Late-Time X-ray Flares during GRB Afterglows Extended Internal Engine Activity

a r X i v :a s t r o -p h /0105386v 2 23 N o v 2001Astronomy &Astrophysics manuscript no.(will be inserted by hand later)X-ray bursts at extreme mass accretion rates from GX 17+2E.Kuulkers 1,2,J.Homan 3,⋆,M.van der Klis 3,W.H.G.Lewin 4,and M.M´e ndez 11SRON National Institute for Space Research,Sorbonnelaan 2,3584CA Utrecht,The Netherlands e-mail:E.Kuulkers@sron.nl,M.Mendez@sron.nl2Astronomical Institute,Utrecht University,P.O.Box 80000,3508TA Utrecht,The Netherlands3Astronomical Institute “Anton Pannekoek”,University of Amsterdam,and Center for High Energy Astrophysics,Kruislaan 403,1098SJ Amsterdam,The Netherlands e-mail:homan@merate.mi.astro.it,michiel@astro.uva.nl4Department of Physics and Center for Space Research,Massachusetts Institute of Technology,Cambridge,MA 02138,USAe-mail:lewin@Received –;accepted –Abstract.We report on ten type I X-ray bursts originating from GX 17+2in data obtained with the RXTE/PCA in 1996–2000.Three bursts were short in duration (∼10s),whereas the others lasted for ∼6–25min.All bursts showed spectral softening during their decay.There is no evidence for high-frequency (>100Hz)oscillations at any phase of the bursts.We see no correlations of the burst properties with respect to the persistent X-ray spectral properties,suggesting that in GX 17+2the properties of the bursts do not correlate with inferred mass accretion rate.The presence of short bursts in GX 17+2(and similar bright X-ray sources)is not accounted for in the current X-ray bursts theories at the high mass accretion rates encountered in this source.We obtain satisfactory results if we model the burst emission with a black body,after subtraction of the persistent pre-burst emission.The two-component spectral model does not fit the total burst emission whenever there is a black-body component present in the persistent emission.We conclude that in those cases the black-body contribution from the persistent emission is also present during the burst.This implies that,contrary to previous suggestions,the burst emission does not arise from the same site as the persistent black-body emission.The black-body component of the persistent emission is consistent with being produced in an expanded boundary layer,as indicated by recent theoretical work.Five of the long bursts showed evidence of radius expansion of the neutron star photosphere (independent of the spectral analysis method used),presumably due to the burst luminosity reaching the Eddington value.When the burst luminosity is close to the Eddington value,slight deviations from pure black-body radiation are seen at energies below ≃10keV.Similar deviations have been seen during (long)X-ray bursts from other sources;they can not be explained by spectral hardening models.The total persistent flux just before and after the radius expansion bursts is inferred to be up to a factor of 2higher than the net peak flux of the burst.If both the burst and persistent emission are radiated isotropically,this would imply that the persistent emission is up to a factor of 2higher than the Eddington luminosity.This is unlikely and we suggest that the persistent luminosity is close to the Eddington luminosity and that the burst emission is (highly)anisotropic (ξ∼2).Assuming that the net burst peak fluxes equal the Eddington limit,applying standard burst parameters (1.4M ⊙neutron star,cosmic composition,electron scattering opacity appropriate for high temperatures),and taking into account gravitational redshift and spectral hardening,we derive a distance to GX 17+2of ∼8kpc,with an uncertainty of up to ∼30%.Key words.accretion,accretion disks —binaries:close —stars:individual (GX 17+2)—stars:neutron —X-rays:bursts2 E.Kuulkers et al.:X-ray bursts in GX17+2(Woosley&Taam1976;Maraschi&Cavaliere1977).Another kind of X-ray bursts was found(together withthe above type of bursts)from MXB1730−355(later re-ferred to as the Rapid Burster),which were suggested tobe due to accretion instabilities.The former and latter kind of bursts were then dubbed type I and type II,re-spectively(Hoffman et al.1978a).The main characteristics of type I bursts(for a review see Lewin et al.1993)are:sudden and short(≃1s)increasein the X-rayflux,exponential decay light curve,durationof the order of seconds to minutes,softening during the decay(attributed to cooling of the neutron star surface),(net)burst spectra reasonably well described by black-body emission from a compact object with≃10km radiusand temperature of≃1–2keV,and total energies rangingfrom≃1039to1040erg.When the luminosity during the burst reaches the Eddington limit(i.e.,when the pressureforce due to radiation balances the gravitational force),theneutron star photosphere expands.Since L b∝R2T eff4, when the radius of the photosphere,R,expands,the ef-fective temperature,T eff,drops,with the burst luminosity, L b,being constant(modulo gravitational redshift effectswith changing R)at the Eddington limit,L Edd.Burstsduring their radius expansion/contraction phase are there-fore recognizable by an increase in the inferred radius with a simultaneous decrease in the observed tempera-ture,while the observedflux stays relatively constant.The emission from a(hot)neutron star is not expectedto be perfectly Planckian,however(van Paradijs1982; London et al.1984,1986;see also Titarchuk1994;Madej 1997,and references therein).This is mainly due to the effects of electron scattering in the neutron star atmo-sphere,deforming the original X-ray spectrum.This re-sults in a systematic difference between the effective tem-perature(as would be measured on Earth),T eff,∞,and the temperature as obtained from the spectralfits,T bb (also referred to as‘colour’temperature,see e.g.Lewin et al.1993).In general,the deviations from a Planckian distribution will depend on several parameters,such as temperature,elemental abundance,neutron star mass and radius.The hardening factor,T bb/T eff,∞,has been deter-mined through numerical calculations by various people and its value is typically around1.7.When the burst lu-minosity approaches the Eddington limit the deviations from a black-body become larger,and so does the spec-tral hardening(T bb/T eff,∞∼2,Babul&Paczy´n ski1987; Titarchuk1988).During extreme radius expansion phases, however,this trend may break down and T bb/T eff,∞<1 (Titarchuk1994).Attempts have been made to determine the spectral hardening from the observed cooling tracks, but conclusions are still rather uncertain(e.g.Penninx et al.1989;Smale2001).As a result,the interpretation of X-ray bursts spectra has remained uncertain and constraints on the mass-radius relationship for neutron stars elusive.Type I X-ray burst theory predicts three differentregimes in mass accretion rate(˙M)for unstable burn-ing(Fujimoto et al.1981,Fushiki&Lamb1987;see also Bildsten1998,2000,Schatz et al.1999,and references therein;note that values of critical˙M depend on metal-licity,and on assumed core temperature and mass of the neutron star):1)low accretion rates;10−14M⊙yr−1∼<˙M∼<2×10−10M⊙yr−1:mixed H/He burning triggered by thermally unstable H ignition2)intermediate accretion rates;2×10−10M⊙yr−1∼<˙M∼<4–11×10−10M⊙yr−1:pure He shell ignition after steady H burning3)high accretion rates;4–11×10−10M⊙yr−1∼<˙M∼< 2×10−8M⊙yr−1:mixed H/He burning triggered by thermally unstable He ignitionH and He are burning stably in a mixed H/He environ-ment for very low and very high values of˙M,i.e.˙M be-low∼10−14M⊙yr−1and above∼2×10−8M⊙yr−1(close to the critical Eddington˙M).During pure heliumflashes the fuel is burned rapidly,and such bursts therefore last only5–10s.This gives rise to a large energy release in a short time,which causes the bursts often to reach the Eddington limit,leading to photospheric radius expan-sion.Bursts with unstable mixed H/He burning release their energies on a longer,10–100s,timescale,due to the long series ofβdecays in the rp-process(see e.g.Bildsten 1998,2000).The Z sources(Hasinger&van der Klis1989)are a group of sources inferred to persistently accrete near the Eddington limit.In a colour-colour diagram they trace out a Z-like shape,with the three limbs of the Z(his-torically)referred to as the horizontal branch(HB),nor-mal branch(NB)andflaring branch(FB),from top to bottom.˙M is inferred to increase from sub-Eddington at the HB,near-Eddington at the NB to super-Eddington at the FB(e.g.Hasinger1987;Lamb1989;Hasinger et al. 1990).According to the burning regimes outlined above these sources should exhibit long(∼>10–100s)type I X-ray bursts,at least on the HB and NB.However,of the Z sources,only GX17+2and Cyg X-2show(infre-quent)bursts(Kahn&Grindlay1984;Tawara et al.1984c; Sztajno et al.1986;Kuulkers et al.1995,1997;Wijnands et al.1997;Smale1998),indicating that most of the ma-terial is burning stably.This is in contrast to the above theoretical expectations for high˙M.Moreover,the bursts in Cyg X-2are short(≃5s),whereas GX17+2shows both short(≃10s)and long(∼>100s)bursts.One of the bursts of Cyg X-2showed a radius expansion phase(Smale1998); this burst clearly bears all the characteristics of a Heflash (regime2),whereas the neutron star is inferred to ac-crete at near-Eddington rates.The short duration bursts in GX17+2also hint to a Heflash origin,whereas the long duration bursts hint to unstable mixed H/He burn-ing(regime3;see also van Paradijs et al.1988).A study of X-ray bursts from Cyg X-2and GX17+2observed by EXOSAT showed no correlation of the burst properties with position in the Z,although the number of bursts observed from GX17+2was small(Kuulkers et al.1995, 1997).E.Kuulkers et al.:X-ray bursts in GX17+23The Proportional Counter Array(PCA)onboard theRossi X-ray Timing Explorer(RXTE)combines highthroughput using a large collecting area(maximum of≃6500cm2)with the ability to label photons down to atime resolution ofµs.This is ideal to study short eventslike X-ray bursts,especially during the start of the burst.Such studies may provide more insight in the propertiesof X-ray bursts which occur at these extreme mass accre-tion rates.Analysis of X-ray bursts in persistent sourcesat(relatively)high inferred mass accretion rates(typi-cally∼>0.2˙M Edd)observed with the RXTE/PCA were pre-sented for one burst seen with Cyg X-2(Smale1998)andone seen in GX3+1(Kuulkers&van der Klis2000).Inthis paper we present thefirst account of ten X-ray burstsfrom GX17+2observed by the RXTE/PCA during the period1996–2000.For a description of the correlated X-ray timing and spectral properties of GX17+2using the same data set we refer to Homan et al.(2001).2.Observations and AnalysisThe PCA(2–60keV;Bradt et al.1993)onboard RXTE observed GX17+2various times during the mission.Up to now a total of657ksec of useful data has been obtained.A log of these observations is given in Table1.During the observations in1996–1998allfive proportional counter units(PCUs)were active,whereas in1999and2000only three units were active.The high voltage settings of the PCUs have been altered three times during the mission (so-called gain changes),which modified the response of the detectors.These changes therefore mark four periods, called gain epochs.The observations were done in two standard modes:one with relatively high spectral resolu-tion(129energy channels covering the whole PCA energy band)every16sec,the Standard2mode,the other hav-ing no spectral information providing the intensity in the whole PCA energy band at a moderate time resolution of 0.125s,the Standard1mode.Additionally,data were recorded in various high time resolution(≤2ms)modes that ran in parallel to the Standard modes,and that recorded photons within a specific energy band with ei-ther low spectral resolution(B-modes or E-modes)or no spectral resolution(SB-modes).The B-,E-and SB-modes used here combined the information from all layers of all active PCUs together.During the1996observations a B-and E-mode were available,giving16and64energy bands, covering channels0–49and50–249,at2ms and125µs,re-spectively.For most of the observations in1997four SB modes covering the total PCA energy range were available. In the1998,1999,and2000observations two SB-modes covering channels0–13and14–17at125µs,and an E-mode giving64energy bands covering channels18–249at 16µs time resolution were available.For the spectral analysis of the persistent emission we used the Standard2data.We accumulated data stretches of96s just before the burst,combining the PCUs which were operating at that time.In order to study the spectral properties of the bursts we used twoapproaches.Fig.1.Standard1light curve of the GX17+2observa-tions during October10,1999,at a time resolution of5s. Time zero corresponds to05:39:27(UTC).No correctionsfor background and dead-time have been applied.Clearly,the source varies on the same time scale(and faster)as the burst which started at T=12680s.The source was in theFB and the lower part of the NB during the observations. Spectra during the bursts were determined every0.25secfor thefirst≃20s of the burst.For the short bursts this means the whole duration of the burst.For the long burstswe also used the Standard2data to create spectra at16s intervals,for evaluating the remainder of these bursts. Since no high time resolution spectral data were available during the1997observations(only4SB-modes),only the Standard2data were used to study the spectral prop-erties of the burst from this observation.All spectra were corrected for background and dead-time using the proce-dures supplied by the RXTE Guest Observer Facility1.A systematic uncertainty of1%in the count rate spectra was taken into account.For our spectralfits we confined ourselves to the energy range of3–20keV,which is best calibrated.The hydrogen column density,N H,towardsGX17+2wasfixed to that found by the Einstein SSS and MPC measurements(2×1022atoms cm−2,Christian&Swank1997;see also Di Salvo et al.2000).In all caseswe included a Gaussian line(see Di Salvo et al.2000)fixedat6.7keV,with afixed line width of0.1keV.One sigma confidence errors were determined using∆χ2=1.Large amplitude,high coherence brightness oscilla-tions have been observed during various type I X-ray bursts in other low-mass X-ray binaries(LMXBs;see e.g. Strohmayer1998,2001).We searched all the bursts fromGX17+2for such ing the high time res-olution modes we performed Fast Fourier Transforms tobook.html.4 E.Kuulkers et al.:X-ray bursts in GX17+2 Table1.RXTE observation log of GX17+2a1996feb0713:27feb0900:025811997feb0219:13feb2703:345911997apr0119:13apr0423:263501997jul2702:13jul2800:334301998aug0706:40aug0823:407111998nov1806:42nov2013:318621999oct0302:43oct1207:0529852000mar3112:15mar3116:31702T90is defined as the time it takes to observe90%of the total background-subtracted counts in an event,starting and the longer events,since the persistent emission varies on the same time scale(or even fatser)as the event itself,see e.g.Fig.1.The rise time of the events was determined as follows.We constructed light curves in the full PCA en-ergy band with a time resolution of1/32sec.We defined t rise as the time for the event to increase from25%to90% of the net peak event rate(see e.g.Muno et al.2000;van Straaten et al.2001).A detailed look at the light curves revealed that the onset of events b1–b4,b8,b10,and f2 consisted of a rapid rise phase and a subsequent slower rise to maximum.In the events b5–b7,b9,f1,and f3the count rate rapidly increased to maximum,with no sub-sequent slower rise.We therefore alsofitted the pre-event phase,fast rise phase(and slow rise phase)with a constant level and one(or two)linear functions,respectively,using the Standard1light curves.Wefind that the total du-ration of the fast rise phase,t fr,is between0.1and0.6sec for events b1–b10and0.4–1s for f1–f4(Table2).In the fits to the decay portion of the light curves for the short events we used the Standard1light curves at0.125sec time resolution,while for the longer events we rebinned these light curves to a time resolution of5sec.For some of the events an exponential does not describe the decay very well;this is probably due to the short term variations in the persistent emission.For these events wefitted only the initial decay(first few seconds for f2and f3,andfirst few100s for burst b9and b10).In Fig.2we show the light curves of the four events f1–f4,at low(∼<7keV)and high(∼>7keV)energies,with the corresponding hardness(ratio of the count rates in the high and low energy band)curves,all at a time resolution of0.125s.Although they have a fast rise and a longer decay(see also Table2),they show small or no variations in hardness.Time-resolved spectral analysis(like done in Sect.3.2.3)confirmed this;no clear cooling during the decay can be discerned.We can therefore not classify these events as type I bursts.Since all of the four events occurred in the FB,we conclude that they must beflares,of which the light curves happen to resemble those of X-ray bursts (such as b1,b3and b5).We will not discuss these fourE.Kuulkers et al.:X-ray bursts in GX17+25 Table2.Bursts and burst-like events(flares)in GX17+2b11996feb0816:17:121510 1.220.35±0.05 1.83±0.08 1.1/130mNB b21997feb0802:36:3435>360 1.190.34±0.08248+4−9 2.5/49mNB b31998aug0713:15:5035100.530.27±0.09 2.55±0.24 1.0/131lNB b41998nov1808:51:26351000 1.340.61±0.04197±2 3.7/147SV b51998nov1814:37:3035100.410.54±0.04 2.06±0.13 1.1/85mNB b61999oct0315:36:3243(0,2,3)16000.410.19±0.04274±3 2.0/242lHB b71999oct0523:41:4343(0,2,4)5000.560.30±0.0377.3±1.2 1.9/104uNB b81999oct0611:10:3343(0,2,3)500 1.660.13±0.0270.2±1.4 1.2/57lHB b91999oct0912:34:2443(0,2,3)5000.130.16±0.0276.4±1.5 3.1/66uNB b101999oct1009:10:4743(0,2,3)7000.720.20±0.04115±3 3.7/57lNB f11996feb0703:39:111510 1.03 1.12±0.07 2.98±0.49 1.3/88mFB f21998nov1914:38:2435100.410.33±0.04 1.72±0.35 1.1/28uFB f31998nov2000:44:4335100.750.45±0.06 1.85±0.25 1.1/55uFB f41999oct1108:55:3043(0,2,3)10 5.0 1.06±0.20 2.18±0.22 1.1/148lFB6 E.Kuulkers et al.:X-ray bursts in GX17+2Fig.3.Same as Fig.2,but for the three short X-ray bursts b1,b3,and b5.The low and high energy ranges are 1–7.2keV and 7.2–19.7keV,respectively,for b1,whereas they are 1.9–6.2keV and 6.2–19.6keV,respectively,for b3and b5.events further in the paper and denote the remainder of the ten events as bursts,since we will show below that they are genuine type I X-ray bursts.In Fig.3we show the light curves of the three short bursts b1,b3and b5,at low and high energies,with the corresponding hardness curves,all at a time resolution of 0.125s.All three bursts show a fast rise (typically less than 0.5s)and an exponential decay with a decay time of ≃2s (see Table 2).During the rise the emission hardens;as the bursts decay,the emission becomes softer.The main difference between the three bursts is the fact that the peak of burst b3is about 25%lower than the other two bursts;it looks like a ‘failed’burst.It also has two peaks,as if some new unstable burning occurred,≃5s after the start of the burst.In Figs.4and 5we show the light curves of the long bursts b2,b4,and b6–b10,at low and high energies,with the corresponding hardness curves,all at a time resolu-tion of 2s.In Figs.6and 7we focus on the start of these bursts,all at a time resolution of 0.125s.Again the rise times are very short (also typically less than 0.5s),but the decay times are much longer,with decay times in the range ≃70–280s (see Table 2).Apart from the fact that the hard burst emission decays faster than the soft burst emission (i.e.spectral softening),there are more pronounced differ-ences between the light curves in the two energy bands.In Figs.4and 5one sees that all the low energy light curves show a kind of spike at the start of the decay.These spikes last for a few seconds in most cases;however,for burst b6it seems to last ≃15s (with an exponential de-cay time of 9±2s).At high energies no such spikes occur(except for burst b10);instead the bursts have more flat-topped peaks,with durations ranging from tens of seconds to ≃200s.Burst b6is the nicest example of this.Zooming in on the start of these long bursts,it becomes clear that the rise is somewhat slower at high energies with respect to low energies (causing the hardening of the emission dur-ing the early phase of the burst).Also,in bursts b4and b6–b9very short (<0.5s)spikes occur during the rise in the high energy light curve,which again are especially ev-ident in burst b6(two spikes!).This causes the hardness to drop on the same time scale in some of these ter we will show that this corresponds to fast radius expansion/contraction episodes.3.2.Spectral behaviour3.2.1.Persistent emission before the burstsThe persistent emission just before the bursts b6and b7can be satisfactorily (χ2red =0.7and 1.2,respectively,for 35degrees of freedom,dof)described by an absorbed cut-offpower-law component plus a Gaussian line.For the per-sistent emission before the remainder of the bursts this is not a satisfactory model (χ2red ranges from 1.4with 35dof for b8to 5.5with 42dof for b3).An additional compo-nent is necessary to improve the fits.We used the F-test to calculate whether the additional component was indeed significant.For the additional component we chose a black body,as is generally used when modeling the X-ray spec-tra of bright LMXBs (e.g.White et al.1986;Christian &Swank 1997;Church &Ba l uci´n ska-Church 2001).Note that more complicated models are necessary to describeE.Kuulkers et al.:X-ray bursts in GX17+27Fig.4.Same as Fig.3,but at a time resolution of2s,for the long bursts b2,b4and b6.Note that b2was interrupted by a South Atlantic Anomaly(SAA)passage as can be seen by the sudden decrease in count rate.The low and high energy ranges are1.9–6.2keV and6.2–≃60keV,respectively,for b2,1.9–6.2keV and6.2–19.6keV,respectively,for b4, and2.1–7.1keV and7.1–19.9keV,respectively,for b6.Table3.Persistent emission spectral parameters ab1 2.0±0.3 1.15±0.0614.1±1.5 1.0±0.1 4.5±0.2 3.8±0.40.013±0.0020.89/47165×10−10b2 2.2±0.5 1.10±0.0322.1±2.20.4±0.2 3.8±0.2 1.5±0.40.008±0.0020.56/40294×10−18b3 1.9±0.3 1.13±0.0319.2±1.50.9±0.1 4.2±0.2 2.6±0.40.013±0.0020.58/40291×10−18b4 2.0±0.3 1.10±0.0420.3±1.80.7±0.1 4.0±0.2 2.3±0.40.014±0.0020.85/40286×10−15b5 2.3±0.3 1.16±0.0517.5±1.70.9±0.1 4.5±0.2 3.2±0.40.010±0.0020.59/40235×10−15b6 2.5±0.1—— 1.03±0.03 5.1±0.1 4.8±0.10.012±0.0020.71/35<7i8b7 2.5±0.1—— 1.17±0.03 4.6±0.17.0±0.20.015±0.002 1.22/35<2i30b8 2.4±0.5 1.08±0.1311.0±4.3 1.0±0.1 5.3±0.3 3.8±0.50.011±0.002 1.13/3361b9 2.4±0.4 1.26±0.0811.6±1.5 1.0±0.1 4.9±0.3 4.3±0.40.010±0.0030.87/33138×10−5b10 1.9±0.3 1.18±0.0515.1±1.4 1.1±0.1 4.5±0.3 3.7±0.50.011±0.002 1.16/33225×10−78 E.Kuulkers et al.:X-ray bursts in GX17+2Fig.5.Same as Fig.4,for the last four long bursts observed in 1999Oct (b7–b10).The low and high energy ranges are 1.9–7.1keV and 7.1–19.9keV,respectively.Fig.6.Same as Fig.4,but 5s before and 10s after the start of the bursts b2,b4and b6,at a time resolution of 0.125s.b7we determined single parameter 95%confidence up-per limits (using ∆χ2=2.71)on the black-body contribu-tion,by including a black-body component in the spec-tral fits,and fixing the temperature to its mean value de-rived for the persistent spectra of the other bursts,i.e.kT bb =1.14keV.The black-body component contributionto the persistent emission varied between less than 2%(burst b7)up to 29%(bursts b2,b3)in the 2–20keV band (Table 3).3.2.2.Persistent emission during the bursts?3.2.2.1Previous EXOSAT resultsUsually it is assumed that the persistent emission is not influenced by the burst and that one can,therefore,study the burst by subtracting the persistent emission from the total source emission.This is referred to as the ‘stan-dard’burst spectral analysis (see e.g.Sztajno et al.1986).However,if the neutron star photosphere contributes sig-nificantly to the persistent emission,this approach is not correct,if the burst emission originates from the same region (van Paradijs &Lewin 1986).In this case the spec-tral fits to the net burst spectra yield a systematically larger black-body temperature,T bb ,and smaller appar-ent black-body radius,R bb ,especially near the end of the burst when the net burst flux is low.In fact,in this case of incorrect subtraction of the persistent emission,the net burst spectrum is not a black-body.Van Paradijs &Lewin (1986)proposed to fit the total source spectrum with a two-component model,a black-body component and a non-black-body component.During the burst the non black-body component is fixed to what is found in the persistent emission,and the black-body component is left free.The black-body component should include all emission from the neutron star photosphere.The underly-ing idea is that the non black-body component arises fromE.Kuulkers et al.:X-ray bursts in GX17+29Fig.7.Same as Fig.5,but5s before and10s after the start of the last four bursts observed in1999Oct(b7–b10),at a time resolution of0.125s.Fig.8.Leftmost panel:Two-component spectralfit results for the total burst emission of burst b4plotted on a logarithmic time scale.Thefilled circles and open squares represent thefit results of the0.25s and16s spectra, respectively.The data have been logarithmically rebinned in time for clarity.The values for the persistent black-body component have been indicated by a dotted line.Right panels:Two-component spectralfit results for the0.25s spectra of the total burst emission of burst b1,b4and b6,see text.The values for the persistent black-body component have been indicated by a dotted line for burst b1and b4.Both panels:from top to bottom:bolometric black-bodyflux,F bb, in10−8erg s−1cm−2,black-body temperature,kT bb,apparent black-body radius,R bb,10,at10kpc,and goodness of fit expressed inχ2red.For bursts b1,b4and b6the number of dof is22,18and16,respectively,for the0.25s spectral fits.For burst b4the number of dof is44for the16s spectralfits.Note the difference in scales ofχ2red in the leftmost panel with respect to the other panels.10 E.Kuulkers et al.:X-ray bursts in GX17+2Fig.9.a :At the top the first 16s spectrum observed after the start of burst b2is displayed.A two-component fit is shown,i.e.a single black-body and cut-offpower-law plus Gaussian line (subjected to interstellar absorption).The parameters of the cut-offpower-law component plus Gaussian line are fixed to the values derived for the persistent emission.At the bottom the residuals after subtracting the best model from the observed spectrum is displayed.The fit is clearly bad (χ2red /dof=24.2/44).b :At the top the same first 16s spectrum at the beginning of burst b2is displayed.Now a three-component fit is shown,i.e.two black-body components and one cut-offpower-law component (subjected to interstellar absorption).The cut-offpower-law and one black-body component parameters are fixed to the values derived for the persistent emission.At the bottom the residuals after subtracting the best model from the observed spectrum is displayed.The fit has clearly improved (χ2red /dof=1.8/44).the accretion process,and is not influenced by the X-ray burst.GX 17+2is a bright X-ray source,and presumably the neutron star contributes significantly to the persistent emission (Van Paradijs &Lewin 1986;Sztajno et al.1986).Since the source is bright,compared to most other burst sources the net flux at the peak of the burst relative to the persistent flux is rather low.Sztajno et al.(1986)found that the black-body component contributed ≃40%to the persistent emission just before the two bursts observed by ing the ‘standard’burst spectral analysis Sztajno et al.(1986)found relatively high black-body tem-peratures (kT bb ≃2–3keV)and relatively small apparent black body radii at a distance of 10kpc (R bb ,10≃3–5km,for isotropic emission)for the short (≃10s)burst.For the long (>5min)burst,kT bb only showed a small change from ≃2.1keV at the peak of the burst to ≃1.7keV near the end of the burst,with R bb ,10decreasing from ≃7km to ≃4km.Their fits were,however,satisfactory,with χ2red of 0.6–ing the two-component model,Sztajno et al.(1986)found that during the short burst the values for T bb were in the range normally seen in type I X-ray bursts,and that the systematic decrease in R bb had disap-peared.The χ2red values for the two-component model fits ranged between 0.6and 1.3,except for one fit,for which χ2red =1.6,i.e.slightly worse than the ‘standard’spectral analysis.They attributed this to the smaller relative er-ror in the total burst data than in the net-burst data,for which the persistent emission was subtracted.Note that in this case Sztajno et al.(1986)found a slight increasein R bb ,10,apparently anti-correlated with T bb ,which they argue was due to the non-Planckian shape of the spectrum of a hot neutron star (van Paradijs 1982;see Titarchuk 1994;Madej 1997,and references therein;see,however,Sect.5.2).3.2.2.2Our RXTE resultsGuided by the results of Sztajno et al.(1986)discussed in Sect.3.2.2.1,we first fitted the total burst data using a black body and a cut-offpower law plus a Gaussian line.The parameters of the absorbed cut-offpower law and Gaussian line were fixed to the values found for the persistent emission before the burst (Table 3).Using this model we obtained good fits to the 16s spectra of bursts b6and b7,for which the persistent emission did not con-tain a significant black-body contribution (χ2red of ≃1for 37dof).However,the 16s spectral fits were bad when-ever the persistent emission spectra contained a black-body component;the fits became worse as the persis-tent black-body contribution became stronger (for the first ∼100s of the burst:χ2red /dof ≃1.5–3/37[burst b8],χ2red /dof ≃2.5–8/37[burst b9],χ2red /dof ≃5–10/37[burstb10],χ2red /dof ≃20–24/44[burst b4],χ2red /dof ≃22–25/44[burst b2]).The worst χ2red occurred near the peak of the bursts.For instance,Fig.8shows the best-fit parameters and χ2red for burst b4,for which the persistent emission had a black-body contribution of ≃28%(2–20keV).An example of a burst spectrum and the result of the two-component fit is shown in Fig.9a for burst b2.The χ2red。

