DAW comments:
>A Possible >200 GeV Counterpart to GRB 970417a
Best resolved in another venue.
>I do not particularly like extensive use of VHE. We use it as if it has a well-
>established meaning, but as indicated in my message of a few moments ago, the
>EGRET GRB paper has muddied its meaning. Furthermore, in writing an archival
>paper, we should anticipate that our audience includes folks 20 or 50 years from
>now. (We can dream, anyway.) From their point of view, we should make explicit
>statements about energy (GeV, TeV) as much as possible, rather than falling back
>on the presently accepted meaning of a piece of jargon (VHE). There are places
>where I think saying "very high energy" is what works best, so I don't feel that
>we need to expunge the phase completely. Those places are where we are speaking
>qualitatively and generally. We should not
>rely on this term to define the spectral band. If we have to define the "VHE"
>acronym, we are probably using it too much. So I do think that any surviving
>VHE should be replaced with the words "very high energy."
In many place VHE has been chnged to TeV where still appropriate,
VHE is changed to very high energy where appropriate.
>sec. 1 (s1), par. 1 (p1): There needs to be a reference for the statement i
>that EGRET has seen 8 bursts.
refs added
>s1,p2: ...intensity of the the sub-MeV emission detected by BATSE. (BATSE
>does not emit anything.)
change made.
>s1,p3: The R. Ong review, Phys. Rep. 305 (1998) 93 should also be cited.
citation added
>s1,p3: ...therefore able to detect small fluxes... (small sounds better i
>than low to me)
change made.
>s1,p6: ...a limited prototype, (COMMA) operated ... to May 1998, (no new sentence)
>during which 54 ...
comma added, and sentence reworded.
>s1,p6: ... but concentrates more specifically on (remove "the significance of")
>one GRB candidate.
change made.
>s2: The PMTs detected... (instead of These sensors detected.... Here clarity
>wins over avoiding repetition)
change made.
>s: ...the interaction OF THE SHOWER with the water transfers--via pair
>production and Compton scattering--the energy of the abundant $\gamma$-rays
>in the air shower to relativistic charged paritcles, ... (the pair production
>phrase is presently misplaced in the sentence)
change made.
>s2: ...detection of the shower particleS incident on the reservoir...
change made.
>s2: ...almost all these trigger WERE caused ... (to match case of rest of paragraph)
change made.
>s2: (remove These ans other tests showers that) The instrument had an angular...
change made.
>s3,p1: A search ... background OF cosmic rays...BATSE $\gamma$-ray burst. (period)
>Bursts with 45...Milagrito WERE CONSIDERED; this ...
"were considered" added. sentence reworded into 3 sentences for clarity.
>s3,p2: ...showers must be estimated, (comma)...
change made.
>s3,p2: ...0.2 apart. EACH BIN IN THE ACTUAL DATA WAS COMPARED TO THE CORRESPONDING
>BIN IN THE BACKGROUND MAP, AND the Poisson...of events in THE (not each 1.6)
>bin ..was calculated. (period. new sentence) The bin with ...
change made.
>s3,p3: ...54 $\gamma$-ray bursts obstaind by this method is given in Figure 1.
>(instead of the phrase about correcting each for the size, in order to emphasize
>that there is not some further correction applied before making the plot)
change made.
>As I get into the second half of the paper, there is a regular tendency to drop
>our guard and start talking about the burst as if we know it to be real. In
>three places this should be fixed:
>1) Sec 3, par. 5: To allow for the positional uncertainty of the TeV \gamma-ray
>CANDIDATE (not detection) during T90...
change made
>2) Sec 4, par. 3: If the Milagrito candidate is indeed a detection of GRB 970417a,
>the the opacity allows us to place an upper limit on the redshift of ~0.3, making
>it a relatively nearby event.
change made.
>3) Sec 5, par. 1: If we have detecting a high-energy counterpart to the burst,
>the fluence above 10 GeV...
changed to: The observed excess corresponds to a fluence above 10 GeV
this languge does not conclude that we have made a detection, but isn't as
wishy washy as saying "if we are right,..." all the time.
>sec 3, par. 4: Of the 54 bursts, one, (comma) GRB970417a,...
change made.
>s3,p4: ...to be a relatively weak burst, but OTHERWISE typical, ... (to
>answer the question "Is it weak or is it typical?")
change made.
>s3,p4: ...and leads to A serach region ...
change made.
>s3,p4: ...both completely consistent with, and much better determined (no comma)
>than, (comma) ...
change made.
>s3,p4: ...2.9 \times 10^{-8} ... (this number does not have three significant
>figures)
change made.
>s3,p4: ..was found by THE Monte Carlo simulation DESCRIBED ABOVE to be 2.8E-5.
change made.
>s3,p5: Although the initial search was limited to T90, UPON IDENTIFYING GRB970417a
>AS A CANDIDATE, longer time intervals ...
change made.
>s3,p5: ...are shown in Figure 3. ("Figure" should be capitalized here
>and elsewhere. There are several other instances which I shall assume everyone
>are capable of finding.)
change made.
>s4,p2: Need to decide whether we are using the scaler data or not. If so, we
>at least need more info disseminated within the collaboration about it. If not,
>this paragraph should be removed now so we can start suturing the wound.
paragraph removed.
>s4,p3: ...at cosmological distances ARE (actually were, but not will be)
>absorbed...
changed made.
>s4,p3: See comment at top about rewording upper limit on z sentence.
comment addressed above.
>s5,p1: The sentence about two independent analyses MUST GO. It sets up the
>conference papers as the original reference for this work, which is not our
>intention. Among other things, it gives careful referee grounds to reject the
>paper as not being original work. This is also a very peculiar place to wedge
>the comment in. If we want to say something along these lines, it belongs at
>the end of section 3. Something along the lines of: "Details of a somewhat
>different analysis which yields consistent results with those reported here,
>as well as more detailed results from the other 53 bursts, will be described
>elsewhere (reference to Isabel's Ph.D. thesis, in preparation)."
change made.
>s5,p1: See comment at top about adding a conditional introductory phrase to the
>fluence result.
changed. see above.
>Fig. 1 caption: no comma after probabilities; period at the very end.
change made.
>Fig. 3 caption: The histogram showS (not showed) ... second intervals. (period)
>(b) THE MILAGRITO DATA integrated... seconds. (period at the very end)
change made.
>Fig. 4: Remove scaler text from plot and replace with a shaded region for the
>excluded zone, together with caption to explain. See my comments on the earlier
>draft. Period at the very end of the caption.
plot remade without the scaler section.
>Ref. 3. Write on MNRAS more fully, in Ap. J. style.
ApJ style does not write out MNRAS
>Ref. 6. Capitalize Physics.
change made.
>Ref. 8.5 Add Ong review article.
ref added.
>Ref. 10. Astrophys. J. **Lett.**; put in full reference
change made.
>Ref. 14 and 15. Remove.
removed.
>The statement in draft D5 (sec. 1, para. 4) that the HEGRA emission was
>in a time interval longer than the sub-MeV emission is not correct.
>The most signficant excess seen by HEGRA was a 4 minute bin starting
>at UT 22:45:21. The WATCH start and T90 for this burst are 22:46:24
>and 282 s (see astro-ph/9708156). The time intervals don't
>overlap completely, but there is the whole business where HEGRA
>implies the WATCH time might be wrong. In any event, the HEGRA
>duration is SHORTER and if anything slightly BEFORE the WATCH burst.
>HEGRA calls it a coincident detection, and I think we should too:
>
>...and one burst showed an excess over background coincident in time with the
>sub-MeV emission. But the observed position was somewhat offset from the
>satellite measurements and was not claimed as a firm detection (12).
>
>(saying the excess was not consistent with the satellite measurements makes it
>sound like the discrepancy was in the flux, rather than the position)
change made.
Begin Cy's comments:
>ABSTRACT
>Milagrito, a detector sensitive to ** very high energy (VHE)***gamma-rays
>monitored the ...
change made.
>The excess **has a*** chance probability ... **"probabilioty"
>is a misspelling
change made.
>No **significant correlations were detected from the other
>bursts." (this is better than "remaining") **
change made.
>1. Introduction
>1st para: Need references to BATSE and EGRET results.
refs added.
>Rewrite 1st sentence: VHE gamma-ray emission may not be observable for
>sources at redshifts much greater than z=0.5 because of pair production
>with infrared extragalactic background photons (5; 6).
change made.
>Recent measurements of several gamma-ray bursts with optical afterglows
>have shown **note typo redsifts** of z > 0.8 with only one exception
>(REFS ?). These measurements indicate that the measured intensity of the
>gamma-ray emission is not well **remove "-"** correlated with
>distance. This suggests that observation of TeV emission from a gamma-ray
>burst, which requires a relatively nearby source, may not be correlated
>with the intensity of the **remove sub-MeV** BATSE emission.
typo fixed, - removed.
"sub-MeV" sets the energy scale for BATSE, and contrasts it
with our TeV energy scale. The authors feel that its inclustion
adds to the paper.
