics, where classical ideas are not relevant, then ideas of engineering which is always in the classical domain (even with non-linear phenomenon) are useless. This is where the engineer is incapable of understanding his limitations of 'predicting' the outcome. So, any amount of precision engineering is not going to help.

2. What the method implementing theory does when suitable system identification is carried out on the system. In many problems this is typically useful to predict the "mean field" or "average" response or effect. The error bars here are varied but are typically good in the vicinty of the parameters where system was identified.

The typical engineer limits himself to the classical domain, and proclaims quite proudly that he measures something to 0.000000001% accuracy, while avoiding the domain where phenomenon are probabilistic in nature.

[NRao]

Hopefully I can add to the confusion.

The purported Paki Pu in NoKo test, is that a test for physicists, engineers or politicians? I think the yield would change from profession to profession. With an (acceptable?) error rate of 50+%.

[sugriva]

....

Bade-Saab,
Unlettered 12th fail gawar only here saar. So please bear for a while with this bumpkin. I don't know much about yields and how they are measured and will admit that I don't follow most of the technical arguments posted on this thread. However I have a simple Rakshak question, the answer to which I seek purely from the POV of physics. If tomorrow India wants to use an allegedly 20kt tactical nuclear weapon on ARS, somewhere near Multan, in the course of war,

1. is there a finite probability that the bomb will not explode at all purely because of the nature of nuclear physics . Assume that all of the engineering, in the classical domain of course, works and product probability that the engineering fails is probably 1/10^googol or some such number

2. is the yield of the bomb probabilistic? say 20kts with probability of 50%, 10 kts with p=90 and 2 kts with p=99.9999999999 %, all of which is because of the nature of nuclear physics, which is not classical

If so what is the use of such a weapon. In the context of the situation that I bring up, perhaps 20kt is what will be required to halt ARS and anything less may mean, breakfast at Jodhpur, lunch at Jaipur and dinner at Delhi for the TSPA.

General question to all rakshaks.
Do the end users, lets say the IA in this case, need to measure the yield of the nuclear weapon that it just used? If so what for and if not why not? If end users do indeed measure yields, post event, how do they assure themselves of any level of certainty in the measurement?

[Bade]

If tomorrow India wants to use an allegedly 20kt tactical nuclear weapon on ARS, somewhere near Multan, in the course of war,

1. is there a finite probability that the bomb will not explode at all purely because of the nature of nuclear physics . Assume that all of the engineering, in the classical domain of course, works and product probability that the engineering fails is probably 1/10^googol or some such number

It has already been mentioned in previous posts, and even in one of the newspaper articles that there is a 10% uncertainty attributed to various nuclear physics processes involved in the chain reaction, like in the measurement of decay channel properties. I think it was in the RR article if I recall correctly. You then also have various engineering uncertainty estimates on top of this due to the complexity of the device ( machining accuracy etc perhaps). So there is a no better than (10+x)% uncertainty which will propagate to the yield estimates...even the most accurate ones. In this context, the overall yield estimate of 50+/-10 kt was seen as in agreement for laymen use, 10kt being around 20% uncertainty.

As for not explode at all, which is the equivalent of a <10% of the value of the expected or design yield. There were even links posted (by Shiv I think) which had numbers as low as 3% of design yield as an accepted definition of a fizzle in the context of non-Indian test yields. So the same should hold for us too.

2. is the yield of the bomb probabilistic? say 20kts with probability of 50%, 10 kts with p=90 and 2 kts with p=99.9999999999 %, all of which is because of the nature of nuclear physics, which is not classical

So if the design is such that the expected yield when used is 50+/-10kt, it means there is 99% probability that the device when exploded will give you at the lower limit something better than 50-3*10=20Kt. Hence if you exploded the device a 100 times, only once will you expect to get it lower than 20kt.

This is not so bad as an engineer thinks, since the probability of a limited number of mountain divisions stopping a PRC/TSP assault on its own is nowhere near even >50%, with so many uncertainties being part of the process.

So if I were a general, I would still be happy with a device which works with 90% probability to explode at a desired yield value(even if lower than the device expected design value), than have a mountain division which will work only with 50% or lower in reality probability of achieving its desired yield in battle effectiveness.

So the context is very important when talking about the usefulness of any scheme, not the absolute value of the uncertainty itself.

[Amber G.]

NRao

Amber,

In R=K*Y^(1/n), n varies from 3.0 to 3.4, Y = yield, K is based on the soil composition and the graph in the above article will throw some "light" on that (beyond that it is guesstimate and approximation), R is computed.

Thanks. I saw that before too, but as I said I was looking for "current best values" (as applicable to S1/2). I n was hoping Ramana would give explicit values..

BTW (By choosing K between 12-14 and n between 3 and 3.4 and R as 40+-4 (36 to 44), Y could be any value between 17KT to 83 KT. ... according to that formula... (Interesting that equations in that paper uses 3 sig figure for yield in one example... )..