AE中particular插件中英文对照（一）Emitter（发射器）Particular/sec (粒子/秒)——每秒钟发射粒子的数量。

Emitter Type（发射器类型）——它决定发射粒子的区域和位置。

Point（点）——从一点发射出粒子Box（盒子）——粒子会从立体盒子中发射出来，（Emitter Size中XYZ是指发射器大小）Sphere（球形）——和Box很像，只不过发射区域是球形Grid（网格）——（在图层中虚拟网格）从网格的交叉点发射粒子Light（灯光）——（要先新建一个Light Layer）几个Light Layer可以共用一个Particular。

Layer——使用合成中的3D图层生成粒子，Layer Grid——同上，发射器从图层网格发射粒子，像Grid一样。

Direction（方向）Uniform（统一）——任一粒子发射类型发射粒子时，会向各个方向移动。

Directional（特定方向）——（如枪口喷射）通过调节X、Y、Z Rotation来实现。

Bi-Directional（相反特定方向）——和Directional十分相似，但是向着两个完全相反的方向发射。

通常为180度。

Disc（盘状）——通常只在两个维度上，形成一个盘形。

Outwards （远离中心）——粒子会始终向远离发射点的方向移动。

而Uniform是随机的。

Direction Spread（方向拓展）——可以控制粒子发射方向的区域。

粒子会向整个区域的百分之几运动。

（即粒子发射方向有多宽） Velocity（速度）——粒子每秒钟运动的像素数。

Velocity Random——每个粒子Veloc ity的随机性会随机增加或者减小每个粒子的Velocity。

Velocity Distribution（）—— Velocity from Motion（速度跟随运动）——粒子在向外运动的同时也会跟随发射器的运动方向运动。

Swift Observations of GRB 060926S.T. Holland (GSFC/USRA), S.D. Barthelmy (GSFC), R. Starling (U.Leicester), L.M. Barbier (GSFC), M. Perri (ASDC), M. Capalbi (ASDC), P. Roming (PSU), K. Page (U.Leicester),J. Nousek (PSU), N. Gehrels (GSFC)for the Swift Team0. REVISIONSCorrected the mistake about which UVOT detections were retracted. In version 1 it was erroneously stated that both the UVW1- and V-band detections were retracted. This is wrong. Only the UVW1-band data point was retracted; the V-band detection still stands. Changed conversion factor in Fig 2 caption.1. INTRODUCTIONBAT triggered on GRB 060926 at 16:48:41 UT (Trigger=231231) (Holland, et al., GCN Circ. 5612). This was a 1.024-sec rate trigger. Swift slewed immediately to this burst, and the XRT & UVOT began making follow-up observations. Our best position is the UVOT location: RA,Dec(J2000)=263.9319,+13.0385 with a 1-sigma error radius of 0.5arcsec. The initial UVOT UVW1-band detection of an afterglow was later retracted. This burst has a spectroscopic redshift of 3.208 by the MISTICI collaboration using the VLT (V. D'Elia et al., GCN Circ 5637).2. BAT OBSERVATION AND ANALYSISUsing the data set from T-119 to T+183 sec, we report further analysis of GRB 060926 (Holland, et al., GCN Circ. 5612). The BAT ground-calculated position is RA,Dec(J2000) = 263.925, 13.039 deg {17h35m 41.9s, 13d 2' 21.5"} ± 1.4 arcmin, (radius, sys+stat, 90% containment). The partial coding was 94% (the bore sight angle was 9.1 deg).The mask-weighted lightcurve (Fig 1) shows a single FRED peak starting at T-1 sec, peaking at T+1 sec, and ending at ~T+10 sec. T90 (15-350 keV) is 8.0 ± 0.1 sec (estimated error including systematics).The time-averaged spectrum from T-0.1 to T+8.6 sec is best fit by a simple power-law model. The power law index of the time-averaged spectrum is 2.54 ± 0.23. The fluence in the 15-150 keV band is 2.2±0.3 x 10-7 erg/cm2. The 1-sec peak photon flux measured from T+0.38 sec in the 15-150 keV band is 1.1±0.1ph/cm2/sec. All the quoted errors are at the 90% confidence level.3. XRT OBSERVATION AND ANALYSISXRT began follow-up observations at T+60 s.We have analyzed the whole XRT data set on the GRB 060926 (Holland, et al., GCN Circ. 5612). A 4.7 ks Photon Counting mode image provides a refined XRT position: RA,Dec(J2000) = 17h 35m 43.93s, +13d 02' 18.4" with an uncertainty of 5.9" (90% containment, including bore sight uncertainties). This is 29.8" away from the center of the BAT refined position (Cummings et al., GCN Circ. 5621). This localization lies 11.3" from the initial XRT position.The 0.3-10 keV X-ray light curve (Fig 2) during the first orbit shows a flare at T+430s. From the second orbit a power-law decline with a temporal index of -1.4±0.3 is observed.The X-ray spectrum covering the time period from T+67s to T+878s is well fit by an absorbed power-law with a photon index of 2.1(±0.3) and column density of (2.2±0.9)e21 cm-2. We note the Galactic column density in the direction of the source is 7.3e20 cm-2.4. UVOT OBSERVATION AND ANALYSISThe UVOT began observing GRB 060926 57 seconds after the BAT trigger (Holland et al., GCN Circ 5612). A source was detected at the 4.8-sigma level in the V filter, located at RA(J2000) = 17:35:43.66 = 263.9319deg, DEC(J2000) = +13:02:18.6 = 13.0385deg, with a 1-sigma error radius of 0.5 arc sec. The source has a magnitude of 19.0+/-0.2 as determined by the Swift analysis tool, uvotsource, (Roming & Holland, GCN Circ 5625). The initial 4.3-sigma level detection of a source in the UVW1 filter was later retracted (Roming & Holland, GCN Circ 5645).No optical afterglow is detected at the 3-sigma level in individual or coadded exposures in the UVOT UVW1-, UVW2-, UVM2-, U-, or B-filters. The 3-sigma limiting magnitudes for the coadded images of the UVOT filters are listed below:Filter T_range(s) Exp(s) Upper Limit (3-sigma)UVW1 487-11748 1337 20.4UVW2 551-6405 432 20.1UVM2 463-10840 1366 20.4U 511-12283 968 20.8B 535-6199 422 20.6where T_range is the start and end times of the coadded exposures. No correction has been made for Galactic reddening along the line of sight (E(B-V) = 0.16).No images were taken in the White-filter by the UVOT since bright stars were in the field-of-view.Fig.1: BAT Lightcurve. The mask-weighted light curve in the 4 individual plus total energy bands. The units are counts/sec/illuminated_detector (note illum_det = 0.16 cm2) and T_zero is 16:48:41 UT.Fig. 2:XRT Lightcurve. There is a flare at T+430 sec. From the second orbit a power-law decline witha temporal index of -1.4±0.3 is observed. The conversion factor from count rate to observed (unabsorbed)0.3-10 keV flux is approximately 1 cnt/s = 4.2x10-11 (6.5x10-11) erg/cm2s.。

optical flares 英文版Optical flares are a visual phenomenon caused by the interaction of light with the camera lens. They are often used in photography and videography to add a sense of realism or artistic effect. The term "optical flare" refers to the patterns of light that appear when a light source is scattered or diffracted within the lens system of a camera. This effect can be seen in everyday life when looking at bright lights, but it is particularly pronounced and sought after in visual media for its dramatic impact.The creation of an optical flare involves several physical processes. When light enters a camera lens, it passes through various elements and coatings designed to focus and direct the light onto the camera's sensor. However, imperfections in the lens elements, such as dust or scratches, as well as the inherent design of the lens, can cause some of the light to scatter. This scattered light creates the streaks and halos characteristic of an optical flare.In addition to natural occurrences, optical flares can be artificially created using software. Visual effects artists often use computer-generated imagery (CGI) to add flares in post-production. This allows for greater control over the appearance and intensity of the flare, enabling artists to match the flare to the mood and style of the scene.One of the most popular tools for creating and customizing optical flares in post-production is the Optical Flares plugin for Adobe After Effects. Developed by Video Copilot, this plugin provides a wide range of options for simulating realistic lens flares. Users can choose from a library of preset flares or create their own custom flares by adjusting various parameters such as brightness, color, position, and complexity.The use of optical flares in film and video can serve multiple purposes. They can be used to simulate the natural reaction of a camera to bright lights, which can help to ground computer-generated elements in reality. Flares can also be used to draw the viewer's attention to a particular part of the frame or to convey a character's point of view. In some cases, flares are used purely for their aesthetic value, adding a sense of dynamism and energy to a shot.While optical flares can enhance the visual appeal of a scene, they must be used judiciously. Excessive or inappropriate use of flares can distract from the narrative and reduce the overall impact of the scene. Therefore, filmmakers and visual effects artists must carefully consider when and how to incorporate flares into their work.In conclusion, optical flares are a powerful tool in the visual artist's toolkit. Whether occurring naturally or crafted in post-production, they can add depth, emotion, and realism to a scene. As technology advances, the ability to create and manipulate optical flares becomes increasingly sophisticated, offering endless possibilities for enhancing visual storytelling. The key to effective use of optical flares lies in the balance between artistic expression and narrative clarity, ensuring that each flare serves a purpose and contributes to the overall vision of the project. 。