>At energies greater than 30 GeV, photon fluxes from most astrophysical
>sources become too low for the satellite-based experiments to detect
>**because of their
>small sensitive areas. Only ground based experiments have areas large
>enough to detect these sources ** by observing ...
change made.
>** VHE gamma-ray emission** from astrophysical sources has been...
>... resulting in a duty cycle of **~10%.**
>...No detections of GRBs have been reported, but the narrow...
change made.
>At higher energies, wide field instruments
>** what is this? You are talking about Tibet and should call this
>an EAS array of widely-spaced scintillators or something**
>have reported **a** possibly significant deviation
change made. Tibet collab. substituted for wide field instruments.
The authors see no reason to describe in detail the Tibet detector's
configuration, as it would later necessitate a simuilar description
for AIROBICC.
>** Two GRBs occurred with the field of view of the HEGRA AIROBICC
>Cherenkov array, one of which showed an excess** ...
>However, the location of the excess was not consistent with satellite
>measurements so this was not claimed ... **
change made.
>Milagrito (11), a limited **what does this mean? I'd remove
>"limited"** prototype ...
>**During this time interval, 54 GRBs detected by BATSE were within the
>field of view of Milagrito. **
limited removed. change made.
>2. The Milagrito Detector
>Milagrito consisted of ** a planar array of 228 8-inch photomultiplier
>tubes (PMTs) submerged in the purified water of a covered reservoir. The
>PMTs were located on square grid with spacing 2.8 m, covering a total
>area of 1500 m**2.**
change made.
>**The PMTs detected the Cherenkov light produced as charged shower**
change made.
>...
>**In addition, Cherenkov light is produced by the relativistic charged
>particles resulting from the interaction of the abundant gamma-rays in
>the air shower with the water via pair production and Compton
>scattering.**
change made.
>...
>The continuous medium allows for the efficient detection of ** shower
>particles incident on the reservoir**
>...
change made.
>** The trigger required > 100 PMTs registering at least one
>photoelectron within a 300 ns time window. **
change made.
>...
> ** almost all of these triggers were caused by **
>...
change made.
>The ability of ** Milagrito** to detect VHE
>...
text changed as response to other comments.
>3. Observations and Results
>A search was conducted for **excess events above those due to **
>isotropic background cosmic rays in the Milagrito **data coincident with
>each** BATSE
change made. "each" not added as it implies that we looked at all
BATSE GRB, and not the subset in out FOV.
>...
>** Milagrito had the best combination of sensitivity and low effective
>energy threshold in this angular region.** ...
>This time period, known as T90, is that in which
>the BATSE flux rose from 5% to 95% of its total **[I doubt if this is
>true. I believe that T90 is the timer interval that contains 90% of
>the gammas - this is NOT where the flux rose as described (The flux both
>rises and falls within T90)].**
>T90 was chosen **a priori** because EGRET detected ...
>**The sentence starting "The size of the search bin" is
>redundant - said earlier**
>
change made. flux is indeed the wrong word. fluence is the correct
word.
>...
>fluctuations, both statistical and systematic, **[what is a systematic
>fluctuation?]**
I don't know either. Others have suggested removing this sentence. I agree
with them. The line was included to address comments that that we
should state that we searched for systematic causes for the excess.
I have changes the sentence to read "systematic errors", not
"fluctuations".
>...
>This distribution was then normalized to the number of events (N T90 )
>detected by Milagrito over the entire sky during T90 (N T90 )**strike the
>2nd redundant T90)**
change made.
>
>The **chance** probability of obtaining the observed significance
>**anywhere** within the entire search region ...
change made.
>The distribution of the **chance probabilities for the 54 GRBs is given
>in Figure 1.**
chance added.
>
>**One of the 54 bursts, GRB 970417a, shows a **large** excess above
>background ** NOTE: ]I DO NOT KNOW WHAT a 4-SIGMA EXCESS IS
>IN THIS CONTEXT.]
change made.
>...
>The low BATSE fluence results in a large **positional uncertainty of 6.23
>deg (1-sigma). The resulting search region for TeV emission has a
>radius of 9.4 deg.**
change made.
>...
>** This location is completely consistent with the position determined by
>BATSE. The uncertainty in this location is approximately 0.5 deg
>(1-sigma), much better than that determined by BATSE.**
change made.
>...
>The Poisson probability for observing a signal at least this large **in a
>given bin** due to a background ...
>
"in a given bin" implies that the prob. does not depdend on where the
excess is positioned. The next line explains what the prob. of finding
an excess this large in the entire search area. We don't think that
this text adds to the paper, and chose to not change the text.
>**The probability of such an observation anywhere within the search
>region is 2.8 x 10**-5. The chance probability of the background
>fluctuating to the level observed for GRB 970417a for any of the 54 GRBs
>is 1.5 x 10**-3. No anomalies were found in the individual events
>contributing to this excess. **
The language that you suggest removing "...found by the Monte Carlo simulation
described above..." was added to address confusion about how the probabilities
were determined. We feel it is better to be redendant than unclear, and
retain the old wording.
>Although the initial search was limited to T90, longer time intervals
>were also examined for GRB970417a.**
Text changed in a similar way to address other comments.
>To allow for the positional uncertainty of the TeV -ray detection during
>T90, the radius of the search bin was increased ** for this study ** to
>2.2 deg. shown by simulation to be optimal for the given parameters. A
>search for TeV gamma-rays integrated over ** delete (long) ** time
>intervals of one hour, two hours and a day after the GRB start time did
>not show any significant excess.
change made.
>
>** Figure 3 shows the arrival times of events observed by Milagrito
>within the 1.6 deg. bin for the 30 seconds around the start of GRB
>970417a. **
>[NOTE: I STRONGLY DISLIKE THE WORK "Lightcurve" AS IT
>IMPLIES SOMETHING WE ARE AVOIDING SAYING. Further, I see no reason
>to show the other time intervals or to restate that we see no
>after-flares]
Your comment is noted. What we say about the "lightcurve" is controversial
to many, and best addressed at a higher level.
>4. Discussion
>If the observed excess of events in Milagrito is indeed associated with
>GRB 970417a then it represents the highest energy photons yet detected
>from a GRB in coincidence with the sub- MeV emission. The ** energy**
>spectrum and maximum energy of emission is difficult to determine from
>Milagrito data (11). Monte Carlo simulations of gamma-ray-initiated air
>showers show that the effective
>area ** of Milagrito** increases smoothly with energy, so that the
>precise energy threshold is undefined. Figure 4 shows the implied fluence
>of this observation as a function of upper cutoff energy for a range of
>power-law input spectra. ** WE NEED SOME STATEMENT ABOUT THE SYSTEMATIC
>ERROR ASSOCIATED WITH THIS**
changes made. Our energy uncertainty from rate matching and the 501 result
strongly constrain systematic energy scale errors to be small (+/-~30%)
compared to the order of magnitude of allowed fluences.
We do not see the need to state our systmatic energy error may be as
high as 30% when we don't measure energies to better than ~1000%.
>
>Some **additional **information about the energies of the observed
>**events ** can be obtained by considering the response of the **
>summed** untriggered counting ** rate ** of the individual PMTs.**
>Detector simulations of the dependence of ** the PMT counting rate to
>gamma-ray induced air- showers indicate that this rate is more sensitive
>than the standard
>shower data at energies below a few hundred GeV(11). ** No
>excess was observed in ** this rate, which implies that the air-showers
>detected by Milagrito were probably due to gamma-rays above several
>hundred GeV.**
>
paragraph removed to address other comments.
>** delete - this is redundant: High energy -rays from sources at
>cosmological distances will be absorbed via electron- positron pair
>production. **
change made.
>** Thus if emission from GRB 970417a has been detected by Milagrito, it
>must be a relatively nearby event. ** [NOTE: I STRONLGY FEEL THAT WE
>SHOULD NOT USE THE WORDS "UPPER LIMIT" REGARDING THE REDSHIFT
>AS IT IMPLIES A DETECTION THAT WE ARE NOT CLAIMING]
we feel it is clear that we ar not claming an upper limit on the redshift.
We clearly couch the conclusion in the context that the excess is due
to gamma-rays from a GRB.
>5. Conclusion
>...
>The **chance** probability that an excess of this significance **or
>greater ** would be observed from the entire sample of 54 bursts is 1:5 x
>10**-3.
>** [Delete the sentence "This result was found simultaneously... -
>This has no place in this paper, unless we are prepared to discuss both
>analyses in detail] ** If emission from GRB 970417a has been
>detected by Milagrito, the fluence ... [change 10**-4 to 10**-3, as this
>is what I get from Fig. 4.] **
change in fluence numbers made. chance added. sentance about second analysis
reworded and moved to an earlier section as suggested by DAW.
>...
>The TeV fluence inferred from this result is at least an order of
>magnitude greater than the sub-MeV fluence **. [delete "and the
>maximum energy of emission extends to hundreds of GeV" as this is
>redundant.] **
>
change made.