[NRao]

Since my foot ball team has the week off, did some comps here (based on the Sunder equation set and my own "curve" - for hard rock and NE):

1) Used MS Excel to build a curve (polynomial) based on the five data points in the Toman paper (hey, I get three of those five on this curve - have a winner)
2) Extrapolated the curve to extend to X=0 - again used MS Excel
3) Had MS Excel provide the equation (Got my monies worth by now for Excel)
4) Here are some prelim value that need to be peer reviewed:

All comps based on 45 Kt:

a) Retarc at 171 meters deep
b) Max radius of 124.89 meters @ 128.36 meters depth
c) 70 meter radius crater at 99.603 meters and then again at 157.59 meters
d) At depth of >200 meters (actually beyond 172 meters I would imagine) I would not expect shaft mouth + A-frame to budge - no reason as far as I can see

I will post the following, but reserve the right to withdraw it:
e) Max kt permissible @ 230 meters without venting (I think) is 326 Kt
f) IF e) holds, then it will form a crater of 218 Meter radius

[shiv]

General question to all rakshaks.
Do the end users, lets say the IA in this case, need to measure the yield of the nuclear weapon that it just used? If so what for and if not why not? If end users do indeed measure yields, post event, how do they assure themselves of any level of certainty in the measurement?

I think all armed forces who use any weapon would be interested in finding out the effects of what they have used and how much it will impact any war. Unfortunately the nature of a nuclear explosion does not lend itself to easy yield measurement after the event other than by gross generalizations.

[Umrao Das]

No system will give you a prob of .95 all four critical subsystems the overall probability of theentire system will work as envisaged is reduced to .8145 i.e 81.45%.

Even RC can not guarantee sizzle or fizzle.

My one reupee that is Two cents equivalent.


http://wws.princeton.edu/webmedia/index.xml

Speaker: Bharat Karnad, a Research Professor in National Security Studies at the Centre for Policy Research in New Delhi Introduction by Aaron L. Friedberg, Professor of Politics and International Affairs, Woodrow Wilson School, Princeton University

[Satya_anveshi]

Please pardon my blasphemy of quoting a paki source in these matters but here it is anyway:

Easy access to N-material for India

In US estimate, Pakistan has 70 to 90 nuclear warheads as against current Indian stock of 200, over and above small tactical nukes.

With free access to nuclear materials from international suppliers group, India will be in a position to produce 40 nukes annually.

[Kanson]

No system will give you a prob of .95 all four critical subsystems the overall probability of theentire system will work as envisaged is reduced to .8145 i.e 81.45%.

Even RC can not guarantee sizzle or fizzle.

Statistics is shitty science unless you know what you are doing. Thats why there is another branch called 'Engineering'. It happens someone gave a ~99% Kill probability for PAD missile system.

[NRao]

Even RC can not guarantee sizzle or fizzle.

Nor has he.

The point being over time the probability of success increases, but never reaches 100%.

However, the good news is that such things apply to all participants. Anyone who builds a missiles and hopes to mate a nuke faces these headaches.

It is said that civilian airplanes went from 4 engines to 2 because the reliability increased. Even in the dual engined config, the assumption is that even if one fails the plane can fly with the other. But there have been cases where both have failed - not because the "other" engine failed, but due to pilot error!!!

[Gagan]

Good to see that this thread is slowly dying its natural death.

[shiv]

I have no idea. I am not a physicist - certainly not a nuclear physicist. But surely even for fusion and the release of fusion neutrons there must be a probability of numbers produced under certain conditions of pressure and temperature. After all it has been stated right here in this thread that fusion has been produced even from conventional explosive with the Russkies causing a yield of 57 milligrams of TNT using conventional explosive alone.

Isn't that what "equations of state" are all about?

Can someone enlighten moderately?

[Sanku]

Good to see that this thread is slowly dying its natural death.

No just a hiatus, it will come back after Deepavali I think.

[NRao]

Are we not expecting fireworks after the three week hibernation? A week or so to go?

[NRao]

A nuclear test would damage India-US ties, warns NYT

NYTimes :: editorial ::Just Say No

[Amber G.]

3) Had MS Excel provide the equation (Got my monies worth by now for Excel)
4) Here are some prelim value that need to be peer reviewed:

All comps based on 45 Kt:

a) Retarc at 171 meters deep
b) Max radius of 124.89 meters @ 128.36 meters depth
c) 70 meter radius crater at 99.603 meters and then again at 157.59 meters
d) At depth of >200 meters (actually beyond 172 meters I would imagine) I would not expect shaft mouth + A-frame to budge - no reason as far as I can see

I will post the following, but reserve the right to withdraw it:
e) Max kt permissible @ 230 meters without venting (I think) is 326 Kt
f) IF e) holds, then it will form a crater of 218 Meter radius

Can you tell what exactly the equation (and parameter values) you were using and how precise /trustworthy/accurate (in your opinion - or best estimations) those values are? Thanks.