Key FeaturesFull HD 1080p ResolutionEdge LED Backlighting with Monolithic Design Motionflow™ 120Hz Technology for Smooth Motion Built-in Wi-Fi®1BRAVIA® Internet Video & Widgets 2BRAVIA Engine™ 3 fully digital video processorLightSensor™ adjusts backlight with room light USB port for photos, music & video playback 3 Energy Saving Switch eliminates standby power Key TechnologiesFull HD 1080p Resolution Experience Full HD 1080p picture quality, the highest at-home resolution,and take full advantage of HD sources like a Blu-ray Disc™ Player or PlayStation®3 gaming console via the 16:9 wide screen panel (1920 x 1080).Edge LED backlight Enjoy a slim design plus amazing dynamic contrast with an Edge LED backlight.Motionflow™ 120Hz Technology Experience smooth motion detail and clarity with Motionflow™120Hz technology. Motionflow™ 120Hz technology doubles the frame rate from the standard 60frames per second and applies Image Blur Reduction technology to deliver a crisp, fluid and realistic viewing experience during fast-action scenes.Built-In Wi-Fi® Enjoy the convenience of built-in wireless connectivity to your home entertainment network and make accessing your online entertainment - including movies, photos, music, videos and more - even easier!1 BRAVIA® Internet Video Instantly stream a wide variety of movies, TV episodes, videos and musicfrom a large selection of entertainment apps including: Netflix®, YouTube™, Slacker® - the list goes on and on.2 BRAVIA® Internet Widgets Personalize your viewing experience by adding information apps toyour screen so you can instantly access the latest local news, weather, sports headlines and Twitter® updates.2 BRAVIA Engine™3 Enjoy Sony's most advanced processing system for superior image detail,exceptionally vibrant, life-like images, reduced noise and optimized contrast and color with the BRAVIA Engine™ 3 fully digital processor.LightSensor™ Technology Automatically increase or decrease the level of brightness of your TV'sbacklight based on a room's lighting conditions with the integrated LightSensor™ technology. Dim the lights to watch a movie and the sensor dims the backlight, which also helps reduce unnecessary power consumption.USB Input Share your photos on the big screen or listen to your favorite music. Simply connectyour digital camera, USB-enabled MP3 player, or USB storage device directly to your HDTV's USB input.3 Energy Saving Switch Eliminate standby power consumption without having to unplug your TV.Simply turn the Energy Saving Switch to off position and the TV no longer draws energy from the power source.Sony Electronics Inc. • 16530 Via Esprillo • S an Diego, CA 92127 • 1.800.222.7669 • w Last Updated: 02/22/2010Please visit the Dealer Network for currentinformation at /dnFeaturesGeneralClosed Captions (CC): Analog, Digital Illuminated Sony Logo: Yes Finish (Surface Color): Black XMB: YesAudio Return Channel: Yes HDMI™ InstaPort: YesDisplayBacklight Type: LED Display Device: LCD Deep Black Panel: YesVideoVideo Processing: BRAVIA Engine 3™ TechnologyCineMotion® Reverse 3:2 Pulldown Technology: Yes Wide Mode: Normal, Zoom, Wide Zoom, Full (Full1-2 for PC), H Stretch Deep Color: Yes 3D Comb Filter: YesPhotoTV HD: YesMPEG Noise Reduction: Yes Auto Wide: Yes 4:3 Default: Yes24p True Cinema™ Technology: YesPicture Mode: Vivid, Standard, Custom, Cinema, Game-Standard, Game-Original, Graphics, Sports, Photo-Vivid, Photo-Standard, Photo-Original, Photo-Custom Advanced Contrast Enhancer (ACE): Yes Motionflow™ Technology: 120Hz Cinema Mode: YesLive Color™ Technology: YesAudioAuto SAP: 20FM/10AMAuto Mute (on no signal): YesSteady Sound® Automatic Volume Control: Yes Audio Out: Variable/Fixed Sony Original Surround: Yes Stereo System: MTS Dolby®: Dolby® DigitalAlternate Audio (Digital): yesSurround Effect: Cinema, Music, Live Sports, Game5.1 Channel Audio Out: YesSound Mode: Dynamic, Standard, Clear Voice, Custom5.1ch Through Out (Handycam): Yes Sound Enhancer: YesClear Voice: YesInvisible Speaker (w/Sound Elevation): YesConvenienceChannel Jump: YesMultiple Language Display: English, French, Spanish Info Banner: TV Guide on Screen, PSIP, XDS Sleep Timer: YesInput Skip: Automatic Channel Surf: YesChannel Block (with password): Yes Edit Channel Labels: Yes Label Inputs: YesOn/Off Timer: YesUSB Application(s): Photo Viewer (jpeg), Photo Viewer (RAW), Music Player (mp3), Video Player (AVCHD), Video Player (AVC), Video Player (MPEG4) Theatre Mode (Theatre Button): YesShow/Hide Channels: YesAuto Shut-off: YesBRAVIA® Sync™: Yes Parental Control: YesBRAVIA® Sync™ Menu Button: Yes (on Remote Control)On-Screen Display for External Input: Yes PIP (Picture in Picture): PC Input(HDMI)/TV Channel&CVBS Channel Labels: YesAuto Photo Mode: via HDMI, via DLNA, via USB On-Screen Clock: YesPicture Adjustments Lock: Yes White Balance: RGB Gain/BiasAuto Picture Setting w/HDMI™: Cinema, Sport, Music, Photo, Graphics, Game Internet Radio: YesScene Select: Auto, Cinema, Sports, Photo, Music, Game, Graphics, OFFTheatre Mode with Cinema Surround: Yes TV Guide On-Screen: YesStand Tilt: YesStand Tilt Range: 2position 0°/ Back 6° On-Screen Operating Instructions: Yes Favorites: YesPicture Frame Mode: Yes Quick Start: YesStand Swivel: YesStand Swivel Range: Left 20°/Right 20° Start Up Sony Logo Display: Yes USB Player: YesWireless LAN: Integrated Upward Style: YesNetworkDLNA: YesBRAVIA® Internet Video: Ready BRAVIA® Internet Widgets: Yes Photo MAP: YesSpecificationsDisplayViewing Angle (Right/Left): 178(89/89) Viewing Angle (Up/Down): 178(89/89) Dynamic Contrast Ratio: InfiniteScreen Size (measured diagonally): 46"Display Resolution: Full HD 1080(1920x1080) Aspect Ratio: 16:9TV SystemTuner (Cable): Clear QAMTuner (Terrestrial): ATSC™ / NTSC Channel Coverage: VHF: 2-13ch UHF: 14-69chCATV (analog): 1-135ch CATV (digital) : 1-135ch TV System (Analog): MNumber of Tuners (Digital): 1(Hybrid)TV System (Digital): ATSC Color System: NTSCVideoVideo Signal: 1080/60p (HDMI™ / COMPONENT), 1080/60i, 1080/24p (HDMI™ ONLY), 720/60p, 480/60p, 480/60iAudioAudio Power Output: 20W (10W x 2) Speakers (Total): 2Speaker Configuration: 2Inputs and OutputsComponent Video (Y/Pb/Pr ) Input(s): 1(1 Side) (Composite input common use))Composite Video Input(s): 2 ( 1 Side (Component input common use)/ 1 Rear) Digital Audio Output(s): Optical Digital Output - 1 (Rear)RF Connection Input(s): 1 (Rear)Analog Audio Input(s) (Total): 3 (2 Side/1 Rear) Audio Out: 1 (Rear)HDMI™ Connection(s) (Total): 4(2 Side/2 Rear) PC In (D-Sub) + Audio In (Stereo Mini): 1 (Side) USB 2.0: 1 (Side)HDMI™ PC Input: Yes (see owner's manual for supported timings) AC Power Input: 1 (Rear)Ethernet Connection(s): 1 (Rear)PowerPower Consumption (in Standby): 0.17W Power Requirements (voltage): AC 120V Power Requirements (frequency): 60HzPower Consumption (in Operation) max.: 129WPower Consumption (in Download Acquisition Mode): Less than 16 WRegulation and Standard ComplianceVESA® Hole Spacing Compatible: Yes VESA® Hole Pitch: 300x300 M6Service and Warranty InformationLimited Warranty Term: 1 Year Labor / 1 Year Parts Online Instruction Manual (Reference Book): YesEnergy Saving & EfficiencyLightSensor™ Technology: Yes Power Saving Modes: Yes Eco Settings: YesDynamic Backlight Control: Yes Backlight Off Mode: YesEstimated Annual Energy Consumption (kWh): 163.3kWh/year Idle TV Standby: YesEnergy Saving Switch: YesRecycled & Recyclable MaterialsProduct Materials: Recycled Carton (more than 40%)DimensionsWeight: 61.3 lbs. (27.8kg) with pedestal; 52.9 lbs. (24kg) without pedestal Measurements: 44.375 x 29 1/2 x 12 5/8 in (1127 x 748 x 320mm) with pedestal; 44.375 x 28 1/4 x 2.625 in (1127 x 716 x 64mm) without pedestalSupplied AccessoriesRemote Control (RM-YD037)Batteries (Type AAA x2) AC Power CordScrews (for pedestal, 4-screws)Table Top Stand (separate, pre-assembled)Optional AccessoriesWallmount Bracket (SU-WL700)UPC Code: 027*********1. Wireless router required (sold separately). Requires a compatible 802.11n access point. Some functionality may require Internet services.2. Connection speed of at least 2.5 Mbps recommended (10 Mbps for HD content). Video quality and picture size vary and are dependent upon broadband speed and delivery by content provider. Select content provided subject to change. Premium content may require additional fees and/or PC registration3. USB device must be formatted FAT-32.4. Dynamic contrast ratio compares the luminance of a 100 IRE (white) signal with that of a 0 IRE (black) signal. When the TV receives a 0 IRE signal the backlight is turned off and the luminance measurement is 0.00 cd/m . The resulting dynamic contrast ratio is infinite. © 2010 Sony Electronics Inc. All rights reserved. Reproduction in whole or in part without written permission is prohibited. Sony, BRAVIA, Motionflow, BRAVIA Engine, LightSensor, make.believe, and the Sony logo, Full HD 1080p logo and BRAVIA logo are trademarks of Sony. PlayStation is a trademark of Sony Computer Entertainment. Blu-ray Disc is atrademark. All other trademarks are trademarks of their respectiveowners. Features and specifications are subject to change without notice. Non-metric weights and measures are approximate and may vary.2。

a r X i v :0806.4243v 1 [a s t r o -p h ] 26 J u n 2008Characterizing X-ray Variability of TeV Blazars ∗Speaker.F2av, a(f)=1n n−1∑j=0F j sin(2πft j),(2.1)where F j is the source count rate at time t j(0≤j≤n−1),T is the data length of the time series and F av is the mean value of the source counting rate.The power due to the photon counting statistics is given byσ2stat.With our definition,integration of power over the positive frequencies is equal to half of the light curve excess variance(e.g.,Nandra et al.1997).To calculate the NPSD of certain data sets,it is recommended to make light curves of two different bin sizes shorter/longer than orbital gaps(e.g.,256and5760sec,respectively).Each light curve is divided into“segments”,which are defined as the continuous part of the light curve.∑[a(t)−a(t+τ)]2,(2.2) N∑UDCF i j(τ),(2.4)Mwhere M is the number of pairs in the bin.The DCF advantages are that it uses all the data points available,does not introduce new errors through interpolation,and calculates a meaningful error estimates.The standard error for each bin(∑[UDCF i j−DCF(τ)]2)1/2.(2.5)M−1As an application,Figure5shows an example to see time-lag in the light curve of PKS 2155−304(Kataoka et al.2000).The data reveal a largeflare at the beginning,followed by lower amplitudefluctuations.The source variability is somewhat different in different energy bands.No-tably,amplitude offlux change is larger at higher photon energies;a factor of2at1.5−7.5keV (red),while it is a factor of1.5in the0.5−1.5keV(blue).Also note that the peak of the light curve in the hard X-ray bands leads that in the soft X-ray bands by∼4−5ksec.This was also suggested by directfitting of the light curves with a simple Gaussian plus constant offset,resulting that a lag of the peaking time by≃4ksec.We therefore computed the cross correlations using the the DCF by dividing the0.5−7.5keV range intofive energy bands and measured the time lag for each light curve compared to the3.0−7.5keV light curve.The results are shown in Figure5(right),again suggesting≃4ksec lags in the X-ray variability of PKS2155−304.As we have seen in§1,reality of this small amount of lag is still matter of debate,due to the periodic gaps(∼6ks)of low-Earth orbit satellites(e.g.,Edelson et al.2001).Meanwhile,it is also suggested that lags on hour-scale can hardly be produced by periodic gaps based on careful simulations(e.g.,Tanihata et al.2001;Zhang et al.2004).To quickly follow their arguments,I have made hundreds pairs of light curves by Monte Carlo simulation,one of which is artificially “lagged”by4ksec.Then the resultant light curves arefiltered by the same window as the actual observation.Figure6(le ft)shows an example pair of light curves thus produced,and Figure6 (right)shows the calculated DCF for10pairs of light curves.It seems that the DCF exhibits large uncertainties but the peak of the DCF is always retained as expected(i.e.,4ksec).Obviously, higher quality X-ray data which are less affected by window sampling is strongly awaited forSource Name Redshift Class Flux(2-10keV)Flux(≥100MeV)[10−12erg/cm2/s][10−5ph/cm2/s] Table1:A list of“VIP”blazars to be simultaneously observed with GLAST and Suzaku in2008/09.Figure7:An expected X-ray sky map for1day exposure with MAXI.an unprecedented accuracy.Coordinated observations between GLAST and X-ray satellites are crucial for further understanding the nature of various types of blazars.Another important mission for future blazar studies will be the Monitor of All-sky X-ray Image (MAXI).MAXI is an X-ray all-sky monitor which is currently scheduled to be attached to the Japanese Experiment Module-Exposed Facility(JEM-EF)on the International Space Station(ISS) in early2009.The MAXI carries two scientific instruments:the Gas Slit Camera(GSC)and the Solid State-slit Camera(SSC).The GSC consists of position-sensitive proportional counters with large collecting area of5350cm2in2−30keV range,while the SSC is utilizing32X-ray CCD chips covering an energy range of0.5-12keV.The MAXI has two sets of GSC and SSC orthogonally oriented,each of which covers a narrow instantaneousfield of view of1.5deg times 160deg that sweeps over the whole sky during every orbit of90minutes.Thus a certain sky area is generally monitored twice in an orbit.The expected detection sensitivity for the GSC is ∼5mCrab in a day and∼1mCrab in one month,which is higher by a factor of5than that of11。