>** will be continued with Milagro, which has significantly increased
>sensitivity to GRBs between 0.1 and 10 TeV.
change made.
>Fig. 1.| ** The distribution of probabilities for each of the 54 bursts
>that the observed signal was a background fluctuation anywhere within the
>search region. The two plots show the same data with a linear and
>logarithmic scale for the x-axis, respectively.**
>
Other suggested caption changes adopted. We hope to the satisfaction of all.
>Fig. 2. **The number of events recorded by Milagrito during T90 in
>overlapping 1.6-degree-radius bins in the vicinity of GRB 970417.
>**
change made.
Fig. 3. ** The number of events/s observed in the bin centered on the
position of the candidate TeV counterpart to GRB 970417a for 30 seconds
surrounding the BATSE trigger time. **
Caption not changed. see above.
Fig. 4. ** The implied fluences of candidate TeV counterpart to GRB
970417a for a range of input spectra and high energy cutoffs. **
change made.
*****************************
Gus:
Abstract:
>Milagrito, a detector sensitive to very high-energy (VHE) gamma rays
>monitored the ...
>With a large field of view and A high duty cycle, this instrument was well
>suited [NO -] to search for ...
In this sentence the well-suited is an adjective describing instrument, doesn't
this mean that the [-] should stay?
>In this paper we report ... with Milagrito for counterparts to GRBs observed
>by ...
change made
>BATSE observed 54 GRBS within the field of view of Milagrito during this
>period.
change made
>The excess had A chance probability OF ... background. The probability
>[SPELLING] OF observing ...
change made
>No significant correlations ... [remove OTHER, you say it later].
change already made
Section 1: Introduction
>P1 One of these BATSE ... sensitive TO gamma rays with energies between 50
>keV ...
change made
>P1 The EGRET experiment ... to 18 GeV. To date, there is no evidence ...
change made - I think the semi-colon here was grammatically correct,
however I personally prefer periods.
>P2 However, for sources ... z=0.5, VHE gamma-ray emission ... (omit the
>"then")
>P2 absorbitive: examinate the sentenciation above.
>P2 Recent measurements of the optical afterglows associated with several
>gamma-ray bursts indicate that all but one of them have redshifts greater
>than 0.8.
I have made this change, but since we started this paper an afterglow with redshift
0.4 has been measured. We should probably reword things here to indicate that i
the redshifts tend to be around one rather than giving a specific lower limit.
>P2 These measurements indicate ... is not well correlated ... (omit the -
>between well-correlated).
already done
>P2 (Next sentence) THESE CONSIDERATIONS suggest that observation of TeV ...
>P3 The limited area of current space-based instruments limits their
>sensitivity to gamma-ray sources above 30 GeV, where the flux is low and
>rapidly falling. [The sensitivity to photons is fine.]
This paragraph has already been rewritten.
>P3 Instead, ground-based experiments, which detect the extensive air shower
>produced by gamma rays, are used to study the VHE sky.
This paragraph has also been rewritten. I think that it still needs
some work. The reader has been told about cascading relativistic
particles before being told about air-showers. We need to tell them
(reader) where the relativistic particles came from.
>P4 These telescopes have a large collection area and can reject the abundant
>cosmic-ray background. These features make them more sensitive to TeV
>gamma-ray sources than current space-based instruments.
I prefer the current phrasing. The inclusion of the actual numbers for
the ACT collection area was made at the request of some other
collaborator (Cy or Guarang I think).
>P4 However, they have ... duty cycle of <10%.
change made
>P4 Searches for gamma-ray bursts ... within a few minutes OF the
>notification ...
change made
>P5 At higher energies, the Tibet collaboration has reported a possibly
>significant deviation of the probability distribution of all of the bursts
>within their field of view (REF 13). However, no single burst had a
>stastically significant excess.
changes made
>P5 But the DIRECTION OF THE observed excess was ... and THE EPISODE was not
>claimed as a firm detection (12). NOTE: we are not claiming a "firm"
>detection, so this sounds a bit funny in context.
This section has been reworded already - I don't like the phrase "THE EPISODE"
>P6 Milagrito, a small, single layer predecessor to Milagro, operated ...
>Don, you are half right. Prototype is not the right word, but it sounds to
>me like "limited prototype" is an oxymoron. The above is the best I can do.
Calling Milagrito a prototype detector (the current phrase) makes perfect sense to us.
>P6 This paper ... , but concentrates on the ... Or if you prefer,
>"concentrates more specifically in a detailed fashion with a close eye upon
>the significance of one GRB candidate".
I am not quite sure what you mean here.
>Section 2. The Milagrito Detector.
>P1 Milagrito consisted ... submerged in a 30mx50m light-tight water
>reservoir.
change made
>The PMTs were positioned at the vertices of a square grid with
>2.8 m spacing. [NOTE: You say PMT stands for photomultiplier tubes, use
>it. You never say what tube stands for.]
This sentence has already been rephrased
>P1 Data were collected ... of 0.9, 1.5, and 2.0 m above the PMTs. (there is
>only 1 layer, use the comma before the and).
change made
>P1 The PMTs detected the Cherenkov light ... [Where did "These SENSORS come
>from? You never talked about sensors.]
already changed
>P1 In addition, the abundant gamma rays in the air shower can be detected
>after interactions in the pond yield relativistic charged particles that
>generate Cherenkov light.
This sentence has already been changed, so I have left it as it is.
>P1 The continuous medium AND large Cherenkov angle (41 degrees) allows for
>the efficient detection of shower particles incident on the reservoir with a
>sparse array of PMTs.
change made
>P1 The trigger requirement for Milagrito was that >100 PMTs had to register
>at least 1/4 of a photoelectron each within 300 ns.
I have not made this change
>P1 The event rate of Milagrito was 300 Hz. Almost all of these triggers
>WERE caused by the cosmic-ray background. [NOTE: BE CAREFUL of mixing
>tenses. It is easy to do. Also, note the -]
changes made
>P1 I still object to calling 1997 a flare. This seems ridiculous. There
>were many flares with quiet times in between. Flaring state, high state,
>increased variability. Chose something else please.
This was a flare. Mrk501 stayed in a consistently high state for an
extended period of time. Shorter timescale flares were superimposed on
this. On no night of observations during 1997 did we ever feel that mrk501 had
gone back to 1995/96 levels.
>P1 These and other tests indicated that Milagrito had an ... (why get cute
>here?)
This has alreadt been reworded.
>Section 3: Observations and Results
>I give up.
>Repeat of a previous comment that went ignored. When you use the word
>figure to denote Figure 1 you must capitalize the word figure. Look in any
>issue of ApJ.
changes made already
Content:
>1.) THE TITLE SUCKS. This was not what we agreed on at the collaboration
>meeting and was certainly not agreed to by the morass of emails. Let the
>spokesmen decide. There is a code in this field. "Search for ..." means I
>looked but did not see, READ NO FURTHER you will be wasting your time.
>"Study of" is the same as "Search for". People will not even read the
>abstract.
This is a collaboration issue
>2.) The author list. I never heard a consensus to change it. Who
>determined that? I mildly object to Spencer, I strongly object to Gwelen.
>If Spencer is included I demand the Biller be included and anyone else that
>worked on Milagro during the early days.
to be resolved elsewhere
>3.) 0.9x0.9 != 0.9 Word in the street is that in BATSEland it does. If
>this is true we need a reference.
We have changed this to:
For each burst, a circular search region was defined
by the BATSE 90\% confidence interval, which incorporates both the
statistical and systematic position errors~\cite{briggs99}.
The Briggs et al paper appeared in 1999, they suggested that the
recommended way to get BATSE position errors was simply to read the
values off the graph. We changed the analysis to use these values for
the search radius rather than the statistical + systematic errors that
we had previously added in quadrature.
> 4.) REMOVE "This radius was derived ..." You repeat that EXACT SAME
>SENTENCE 4 SENTENCES LATER. In the next version of the sentence remove the
>"was selected prior to the search". Really we didn't cheat guys.
I am happy to leave in 'selected prior to the search' . However, I
agree that it is unnecessary.
>5.) TALLIED?
What do you suggest? We think that tallied is ok.
>6.) The "duration of the burst is NOT T90." T90 is a subset of the
>duration.
This has been changed to:
The number of events falling within each of the $1.6^{\circ}$ bins was
summed for the duration of the burst defined as the T90 interval
reported by BATSE. This time period is that in which the BATSE
fluence rose from 5\% to 95\% of its total.
>7.) Understanding statistical fluctuations is not rocket science. Why don't
>we say, "Before performing the search we had to look up the Poisson
>distribution in Eadie because we forgot it."
I found understanding the statistics involved in finding the trials factor
for an binned oversampled grid search on a non-uniform background to
be non-trivial. We needed to work on understanding the situation and
chose to do so by montecarlo simulation. We don't find any problem
with the above statement.