[hnair]

NYTimes :: editorial ::Just Say No

Indian nuclear scientists are trying to bully their government into testing a nuclear weapon. That would be a huge setback — for India’s relations with Washington, for the battle against terrorists, and for global efforts to halt the spread of nuclear weapons.

This is a nonsense article by the establishment mouthpiece

1) how does it affect India's battle against terrorists? Before POKII in 1998, there were terrorists attacking Indians. And we fought them then as we fought them now. They did not change, neither did we. The war in Afghanistan is not **our** war, as we are not involved on the ground.

2)

far more important battle against the Taliban and other extremists inside their country and along their border with Afghanistan

Who decides which is the more important battle for Indians? Neither Indians (nor Pakistanis) believe this particular is anything important, other than an opportunity for Pak Army to extract money out of American taxpayers in these troubled times.

3) Where is this sort of bellyaching, let alone some serious action on the new plutonium facilities of Pakistan? Where is the relationship between those new facilities of Pakistan and concerns of Indian scientific community?

4)

He insisted that his agency has confidence in its ability to get the weapons data it needs by conducting simulated tests. He should keep insisting.

NYTimes should stop using Shree Kakodkar's specific rejoinder to Shree K Santhanam. His answer in this quote does not mean he loves your policies and currently agrees with your CTBT POV. This editorial is an effort to divide the house by hitching their proliferation-soiled name with one side in an internal dissension of another country.

This editorial is only a step away from "Indian nuclear scientists are envious of the freedoms we enjoy" and the final "What would Jesus say about India's behavior?" .

[Austin]

The article is an indication about the nervousness that the US Establishment is now under , perhaps they have some intel on what GOI is thinking , and in a way striking back on the implications of testing , sanctions and impact on (US GWOT ? ) for India.

I think Santy article got some one thinking and thinking hard ,and finally lets keep Jesus away from this debate

[NRao]

Amber,

If you do not mind I would like to take it off-line. It would be a good "peer review" too.

indicgroup at netscape . net

[Amber G.]

Sent a mail - Thanks,
(Peer-review would be hard as I don't have any back ground in yield measurements ...Nevertheless curious to see and learn )

[Gagan]

I am thinking, that MMS is at least considering a test series.
There is a suspicion that KS and the regular establishment are really a tag team who aim to soften international opinion, test the waters so to say, for an upcoming test series by India.

Hnair saar,
India IS involved on the ground in the war on terror. Have you forgotten the 100,000 embassies that India has in afpak?

What this article is suggesting, is that India had better behave, otherwise washington will not hold pakistan's hands should it decide to do another mumbai. That is a poor choice of words and will be of not much deterrence value to any government in India.

I think that India calculates that despite the Noble to Pres. Obama, and his eagerness henceforth to strictly go after his UN N weapons reduction plan, there is a sufficient window for India to complete its strategic weapons program.

I highlight the reasons I think are important, please share your opinions:
1. Obama is a peacenick, and righteous individual. Although he will say he is deeply disappointed at an Indian nuclear test, I feel that he will also understand India's unique situation. He will ultimately be the person who will ultimately limit the scope of any sanctions on India.
2. Obama's peace plan talks of "no nuclear testing at all by anyone after the end of 2010. That'll include the US.
3. The atlanticists and the NPA genuinely fear a world where there is no nuclear testing. What better oppertunity than India breaking out, taking all the blame for a resumption of testing. The P-5 are not about to break out into testing any time soon, there is no sufficient threat to any of them at present. The US, Russia, and China are eager to test, the french won't mind another final round themselves.
4. We have already debated to death, why any sanctions on India will not be very damaging or their effects long lasting. The NPA's tried their level best at a time when their star was on the ascendent in the US - and failed spectacularly in putting India in the same doghouse they had india in after '74 or they have had Iraq, Iran and NoKo all these years. The bottom line seems to hold true:

Nations the size of India don't get really sanctioned, they get accommodated.

[Amber G.]

OPed From NY times:
Just Say NO

Indian nuclear scientists are trying to bully their government into testing a nuclear weapon. That would be a huge setback — for India’s relations with Washington, for the battle against terrorists, and for global efforts to halt the spread of nuclear weapons
<snip>
So far, New Delhi does not seem to be taking Mr. Santhanam’s bait. “India does not need to carry any more nuclear tests,” the Indian Atomic Energy Commission chairman, Anil Kakodkar, said last month. He insisted that his agency has confidence in its ability to get the weapons data it needs by conducting simulated tests. He should keep insisting.

[pgbhat]

OPed From NY times:
Just Say NO

Comments are more interesting rather than the article itself.