Reliability matters.Image quality matters.Performance matters.UPTIME OR DOWNTIME YOUR RESULTS MATTER• F rees you from having to leave the site to send images and/or reports •S end reports wirelessly when they’re needed, where they’re needed• C omplete more inspections in a day • On-site analysis• G et instant feedback from others or next steps approved immediately• R eal-time report previewing— instant gratification • U ser interface is optimized for each mobile device (iOS, iPhone ® and iPad ®)SmartView ® MobileFluke CNX ™ Wireless System• Capture up to five additional measurements with CNX wireless modules • M ultiple tools report to your CNX enabled Fluke infrared camera • Q uicker readings means less time finding problems and more time solving them • Capture measurements from as far as 20-meters away• T he list of Fluke test tools that can connect wirelessly continues to growYOUR WORLD. YOUR TOOLS.CONNECTED.Sending a comprehensive report to a supervisor’s or customer’s mobile phone… Analyzing and reporting from the field without having to go back to the office… Multiple tools that report to you simultaneously… This is the world of SmartView ® Mobile app and CNX ™ Wireless System. Available only from Fluke—where your results matter.FOCUS is the single most important thing to ensure when conducting an INFRARED INSPECTION.Many inspection sites are difficult for certain auto focus systemsPassive auto focus systems often only capture the near-field subject, in this case the chain link fenceFluke LaserSharp ™ Auto Focus clearly captures what you want to inspect. Every. Single. Time. The red dot from the laser confirms what the camera is Without an in-focus image, temperature measurements may not be as accurate (sometimes as much as 20 degrees off) and you could miss a problem.Fluke provides customers with two superior focusing solutions—LaserSharp ™ Auto Focus (see page 5) and IR-OptiFlex ™ Focus System (see page 7) and still gives you the flexibility of using manual if you wish.Ti400 Ti300 Ti200Ti400Ti300Ti200ACCURACY MATTERSOptimized for Industrial, Electrical and Building ApplicationsA new generation of tools with next generation performance.Technology changes. The last thing we want is for you to feel like you’re missing out on critical innovations, so Fluke has engineered all three new infrared cameras to adapt to change. Being future-ready is part of their DNA. You can test and measure with wireless speed and ease, and connect with other wireless devices. If there’s an infrared camera in your future, make sure it’s one with a future.Your confidence level is about to go up a notch. With precision laser technology, you can focus on your target with pinpoint accuracy and know you’re getting the correct image and temperature measurements you need. Troubleshooting has never been easier. This isn’t hit-and-miss technology. This is point-and-shoot-and-get-it-right every single time performance.Fluke introduces the only infrared cameras withLaserSharp ™ Auto Focus for consistently in-focus images.EVERY . SINgLE. TIME.IR-PhotoNotes ™ Annotation SystemGet an exact reference to your problem area by capturing multiple photos per file. Add images of equipment, motor nameplates, workroom doors or any other useful or critical information.Multi-mode video recordingTroubleshoot with the industry’s only infrared camera that offers the proprietary IR-Fusion ® Technology and records focus-free video in visible light and infrared. Monitor processes over time, easily create infrared video reports, and troubleshoot frame-by-frame. Easily download to PCs for video viewing and analysis.Ti125TiR125Ti110TiR110Ti105TiR105Ti100Ti125Ti110Ti105Ti100TiR125TiR110TiR105SIMPLICITY MATTERSBuilding ApplicationsIndustrial/Electrical Fluke innovation makes it easier to do more in less time.EASY TO CHOOSE. EASY TO USE.HARD TO BEAT .When you’re budget-conscious (and who isn’t these days?), the fact that you can get Fluke quality at an affordable price means you can breathe a sigh of relief. At Fluke, ‘affordable’ doesn’t mean sacrificing quality to give you a lower price. It means we’ve found a way to give you the most camera for your money. In this case, a suite of the lightest, most rugged, easiest-to-use professional infrared cameras you can buy.IR-Fusion ® TechnologyEnjoy the industry’s only point-and-shoot IR-Fusion infraredcameras that provide five different user-selectable modes for greater clarity. Our patented technology blends digital and infrared images into a single image to precisely document problem areas. Fluke exclusive AutoBlend ™ Mode generates a partially transparent image to make problem detection and communication fast and easy.R ugged one-hand operationExperience the most rugged and reliable, lightweight professional infrared camera around. One-touch focus, laser pointer, and torch. Point-and-shoot simplicity and the ergonomic design details that matter.Electronic compassMake sure you and others know the location of the problem. Compass readings easily appear in images and reports.IR-OptiFlex ™ Focus SystemDiscover issues significantly faster with Fluke’s revolutionary, ultra-rugged focus system. The IR-OptiFlex ™ Focus System gives you optimum focus by combining focus-free ease-of-use with the flexibility of manual focus on the same camera!For more than 65 years, Fluke isDesigned better. Built tougher.Superior image qualityThere’s a reason Fluke is so passionate about image quality. Clearer, cleaner, crisper images result in better information and more informed solutions. The better the image, the better you look when you show the images to your managers and customers. Our newest models of infrared cameras are the only ones where you can find IR-Fusion ® Technology and LaserSharp ™ Auto Focus. The Ti400, Ti300 and Ti200 also come fully loaded with a 5 MP digital camera, a HDMI video output, and a 640x480 high resolution LCD display.Legendary ruggednessand reliabilityFluke has earned their reputation as a tool of choice for electrical, industrial and building professionals. Whatever the job andwherever you work, when there’s a Fluke infrared camera in your hand, you’re prepared for the worst and ready to do your best. Fluke infrared cameras are designed to withstand a 2 meter drop (6.5 ft) and engineered to resist water and dust (IP54 Rating) so that your camera works without compromise..5 m 1 m 1.5 m 5 ft3.25 ft1.6 ft2 m6.5 fthow qUALITY IS MEASUREDBecause your results matter ™.Ease of useOur customers would rather spend time preventing and solving issues—not figuring out how their infrared camera works. We’ve gained a few other insights after spending thousands of hours in the trenches with them. That time and knowledge has allowed our engineers to develop breakthroughs in design, like buttons you can use when you’re wearing work gloves, and simple-to-use, on-camera functions such as voice annotation, so that you don’t have to stop to take notes with pen and paper. More recent innovations include:•L aserSharp ™ Auto Focus to ensure the best focus every single time •C NX ™ Wireless System to allow your CNX test modules to communicate additional measurements to your camera •I R Fusion ® Technology with Auto Blend ™ Mode to more easily locate, understand and report what the problem could be • C onnectivity to wirelessly transfer images to your PC, Apple ® iPad ® and iPhone ®All of these innovations can help you quickly understand what the current state is, create a report, determine next steps or begin a preventive maintenance program; all while the factory and processes are still up and running.Innovation that works for youFluke engineers know you’re not interested in the bells and whistles other manufacturers like to tout, so they focus solely on features you really need to help you work better, faster, and smarter.The groundbreaking features that you’ve come toknow, like IR-Fusion ® Technology, AutoBlend ™ Mode, voice annotation, IR PhotoNotes ™ Annotation System, and now LaserSharp ® Auto Focus help you get better results faster and easier. Get into the best position possible to get the results that matter to you and your customers with SmartView ® Software and SmartView ® Mobile.Ti400Ti300Ti200Ti125 Product Specifications Optimized for Industrial, Electrical and Buildings InspectionsTemperature measurement range (not calibrated below -10 °C) -20 °C to +1200 °C(-4 °F to +2192 °F) -20 °C to +650 °C (-4 °F to +1202 °F)-20 °C to +350 °C(-4 °F to +662 °F)Detector type 320 x 240 pixels240 X 180 pixels200 X 150 pixelThermal sensitivity (NETD)≤ 0.05 °C at 30 °C target temp (50 mK) ≤ 0.075 °C at 30 °Ctarget temp (75 mK)Field of view24 ° x 17 °Spatial resolution (IFOV) 1.31 mRad 1.75 mRad 2.09 mRadCustomizable logo options Users can brand their infrared images with a Fluke logo,upload their own company logo or no logo.Primary focusing system LaserSharp™ Auto Focus IR-OptiFlex™ Focu Manual focus YesIR-Fusion® Technology YesCNX™ Wireless enabled (Availableas country certification areapproved—notifications made viaSmartView® Software)Voice annotation60 seconds maximum recording time per image; reviewable playback on imagerIR-PhotoNotes™Yes (5 images)Yes (3 images)Wi-Fi® connectivity Yes, to PC and Apple® iPhone® and iPad®Streaming video Via USB to PC and HDMI to HDMI compatible device Streaming USB-to-PCvideo outputMulti-mode video recording*Yes (fully-radiometric .IS3 and standard MPEG-encoded .AVI)Yes (fully-radiometric.IS3 and standard MPEG-encoded .AVI) M8-point cardinal compass* Yes YesRuggedized touchscreen display (capacitive)8.9 cm (3.5 in) diagonal landscape color VGA (640 x 480)LCD with backlightSoftware SmartView® full analysis and r Warranty11Ti110Ti105Ti100TiR125TiR110TiR105Optimized for Industrial and Electrical InspectionsOptimized for Building Inspections -20 °C to +250 °C (-4 °F to +482 °F)-20 °C to +150 °C (-4 °F to +302 °F)160 X 120 pixels≤ 0.10 °C at 30 °C target temp (100 mK)≤ 0.08 °C at 30 °C target temp (80 mK)22.5 °H x 31 ° V 3.39 mRad—™ Focus System Focus-free 1.2 m (4 ft) and beyondIR-OptiFlex ™ Focus SystemFocus-free 1.2 m (4 ft)and beyond—Yes——YesYes—60 seconds maximum recording time per image; reviewable playback on imager ——Yes (3 images)———Streaming USB-to-PCvideo output—Yes (Standard MPEG-encoded .AVI)——Yes (fully-radiometric.IS3 and standardMPEG-encoded .AVI)Yes (StandardMPEG-encoded .AVI)—Yes—YesYes——nd reporting software included with free download of SmartView ® Mobile app2 years, Instrument Care Plans are also available.* Features marked with an asterisk are coming soon in a firmware download from SmartView ® software.1.800.868.7495********************Fluke -Direct .caFor more information call:In the U.S.A. (800) 443-5853 or Fax (425) 446-5116In Europe/M-East/Africa +31 (0) 40 2675 200 or Fax +31 (0) 40 2675 222In Canada (800)-36-FLUKE or Fax (905) 890-6866From other countries +1 (425) 446-5500 or Fax +1 (425) 446-5116Web access: ©2013 Fluke Corporation.Specifications subject to change without notice.All trademarks are the property of their respective owners. Printed in U.S.A. 08/2013 2674264M_ENFluke CorporationPO Box 9090, Everett, WA 98206 U.S.A.Fluke Europe B.V.PO Box 1186, 5602 BD Eindhoven, The NetherlandsModification of this document is notpermitted without written permission from Fluke Corporation.Dedicated supportquestions? Call 1-800-760-4523 or contact us via our chat function on our website at /thermography to request your free product demonstration. We’ll be happy to answer your questions, ship a unit for you to test for a week or send out a representative if you need on-site support.Fluke accessoriesEnhance your infrared camera’s performance with Fluke accessories. Choose car chargers,additional smart batteries or smart battery chargers to keep you up and running in the field. For special applications select optional lenses, a visor for outside inspections or a tripod mounting accessory.Fluke also offers specialized instrument CarePlans—ask your Fluke representative or distributor for additional information.Fluke trainingGet additional information and training at the Fluke Training web page. Take advantage of free on-line seminars and for those who seek more advanced training and professional mentoring, contact our Fluke training partner, The Snell Group, the most respected name in infrared education.Fluke authorized training is provided by our partner,1.800.868.7495********************Fluke -Direct .ca。

3continuous range from 1 mHz to 20 MHz.Triggered and manual sweep operations also can be performed. Seven sweepmodes, with linear and logarithmic sweep spacing, a TTL level sweep marker, and a ramp output give you the flexibility you need. There’s even a pen lift function for use with chart recorders.Modulation Source . If you do communi-cation or audio design and test, Model 395offers internally generated amplitude modulation and frequency modulation, as well as externally controlled amplitude modulation in two modes.Convenience and Versatility . Wavetekdesigned Model 395 for user convenience.The user screens are tailored to theparticular jobs you want to perform, such as setting up a pulse generator or a noise function.From any screen, you can access help screens that guide you in using theinstrument’s extensive capabilities. And you can store at least 10 instrument set-ups so you don’t have to spend valuable time duplicating past effort.Low cost of ownership is assured by the high reliability and ease of calibration of the Model 395. Calibration is performed with covers on in less than 15 minutes,under front panel or remote control.SpecificationsSpecifications apply within the specified environmental conditions after a 20 minute warm-up.AmplitudeRange: 10 mVp-p to 10 Vp-p into 50 Ω.Resolution: 3.0 digits.Accuracy: 25 ± 10° C: ± (1% + 2 mVp-p).OffsetRange: ± 5 V into 50 Ω.Resolution: 3 digits.Accuracy: 25 ± 10° C: ± (1% + 20 mV).Standard WaveformsSine, square, triangle, pulse, pulse trains, DC,positive/negative ramp, positive/negative haversine,(sin x)/x, and five noise functions.Frequency (Sine and Haversine)Range: 1 µHz to 40 MHz.Resolution: (Resolution limited by 1 µHz.).≤ 20 MHz: 10 digits; ± 30 ppm.> 20 MHz: 4 digits; ± 100 ppm.Frequency (Square)Range: 1 µHz to 50 MHz.Resolution: 4 digits; ± 100 ppm.Frequency (Triangle)Range: 1 µHz to 10 MHz.Resolution:≤ 100 kHz: 10 digits; ± 30 ppm.> 100 kHz: 4 digits; ± 100 ppm.Frequency (Ramp)Range: 1 µHz to 2 MHz.Resolution:≤ 100 kHz: 10 digits; ± 30 ppm.> 100 kHz: 4 digits; ± 100 ppm.Frequency (Sin (x)/x)Range: 1 µHz to 1 MHz Resolution:≤ 100 kHz: 10 digits; ± 30 ppm > 100 kHz: 4 digits; ± 100 ppm Waveform QualitySquare Transition Time: < 8 ns.Square Aberrations: < (5% + 20 mV).Sine Distortion:< 100 kHz: 0.15% (-56 dBc).< 5 MHz: No harmonic > -35 dBc.Arbitrary WaveformsSampling FrequencyRange: 100 mS/s to 100 MS/s.Resolution: 4 digits.Accuracy: ± 100 ppm.Waveform Memory Size64 k points; 256 k points optional.Minimum Waveform Size: 10 points.Vertical Resolution : 12 bits.Output Filters (Selectable): 20 MHz Elliptic (8pole), 40 MHz Elliptic (8 pole), 10 MHz Bessel (2pole), no filter.Waveform Sequencing : Up to 4 waveforms can be linked. Each waveform can have a repeat (loop)count of up to 65,535 or run continuously,conditional upon an external trigger event (repeat until event true). Additionally, a sequence ofwaveforms can be repeated up to 524,287 times or run continuously.Pulse WaveformsUp to 10 pulses may be independently programmed in a pulse pattern. Parameters that can beindependently programmed for each pulse are rise time, fall time, width, delay, and amplitude.For Periods ≤ 655 µs:Range: 100 ns to 655 µs.Resolution: 20 ns.Accuracy: ± 100 ppm.Rise/Fall:Fixed: 8 ns.Variable: 50 ns to 500 µs.Resolution: 8 ns.Accuracy: ± 0.1% ± 5 ns (< 8 ns for fixed rise/fall).Delay:Range: -600 to +600 µs.Resolution: 10 ns.Accuracy: ± 0.1% ± 5 ns.Width:Range: 10 ns to 655 µs.Resolution: 10 ns.Accuracy: ± 0.1% ± 5 ns.For Periods > 655 µs:Range: 655 µs to 10 s.Resolution: 4 digits.Accuracy: ± 100 ppm.Rise/Fall: 0.1% to 79% of period (or < 8 ns).Delay: -99.9% to +99.9% of period.Width: 0.002% to 99.9% of period.NoiseWhite (Analog) Noise:Uniform frequency distribution with programmable noise bandwidth.Noise BW Range: 10 mHz to 10 MHz.Sequence Length:Standard: 2n - 1, n = 6 to 16.With Option 002: 2n - 1, n = 6 to 17.Model 395Digital Noise:Digital noise provides a random 0,1 pattern with programmable sequence length.Clock Range: 10 mHz to 100 MHz.Sequence Length:Standard: 2n - 1, n = 6 to 16.With Option 002: 2n - 1, n = 6 to 17.Comb:Uniformly distributed frequency spectra within a well-defined frequency band.Start/Stop Range: 1 Hz to 10 MHz.Number of Lobes: 3 to 256.Signal-Plus-Noise, Signal-Plus-Comb: Adds analog noise or comb to any standard or arbitrary waveform with precise, controlled noise-to-signal ratio.N/S Ratio: 1% to 99% Vp-p.Resolution: 1%.Operational ModesContinuous: The selected waveform is output continuously at the programmed frequency. Gated: The selected waveform is output continuously at the programmed frequencywhile the selected trigger signal is true. Triggered: Upon transition of the selected trigger from false to true, the number of cycles specified by the count is output at the specified frequency. Burst count is programmable from 1 to 1,048,575. (One to 524,287 for waveform sequence operation.) Sweep: Frequency sweep.TriggeringTrigger Sources: 4 trigger sources: External TRIG IN BNC, internal trigger generator, front panel manual trigger key, and remote trigger command. Trigger Level: The trigger level at the TRIG IN BNC is programmable.Range: -10 V to +10 V.Trigger Slope: Positive or negative.Internal Trigger SourceRange: 200 ns to 1000 s.Resolution: 100 ns limited by 6 digits.Sync OutputSync output can be selected from among the following 7 sources: waveform sync, trigger signal, burst done, loop done, sweep marker,position marker, pen lift.ModulationFor both standard and arbitrary waveforms. Internal Frequency ModulationCarrier SignalSource: Sine Wave Center Frequency Range: 0.01 Hz to 40 MHzDeviation Frequency Range: 0.01 Hz to 40 MHz.Note: Center frequency plus deviation frequencymust be ≤ 40 MHz.Modulating SignalSource: Any waveform except noise, AM, FM, orpulse.Modulation Frequency Range: 0.01 Hz to 40 MHz.Internal Amplitude ModulationModesAM: 0 to 200% modulationSCM: 200% modulationCarrier SignalSource: Sine waveCarrier Frequency Range: 0.01 Hz to 40 MHzModulating SignalSource: Any waveform except noise, AM, FM, orpulse.Modulation Frequency Range: 0.01 Hz to 40 MHz.External Amplitude ModulationNormal AM: 0 to 100% modulation.Suppressed Carrier Modulation (SCM):± 100% modulation.Signal SummingExternal signals can be summed directly to theModel 395 output through the SUM IN BNC.SweepStandard and arbitrary waveforms can be swept.Sweep Start/StopRange: 1 mHz to 20 MHz.Resolution: 4 digits limited by 1 mHz.Sweep TimeRange: 30 ms to 1000 s.Resolution: 1 ms.Sweep Types:Sweep off, continuous, continuous with reverse,triggered, triggered with reverse, triggered withhold, triggered with hold and reverse, and manual.Sweep Spacing:Linear and logarithmic.OutputsReference Output (50 Ω):TTL level into open circuit; > 1.2 Vp-p .Main Output (50 Ω):Output may be selected on or off.A M Input (2.5 kΩ): ± 2.5 V.Sweep Output (1 kΩ): 0 to 10 V rampproportional to completion of sweep.Sync Output (50 Ω)Low Level: < 0.4 V into 50 Ω.High Level: > 2.0 V into 50 Ω.Rise/Fall Time: < 7 ns.InputsTrigger Input (2 kΩ)Level: ± 10 V (programmable).Maximum Frequency: 10 MHz.Sum Input (600 Ω)Level: ± 5 Vp-p max.Bandwidth: > 30 MHz.Protection: Over-voltage to ± 10 V.Reference Input (5 kΩ)Level: 1 Vp-p minimum, 10 Vp-p maximum;50 Vdc maximum.Frequency: 10 MHz ± 5%.GeneralRemote OperationRS-232 interface is standard. IEEE-488.2 (SCPIcompatible) GPIB interface is optional.EnvironmentDesigned to MIL-T-28800C Class 5.Temperature Range: Operates from 0° to+50°C: -20° to +70°C for storage.Dimensions: 35.6 cm (14.00 in) wide,13.3 cm (5.22 in) high, and 39.4 cm (15.5 in) deep.Weight: Approximately 7.7 kg (17 lb) net;10.0 kg (22 lb) shipping.Power: 90 to 132, 198 to 252 volts rms;48 to 440 Hz; 1 phase; < 80 VA.Ordering InformationModel 395: 100 MHz synthesized ArbitraryWaveform Generator with serial cable and QuickStart Demo Disk.Option 001: IEEE-488 Interface/Direct DSOWaveform TransferOption 002: 256k Extended MemoryOption 004: Rack Mount KitModel 485: WaveForm DSP, Windows-basedsoftware for creating and editing complex waveforms.Model 3954。