>8.) The paragraph beginning with "The probability of obtaining .." is still
>a little unclear. I think this may confuse the typical reader.
The problem here is that the analysis does require some thought to be
understood. I am not sure how easy this is to rewrite in a clearer way.
>9.) Of the 54 bursts, one GRB 970417a, shows a 4 sigma excess ... (Never
>mind the wording and punctuation), but to what does 4 sigma refer? The
>pre-pretrial, the pretrial, or the final probability?
This has been changed to:
One of these bursts, GRB 970417a, shows a large excess above
background in the Milagrito data.
>10.) What does "weak, but typical" mean? Was it typical or was it weak?
This has been changed to:
The BATSE detection of this burst shows it to be relatively weak burst
with a fluence in the 50--300 keV energy range of $1.5 \times 10^{-7}$
ergs/cm$^2$ and T90 of 7.9 seconds.
>11.) "The individual events contributing ..." Please read this sentence
>slowly to yourself. I think I know what you are trying to say, but you
>ain't saying it. I also do not understand the point of it. Why not insert
>a sentence that indicate we all love our mothers and apple pie? We would
>not be publishing the paper if we found a problem with the data. And every
>experimenter looks for such effects, and only publishes if they don't find
>any. It is no guarantee that we won't find some such effect in the future.
>Perhaps we should say that "We have done our best to search the data for
>systematic or instrumental problems. While we have not found any such
>problems to date, we will continue to look for them. The publication of
>this paper should not be taken as a guarantee of the veracity of the data
>described herein. The authors' liability in the case of a reader damaging
>their career by taking this result to represent an actual detection of TeV
>emission from a gamma-ray burst is strictly limited to the cost of the paper
>on which it is written. Any other guarantees either express or implied are
>null and void."
This sentence was included as a response to two requests. Several suggestions to
remove this sentence have been made. We agree with them.
>12.) I do not think that the opening of the bin is the most sensitive way to
>search for post T90 emission. While likelihood is the way to go, I see that
>this won't happen. But I still believe that a heavy oversampling in the
>vicinity is more sensitive. So we should not call this search "optimal", it
>is not.
For the search method described, the radius used were found by monte
carlo to be optimal. We feel that the sentence is required to indicate to the
reader that the choice of binsize was not as a result of tuning.
we have changed ...given parameters... to
...for the Milagrito
point spread function and the number of background events.
>13.) While we're talking about the after bursts, "statistically compelling"
>is a slow retreat to reality. Try "statistically significant". Also we
>should insert a disclaimer that we can not guarantee that there was no post
>burst emission. Because of the trials our sensitivity is not very great, we
>should elucidate this point.
'compelling' changed to 'significant'
I have no objection to putting such a disclaimer in. However, I feel
that it is unnecessary - we cannot exclude the possiblility of TeV
emission below our flux threshold for any of the 54 bursts - trials
pay a role here. Neither can we exclude afterbursts below the
detection threshold of the search method we used. I think that this
point is relatively obvious and does not need to be explicitly
made. The word preliminary in the sentence leaves us free to
perform/develop/present a more detailed analysis in the future.
>14.) My comments on the scaler stuff were completely ignored. I won't
>bother to rewrite them. Read what I said last time.
scaler paragraph has been removed.
>15.) Discussion, Conclusion, what gives? Why is a topic introduced in the
>"Discussion" section? The energy is a "Result". Then combine the
>"Discussion" and the "Conclusions".
Now that we have removed the scaler paragraph, I think that the rest
of this section is appropriately labeled discussion.
>16.) To satisfy the "Monte Carlo may be wrong" crowd. Everyone in this
>field knows that the Monte Carlos are not correct. Yet it is impossible to
>assign any statistically meaningful error to the effective area plots. We
>should just insert a statement to the effect that our response is determined
>via simulations and that there may be as much as a 30% uncertainty in the
>absolute energy scale. Since satellite people do not have this problem we
>need to make this clear.
change made
>17.) The distance. Since several studies say opacity =1 at 200 GeV for z=.3
>our upper limit on z is .3 What upper limit? Absolute? If all we are
>doing is rapidly waving our hands, we should say so. Do not give the result
>a veneer of respectability by saying upper limit. Or better yet, get
>precise. Use a model, and give a 90% CL UL based on that model of the IR.
I think that we may have problems deciding which model to pick, so instead we
have made the statement more vague.
If the Milagrito candidate is indeed a detection of GRB 970417a, the
opacity allows us to place an upper limit on the redshift of
$\sim$0.3, making it a relatively nearby event.
changed to
Thus, if Milagrito has indeed detected high energy photons from
GRB 990417a, it must be from a relatively nearby object.
>18.) "This result was found, ALMOST SIMULTANEOUSLY by two independent
>analyses." Whoop de fuckin doo! (Excuse my French.) What are the odds of
>that!? This really sounds inane and immature. We either believe the result
>is valid or we don't in which case we don't publish the paper. And then
>referencing our conference proceedings? This is OUR paper. OUR paper. We
>are ONE. All together now.
We have changed this to reference Isabel as per DAW suggestion (see above).
Begin Gaurangs comments:
>3rd paragraph of Introduction. He suggests we change the "Instead, ground
>based..." sentence a little to read something like "Instead, ground based
>experiments detect these gamma-rays by observing the extensive air showers
>they induce in the atmosphere, thus giving them a much larger effective
>area at higher energies." This explicitly works in the point that the
>ground based expts. have big Aeff.
change made
>remove the "the" in "as the charged shower" to read "as charged shower" in
>sec. 2. He also made the general comment that there were many "the"'s in
>places they didn't belong.
change made
>4th paragraph of "observation and results" section. He suggests we add a
>"(see Fig. 1)" just after the 2.8x10^-5 near the end of this paragraph.
change made
>He points out that the discussion of scalars need to be clarified. I had
>implicitly assumed this paragraph was still to be re-written. Is this not
>true? If it is not to be rewritten, then indeed we need to discuss the
>scalar stuff in somewhat more detail, e.g. the instrumentation, the
>dependence of the relative shower/scalar sensitivity on the assumed
>spectrum and angular resolution, etc.
we have removed the discussion of the scalers.
Begin Stefans comments
>1) the new version, including the title, is fine with me.
Wow, someone who does not hate the paper!
>2) I agree with Brenda that "TeV" in the title makes more sense
> than "VHE" -- I've suggested that before, and the Mrk501-paper
> uses it, too
This has been addressed with Davids comments (see above)
>3) can we somehow get the references right ?
They are getting there...
>Some minor remarks:
>page 2, paragraph 3: omit "...beacuse their detectors...area."
This has already been rewritten
> paragraph 5: mention "Tibet"
change made
>page 3, section 2: the water depth thing is very confusing. Let's
> just say 0.9 m (the depth at burst time) and
> (optional) mention that we increased the water
> level LATER to study the performance
The inclusion of different water depths was included as a response to
someone else comment. This sentence does not seem too confusing to us
>page 4, paragraph 1: NT90 is mentioned twice, but the 2195 events
> aren't -- that looks like a typo
this has been fixed.
>page 4, paragraph 3: give the RA and DEC with the same precision, that
> is 289.9 and 54 (which I'd prefer, given the 0.5
> degree error), or two digit after the decimal point
> for both -- I by the way find the BATSE "accuracy"
> of two digits with a 6.23 (!!) error pretty funny.
changed 289.89 to 289.9
>page 5, last section of "Discussion": I do not think we can set an upper
> on z -- that needs more explaination of what model
> is actually used and will get us into unwanted
> discussions. Let's omit that paragraph.
we have rephrased this (see above)
>I am not trying to take away well deserved credit from Julie and Isabel,
>but the ICRC and TeV workshop papers should not be in the references
>as they do not add significant information. There is no reference
>to Andy's Portugal paper in the Mrk501 paper, either.
This has been changed to reference Isabels PhD thesis (see above)
Begin Todds comments:
>--The title is PERFECT!!!! Let's leave it alone.
I assume that this will be decided elsewhere.
>--The author list PISSED OFF at least one local, who correctly noted that
>the Reps. committee has not bought off on the addition of Paglia & Klein.
I assume that this will be decided elsewhere.
>--The institution list. It is common practice to list the
>department/division of each participant, not just the
>lab/university. Please list LANL as Physics Division, P-23, Los Alamos
>National Laboratory....
change made
>--We need to add acknowlegements. Authors ought to put one together
>modeled on that in the AGN paper and then float it by the Inst. Reps.
We have added the acknowledgments from the 501 paper
>--The references still need to be cleaned up. In particular, we need to
>get then ordered in numerical order, and get the numbers on them.
ApJ does its references in Author date style. I forgot to add the key
at the begining of each reference.
>--The abstract:
> needs a comma after (VHE)
This sentence has been changed in response to earlier comments.