(1)Ying Lungan, Zhou Tie, "Long-time Asymptotic Behavior of Lax-Friedrichs Scheme", J. PartialDiff. Eqs., Vol.6, No.1 (1993), 39-61(2)Long-an Ying, Tie Zhou, "Nonlinear Stability of Discrete Shocks", Proceedings of the Japan-ChinaSeminar on Numerical mathematics, Lecture Notes in Numerical and Applied Analysis, Vol.14 (1995), pp227-239.(3)Tie Zhou, "A Remark on the Nonlinear Stability of Discrete Shocks for Single HyperbolicConservation Laws", Acta Scientiarum Naturalium Universitatis Pekinensis, Vol.33, No.5 (1997), pp554-560. (EI)(4)Han-song Tang, Tie Zhou, "Why nonconservative interface algorithms may be applicable: analysisfor Chimera grids", Proceedings of 7th International Symposium on CFD(1997), pp336-341. (5)Han-song Tang, Tie Zhou, "On Nonconservative Conditions at Grid Interfaces", SIAM Journal onNumerical Analysis, V ol.37, No.1 (1999), pp173-193. (SCI)(6)李荫藩，宋松和，周铁，“双曲型守恒律的高阶、高精度有限体积法”，力学进展, V ol. 31,No.2 (2001), pp245-263.(7)Tie Zhou, Yin-fan Li, Chi-Wang Shu, "Numerical Comparison of WENO Finite Volume and Runge-Kutta Discontinous Galerkin Methods", Journal of Scientific Computing, Vol.16, No.2 (2001), pp145-171. (EI)(8)Tie Zhou, Yan Guo, Chi-Wang Shu, "Numerical Study on Landau Damping", Physica D, V ol.157(2001), pp322-333. (SCI)(9)Tie Zhou, Shulin Zhou, Ming Jiang, Danfeng Lu, "Theory of continuous multiscale homomorphicfiltering and applications", XI-th International Congress for Sterology Beijing Conference, inconjunction with Xth Chinese National Symposium for Sterology and Image Analysis, November 4-8, 2003, Beijing.(10)Yan Guo, Chi-Wang Shu, Tie Zhou, "The dynamics of a Plane Diode", SIAM Journal onMathematical Analysis, V ol. 35, Number 6 pp. 1617-1635. 2004. (SCI)(11)Jinghua Wang, Hairui Wen, Tie Zhou, "On large time step schemes for hyperbolic conservationlaws", COMM. MATH. SCI. , V ol. 2, No. 3, pp. 477-495. 2004.(12)Jiansheng Yang, Qiang Kong, Tie zhou, Ming Jiang, "Cone Beam Cover Method: An Approach toPerforming Backprojection in Katsevich’s Exact Algorithm for Spiral Cone Beam CT", Journal of X-ray Science and Technology，V ol.9 , 1-16, 2004. (EI)(13)蔚喜军，周铁，“流体力学方程的间断有限元方法”，计算物理，V ol. 22, No. 2 (2005), pp108-116. (EI)(14)DanFeng Lu, HongKai Zhao, Ming Jiang, ShuLin Zhou, Tie Zhou, "A Surface ReconstructionMethod for Highly Noisy Point Clouds", N. Paragios et al. (Eds.): VLSM 2005, LNCS 3752, pp.283–294. Springer-Verlag Berlin Heidelberg 2005. (SCI)(15)郭晓虎，杨建生，孔强，周铁，姜明，“锥束覆盖方法的并行实现及性能分析”，中国体视学与图像分析，V ol.10, No.3, 165-169, 2005.(16)Jiansheng Yang, Xiaohu Guo, Qiang Kong, Tie Zhou, Ming Jiang, "Parallel Implementation of theKatsevich's FBP Algorithm", International Journal of Biomedical Imaging, Special Issue on"Development of Computed Tomography Algorithms", 2006, Article ID 17463.(17)Ge Wang, Ming Jiang, Jie Tian,Wenxiang Cong, Yi Li, Weimin Han, Durai Kumar, Xin Qian, HaiouShen, Tie Zhou, Jiantao Cheng, Y ujie Lv, Hui Li, Jie Luo, "Recent development in bioluminescence tomography", 2006 3RD IEEE INTERNA TIONAL SYMPOSIUM ON BIOMEDICAL IMAGING: FROM MACRO TO NANO, VOLS 1-3, IEEE International Symposium on Biomedical Imaging, pages: 678-681, 2006. (SCI)(18)Ming Jiang, Tie Zhou, Jiantao Cheng, Wengxiang Cong, Durairaj Kumar, Ge Wang, "ImageReconstruction for Bioluminescence Tomography", RSNA 2005.(19)Ming Jiang, Tie Zhou, Jiantao Cheng, Wenxiang Cong, Ge Wang, "Development of bioluminescencetomography", art. no. 63180E, Developments in X-Ray Tomography V, PROCEEDINGS OF THE SOCIETY OF PHOTO-OPTICAL INSTRUMENTATION ENGINEERS (SPIE), vol. 6318, pages: E3180-E3180, 2006. (SCI)(20)Jinxiao Pan, Tie Zhou, Yan Han, Ming Jiang, “Variable Weighted Ordered Subset ImageReconstruction Algorithm”, International Journal of Biomedical Imaging, V olume 2006 (2006), doi:10.1155/IJBI/2006/10398 Article ID 10398, 7 pages(21)Seung Wook Lee, Jinxiao Pan, Chunhua Liu, Tie Zhou, Cheul-Muu Sim, Ming Jiang, A PreliminaryStudy of Iterative Reconstruction Algorithms for Neutron Tomography, The 8th World Conference on Neutron Radiography, National Institute of Standards and Technology, Gaithersburg, MD, 16 - 19 October, 2006. (SCI)(22)G. Wang, X. Qian, W. Cong, H. Shen, Y. Li, W. Han, D. Kumar, M. Jiang, T. Zhou, J. Cheng, J. Tian,Y. Lv, Hui Li, J. Luo, "Recent development in bioluminescence tomography", Current MedicalImaging Reviews, V olume 2, Number 4, November 2006. (SCI)(23)Tony Chan, Haomin Zhou, Tie Zhou, "Error Analysis for H^1 Based Wavelet Interpolations", ImageProcessing Based on Partial Differential Equations: Proceedings of the International Conference on PDE-Based Image Processing and Related Inverse Problems, Editors: X.-C. Tai, K.-A. Lie, T.F.Chan, and S. Osher, Series: Mathematics and Visualization, Springer Verlag, 2007.(24)Jiang, Ming; Louis, Alfred K.; Wolf, Didier; Zhao, Hongkai; Daul, Christian; Zhang, Zhaotian;Zhou, Tie, Mathematics in biomedical imaging: Editorial, International Journal of BiomedicalImaging, v 2007, Article number: 64954. 2007. (EI)(25)M. Jiang, T. Zhou, J. Cheng, W. Cong, G. Wang, "Image reconstruction for bioluminescencetomography from partial measurement", OPTICS EXPRESS, V ol. 15, No. 18, September 2007.(SCI)(26)Wenlei Ni, Ming Jiang, Shulin Zhou, Tie Zhou, “On Reconstruction Algorithms of X-ray PhaseContrast CT by Holographic Measurements” (invited), An Interdisciplinary Workshop onMathematical Methods in Biomedical Imaging and Intensity-Modulated Radiation Therapy (IMRT), Centro di Ricerca Matematica Ennio De Giorgi, Scuola Normale Superiore di Pisa, Italy, October 15 - 20, 2007.(27)Ni Wen-lei, Zhou Tie, "Algorithm for phase contrast X-ray tomography based on nonlinear phaseretrieval", Applied Mathematics and Mechanics (English Edition), Vol.29, No.1, pp101-112, 2008.(SCI)(28)Jiantao Cheng, Tie Zhou, "A Variational EM Method for the Inverse Black Body RadiationProblem", Journal of Computational Mathematics (English Edition), V ol.26, No.6, pp876–890, 2008.(SCI)(29)Tie Zhou, Jiantao Cheng, Ming Jiang, Bioluminescence Tomography Reconstruction by Radial BasisFunction Collocation Method, Industrial and Applied mathematics in China (Series in Contemporary Applied Mathematics CAM 10), pp229-239, Higher Education Press, 2009.(30)Caifang Wang, Tie Zhou, “Local convergence of an EM-like image reconstruction method fordiffuse optical tomography”, Journal of Computational Mathematics (English Edition), Vol.29, No.1, pp.61-73, 2011. (SCI)(31)Caifang Wang, Tie Zhou, “The order of convergence for Landweber Scheme with α,β –rule”, InverseProblems and Imaging,V ol.6, No.1, pp.133-146, 2012. (SCI)(32)周宇，周铁，姜明，“X射线相位相位衬度层析成像的数学模型”，数学建模及其应用，第1卷，第2期，pp.12-18，2012.(33)Yu Zhou, Tie Zhou, Ming Jiang, “An alternative derivation for Bronnikov’s formula in x-ray phasecontrast tomography”, World Congress on Medical Physics and Biomedical Engineering, ofInternational Federation for Medical and Biological Engineering (IFMBE) Proceedings, vol.39,pp.1038-1040, 2012.(EI)(34)Yanbin Lu, Jiansheng Yang, John W Emerson, Heng Mao, Tie Zhou,Y uanzheng Si, Ming Jiang,“Cone-beam reconstruction for the two-circles-plus-one-line trajectory”, Physics in Medicine and Biology”, Vol.57, No.9, pp.2689-2707, 2012.(SCI)(35)S. Luo, H. Zhou, T. Zhou, “An Improved Color Image Demosaicking Algorithm”, 2012 5thInternational Congress on Image and Signal Processing (CISP 2012), 2012-10. (EI)(36)Yu Zhou, Alfred K Louis, Tie Zhou, Ming Jiang, “Partial Coherence Theory for X-ray PhaseContrast Imaging Technique with Gratings”, Optics Communications，V olume 285, Issue 24, Pages 4763–4774, 2012-11. (SCI)(37)Seung Wook Lee, Yu Zhou, Tie Zhou, Ming Jiang, Jongyul Kim, Chiwon Ahn, Alfred K. Louis,Visibility studies in grating-based neutron phase contrast and dark-field imaging by partial coherence theory, Journal of Korean Physical Society, V ol. 63, NO. 11, pp. 2093 - 2097, 2013.(38)Shousheng Luo, Jiansheng Yang and Tie Zhou, “Moment-based cosh-Hilbert Inversion and ItsApplications in Single-photon Emission Computed Tomography”, CHINESE JOURNAL OFCOMPUTATIONAL PHYSICS（计算物理）, Vol. 30, No. 6, pp.799-807, Nov. 2013.（核心期刊）(39)Shousheng Luo, Tie Zhou, “Superiorization of EM Algorithm and Its Application in Single-PhotonEmission Computed Tomography(SPECT) ”, Inverse Problems and Imaging , Vol 8, Issue 1, Pages: 223 - 246, February 2014.(SCI)(40)Tangjie Lv, Tie Zhou, “V ARIATIONAL ITERA TIVE ALGORITHMS IN PHOTOACOUSTICTOMOGRAPHY W ITH V ARIABLE SOUND SPEED”, Journal of Computational Mathematics, Vol.32, No.5, pp.579-600, Aug. 2014. (SCI)(41)毛珩，李宣成，李海文，TAO Louis，周铁，“基于主动轮廓模型的运动线虫中心线定位算法”，中国科学：数学46(7)，pp.1005-1016，2016.(42)Ji Li and Tie Zhou, “On gradient descent algorithm for generalized phase retrieval problem”,Proceedings of IEEE Conference on Signal Processing (ICSP 2016).(EI)(43)Ji Li and Tie Zhou, “On Relaxed Averaged Alternating Reflections (RAAR) Algorithm for PhaseRetrieval from Structured I lluminations”, Inverse Problems, 33 (2), 025012(20pp), 2017. (SCI) (44)Ji Li and Tie Zhou, “NUMERICAL OPTIMIZATION ALGORITHMS FOR WA VEFRONT PHASERETRIEV AL FROM MULTIPLE MEASUREMENTS”, Inverse Problems and Imaging 11(4),pp.721-743, 2017.(SCI)(45)Ji Li, Tie Zhou and Chao Wang, “On global convergence of gradient descent algorithms forgeneralized phase retrieval problem”, Journal of Computational and Applied Mathematics, Volume 329, Pages 202-222, 2017.(SCI)(46)Chao Wang, Tie Zhou, “On Iterative Algorithms for Quantitative Photoacoustic Tomography in theRadia tive Transport Regime”, Inverse Problems, 33(11), 115006 (25pp) , 2017. (SCI)(47)Chao Wang, Tie Zhou, “A Hybrid Reconstruction Approach for Absorption Coefficient byFluorescence Photoacoustic Tomography”, Inverse Problems, 35 (2019) 025005, 2019. (SCI) (48)S. Luo, Y. Zhang, T. Zhou and J. Song , “XCT Image Reconstruction by a Modified SuperiorizedIteration and Theoretical Analysis”, Optimization Methods and Software, online,https:///10.1080/10556788.2018.1560442, 2019. (SCI)。

a rXiv:as tr o-ph/46250v19J un24GLAST:UNDERSTANDING THE HIGH ENERGY GAMMA-RAY SKY Julie E.McEnery ∗NASA/Goddard Space Flight Center Code 661,Greenbelt,MD 20771,USA mcenery@ Igor V.Moskalenko ∗NASA/Goddard Space Flight Center Code 661,Greenbelt,MD 20771,USA imos@ Jonathan F.Ormes NASA/Goddard Space Flight Center Code 660,Greenbelt,MD 20771,USA Jonathan.F.Ormes@ Abstract We discuss the ability of the GLAST Large Area Telescope (LAT)to identify,resolve,and study the high energy γ-ray pared to previous instruments the telescope will have greatly improved sensitiv-ity and ability to localize γ-ray point sources.The ability to resolve the location and identity of EGRET unidentified sources is described.We summarize the current knowledge of the high energy γ-ray sky and discuss the astrophysics of known and some prospective classes of γ-ray emitters.Besides,we describe the potential of GLAST to resolve old puzzles and to discover new classes of sources.IntroductionOur rudimentary understanding of the GeV γ-ray sky was greatly ad-vanced with the launch of Energetic Gamma-Ray Experiment Telescope ∗JCA/University of Maryland,Baltimore County,Baltimore,MD 21250,USA2(EGRET)on the Compton Gamma-Ray Observatory(CGRO)in1991. The number of previously known GeVγ-ray sources increased from1–2 dozen to the271listed in the3rd EGRET Catalog.Theγ-ray sky was shown to be dominated by time-varying emitters.The science returns from these observations exceeded prelaunch expectations.Among them were the discoveries that for many blazars,intenseγ-rayflares are seen, that many blazars and some pulsars have peak luminosity at GeV ener-gies,and that the spectrum of GRBs extends to at least GeV energies. However,of this multitude of sources,only101have been definitively associated with known astrophysical objects.Thus,most of theγ-ray sky,as we currently understand it,consists of unidentified objects.This leaves intriguing puzzles for Gamma-ray Large Aperture Space Telescope (GLAST),the next generation GeVγ-ray instrument,to uncover. One of the reasons that such a small fraction of the sources were iden-tified is because the size of the typicalγ-ray error box from EGRET was about1sq degree,an area that contains several candidate sources preventing a straightforward identification.Most of the identified GeV sources have distinct temporal features that allowed them to be associ-ated with a known object.For example,the observation of pulsation al-lowed unambiguous association between radio pulsars andγ-ray sources, and the observation of correlated radio,optical or X-ray emission with γ-rayflares allowed the identification of several active galactic nuclei (AGN).This is why most of the high energy sources identified to date are either pulsars or AGN.The largefield of view and improved angular resolution of the LAT instrument on the GLAST mission will improve this situation dramat-ically.The angular resolution and“tails”on the point spread function are improved compared to EGRET so we will be able to refine maps of theγ-ray sky in crowded regions.The higher effective area allows flares from AGN and pulsations from pulsars and binary systems to be measured to correspondingly lowerflux levels.We begin with a brief description of the LAT as the next generation instrument in the GeV energy range.We will then discuss in Section1.2 the questions raised by the EGRET results and the prospects for LAT studies of known and potential classes ofγ-ray sources and speculate on their eful and up to date information can be found in GLAST Science Brochure1.GLAST:Understanding the High Energy Gamma-Ray Sky3 1.Instrument description1.1HardwareGLAST is a major space mission to explore the high energyγ-ray universe.There are two instruments on board.The primary instrument is the GLAST Large Area Telescope(LAT),which is sensitive at energies from20MeV to300GeV.The secondary instrument is the GLAST Burst Monitor(GBM)2to detectγ-ray bursts and provide broad–band spectral coverage of this important phenomenon.The discussion of GBM is given elsewhere.Like EGRET,the LAT will detectγ-rays through conversion to electron-positron pairs and measurement of their direction in a tracker and their energy in a calorimeter.It uses a segmented plastic scintillator anti-coincidence system to provide rejection of the intense background of charged cosmic rays.A schematic of the LAT is shown in Fig.1. Along with increased area with respect to EGRET,the design of each element has been refined to improve the sensitivity and angular resolu-tion and to extend the energy range to higher energy.In the tracker,incident photons convert to electron-positron pairs in one of16layers of lead converter and are tracked by single-sided solid-state silicon strip detectors(SSD)through successive planes(the strips are centered every228microns).Tracking the pairs allows the recon-struction of the incident photon’s arrival direction.The technology is the same as that used for precision measurements at the collision vertex in modern high-energy physics particle investigations.Multiple Coulomb scattering limits the angular resolution in the30to300MeV band,but the geometry of the SSD layers results in an important improvement over EGRET.At higher energies(>1GeV)the precision of the EGRET spark chambers limited the ability of that instrument to take advantage of the intrinsic reduction in multiple scattering as energy increases.The very high measurement resolution of the SSD being used in the LAT means that the angular resolution is dominated by measurement uncer-tainty due to multiple scattering only above energies of100GeV.In determining source positions,each photon can be weighted in propor-tion to its energy,higher weight being given to higher energy particles whose arrival direction is better known.This results in an improvement in angular resolution that can reduce the source location error boxes by as much as a factor of100depending on the energy spectrum of photons detected and the localγ-ray background.Furthermore,LAT will have 10–20microsecond deadtime to enable for far better efficiency for closely spaced(in time)γ-ray events during intense phases ofγ-ray bursts and solarflares.4Figure1.The GLAST Large Area Telescope.A further subtlety in the LAT design can be found in the distribution of the thickness of the converter plates and SSD tracker planes.There are a total of36SSD planes,18(x,y)-pairs.Thefirst12pairs(the front section of the tracker)are covered by tungsten plates,each of0.03 radiation lengths(r.l.)thick,and the next4are covered by converters 0.18r.l.thick,and thefinal2planes have no converter.In addition, there are about0.405r.l.in silicon detector and structural materials. The converter is thus1.485r.l.thick in total.The photons that convert in the front section(0.36r.l.)give the highest angular resolution,while the back section assures that the overall photon detection efficiency is asGLAST:Understanding the High Energy Gamma-Ray Sky5 high as practical.The last two layers make sure we know the entry point of the particles into the calorimeter,where the tracking in the crystal layers,described below,continues.The calorimeter consists of a segmented arrangement of1536CsI(Tl) crystals in8layers,giving both longitudinal and transverse information about the energy deposition pattern.This has several advantages over the shallower,monolithic calorimeter used in EGRET.The segmentation allows for the shower development in the calorimeter to be reconstructed resulting in a method of correction for energy leaking out the bottom of the bined with the greater depth of the calorimeter this provides much the required energy resolution at high energies up to several hundred GeV.For normal incidence photons,shower maximum is contained up to100GeV.In addition,the segmentation allows the calorimeter to provide direction information of theγ-ray photon which can be used if theγ-ray photon did not convert in the tracker and provide important pattern recognition inputs for the rejection of background. The AntiCoincidence Detector(ACD)provides most of the rejection of charged particle backgrounds.It consists of an array of89plastic scintillator tiles each with charged particle detection efficiency of0.9997 read out by wave-shiftingfibers and photomultiplier tubes.The seg-mentation prevents loss of effective area at high energy due to self-veto caused by low energy photons coming from the cascade showers in the calorimeter.This was an issue in EGRET that occured when albedo from showers in their calorimeterfired the monolithic ACD causing a 50%reduction in effective area at10GeV.The ACD is being designed so the LAT will have at least80%efficiency for300GeV photons.The GLAST mission will beflown in low-Earth orbit and will operate in both a“rocking”mode and a“stare”mode.In rocking mode the instrument moving35degrees north of zenith and then35degrees south about the orbit plane on alternate orbits.The instrument has a huge field of view;∼20%of the sky at a time,and in rocking mode covers ∼75%of the sky every orbit.This is illustrated in Fig.2which shows the amount of sky coverage and sensitivity in Galactic coordinates for 3different integration times while the instrument is in rocking mode. The top panel is for a100second observation.In rocking mode,the instrument does not move significantly during this time,so the extent of the non-black region indicates the size of the region of the sky that will be visible to the LAT at any instant.The middle panel shows the sensitivity after a one orbit(90minute)observation.After two orbits (one with GLAST rocked north,and one rocked south)there is complete sky coverage.For a one day observation(bottom panel)the exposure on the sky becomes fairly uniform,the variations in point source sensitivity6Figure2Sensitivity andsky coverage of GLAST(top to bottom)for100s,one orbit,and one day ob-servations.are dominated by the distribution of the background due to the Galactic diffuse emission.The sensitivity achieved after a single day observation is similar to the point source sensitivity of EGRET for the entire mission.1.2Instrument capabilitiesThe capabilities of GLAST’s LAT are summarized in Table1.The improvement in angular resolution will result in a much greater ability to localize sources.The area of source location error boxes will be reduced, compared to those of EGRET,by at least a factor of three depending on the source spectrum.This means that GLAST will be much more capable of resolving structure in the interstellar emission and separating point sources from extended or diffuse sources of emission.This will be of particular interest in,for example,locating and separating hot spots in supernova remnants from possible central point sources or resolving the positions of giant molecular clouds.Fig.3shows a simulation of theGLAST:Understanding the High Energy Gamma-Ray Sky7Figure3(a)Integratedintensity of the2.6-mm COline in Orion(Maddalenaet al.1986),showing thewell-known Orion A andOrion B molecular cloundsand the Mon R2cloud.(b)Intensity ofγ-rays with en-ergies>100MeV observedby EGRET and analyzedby Digel et al.(1999).(c-e)Simulated intensity mapsfrom the GLAST sky sur-vey for the energy rangesindicated.To smoothstatisticalfluctuations,theEGRET data were con-volved with a Gaussianof FWHM 1.5◦,and theGLAST intensities withFWHM0.75◦.Crossesmark the positions of back-ground sources withfluxesgreater than10−8cm−2s−1(>100MeV).The in-tensity scale refers to(b-e)with scale factors as notedfor(d-e)(Digel et al.2001). Orion region where point sources and extended diffuse emission may be intermingled.GLAST will also fare far better in regions that were beset by problems of source confusion(where the sensitivity becomes limited by background due to unresolved sources)in the EGRET all-sky survey.Fig.4shows a8Table1.Properties of the LAT compared to EGRET.Energy range20MeV–30GeV20MeV–300GeV Energy resolution10%9%Effective area1500cm210000cm2 Angular resolution 5.8◦–0.3◦ 3.4◦–0.09◦Field of view0.5sr 2.4srGLAST:Understanding the High Energy Gamma-Ray Sky9 Figure5.Sensitivities of past,current and future high energyγ-ray detectors(Morselli et al.2000).sources and thus determine which are pulsars.The LAT will have the potential to detectγ-ray variability from new classes of putativeγ-ray bright objects such as Galactic microquasars or dim,decaying signals fromγ-ray burst afterglows.The extended energy range opens important new discovery potential for the LAT.The EGRET effective area fell offrapidly above a few GeV; the LAT has been carefully designed to maintain its area up to at least 300GeV.This new high energy capability will provide significant overlap with the next generation ground-basedγ-ray telescopes,as illustrated in Fig.5.It shows the integral sensitivities of past,current and future high energyγ-ray instruments.This will provide complete spectral coverage from MeV to TeV energies for thefirst time.It covers the band over which we expect to see breaks in the spectra of extragalactic objects due to theγγ-absorption on low energy photons in intergalactic space.At the same time,the calorimeter and its energy resolution capability allow searches forγ-ray lines or other spectral features in the energy band where the annihilation signatures of the dark-matter candidate particles might be found.10This combination of GLAST’s operational capability of scanning and Earth avoidance and the LAT’s increasedfield of view,effective area, angular resolution and extended energy range in combination yield a sensitivity about two orders of magnitude improvement in sensitivity compared to EGRET.This capability will be important to do population and broadband spectral studies of such sources as blazars,pulsars and gamma ray burst afterglows.2.Prospects:known and potentialγ-ray sources There are only a few elementary processes capable of producing high energyγ-rays(see Chapter2for more details),but uncovering their rela-tive efficiency provides us with important information about the physical conditions and reactions in distant places that can not be obtained by any other means.The intensity of theγ-ray emission viaπ0-decay and bremsstrahlung depends on the target gas density,while the intensity of inverse Compton(IC)scattering depends on the density of photons. Measurements of theγ-rayflux from an object provides information such as the spectra and distributions of accelerated particles,magnetic and radiationfields and gas density and distribution.This information is vital for studies of most astrophysical environments.The nucleonicγ-rays are generated through the decay ofπ0-mesons produced in nucleus-nucleus interactions of accelerated particles with gas.Nucleonicγ-rays have a spectrum with maximum at approximately mπc2/2≈70MeV.For Eγ≫mπc2/2the spectral index resembles the index of the ambient energetic nucleons,αp.The leptonicγ-rays can be produced via bremsstrahlung and IC scat-tering of cosmic microwave background(CMB)photons,diffuse Galactic photons,and local radiationfields.Bremsstrahlung spectral index is ap-proximately the same as lepton’s in the whole range,αe.The spectrum of high energy photons produced in IC scattering in the Thomson regime isflatter than the spectrum of electrons,αγ=(αe+1)/2.The LAT is particularly well suited for observations ofγ-rays pro-duced by bremsstrahlung,IC scattering,andπ0-decay since its energy range covers the regions where these processes play a major role.This orbital instrument together with a new generation of ground-based tele-scopes will measure theγ-rayfluxes and spectra with the high accuracy, required to provide insights into many different types of astrophysical objects,and thus decipher long-standing puzzles in cosmic-ray andγ-ray astrophysics.GLAST:Understanding the High Energy Gamma-Ray Sky11 This Sectionfirst describes the classes of knownγ-ray sources and the major problems to be addressed by LAT,and then speculates on the LAT potential to discover the objects yet to be detected inγ-rays.2.1Galactic diffuseγ-ray emissionThe diffuse continuum emission in the range50keV–50GeV has been systematically studied in the experiments OSSE,COMPTEL,EGRET on the CGRO as well as in earlier experiments,SAS2and COS B.A review of CGRO observations was presented by(Hunter et al.1997). New models of the diffuse emission are being developed by the LAT team.A new model will be compared to the higher resolution data provided by the LAT and important scientific understanding,described below,will result.The diffuseγ-ray emission is the dominant feature of theγ-ray sky. The diffuse continuumγ-rays are produced in energetic interactions of nucleons with gas viaπ0production,and by electrons via IC scattering and bremsstrahlung.In the plane of our Galaxy,the emissions are most intense at latitudes below5◦and within30◦of the Galactic center and along spiral arms.Each of the emission processes are dominant in a different energy range,and therefore theγ-ray spectrum,can provide information about the large-scale spectra of nucleonic and leptonic com-ponents of cosmic rays(see Fig.6).In turn,an improved understanding of the role of cosmic rays is essential for the study of many topics in Galactic and extragalacticγ-ray astronomy.It is worth noting that an understanding of the spatial distribution and spectrum of the diffuse emission is also important for studies of discrete sources.A self-consistent model of Galactic diffuseγ-ray emission should in-clude cosmic-ray transport as thefirst step.Knowing the number density of primary nuclei from satellite and balloon observations,the production cross sections from the laboratory experiments,and the gas distribution from astronomical observations,one can calculate the production rate of secondary nuclei.The observed abundance of radioactive isotopes de-termines then the value of the diffusion coefficient,halo size and other global parameters(Strong and Moskalenko1998).The modeling of cosmic-ray diffusion in the Galaxy includes the so-lution of the transport equation with a given source distribution and boundary conditions for all cosmic-ray species.The transport equa-tion describes diffusion,convection by the hypothetical Galactic wind, energy losses,and possible distributed acceleration(energy gain).Elec-trons lose energy due to ionization,Coulomb scattering,synchrotron emission,IC scattering,and bremsstrahlung.The study of transport12Figure6.Multiwavelength spectrum of Galactic diffuseγ-ray emission.Hard X-ray/γ-ray Galactic diffuse emission as measured by HIREGS(triangles),RXTE, OSSE/CGRO(diamonds),COMPTEL/CGRO,and EGRET/CGRO.Also shown is the calculatedflux(solid),and separate components:bremsstrahlung(dash-dot),IC (short dash),and neutral pions(triple dot-dash).Adapted from Boggs et al.(2000). of cosmic-ray nuclear component requires consideration of nuclear spal-lation and ionization energy losses.The most sophisticated analytical methods include disk-halo diffusion,the dynamical halo Galactic wind, the turbulence and reacceleration(by second order Fermi acceleration and by encounters with interstellar shock waves).The observation of diffuseγ-rays provide the most direct test of the proton and electron spectra on the large scale.The EGRET observa-tions have confirmed the main features of the Galactic model derived from locally observed cosmic rays;however,they also brought new puz-zles.Theγ-rays revealed that the cosmic ray source distribution required to match the cosmic ray gradient in the Galaxy should be distinctlyflat-ter(Strong and Mattox1996)than the(poorly)known distribution of supernova remnants(Case and Bhattacharya1998),the conventional sources of cosmic rays.The spectrum ofγ-rays calculated under the assumption that the proton and electron spectra in the Galaxy resemble those measured locally reveals an excess at>1GeV in the EGRET spec-trum(Hunter et al.1997).This may indicate that the cosmic ray proton and/or electron spectra in the vicinity of the Sun are not representative of the Galactic average.GLAST:Understanding the High Energy Gamma-Ray Sky13 Since theγ-rayflux in any direction is the line of sight integral,at-tempts were made to explain the observed excess by a harder nucleon spectrum in the distant regions(Mori1997;Gralewicz et al.1997).How-ever,it seems that the harder nucleon spectrum is inconsistent with other cosmic-ray measurements(Moskalenko et al.1998)such as secondary antiprotons and positrons,which are also produced in pp-interactions. While the large deviations in the proton spectrum are less probable, the electron spectrum mayfluctuate from place to place.The rate of energy loss for electrons increases with energy,it is thus natural to as-sume that the electron spectra are harder near the sources producing more high-energyγ-rays(Porter and Protheroe1997;Pohl and Esposito 1998;Strong et al.2000).IC scattering of Galactic plane and CMB pho-tons offelectrons provide a major contribution to the Galactic diffuse emission from mid-and high-latitudes.The effect of anisotropic scatter-ing in the halo(Moskalenko and Strong2000)increases the contribution of Galacticγ-rays and reduces the extragalactic component.New measurements of the spectrum of diffuse continuum Galacticγ-rays by GLAST will address several long-standing problems.The large collection area and efficient operating mode will permit spectra to be derived for much smaller area bins than was possible with EGRET.This will allow for much better measurements of the latitude and longitude distributions of the diffuse emission allowing better constraints to be placed on its origin.The extension of the energy reach of the LAT will allow the excess above1GeV found by EGRET to be confirmed and characterized.This will greatly improve our understanding of the character of cosmic-ray diffusion and acceleration in our Galaxy and galaxies nearby.This in turn will allow the determination of a better background model for both point source and extended source studies. More detailed discussion of Galactic diffuse emission may be found in Chapter11.2.2Pulsars and plerionsPulsars make up the second most numerous class of identified sources in the EGRET catalog which includes six confirmed and three candidate γ-ray pulsars(see Chapter7for more details).Many of the unidentified sources may be pulsars,however in many cases EGRET was not able to collect enough photons per source to perform independent period searches to detect these.The lightcurves of all the EGRET pulsars show a double peak.Aside from the Crab the shapes of the radio andγ-ray profiles are often quite different with the peaks falling at different pulse phases.This implies14that low-and high-energy photons are most probably emitted from dif-ferent regions and thus that their origin is different.Different mech-anisms may be even responsible for emission of low-and high-energy γ-rays(McLaughlin and Cordes2000),with the efficiency of converting the spin-down energy to high-energyγ-rays increasing with pulsar age. In the case of Geminga pulsar,one of the brightestγ-ray sources,most of its energy is emitted in GeVγ-rays(Jackson et al.2002)while its radio emission has not yet been detected.The spectra of these objects are very hard;pulsed emission above5 GeV was seen by EGRET for all six confirmedγ-ray pulsars.However, spectral breaks are seen in most of these objects.PSR1706–44exhibits a break from a power-law spectrum with differential index of–1.27to a power-law spectrum with index–2.25above1GeV.Vela,Geminga, and the Crab all show evidence for a spectral break at around one GeV. Stringent upper limits on PSR1951+32and PSR1055–52at a few hundred GeV imply a spectral break for these objects(Srinivasan et al.1997).However,there were insufficient detected photons to allow a determination of the shape of these spectral breaks.Searches for pulsed emission above a few hundred GeV by ground-basedγ-ray detectors have so far only resulted in upper limits.Two main types of models,polar cap and outer gap,have been pro-posed to explain the pulsarγ-ray emission(see Chapter3for more de-tails).These have been succesful in explaining some features ofγ-ray emission,but there is no model explaining all observations.The polar cap model(Daugherty and Harding1996)explains the observedγ-rays in terms of curvative radiation or IC scattering of charged particles ac-celerated in rotation-induced electricfields near the poles of the pul-sar.The superstrong magneticfield in the pulsar magnetosphere and the dense low-energy photon environment make it highly probable that high-energyγ-rays are converted into e±-pairs.To escape,γ-ray pho-tons should be directed outwards along thefield lines.Another model, the outer gap model(Zhang and Cheng1997),considers curvatureγ-ray production by e±-pairs in the regions close to the light cylinder.The pairs are created inγγ-interactions of high-energy photons with thermal X-rays from the pulsar surface.The high energy spectra of these two models are quite different,both predict a spectral break at high energies,but the phase resolved spec-trum of the polar cap model falls offmuch more rapidly at high energies (super-exponential)compared to the outer gap model(exponential cut-off).There is also the possibility of a second,higher energy,pulsed component due to IC scattering expected in some outer gap models. Fig.7illustrates how the LAT high-energy response and spectral reso-data.Dotted line:outer gap model.Dashed line:polar cap model.Error bars shown on the models are those expected from the GLAST mission in a one-year sky survey (Thompson2001,and references therein).lution will enable these two models to be easily distinuished for bright pulsars.The two types of models also predict different ratios of radio-loud and radio-quietγ-ray pulsars.Polar cap models predict a much higher ratio of radio-loud to radio-quietγ-ray pulsars,because in these models the high-energy and radio emission both originate in the same magnetic po-lar region.Thus a measurement of the ratio of radio-loud to radio-quiet γ-ray pulsars provides an important discrimination between emission models.GLAST will detect many more pulsars and thus will measure this ratio.The number of pulsars that GLAST will see depends on the emission mechanism and on the distribution of these sources on the sky. An empirical estimate made by extrapolating a log N−log S curve of the known pulsars suggests that GLAST might expect to detect between30 and100γ-ray pulsars.Fig.8shows one of the classic measures of pul-sar observability,the spin-down energy seen at Earth.Six of the seven pulsars with the highest value of this parameter areγ-ray pulsars.The GLAST sensitivity will push the lower limit down substantially. Understanding the physics of pulsarγ-ray emission may be important to determine the nature of the unidentified EGRET sources.73out of16Figure8.Gamma-ray pulsar observability as measured by the spin-down energy seen at Earth(Thompson,Chapter7).170unidentified sources are located in or near the Galactic plane,and are likely to be pulsars or yet unknown source population.Some of them could be possibly associated with the Gould Belt,a massive star Galactic structure surrounding the sun.A new population of low-flux sources at an approximate distance100–300pc(Gehrels et al.2000)may be misaligned young pulsars(Harding and Zhang2001). Observations with the LAT will improve our understanding of pulsars in several ways.The high-energy response and energy resolution will allow a determination of the shape of the high energy cutoffs in the bright pulsars.The increased collection area and thus improved photon statistics will allow a search for periodicities on timescales of millseconds to seconds in sources as faint as∼5×10−8without prior knowledge from radio data,allowing for populations studies with these sources. Studying the pulsar driven nebulae is another way to understand the physics of particle acceleration by pulsars.The Crab Nebula has been detected at TeV energies by several groups(Konopelko et al.1998).。