> needs probability spelled correctly
change made
>--Introduction
> --1st paragraph
> The last two sentences ("A crucial question..." and "The
>EGRET experiment") are out of logical order.
change made
> --2nd paragraph
> You discuss "Recent [GRB] measurements ...indicate that
>the measured intensity of the gamma-ray emission is not well correlated
>with distance." Observation of AGNs, esp. at TeV energies, also indicates
>that gamma-ray emission is not well correlated with low-energy
>luminosity. I'd like to ask the authors to consider inclusion of this
>point as part of the intro. to why we might not only want to look at the
>bright BATSE bursts.
I do not agree with the above statement on AGN. The high energy
gamma-ray flux from blazars is correlated with the flux at lower
(radio/optical) energies. Two exceptions are Mrk501 and Mrk421 which
are weak at radio/optical energies, but are two of the brightest agn
at g-ray energies. I suspect that the point here is that with smaller
intrinsic flux at low energies, there may be evidence for less cooling
of the relativistic particles in the jet - allowing them to be
accelerated to higher energies, and also less photon-photon absorption
of the gamma-rays. These are arguements commonly used when discussing
blazars, but I think might not scale up to the very high intrinsic
luminosities of a GRB.
> --3rd paragraph
> Isn't it true that at 30 GeV photon fluxes from ALL (not
>most) astrophysicsal sources become invisible for satellite expts.?
No, EGRET sees 100 GeV photons from the galaxy.
> --4th paragraph
> In the last sentence of this paragraph, I still don't
>think any lay scientific reader (like ME!!!) knows what "does not allow the
>calculation of stringent upper limits" means. Upper limit to what? why
>not? Say what you mean here please--let 'em have it. What you want to say
>is "Even if (all?) GRBs emit in the TeV region, because of the difficulties
>(narrow FOV, etc.) with the observations, that Whipple et al have not
>detected TeV emission does not allow us to conclude that GRBs do not emit
>TeV gamma-rays."
This sentence has been rewritten already.
> --5th paragraph
> 1st sentence needs work on english. Frankly, I suppose it
>is just too long and so should be broken into several shorter sentences. I
>would happily write a version for you if you wish.
This has already been rewritten
> More generally, I still don't understand the point of this
>discussion. You say the others "have reported a possibly significant
>deviation". Its either significant, its not significant, or we can make no
>conclusion. In either of the latter two cases, we don't need to discuss
>it. You then apparently discount their "possibly significant deviation" as
>being immaterial as "no single burst showed a convincing signal"
>(BTW--being a smart ass, I might ask if OUR burst shows a convincing
>signal?). You then talk specifically about a HEGRA burst, then discount it
>as "....not a firm detection". If it is not claimed as a detection, why do
>we discuss it? I have personally not claimed detection of many phenomena
>(e.g. gold at the end of rainbows, even Santa Claus!), yet I don't feel the
>need to discuss them. On the other hand, if we are trying to present a
>fairly inclusive summary of previous work on GRB detection at other
>wavelengths (which is a VERY GOOD thing to do in the introduction), then we
>ought to present an inclusive list (e.g. Schnee's work, other's work at
>VHE/UHE energies, our work with neutrinos?, etc. etc.).
This was not intended to be a complete review of previous work, but a summary
of interesting results (not entirely null) which are in refereed journals.
>--The Milagro Detector section
> I don't agree with Stefan that we should remove the mention of
>different depths, but we should mention the depth at the time of 970417
change made
> In the 3rd sentence, say "These PMTs", not "These sensors" as
>you've not used sensors to describe th ePMTs.
This has been changed
> Toward the end, change Hz to second^-1, as Hz is a unit of cycles
>per second, a periodic frequency not the rate of a random occurance.
change made
>--Observations and results section
> --1st paragraph
> This comment relates to the 1st and 2nd paragraphs of this
>section. I worry that we will loose the lay reader when we discuss the
>"isotropic background OF (add of here) cosmic rays" and later talk about
>"the angular distribution of background events on the sky was characterized
>by 2 hours of data..." You see, you've already characterized the
>background distribution by saying it is isotropic. Pls. work on wording to
>clarify this point--that the background above the atmosphere is
>(approximately!!! there is a variation) isotropic, but the atmospheric
>attenuation of the air-shower causes a peak in the observed distribution
>near zenith.
"isotropic" removed,
> Is it clear when we say "....coincident with some BATSE
>gamma-ray burst within 45 degrees of the zenith of Milagrito", what the 45
>degree cut is referring to? Frankly, I don't know, though I assume it is a
>cut at 45 degrees based on the "best position" of BATSE.
I think that this is clear, I hesitate to add text to clarify it, as I
think that the result may be cumbersome.
> What does it mean when we say "this angular region is
>where Milagrito had the best combination of sensitivity and low effective
>energy threshold"? I sort-of know what you're trying to imply, but what
>are you really trying to say here? If it is the "best combination" there
>should be some sort of metric that defines "best". What is it? If there
>isn't, then we shouldn't imply that this selection was derived in some sort
>of calculation, but rather should simply say it was CHOSEN to reflect some
>(guess?) about a reasonable cut.
The phrasing was an editorial board suggestion/requirement. I agree
with Todd that we are not saying what we mean. I presume that the
point of this addition was to justify the choice of 45 degrees. We
chose it because we felt that below 45 degrees we had negligable
sensitivity.
This section now reads:
Bursts detected by BATSE within 45$^{\circ}$ of the zenith of
Milagrito were considered. This angular range was chosen because the
sensitivity of Milagrito fell rapidly with increasing zenith angle.
> Clarify what you mean by "90% confidence interval" when
>you talk about combining the BATSE errors. And please put in a reference
>here!!!
This has been changed.
> You talk about "the size of the search bin, 1.6
>degrees,...." twice!!! I think once is enough.
change made.
> --2nd paragraph
> In the 1st sentence of this paragraph, say "...to
>establish the STATISTICAL significance".
This section of the sentence has been removed.
> Again, we've said the background is isotropic, yet here we
>talk about the non-isotripicness of it.
I added (due to the variation of the sensitivity of Milagrito as a
function of zenith angle)
> We discuss NOWHERE the systematic fluctuations in the
>background estimate. Yet, the sentence implies that we have understood the
>systematic fluctuations. Moreover, for most people, when they hear about
>the fluctuations of a systematic error, they know that what they're being
>fed is a bag of lies. Just remove that statement about systematics, or
>let's show that we understand them.
The all-sky/galactic plane analysis or the untriggered burst search
presumably both give us a handle on systematic effects in the background.
We could try and come up with a suitable sentence to quantify the level
of systematics in the background calculations and maybe point them to a
reference.
> You define NT90 twice!!!
changed
> I would change "The resulting background data..." sentence
>around a bit to read "The resulting background data were also binned in the
>same 1.6 degree overlapping bins as the initial data." or similar wording.
change made
> What you mean by "The Poisson probability" isn't clear at
>all. What I think you mean is the Poissin probability that the observed
>data would exceed the background estimate was calculated.
> At the end of "The Poisson probability" sentence, you
>might want to say "...candidate position of a VHE gamma-ray counterpart to
>that PARTICULAR BATSE burst" or similar phrasing to make it clear that by
>"candidate VHE gamma-ray counterpart" you mean that each burst has a
>candidate, not that we have one candidate burst (970417).
change made
> --4th paragraph
> the first sentence has some comma problems and should be
>"Of the 54 bursts, one (GRB 970417a) shows a four sigma excess..."
this sentence has already been rewritten.
> in the 2nd sentence, I would phrase as "...it to be a
>typical relatively weak burst, with a fluence...", removing the ", but
>typical," as it looks like you're trying to make some excuse that it is
>weak, but at least it is typical.
This sentence has been reworded to:
The BATSE detection of this burst
shows it to be relatively weak burst with a fluence in the 50--300 keV
energy range of $1.5 \times 10^{-7}$ ergs/cm$^2$ and T90 of 7.9
seconds.
> In the "the low BATSE fluence..." sentence, you want to
>change leads to led.
changed to
The low BATSE fluence
results in a large positional uncertainty of $6.23^{\circ}$
(1$\sigma$).
> I know (and have argued myself for years) what you're
>trying to say by the phrasing "...it is both completely consistent with,
>and much better determined, than that of the BATSE burst". But come on,
>this isn't a proposal here! First, don't use "completely", just say
>"consistent with".
change made
>Next, we missed the important point early on in the
>introductory material--that is that the directional uncertainty of EACH
>air-shower is much better determined than the direction of a typical BATSE
>GRB. This leads to us HAVING TO TILE the BATSE error circle with our much
>smaller "resolution elements". This puts the point where it needs to be
>said, thusly motivating the need to search throughout the BATSE region for
>the GRB with a bin best for our resolution. We need to say this point
>early on.