that identifies and locates the teat posi-tions on the cow’s udder as it steps into the milking box. The information derived from the sensor is used to guide a robot arm to attach the milking cups to the teats. The entire process of locating teats and attaching cups must be robust, fast, accurate and safe, without disturbing or stressing the cow.Sensor Challenge in the barnGEA Farm Technologies, formerly GEA Westfalia-Surge, offers worldwide lead-ing innovations and whole product solu-tions for dairy farming. When GEA decided to develop a new generation of automated milking systems, theyTo survive in the harsh uncontrolled milking machine environment, the Tracker 4000 is housed in a sealed packageThe output of worldwide agricultural milk production in 2007 has been 560.5 million tons with an 83.5% share of cow’s milk. The biggest milk producers have been uSA, India and China. In the european Community 150 million tons are produced annually, the e-15 is the biggest market for milk-based products. The production of milk is an industry, so it comes as no surprise that here as well automation technologies, especially in the area of auto-mated milking, are in high demand.Automated milking in dairy farms is rap-idly expanding. The automated process not only frees up valuable time for the farmer, but also makes happier cows that produce higher yields of better quality milk. There are only a select few compa-nies in the world who specialize in these technologies as this is a very difficult and demanding application of machine vision and robots.A key enabling component in auto-mated milking systems is the 3D sensorapproached LMI Technologies to create the vision guidance sensor. GEA recog-nized LMI as a leading edge 3D sensing technology solutions provider, with par-ticular expertise in implementing sen-sors in challenging applications in ex-tremely harsh environments.One of the challenges in automating the milking process is reliably guiding the milking robot to attach the milking cups to the cow’s teats. Cows are highly individual animals in both behavior and appearance. The sensor must cope with significant variations in target shape and colour, moving animals, in a naturally dirty environment cluttered with mud, straw, water and other uncontrolled ob-jects that interfere with reliable opera-tion of vision sensors. The sensor must also ignore sections of the 3D images caused by other components within the field of view, such as the cow’s legs or swinging tails.Ultrasonic ranging and laser triangu-lation techniques had previously been applied to provide robot guidance for au-tomated milking applications, but had drawbacks, including the need for mov-ing parts and/or laser safety concerns.unique Time of Flight Imaging SolutionLMI is one of only a few companies in the world that have expertise in successfully implementing machine vision technolo-Good milk from Happy CowsTime of Flight Imaging enables Automated milking© a d a m G r y k o /f o t o l i a .c o m8 I nspect 11/2009 C o V e R S T o R ygies for applications such as robot guid-ance in extremely difficult environments. Tasked with finding a better solution for guiding milking robots, LMI determined that implementing innovative time of flight 3D imaging would provide a unique and dramatically improved solution to this guidance application. The sensor based on this technology and developed by LMI in conjunction with GEA, is named the Tracker 4000.TOF imaging cameras have a 2D ar-ray of pixels, with each pixel capable of returning time of flight information as well as intensity. The TOF information produces 3D images of a scene where the brightness of each pixel is proportional to the distance from the sensor to the ob-ject, creating an image similar to a topo-graphical map.TOF imaging provides many unique advantages when applied to guiding milking machine robots. The 3D image field of view includes the entire udder as well as the milking cups. The 3D image information from the sensor is analyzed to determine the position and angular orientation of each individual teat, as well as the milking cups. The location in-formation is transmitted to the robot con-troller through Modbus over Ether-net. Use of industry standard protocols simplifies the integration effort for the machine builder.The sensor currently operates at a frame rate of 8.3 Hz, with faster opera-tion in development. Multiple images are taken as the cups are guided to the teats, which track movement of the cow during the attachment process.Determining the locations of both in-dividual teats and milking cups enable differential guidance, where the offsets to guide the robot to the teat locations are determined as the differences in lo-cation of the teats relative to the location of the milking cups on the robot. The re-sult is improved reliability in guidance, as well as simplification of absolute cali-bration requirements for both the sensor and the robot over the full field of view.3D Images with no moving PartsThe TOF sensor is small in size and low in weight, allowing it to be easily mounted on the robot arm, so its position can bechanged by the robot. This allows the viewing angle to be changed, very useful to obtain an unobstructed view of all teats, particularly if two teats are seen to overlap from one viewing angle. Obtain-ing 3D data from a single TOF camera provides a much smaller sensor package than would be required for two camera stereo imaging, which also requires ex-tensive image analysis software to create a 3D image.The TOF principle provides complete 3D images with no moving parts and no laser lines or spots. Traditional laser scanners require use of a mechanical scanning device to capture a full 3D im-age, which increases time required to capture an image, and adds complexity while reducing reliability.The Tracker 4000 sensor is imple-mented with infrared LED illumination integrated into the sensor housing, which does not distract the cows in the milking station. Also, LED illumination eliminates laser safety concerns and related regula-tory documentation, an issue with laser triangulation based sensing.To survive in the harsh uncontrolled milking machine environment, the Tracker 4000 is housed in a sealed pack-age, with a sealed watertight cable con-nector. The mounting bracket covers the top and sides of the sensor to protect from cow kicks and dirt.Award-winning Teat locationThe Tracker 4000 is implemented with the field proven FireSync platform, de-veloped by LMI to simplify the often com-plex task of integrating and synchroniz-ing the many components of a 3D sensor system. FireSync is a synchronized, scal-able distributed vision processing archi-tecture for building reliable high per-formance systems. Real-time image processing algorithms running in the FireSync processor located inside the sensor use proprietary software to ex-tract teat and milking cup locations in the images, ignoring other objects in the field of view, like a cow’s leg or a swing-ing tail. The final result, coordinate posi-tions for teats and milking cup locations in a predefined coordinate system, is de-livered to the robot controller via an Eth-ernet connection. The FireSync platform is used in all of LMI’s new products.LMI’s Tracker 4000 sensor is a tech-nologically advanced device that will cre-ate significant improvements to yield performance and farm productivity, live-stock well-being, enhance reliability, in-creased speed of farm operations and profitability, and improve product quality in the milking process. A short video demonstrating the Tracker 4000 sensor guiding a robotic milking machine to at-tach suction cups onto a cow’s teats can be viewed at: /en/webcasts/time-flight-imaging- enables-automated-milking.In 2008, GEA Farm Technologies (GEA) was awarded the prestigious silver medal for “New Innovations” presented at the EuroTier 2008 tradeshow in Hanover, Germany for developing the innovative milking robot system with the LMI Tracker 4000 sensor.This implementation of TOF technol-ogy is so unique that in February 2009, LMI Technologies was awarded US Pat-ent 7,490,576 B2 from the United States Patent and Trademark Office for the use of Time of Flight sensors in livestock management.lMi Technologies inc., Delta, BC, Canada The milking robot does not only free up valuable time for the farmer, but also makes happier cows that produce higheryields of better quality milkThe ToF sensor is small in size and low in weight, allowing it to be easily mounted on the robot arm I nspect 11/20099C o V e R S T o R y。

a r X i v :a s t r o -p h /0412129v 1 6 D e c 2004Gamma-Ray Bursts and Afterglow Polarisation S.Covino ∗,E.Rossi †,zzati ∗∗,D.Malesani ‡and G.Ghisellini ∗∗INAF /Brera Astronomical Observatory,V .Bianchi 46,22055,Merate (LC),Italy †Max Planck Institute for Astrophysics,Garching,Karl-Schwarzschild-Str.1,85741Garching,Germany ∗∗JILA,University of Colorado,440UCB,Boulder,CO 80309-0440,USA ‡International School for Advanced Studies (SISSA-ISAS),via Beirut 2-4,I-34014Trieste,Italy Abstract.Polarimetry of Gamma-Ray Burst (GRB)afterglows in the last few years has been considered one of the most effective tool to probe the geometry,energetic,dynamics and the environment of GRBs.We report some of the most recent results and discuss their implications and future perspectives.INTRODUCTION Polarimetry has always been a niche observational technique.It may be difficult to apply,requiring special care for the instruments,data reduction and analysis.Indeed,for real astronomical sources,where often the polarisation degree is fairly small at the level of a few per cent,the signal to noise required to derive useful information has to be very high.However,the amount of information that can be extracted by a polarised flux is also very high,since polarisation is an expected feature of a large number of physical phenomena of astronomical interest.This is particularly true for unresolved sources as GRB afterglows,where polarimetry offers one of the best opportunity to infer on the real geometry of the system.In particular,time resolved polarimetry can in principle give fundamental hints on the jet luminosity structure and on the evolution of the expanding fireball.This would provide reliable tools to discriminate among different scenarios.Finally,it has been recently realised that polarimetry of GRB afterglows can offer adirect way to study the physical condition of the Inter-Stellar Medium (ISM)around the GRB progenitor.GRB polarimetry,thus,becomes a powerful probe for gas and dust in cosmological environments,a valuable research field by itself.In the following of this contribution we want to briefly comment on the most recent advancement in the field and discuss the likely future perspectives that are now open by the advent of the GRB dedicated Swift satellite with its unprecedented rapid localisation capabilities [1].SYNCHROTRON AND BEAMING?The first pioneeristic attempts,culminated with the successful observation of a ∼1.7%polarisation level in GRB 990510[2,3],were driven by the hypothesis that the afterglowFIGURE1.Possible different jet structures.From Rossi et al.[15].emission were due to synchrotron radiation[4,5,6].GRB990510was also a perfect case for testing the hypothesis of a geometrically beamedfireball.Indeed,the detection of an achromatic break in the optical light curve[7,8],together with the observed degree of polarisation,gave support to this scenario.Shortly after this result,it was realised that a jetted ultra-relativistic outflow would produce a characteristic time evolution of the polarisation degree and position angle[9,10].The detailed shape of the polarisation curves depends on the dynamical evolution.Testing this model against data is thus a powerful diagnosis of the geometry and dynamics of thefireball.A large number of polarimetric observations has been carried out since GRB990510.A review of these data has been compiled by Covino et al.[11]and Björnsson[12]. However,until recently,the detection of a low level of polarisation required strong observational efforts.This prevented a satisfactory time coverage of the afterglow decay and,in turn,a convincing test for the model predictions.HOMOGENEOUS,STRUCTURED AND MAGNETISED JETS Lacking strong observational constraints,an improvement of the reference models was achieved considering more physical descriptions for the GRB afterglow jets.In the basic model the energy distribution is homogeneous,making the jet a single entity.More complex beam and magneticfield patterns(Fig.1),reflecting a physically more plausible scenario,were studied in several papers[13,14,15]showing that the light curve is barely affected by this parameter,while the polarisation and position angle evolution changes substantially,providing a further diagnostic tool Fig.2.The universal structured jet model predicts that the maximum of the polarisation curve is at the time of the break in the light curve.The position angle remains constant throughout the afterglow evolution.On the contrary,the homogeneous jet model requires two maxima before and after the light curve break and,more importantly,the position angle shows a sudden rotation of90◦between the two maxima,roughly simultaneouslyFIGURE2.Light curve and polarisation evolution for different jet structures.SJ stands for structured jet,HJ homogeneous jet,GJ for Gaussian jet.Thefigure shows the similarity of the predicted light curves for the various models while the polarisation changes considerably.Negative polarisation degrees mark a 90◦rotation for the position angle.From Rossi et al.[15].to the break time of the light curve.At early and late time the polarisation should be essentially zero(Fig.2).This last result is substantially modified if it is assumed that a large-scale magnetic field is driving thefireball expansion.The topics has been widely discussed in the con-text of polarimetry by Granot&Königl[13],Lazzati et al.[14]and[15].Magnetised jets can be both homogeneous and structured.We do not discuss here the details of this recent research branch.However,we note that,at early times,a large-scale ordered magneticfield produces a non negligible degree of polarisation,contrary to the purely hydrodynamical models.Polarimetry may therefore be the most powerful available di-agnostic tool to investigate thefireball energy content and its early dynamical evolution.Dust Induced PolarisationThe observed low polarisation level from GRB afterglows is often comparable to the expected polarisation induced by dust.Dust grains are known to behave like a dichroic, possibly birefringent,medium[16].Significant amounts of dust are expected to lie closeto the GRB site,as a consequence of the observation of a supernova(SN)component inFIGURE3.Assuming as a reference a typical polarisation curve with a homogeneous jet,the presence of some dust along the line of sight deeply modify the observed time evolution if the dust-induced polarisation is comparable to the intrinsic one,as it seems to be the rule for GRB afterglow at least at rather late time after the high-energy event[11].Depending on the relation between the position angle of the dust-induced polarisation and of the intrinsic GRB afterglow polarisation,the typical shape of the curve can be removed or even enhanced.From Lazzati et al.[16].a few GRBs.The measured polarisation will be modified by the propagation of radiation through dusty media.This effect is,contrary to the intrinsic afterglow polarisation,wave-length dependent.The different wavelength dependence open the interesting possibility to study the polarisation signature from the afterglow to study the physical character-istics of dust in cosmological environments:probably the only way to study dust close to star formation regions at high redshift.Even assuming that dust properties close to GRB formation sites are comparable to what we know in the Milky Way(MW),it is important to take into account this component once information from time evolution po-larimetry are derived.The superposition of the intrinsic time evolution to dust-induced components for the GRB host galaxy and the MW may substantially alter the expected behavior(Fig.3).OBSERV ATIONS VS.THEORYSo far,a rather satisfactory coverage of the polarisation evolution of a GRB afterglow has been obtained for three events only:GRB021004[17,16,18,19],GRB030329 [20,21],and GRB020813[22,14].However,firm conclusions from the analysis could have been derived for the last case only.GRB021004and GRB030329showed some remarkable similarities given that their light curves were characterised by a large num-ber of“bumps”or rebrightenings.Several different possibilities has been proposed to model the irregularities in the light curve invoking clumping in the external medium [23];a more complex and not axi-symmetric energy distribution in thefireball[18]or delayed energy injections[19].It was soon clear[16]that the standard models for polar-FIGURE4.Polarisation data for GRB020813[22].Different curves refer to different models.From Lazzati et al.[14].isation could not be applied in these conditions,since they are all derived in cylindrical symmetry.Even for GRB030329,for which a remarkable dataset was obtained[20],no convincing explanation of the polarization and light-curve erratic behaviors has so far been obtained.It is not clear yet to what extent GRB021004and GRB030329belong to the same population of long GRBs.It is argued however that the failed detection of this erratic behavior in other afterglows(such as GRB020813)is not due to a coarser sampling of the light curve.GRB020813was the best case for model testing.Its light curve was remarkably smooth[24],in several optical/infrared bands,and a break in the light curve was clearly singled out.A few polarimetric observations have been carried out providing for the first time polarisation data before and after the light curve break time[22].Lazzati et al.[14]applied to this event a more quantitative approach not limited,as usually done in the past,to the bare qualitative search of features in the polarisation curve (i.e.rotation of the position angle,etc.).A formal analysis was carried out,taking into account the GRB host galaxy and MW dust induced polarisation and the intrinsic GRB afterglow polarisation.All current jet models were considered,including homogeneous and structured jets,with and without a coherent magneticfield.The dataset,did not allow us to strictly derive a bestfitting model.The main result was to rule out the basic homogeneous jets model at a confidence larger than3σ,mainly because of the lack of the predicted90◦position angle rotation.Again the role of the MW dust induced polarisation is significant.All magnetized models and structured jetsfit satisfactorily the data,the ambiguity being mainly due to the lack of early time measurement,i.e. where magnetised or not magnetised models mostly differ(see Fig.4).The debate is still far from being settled.Recently,for GRB030226Klose et al.[25]a quite low upper limits(∼1%)was reported,in rather strict coincidence with the break time,therefore close to the maximum for the polarisation curve if we assume a structured jet model.With one only measurement it is difficult to drawfirm conclusions,since this null polarisation measurement may well be due to dust induced polarisation superposed destructively to the intrinsic,if any,GRB afterglow polarisation.It isfinally worth,even though tautological,to report that,as soon as Swift will be fully operational,distributing routinely prompt localisations,a new era will be open even for GRB polarimetry.It will allow us to carry out more stringent tests to the available models and therefore strictly constraint geometry,energetics and dynamics of thefireball.REFERENCES1.Gehrels,N.,Chincarini,G.,Giommi,P.,et al.2004,ApJ611,10052.Covino S.,Lazzati D.,Ghisellini G.,et al.1999,A&A348,13.Wijers R.A.M.J.,Vreeswijk P.M.,Galama T.J.,et al.1999,ApJ523,1774.Paczy´n ski B.,Rhoads J.E.1993,ApJ418,55.Mészáros P.,Rees M.J.1997,ApJ476,2326.Sari R.,Piran T.,Narayan R.1998,ApJ497,177.Israel G.L.,Marconi G.,Covino S.,et al.(1999),A&A348,58.Harrison F.A.,Bloom J.S.,Frail D.A.,et al.(1999),ApJ523,1219.Ghisellini G.,Lazzati D.(1999),MNRAS309,710.Sari R.(1999),ApJ524,4311.Covino S.,Ghisellini G.,Lazzati D.,Malesani D.2004,ASP Conf.Ser.312,16912.Björnsson G.(2003),astro-ph/030217713.Granot J.,Königl A.(2003),ApJ594,83zzati D.,Covino S.,Gorosabel J.R.,et al.(2004),A&A422,12115.Rossi E.M.,Lazzati D.,Salmonson J.D.,Ghisellini G.(2004),MNRAS354,86zzati D.,Covino S.,di Serego Alighieri S.,et al.(2003),A&A410,82317.Rol E.,Wijers R.A.M.J.,Fynbo J.P.U.et al.(2003),A&A405,2318.Nakar E.,Oren Y.(2004),ApJ602,9719.Björnsson G.,Gudmundsson E.H.,Jóhannesson G.(2004),ApJ615,7720.Greiner J.,???,et al.(2003),Nature426,15721.Klose S.,Palazzi E.,Masetti N.,et al.(2004),A&A420,89922.Gorosabel J.,Rol E.,Covino S.,et al.(2004),A&A422,113zzati D.,Rossi E.,Covino S.,Ghisellini G.,Malesani D.(2002),A&A395,524.Covino S.,Malesani D.,Tavecchio F.et al.(2003),A&A404,525.Klose S.,Greiner J.,Rau A.et al.(2004b),AJ128,1942。