**We could add a sentence stating that the BATSE 90% confidence intervals
**were at least 4 degrees radius, requiring a a search in milagrito
**for a point source within that region.
> "The probability....was found by the METHOD DESCRIBED
>ABOVE to be 2.8x10^-5", rather than "was found by Monte Carlo".
this has been changed to:
The probability of such an observation within
the search region for this burst was found by the Monte Carlo
simulation described above to be $2.8 \times 10^{-5}$ (see
Figure~\ref{fig:prob_mil}
> --5th paragraph
> I'm afraid that "Lightcurves" is simply jargon. A lay
>astrophysics reader wont' know what you're talking about.
changed to:
Histograms of shorter time
intervals where the data are binned in intervals of one second
and of T90 (7.9 s) are shown in Figure~\ref{fig:lc}
> Ok, so we're honing on the proper wording for the lack of
>statistically significant afterbursts. I don't mind the term "compelling",
>but frankly we don't need it. I do have a problem with using "preliminary"
>here though as we ought not to be publishing preliminary results in a
>refereed journal. Get rid of preliminary, keep compelling if you want.
I like using preliminary (see above) how about changing this to:
Initial analyses of these data reveal no statistically significant evidence for
TeV after-flares.
>--Discussion section
> --1st paragraph
> "If the observation...is associated, then it represents
>the highest energy photons yet detected from a GRB" PERIOD. Remove the
>"in coincidence with the sub-Mev" part.
change made
> --3rd paragraph
> We do not have an upper limit on the redshift. Sorry to
>be a hard-ass here, but its not a limit if it is very model dependent, or
>very assumption dependent, which this is both. I would prefer a phrasing
>like "We thus know this to be a relatively nearby event". You've already
>indicated that "nearby" means z ~0.3 or less.
This has already been changed
>--Conclusion section
> --1st paragraph
> remove "almost simultaneously", keep the references.
We have changed this to reference Isabels thesis. The whole point of this
was to reference Isabel I presume - not because we feel a need to convince
the community that we have bug-free code.
> We've said twice "the maximum energy of emission extends
>to hundreds of GeV" in this paragraph.
We have removed both statements (along with the scaler analysis)
>--Figures
> --fig 4. I still await some discussion of the systematics on
>fluence. But you've said in the text and the figure caption the phrase
>"input spectra", when what you mean is "assumed spectra".
change made
>Also you need to
>talk about what "high energy cutoffs" means--how was that calculated?
I just stopped the integration when I reached the upper cutoff.
>--specific references
> I insist that we refer to Gould and Schraeder's original papers on
>the gamma-ray cutoff!!! Stecker and Salamon aren't honest enough to do so,
>but we must. I won't insist on the reference to my Rome ICRC paper, but
>would like it considered as that is the only place where some mention of
>the cutoff specific to Milagro is discussed, and it is discussed in
>reasonable detail considering it is an ICRC report.
We have refereed to Gould and Schraeder and Jelley.
Don's ED board comments:
>Abstract:
>s1: gamma-rays of very high energies -> very high energy (VHE) gamma-rays
change already made. As we seldom use VHE, "(VHE)" was not added,
and not used.
>s2: "well-suited" -> "well suited"
removed.
>s2: VHE -> TeV
already done most everywhere.
>s3: "...counterparts to bursts..." -> "...counterparts to GRBs..."
already changed to "...counterparts to gamma-ray bursts (GRBs)"
>s4: "...within the Milagrito field of view.." -> "...within the
field of view of Milagrito..."
change made.
>s4: "...the above period." -> "this period."
change already made.
>s6: "The excess had a chance prob..." -> "the excess has a chance prob..."
change made.
>s7: "No other signif...from the remaining bursts" -> "No significant... from
the other bursts"
change already made.
>s8: "The ststistical aspects..." remove aspects.
changed to
"The analysis is presented and the implications are discussed."
to address comments by others.
INTRODUCTION:
>p1s3: BATSE needs ref
reference added.
>p1s4: EGRET needs ref
reference added.
>p1s4: "A crutial question is whether GRB spectra extend to TeV energies.
> The EGRET experiment on board CGRO detected 8 gamma-ray bursts
> with photon..." : reverse order of sentences. add ";" between(?).
change already made.
>p2s1: remove "However"
already removed.
>p2s1: "for sources at redshifts much greater than z=0.5, TeV gamma-ray
> emission may not be observabele because of absorpitive pair-production..."
> -> "TeV gamma-ray emission may not be observabele for sources at ...
> because of absorptive pair-production..."
change made minus the word absorptive.
>p2s2: "Recent measurements of several..." -> "Recent measurements of the
> optical afterglows associated with several... indicate that all but
> one ...(add ref)"
ref already added.
change already made.
>p2s3: "well-correlated" -> "well correlated"
change already made.
>p2s4: "This suggests" -> "These considerations suggest"
change already made.
>p2s4: "...sub-MeV BATSE emission." -> "emission detected by BATSE."
already changed to ...sub-MeV emission detected by BATSE."
The authors feel that it is important to state the energy scale
of BATSE.
>p3s1: "At energies greater than 30 GeV, photon fluxes from most astrophysical
> sources become too low for the satellite-based experiments to detect."
> ->
> "The limited area of current space-based instruments limits their
> sensitivity to gamma-ray sources above 30 GeV, where the flux is low
> and rapidly falling.
Authors prefer:
"At energies greater than 30 GeV, $\gamma$-ray fluxes from most
astrophysical sources become too small for the satellite-based
experiments to detect because of their small sensitive areas."
>p3s2: "Instead, ground based experiments detect these... "
> ->
> "Only ground-based experiments have large enough areas to detect..."
already changed.
>p3s1: "Detection of VHE gamma-rays from ..." -> "Detection of TeV
> gamma-ray emission from ..."
Changed to "TeV $\gamma$-ray emission from astrophysical
sources has been detected..."
as sugessted by Cy.
>p3s2: "Thes telescopes have extreamly large collection areas and
> good cosmic ray rejection, and therefore able to detect
> low fluxes of gamma-rays."
> ->
> "These telescopes have large collection areas and can reject the
> abundant cosmic-ray background, making them more sensitive to
> gamma-ray sources than current space based instruments"
The authors prefer the inclusion of "extremely", because the collection
areas for ACTs are not only large because of the ability of people to
build large instruments on the ground, but also the because of the
extremely large lateral extent of the Cherenkov light produced
by air showers. The suggested text "...making them more sensitive..."
is not correct. The sensitivity of Whipple and EGRET to sources
is highly dependent on the spectrum and angular size of the source.
To address the comments of others, the text was changed to:
"These instruments have extremely large collection areas
($\sim 10^{5}$ m$^{2}$) and good hadronic background rejection and are
therefore able to detect low fluxes of $\gamma$-rays."
>p3s3: specify FOV for ACTs. add comma after nights.
already changed.
>p3s4: "...few minutes after..." -> "...few minutes of.."
change already made.
>p3s5: "No detections were..." -> "No detections have been..."
changed as suggested.
>p3s5: "...reported, and the narrow field, coupled with the delay in slewing
> to the correct position does not allow the calculation of stringent
> upper limits."
> ->
> "...reported; even upper limits are difficult to calculate because these
> telescopes are observing at different times, and probably different
> positions, than the original burst"
The authors prefer:
"...reported, but the narrow field, coupled with the delay in slewing to the correct
position has resulted in no prompt TeV observations at the GRB location."
>p4s1: "At higher energies, wide field instruments..."
> ->
> "At higher energies, wide field instruments have reported evidence of gamma-ray
> burst emission. The Tibet extensive air shower array reported a burst..."
The aurhors prefer the similar text suggested by others:
"At higher energies, the Tibet collaboration reported a possibly
significant deviation of the probability distribution from the
background of all the bursts within their field of view. "
>p4s3: "..view, and one..." -> "...view. One..."
changed as suggested.
>p4s3: "...in a time interval longer than the sub-MeV emission.
> But the observed excess wes not consistent with satellite
> measurements and was not claimed as a firm detection."
> ->
> ,...but with temporal and positional differences from the
> original burst large enough that the excesswas not claimed
> as a detection.
The authors prefer the collaborator suggested rewording:
"...in a time interval not entirely consistent with the sub-MeV
emission. The location of the excess was somewhat offset from
the satellite measurements so this was not claimed as a firm
detection~\cite{padilla98}."
>p5s1: "...over..." -> "...greater than..."
suggested change made.
>p5s2: "..., a limited prototype..." ->
> "..., an earlier stage of Milagro..."
The consensus and the authors have no problem with "prototype".
The text now reads: "A prototype detector, Milagrito..."
>p5s2: "...May 1998. During this time interval..."
> -> "...May 1998, during which ..."
suggested change made.
>p5s3: "...were within Milagrito's field ..."
> -> "...were within Milagrito's useful field..."
We don't feel that this adds anything. Gus's uses
data out to the horizon for his analysis. 45 deg
was chosen for this analysis. We feel that we should
shy away from concluding that 45 deg is really a
special angle for our detector.
>p5s4: "...,but concentrates more specifically on the
> significance of ..."
> ->
> "..., with specific attention to..."
The authors prefer the wording of the former.
THE MILAGRITO DETECTOR
>p1s1: "Milagrito consisted of 228 ..."
> -> "Milagrito consited of a planar array of 228..."
change already made.
>p1s1: "...water of a 30x50 m covered ..." -> "...water of a covered..."
suggested change made.
>p1s1: "...reservior with the tubes at the verticies of a square grid
> with 2.8m spacing."
> ->
> "...;the PMTs were positioned at the vertices of a square grid
> with 2.8m spacing, covering a total area of 1500 m*2"
The authors prefer the suggested text above:
"The PMTs were located on a square grid with spacing 2.8m, i
covering a total area of 1800 m"
2.8x2.8x228 = 1788!! The actual "sensitive" area is actually larger.
>p1s2: "above the layer" -> "above this layer"
clause removed to address previous comment.
>p1s3: "...as the charged..." -> "...as charged..."
change already made.
>p1s4: "In addition, the interaction with the water transfers the
> energy of the abundent gamma-rays in the air shower via pair
> production and compton scattering of relitivistic charged
> particles that generated Cherenkov light"
> ->
> "In addition, abundant gamma-rays in the air shower could
> be detected after their interactions in the water via
> pair production and compton scattering yielded relitivistic
> charged particles, that generated Cherenkov light."
The authors prefer the alternative wording:
"In addition, Cherenkov light is produced by relativistic
charged particles resulting from the interaction of the
abundant $\gamma$-rays in the air shower with the water
via pair production and Compton scattering."
>p1s5: "...medium allows for the efficient..." ->
> "...medium allowed efficient"
The authors prefer the present tense. Milagrito is in the
past, but the water Cherenkov detection method still works.
"for the" already removed.
>p1s5: "...sensors are not..." -> "...sensors were not..."
clause removed as suggested by others.
>p1s7: "...angle..." -> "...direction.."
already changed.
>p1s8: "Milagrito triggered when 100 PMTs or more in the reservoir each
> detected at least..."
> ->
> "The trigger required $>$100 PMTs registering at least one..."
change already made.
>p1s9: "Hz" -> "1/s"
change already made.
>p1s10: "...triggers were causes..." -> "...triggers are caused..."
change already made.
>p1s10: remove isotropic.
changed to hadronic.
>p1s11: "The ability of this instrument..." ->
> "The capability of Milagrito"
change already made.
>p1s11: "...during its 1997 flare..." -> "...during its
increased activity in 1997."
The authors prefer flare.
>p1s12: "These and other tests..." -> "This and profiles of
the moon shadow..."
We have never really deduced our angular resolution from the
moon shadow.
The author's prefer the suggested rewording:
"The instrument had an angular resolution of about $1^{\circ}$."
OBSERVATIONS AND RESULTS
>A search was conducted for excesses of events above the isotropic
>background cosmic rays in the Milagrito data and coincident with some
>BATSE $\gamma$-ray burst within 45$^{\circ}$ of the zenith of
>Milagrito; this angular region is where Milagrito had the best
>combination of sensitivity and low effective energy threshold.
>
> -->
>A search was conducted for an excess of events above that caused by
>background cosmic rays and coincident with a BATSE GRB within 45 degrees
>of the zenith of Milagrito; in this angular region Milagrito had the
>best combination of sensitivity and low effective energy threshold.
In response to earlier comments this has been subdivided into
3 for clarity and changed to:
A search was conducted for an excess of events above those due to the
isotropic background of cosmic rays in the Milagrito data coincident
with BATSE GRBs. Bursts detected by BATSE within 45$^{\circ}$ of the
zenith of Milagrito were considered. Milagrito had the best
combination of sensitivity and low effective energy threshold in this
angular region.
> s3p1 add reference to BATSE 90% confidence interval sentence
sentence changed and ref added in response to earlier comments
>This radius was derived from the measured angular resolution of
>Milagrito and was selected prior to the search.
> -->
>This radius was derived from the measured angular resolution of
>Milagrito.
change made.
>The number of events falling within each of the $1.6^{\circ}$
>bins was tallied for the duration of the burst reported by BATSE.
> -->
>The number of events falling within each of the $1.6^{\circ}$
>bins was summed for a subset of the duration of the burst reported
>by BATSE.
tallied changed to summed.
The absolute duration of the burst is ill-defined. BATSE report two
durations T90 and T50. We chose to use the T90 duration. In response
to earlier comments we have clarified this to read:
The number of events falling
within each of the $1.6^{\circ}$ bins was summed for the duration of
the burst defined by the T90 interval reported by BATSE.
>T90 was chosen because EGRET detected $\gamma$-rays above a GeV at
>higher fluxes during T90 than for longer durations
> -->
>T90 was chosen, a priori, because EGRET detected $\gamma$-rays above a GeV at
>higher fluxes during T90 than for longer durations.
this has been changed to:
T90 was chosen a priori because the EGRET detections were much
more significant during T90 than when the data were integrated over
longer time intervals~\cite{hurley94}.
>In order to establish the significance of an excess, the background
>due to cosmic ray-induced showers must be estimated and fluctuations,
>both statistical and systematic, must be understood.
> -->
>In order to establish the statistical significance of an excess,
the background due to cosmic ray-induced showers must be estimated
>and, fluctuations, both statistical and systematic, must be understood.
sentence removed - it does not add anything to the paragraph.
> number of events (NT90) detected by Milagrito during T90 (NT90) -
> remove second (NT90)
change already made.
>The Poisson probability that the number of events in each
>$1.6^{\circ}$ bin was due to a background fluctuation was calculated
>and the bin with lowest such probability was then taken as the candidate
>position of a VHE $\gamma$-ray counterpart to the BATSE burst.
> -->
>Each bin in the actual data was compared to the corresponding bin in
>the background map, and the Poisson probability (P_bin) that the number of
>events in the bin was due to a background fluctuation
>was calculated and the bin with lowest such probability (P_min) was then
>taken as the candidate position of a TeV $\gamma$-ray counterpart to the
>BATSE burst.
Most of this sentence has already been changed as per collaborators
suggestion.
each 1.6 degree bin changed to the bin
VHE changed to very-high-energy
P_bin and P_bin added, I do not feel that they aid in clarifying this
sentence. I think that it is the concept and language that is confusing
not really the confusion between different probabilities
>s2p2 The probability --> the probability (P_s)
>s2p2 The probability --> the probability (P_min)
P_i's added where requested
>The distribution of the probabilities for the 54 $\gamma$-ray bursts
>after correcting each for the size of its search area is given in
>figure~\ref{fig:prob_mil}.
> -->
>The distribution of the chance probabilities for the 54 GRBs obtained
>by this method is given in Figure 1.
change already made.
> s2p3 Of the 54 bursts, one GRB 970417a, shows a four sigma excess above
>background in the Milagrito data.
> -->
>Of the 54 bursts, one, GRB 970417a, shows a four sigma excess (in P_s) above
>background in the Milagrito data.
in response to earlier comments this sentence was changed to:
One of these bursts, GRB 970417a, shows a large excess above
background in the Milagrito data.
>The BATSE detection of this burst shows it to be relatively weak
>burst, but typical, with a fluence in the 50--300 keV energy range of
>$1.5 \times 10^{-7}$ ergs/cm$^2$ and T90 of 7.9 seconds.
> -->
>The BATSE detection of this burst
>shows it to be typically weak burst, with a fluence in
>the 50--300 keV energy range of $1.5 \times 10^{-7}$ ergs/cm$^2$ and
>T90 of 7.9 seconds.
This makes it sound like it is typical for bursts to be weak.
In reponse to earlier comments we had already changed this to:
The BATSE detection of this burst
shows it to be a relatively weak burst with a fluence in the 50--300 keV
energy range of $1.5 \times 10^{-7}$ ergs/cm$^2$ and T90 of 7.9
seconds.
>The low BATSE fluence results in a large 1 $\sigma$ positional
>uncertainty of $6.23^{\circ}$ (1-sigma) and leads to search region for
>TeV emission with radius $9.4^{\circ}$.
> -->
>The low BATSE fluence
>results in a large positional uncertainty of $6.23^{\circ}$
>(1-sigma) and leads to a search region for TeV emission with radius
>$9.4^{\circ}$.
This had already been changed in response to earlier comments to:
The low BATSE fluence results in a large positional uncertainty of
$6.23^{\circ}$ (1$\sigma$). The resulting search region for TeV
emission has a radius of $9.4^{\circ}$.
We think this reads better than Don's EdBoard suggestion.
>s2p4 the bin with largest excess --> ...largest excess (P_min)
We think that the original phrasing was unambiguous
>s2p4 289.89 --> 289.9
change made
>s2p4 54.0 --> 54
we actually know dec with higher precision than ra, we see
no reason to quote it with lower precision.
>s2p4 ...so it is completely consistent... --> so it is consistent
change already made earlier
>s2p4 ...,and much better determined, than that
> -->
> ...,and much better determined than, that..
This sentence had been broken into two in response to earlioer comments.