a r X i v :a s t r o -p h /0007255v 1 18 J u l 2000Subm to ApJL,7/18/2000Delayed X-Ray Afterglows from Obscured Gamma-Ray Burstsin Star-Forming RegionsPeter M´e sz´a ros 1,2,3and Andrei Gruzinov 21Institute of Astronomy,Madingley Road,Cambridge CB30HA,England,U.K.2Institute for Advanced Study,School of Natural Sciences,Princeton,NJ 085403Pennsylvania State University,525Davey Lab.,University Park,PA 16802ABSTRACT For Gamma-Ray Bursts occurring in dense star-forming regions,the X-ray afterglow behavior minutes to days after the trigger may be dominated by the small-angle scattering of the prompt X-ray emission offdust grains.We give a simple illustrative model for the X-ray light curves at different X-ray energies,and discuss possible implications.A bump followed by a steeper decay in soft X-rays is predicted for bursts which are heavily obscured in the optical.Subject headings:Gamma-rays:Bursts —Radiation Mechanisms 1.Introduction Most of the X-ray fluence of a Gamma-Ray Bursts (GRB)is emitted during the prompt X-ray flash,the afterglow emission being smaller by a factor ∼10(Costa 1999,Piro 2000,van Paradijs,et al 2000).For a GRB in a large star forming region,a significant fraction of the prompt X-ray emission will be scattered by dust grains.Since the dust grains scatter the X-rays by a small angle,time delays of the scattered x-rays will be small (minutes to days,depending on the X-ray energy and the grain size).If the X-ray scattering opacity is substantial,the intermediate timescale,softer part of the X-ray afterglow will be dominated by the dust scattering,the direct X-ray emission from the blast wave being weaker.This intermediate X-ray light curve will then generally be steeper than the original unscattered afterglow would have been.The optical afterglow will be undetectable,due to the high obscuration,but a near-infrared source at the µJy level should be present over timescales of months.When a database of X-ray afterglows in different energy bands becomes available,through missions such as HETE-2and Swift,one should be able to determine whether some GRBs indeed occur in large dusty regions.2.A specific exampleTo illustrate the phenomenon,we consider as a starting point a typical GRB,whose unscattered X-ray light curve is parametrized in a simplified manner by two asymptotic power laws with a peak or break at about100s,decaying at late times as t−1.3,1+(t/100s)F0(t)=1keV −2.(2) The last two assumptions are adopted from the interpretation of the dust-grain scattering observations given by Mitsuda,Takeshima,Kii,&Kawai(1990).Of course,there is no reason to believe that the dust in a distant GRB host galaxy will be similar to the dust here. We just use these assumptions as a nominal example.It follows from assumption(1)that a GRB going offin such a star-forming region will have no detectable optical afterglow,but the gamma-rays should be able to penetrate through.However the behavior of the X-ray afterglow will be dependent on the specific energy band being considered.At X-ray optical depths less than few,dust grains of size a will scatter X-rays of energy ǫby an angleθ∼0.2λ/a,whereλis the X-ray wavelength(Overbeck1964).We parametrize this asθ(ǫ)=4.13×10−3 a1keV −1.(3) The corresponding time lag is t∼Rθ2/2c,ort(ǫ)∼8.8×104s a1keV −2 R0246-6-4-2log time(s)Fig.1.—Dust-scattered X-ray afterglow.Thin line:unscattered X-ray flux.Thin dashed line:scattered X-ray flux.Thick line:total flux.The flux normalization is arbitrary,while the relative fluxes correspond to the example discussed in the text for an energy of 2keV.The X-ray telescope (XRT)on the Swift afterglow space mission scheduled to be launched in 2004(/xray/swift/xrt/)will be sensitive over an energy range 0.2≤ǫ≤10keV,corresponding to GRB rest-frame energies 0.5to 25keV for a redshift z =1.5.Let us consider the dust contribution at several GRB-frame X-ray energies.At 10keV,the X-ray optical depth given by equation (2)is much less than ubity,so the X-ray afterglow is not affected by the dust scattering.The light curve will then just be the usual unscattered time dependent X-ray afterglow,which in this example is parametrized through equation (1).At 3keV,the optical depth is τ∼0.3.The time lag,equation (4),is t ∼104s.Since the total fluence is f ∼ dtF 0∼500s,the scattered flux is F s ∼τf/t ∼0.01.The unscattered flux at 104s is given by equation (1),F 0∼3×10−3.Thus scattering has only a minor influence on the afterglow.At 2keV,the optical depth is τ∼1.The time lag is t ∼2×104s.The scattered fluxis F s∼τf/t∼0.03.The unscatteredflux at2×104s is F0∼10−3.In the time interval from hours to weeks,the dust scattering dominates the afterglow,and,as shown in Fig.1, the afterglow is approximately a power law F∝t−1.75.This is because dust grains of radius a<0.06µm will scatter the prompt emission with longer time lags,t∝a−2,and with smaller optical depthsτ.To calculateτ,we take a standard dust grain size distribution where the number of grains of size of order a is∝a−2.5(Mathis,Rumpl,&Nordsieck1977).For a scattering cross section∝a4(Overbeck1964),the optical depth isτ∝a1.5∝t−0.75,so the flux F∝t−1.75.At1keV,the optical depth is∼3,and the dust grain scattering dominates the afterglow at t>10min.Multiple scatterings will be important,the net deflection angle being,inθ(ǫ)∝τ1/2(ǫ)θ(ǫ)∝ǫ−2,so the time delay is the small-angle deflection regime,¯θ∝N1/2sct(ǫ)∼¯θ2(R/2c)∝ǫ−4.This decreases at early times the ratio of the scattered to the unscatteredflux,compared to the values in the single scattering regime at2keV,and it increases the time after which the scatteredflux becomes dominant,the late time decay having the same time exponent of-1.75.3.DiscussionThe specific example that we have described shows that highly obscured star-forming regions should lead to specific signatures in the X-ray light curve of gamma-ray bursts occurring in them.This consists of a secondaryflattening or bump in the light curve at X-ray energiesǫ∼2−3keV.Depending on the size and dust column density of the region, this bump occurs hours to days after the initial“canonical”X-ray decay has been going on, followed by a steeper F x∝t−1.75decay.At lower energies(e.g.in the0.2-0.5keV range) the X-ray bump in the light curve should appear at increasingly later times and contribute a decreasing fraction of the total energy.The presence of this X-ray signature is expected to be associated with bursts which do not produce a detectable transient at optical wavelengths(OT).There are at least three bursts so far for which an X-ray decay index is known and an OT was not detected:GRB 970204,970828and991214.The X-ray decay indices of these were-1.4,-1.56and-1.00,based on GCN notices(e.g.Greiner,2000).The third is clearly a canonical decay,and should be the unscattered component,but the other two could be the scattered component,considering the uncertainties in the grain size distribution and cross sections that determine the decay rate.There is a larger number of bursts for which an X-ray afterglow was reported,while an OT was not found.However,since optical observationsfirst require an X-ray position, which so far has been possible only after hours of delay and with arc-minutes accuracy,thereis currently a possible bias againstfinding OTs.Stronger constraints will have to await the faster coordinate alerts and smaller X-ray error circles expected from dedicated GRB afterglow missions such as HETE-2and Swift.An infrared source is expected to be associated with such obscured,X-ray peculiar GRBs,since the dust will re-radiate the UV and soft X-ray radiation of the absorbed source. Thermal reemission and scattering outside the sublimation radius is expected to cause de-layed IR emission even when the optical transient is only partially absorbed(Waxman& Draine2000,Esin&Blandford2000),and the same is expected here,except for the optical transient being compeletely obscured.As a numerical example,for an isotropic equivalent total burst energy E∼1053erg at a redshift z∼1the normalization of the X-rayflux for the burst of Figure1would be F x∼10−9erg cm−2s−1keV−1for t∼<100s,in the usual range of X-ray afterglowfluxes detected by Beppo-SAX(Costa,et al,1999).The dust reradiation occurs beyond the sublimation radius R s∼10L1/249pc at wavelengthsλ∼>2(1+z)µm,where 1049L49erg/s is the early UV component of burst afterglow(Waxman&Draine2000).The time delay associated with the reradiatedflux is t IR∼(R s/2c)θ2j whereθj=10−1θ−1is a typical collimation half-angle of the burst radiation.At z∼1the corresponding infrared flux at2.2µm would be F2.2µm∼L49θ2j/[4πD2L(R s/2c)θ2jν]∼0.3L1/249µJy,independent of θj,or m K∼23.3compared to Vega,approximately constant for a time t IR∼5×106θ2−1L1/249 s.The IRflux of the host galaxy at that redshift could exceed this value,but8-m class telescopes in good seeing conditions or with adaptive optics would resolve the galaxy,facil-itating detection of the point-like IR afterglow.If the X-ray to IRfluxes can be calibrated for a sample of sources at z∼1,suchγ-ray detected GRBs with anomalous X-ray afterglow behavior and no OT may be used as tracers of massive stellar collapses.It may thus be possible to detect star-forming regions out to redshifts larger than those detectable with optical or infrared techniques,since typical GRBγ-ray and X-rayfluxes can in principle be measured out to z∼10−15.We thank Andrew Blain,Bruce Draine,Masataka Fukugita,David Hogg,Kevin Hurley, Richard McMahon,Martin Rees and Eli Waxman for useful discussions.PM was supported by NASA NAG5-9192,the Guggenheim Foundation,the Sackler Foundation and the In-stitute for Advanced Study.AG was supported by the W.M.Keck Foundation and NSF PHY-9513835.REFERENCESCosta,E.1999,A&ASS,138,425Esin,A&Blandford,R.D.,2000,ApJL in press(astro-ph/0003415) Greiner,J.,2000,http://www.aip.de:8080/∼jcg/grbgen.htmlMathis,J.S.,Rumpl,W.,&Nordsieck,K.H.1977,ApJ,217,425 Mitsuda,K.,Takeshima,T.,Kii,T.,&Kawai,N.1990,ApJ353,480 Overbeck,J.W.1964,ApJ,141,864Piro,L.2000,astro-ph/0001436van Paradijs,J,Kouveliotou,C.&Wijers,R.A.M.J.,2000,ARAA,in press Waxman,E&Draine,B.,2000,ApJ,in press(astro-ph/9909020)。

小学上册英语第三单元测验试卷英语试题一、综合题(本题有100小题，每小题1分，共100分.每小题不选、错误，均不给分)1. c region is known for its ________ (极昼和极夜). The Azte2. A newt is similar to a ______ (青蛙).3. A _____ (植物故事分享) can inspire others to connect with nature.4.I like to have fun with my ____.5.My hamster has a _______ (舒适的) home.6.The capital of Nigeria is __________.7.He is a great ___. (singer)8.The ________ loves to climb trees and explore.9.The ______ protects the seeds of a plant.10.The ______ (花期) of a plant varies with species.11.I have a ___ (big) imagination.12.What do we call a musical performance by a group of people?A. SoloB. BandC. OrchestraD. ChoirC13.Certain plants are prized for their ______ and use in landscape design. (某些植物因其美丽和在景观设计中的应用而受到赞赏。

)14.The cookies are _______ (baking) in the oven.15. A ______ has a unique pattern on its fur.16.What is the term for a baby horse?A. ColtB. FoalC. CalfD. KidB17.My _____ (家人) supports me all the time.18.The _____ (草) is very green.19.What is the name of the large body of saltwater?A. LakeB. RiverC. OceanD. PondC20.I enjoy watching __________ with my family. (电影)21.The ____ has large wings and can glide through the air.22. A snake is long and ______.23.What is the name of the first man to walk on the moon?A. Yuri GagarinB. Neil ArmstrongC. Buzz AldrinD. John GlennB24.What is the term for the speed of light?A. 300,000 kilometers per secondB. 150,000 kilometers per secondC. 100,000 kilometers per secondD. 1,000 kilometers per secondA25.The owl can turn its head almost _______ (完全) around.26.What is the currency used in the USA?A. DollarB. EuroC. PoundD. Yen27.What do we use to measure time?A. RulerB. ClockC. ScaleD. CompassB28.The _____ (狼) is an important part of its ecosystem.29.The ______ (小鹿) runs quickly from danger.30.What is the name of the largest reptile in the world?A. Komodo DragonB. Saltwater CrocodileC. Green AnacondaD. Leatherback TurtleB31.My brother is an adventurous __________ (旅行者).32.My mom is my caring _______ who loves to hug me.33.The rabbit is ________ carrots.34.Honey is made by ______.35.I usually eat ______ for breakfast.36.The __________ (印加帝国) was located in South America.37.My __________ (玩具名) is very __________ (形容词) and fun.38.All living things need ______ to survive.39.The chemical symbol for silver is ______.40.The ____ is known for its striking colors and patterns.41.What do we breathe in?A. WaterB. AirC. FoodD. LightB42.We have lunch at _____ (noon/midnight).43.I see a _______ (whale) in the ocean.44.The fox is cunning and known for its ______ (智慧).45.What is the name of the first woman to fly in space?A. Sally RideB. Valentina TereshkovaC. Mae JemisonD. Eileen Collins46.What do we call the study of ancient civilizations?A. ArchaeologyB. AnthropologyC. HistoryD. SociologyA47.The stars are _______ (shining) brightly tonight.48.I can ______ (swim) very well.49.Which shape has three sides?A. SquareB. TriangleC. CircleD. Rectangle50.I like to ___ with my family. (spend time)51.The cat is _______ on the couch.52.The _____ (竹子) grows very quickly.53.My brother enjoys __________ (运动) after school.54.What do we call a person who performs in plays?A. ActorB. DirectorC. ProducerD. StagehandA55.What is the main purpose of a toothbrush?A. To wash handsB. To brush teethC. To comb hairD. To clean shoes56.The chemical formula for -pentanol is ______.57.The __________ (历史的共鸣) resonates across cultures.58.She has a new ________.59.I enjoy _______ (参加) music festivals.60.What is the name of the famous landmark in Agra, India?A. Taj MahalB. Red FortC. Qutub MinarD. Hawa MahalA61.The capital of Scotland is __________.62.What is the opposite of "light"?A. BrightB. HeavyC. SoftD. HardB Heavy63. A rabbit's favorite food is ______ (胡萝卜).64.I can _______ (游泳) very well.65.The ________ (城市更新) revitalizes neighborhoods.66.The cake is _____ (sweet/sour).67.Some _______ are known for their strong scents.68.I help my __________ with yard work. (爸爸)69.I have _____ (one/two) sister(s).70. A hydrocarbon consists only of hydrogen and ______.71.What type of tree produces acorns?A. PineB. MapleC. OakD. BirchC72.The ________ (chick) is cute and fluffy.73.What is the name of the famous landmark in Egypt?A. Great WallB. Eiffel TowerC. PyramidsD. Colosseum74.We celebrate our ________ (achievements) together.75.在中国，________ (history) 中的许多人物都是民族的英雄。

英文我最爱的歌曲作文英文：One of my all-time favorite songs is "Bohemian Rhapsody" by Queen. This iconic masterpiece holds a special place in my heart for several reasons. Firstly, its unique structure and musical arrangement captivate me every time I listen to it. The blend of rock, opera, and ballad elements creates an unforgettable auditory experience thattranscends genres. From the hauntingly beautiful intro tothe thunderous climax, every section of the song tells a story and evokes powerful emotions.The lyrics of "Bohemian Rhapsody" are another aspectthat I adore. They are enigmatic and open to interpretation, allowing listeners to find personal meaning in them. The narrative unfolds like a mini-drama, following the protagonist through a journey of self-discovery, confession, and redemption. Lines like "Is this the real life? Is this just fantasy?" and "Nothing really matters, anyone can see"resonate deeply with me, as they reflect the universal human experience of grappling with reality and existential questions.Moreover, the vocal performance by Freddie Mercury is simply unparalleled. His dynamic range, emotional delivery, and theatrical flair elevate the song to a level of brilliance that few artists can achieve. The way he seamlessly transitions between different vocal styles showcases his incredible talent and versatility. Whenever I hear his voice soar during the operatic section or the raw vulnerability in the quieter moments, I am reminded of why he is considered one of the greatest singers of all time.Furthermore, "Bohemian Rhapsody" holds nostalgic value for me. It reminds me of cherished memories associated with family gatherings, road trips, and late-night karaoke sessions with friends. It's a song that brings people together, transcending generations and cultural boundaries. Whether I'm belting out the lyrics at the top of my lungs or simply swaying along to the music, it never fails to uplift my spirits and fill me with joy.In conclusion, "Bohemian Rhapsody" is more than just a song to me; it's a timeless masterpiece that continues to inspire and enchant listeners around the world. Its enduring popularity and cultural significance serve as a testament to the genius of Queen and the enduring legacy of Freddie Mercury. Whenever I hear those iconic opening chords, I know that I'm about to embark on a musical journey unlike any other.中文：我最喜欢的歌曲之一是Queen的《波西米亚狂想曲》。

How Radiation Threatens Health如何辐射威胁健康As worries grow over radiation leaks at Fukushima, is it possible to gauge the immediate and lasting health effects of radiation exposure? Here's the science behind radiation sickness and other threats facing Japan。

由于担心福岛核电站辐射泄漏,有可能测量辐射的直接和持久的健康影响?这是背后的科学面临辐射病和其他威胁日本。

The japan earthquake and tsunami on match 11,a powerful,magnitude 9.0 quake hit northeastern japan,triggering a tsunami with 10-meter-high waves that reached the U.S. West coast.Here’s the science behind the disaster.在比赛11日日本地震和海啸,强大,日本东北部发生9.0级地震,地震引发的海啸与西方10-meter-high波,达到美国coast.Here灾难背后的科学。

The developing crisis at the Fukushima Daiichi nuclear power plant in the wake of the March 11 earthquake and tsunami has raised concerns over the health effects of radiation exposure: What is a "dangerous" level of radiation? How does radiation damage health? What are the consequences of acute and long-term low-dose radiation? 发展危机福岛第一核电站在3月11日地震和海啸后引发了担忧辐射对健康的影响:“危险”水平的辐射是什么?如何辐射损伤健康?急性和长期低剂量辐射的后果是什么?Though radioactive steam has been released to reduce pressure within the wrecked complex's reactors and there has been additional radiation leakage from the three explosions there, the resulting spikes in radiation levels have not been sustained. The highest radiation level reported thus far was a pulse of 400 millisieverts per hour at reactor No. 3, measured at 10:22 A.M. local time March 15. (A sievert is a unit of ionizing radiation equal to 100 rems; a rem is a dosage unit of x-ray and gamma-ray radiation exposure.) The level of radiation decreases dramatically as distance from the site increases. Radiation levels in Tokyo, about 220 kilometers to the southwest, have been reported to be only slightly above normal.尽管放射性蒸汽被释放减少受损的反应堆内压力和有额外的辐射泄漏的三次爆炸,产生的辐射水平激增并不持久。