The relevant sentence now reads:
The uncertainty in the location is approximately $0.5^{\circ}$
(1$\sigma$), much better than that determined by BATSE.
>s2p4 The Poisson probability for observing a signal at least this large...
> -->
>The Poisson probability (P_min) for observing a signal at least this large...
change made
>s2p4 2.89 --> 2.9
change already made
>s2p4 the probability.. the probability P_s...
>...of such a detection within the entire $9.4^{\circ}$
>search region for this burst was found by Monte Carlo simulation to be
>$2.8 \times 10^{-5}$.
> -->
>...of such a detection within the entire $9.4^{\circ}$
>search region for this burst was found by the Monte Carlo simulation
>described above to be $2.8 \times 10^{-5}$ (see Figure 1).
This sentence had already been changed in response to several
collaborators comments to:
...of such an observation within the search
region for this burst was found by the Monte Carlo simulation
described above to be $2.8 \times 10^{-5}$ (see
Figure~\ref{fig:prob_mil}).
>Because 54 bursts were examined, the chance probability of background
>fluctuating to the level observed for GRB 970417a for any of these
>bursts is $1.5\times 10^{-3}$.
> -->
>For 54 bursts, the chance probability (P_final) of background
>fluctuating to the level observed for GRB 970417a for at least one of
>these bursts is $1.5\times 10^{-3}$.
change made
>The individual events contributing to this excess were examined for
>systematic or instrumental causes of this excess and none were found.
> -->
>Individual events contributing to this excess were examined for
>systematic or instrumental peculiarities of this excess and none
>were found.
change made. We feel however, that this statement is unnecessary. It
conveys no useful information and is self evident. Obviously if we had
found an instrumental cause of this effect we would not be publishing
the result.
>Although the initial search was limited to T90, for GRB 970417a longer
>time intervals were examined.
> -->
>Although the initial search was limited to T90, longer time intervals
>were examined for the candidate corresponding to GRB970417a.
We had altready changed this, and prefer the collaborator suggested
wording.:
Although the initial search was limited to T90, upon identifying GRB
970417a as a candidate, longer time intervals were examined.
>To allow for the positional uncertainty of the TeV $\gamma$-ray
>detection during T90,
> -->
>To allow for the positional uncertainty of the TeV GRB candidate
>found during T90,
This had already been changed in response to earlier comments.
>s3p6 ..did not show any significant excess --> did not show any
>significant excesses.
change made.
>Lightcurves of shorter time interval searches where the data are
>binned in intervals of one second and of T90 (7.9 s) are shown in
>figure~\ref{fig:lc}.
> -->
>Arrival times of photons (lightcurves) for shorter time interval
>searches where the data are binned in intervals of one second and
>of T90 (7.9 s), are shown in figure~\ref{fig:lc}.
This had already been changed in response to earlier comments to:
Histograms of shorter time intervals where the data are binned in
intervals of one second and of T90 (7.9 s) are shown in
Figure~\ref{fig:lc}.
We feel that this wording is clearer.
>A preliminary analysis of these data reveal no statistically
>compelling evidence for TeV after-flares.
> -->
>An analysis of these data also revealed no statistically
>significant evidence for TeV after-flares.
change made.
DISCUSSION
>If the observed excess of events in Milagrito is indeed associated
>with GRB 970417a then it represents the highest energy photons yet
>detected from a GRB in coincidence with the sub-MeV emission.
> -->
>If the observed excess of events in Milagrito is indeed associated
>with GRB 970417a then by arguments to follow it represents the
>highest energy photons yet detected from a GRB.
'in coincidence with the sub-MeV emission' already removed in
response to earlier comments.
We have not added the 'arguments to follow' clause. We have removed
the scaler analysis from the paper, so the following arguments do not
address the question of the energies of these photons.
> ...maximum energy of emission is --> ...maximum energy of emission are
change made
>... so that the precise energy threshold is undefined.
> -->
>...so that the precise energy threshold is undefined. In addition there
>may be as much as 30% uncertainty in the absolute energy scale.
change already made.
>s4p2 ...cosmological distances will --> ...cosmological distances are
this sentence has been removed in response to earlier comments.
>This allows us to place an upper limit on the redshift of GRB 970417a
>to be less than $\sim$0.3, making this a relatively nearby event.
> -->
>If the Milagrito candidate is indeed a detection of GRB 970417a, then
>the opacityallows us to place an upper
>limit on the redshift of GRB 970417a to be less than $\sim$0.3, making
>this a relatively nearby event. (explain details of models used and
> what C.L. it is)
This had already been changed in response to earlier comments to:
Thus, if Milagrito has indeed detected high energy photons from GRB
970417a, it must be from a relatively nearby object.
CONCLUSION
P_s and P_final added in several places.
>This result was found, almost simultaneously by two independent
>analyses~\cite{mcenery99,leonor99}.
>--
> (remove sentence)
change already made earlier
>The fluence above 10 GeV is in the range $10^{-4}$ to $10^{-6}$
>ergs/cm$^{2}$ and the spectrum must extend with no cutoff to at least
>a few hundred GeV.
> -->
>If we have detected a high-energy counterpart to the burst, its
>fluence above 10 GeV is in the range $10^{-4}$ to $10^{-6}$
>ergs/cm$^{2}$ and the spectrum must extend with no cutoff to at least
>a few hundred GeV.
This had already been changed in response to earlier comments to:
The observed excess corresponds to a fluence above 10 GeV is in the
range $10^{-3}$ to $10^{-6}$ ergs/cm$^{2}$ and the spectrum must
extend with no cutoff to at least a few hundred GeV.
We prefer this wording.
>The TeV fluence inferred from this result is at least an order of
>magnitude greater than the sub-MeV fluence, and the maximum energy of
>emission extends to hundreds of GeV.
> -->
>The TeV fluence inferred from this result is at least an order of
>magnitude greater than the sub-MeV fluence, and the maximum energy of
>emission could be as large as hundreds of GeV.
This is a strange way to phrase a lower limit. However, this is the
second time it has come as a EdCom requirement - I ignored it the
first time thinking that it was a mistake (inviting the wrath of
Don). We have already removed the offending part of this sentence in
response to earlier comments. We hope that this satisfies everyone.
>s5p2 ...then a new class of $\gamma$-ray bursts bright at TeV
>energies has been observed.
> -->
>...then evidence for a new class of $\gamma$-ray bursts bright at
>TeV energies has been obtained.
change made, I also changed 'has' to 'may have' The fact that we only
saw one in 54 GRBs does not necessarily mean that we have seen a new
class. The observed range of redshifts and intrinsic sub-MeV
luminosities. It is possibly that all bursts are bright at TeV
energies, but that only one was close and bright enough for us to
see. This does not make it a new class. It is simply a selection
effect on our part.
>...the current instrument, Milagro, which has increased...
> -->
>...the current instrument, Milagro, which is projected to have increased...
Milagro is our currect instrument, and is now running. The present
and not future tense is appropriate here.
FIGURES
> figure 1
>The distribution of probabilities, for each of ... (remove first comma)
change already made
> ...scale for the x-axis (add period)
change already made
>Figure 2
>...region of GRB 970417, each bin --> ...region of GRB 970417a; each bin
This caption had already been changed in response to earlier comments to
Number of events recorded by Milagrito during T90 in
overlapping 1.6$^o$ radius bins in the vicinity of GRB 970417a.
>Figure 3
>Lightcurves for GRB 970417a: (a) The crosses indicate the
>arrival time of events from the position of the candidate TeV
>counterpart for 30 seconds surrounding the BATSE trigger time. The
>histogram showed the same data binned in 1 second intervals,(b)
>Integrated in 1 sec intervals for 200 seconds surrounding the burst
>and (c) integrated in bins of 7.9 seconds (T90) for 2000 seconds
> -->
>Lightcurves for the candidate counterpart to GRB 970417a: (a) Discrete
>(crosses), and in 1-second bins, for +-15 s around the start of T90 (b)
>In 1 sec bins for +- 100 s around the start of T90 and (c) in 7.9 second
>bins (T90) for +-1000 s.
changed to:
GRB 970417a: (a) The crosses indicate the arrival time of events from
within a 2.2 radius of the candidate TeV counterpart for $\pm$15 s
around the start of T90. The histogram shows the same data binned in 1
second intervals. (b) The Milagrito data integrated in 1 sec intervals
for $\pm$100 s around the start of T90 and (c) integrated in bins of
7.9 seconds (T90) for $\pm$1000 s.
REFERENCES
>MNRAS (write out)
In all examples of reference to MNRAS in ApJ, none had this written out.
>nature --> Nature
change already made
> Astroparticle physics --> Astroparticle Physics
change already made
> add reference to Ong
Already added.
> R. Atkins et al., {\it Nucl. Inst. and Methods} (1999) (submitted)
> -->
> R. Atkins et al., {\it Nucl. Inst. and Methods} (1999) (give full)
I don't know the complete reference for the NIM paper, could someone
please send it to me.