1: \documentclass[12pt,preprint]{aastex}
2: \shorttitle{A Brightness Enhancement in Q0957+561}
3: \shortauthors{Oscoz et al.}
4: \begin{document}
5:
6: \title{A Large Brightness Enhancement of the QSO 0957+561 A
7: Component}
8:
9: \author{A. Oscoz\altaffilmark{1}, D. Alcalde\altaffilmark{1},
10: M. Serra--Ricart\altaffilmark{1}, E. Mediavilla\altaffilmark{1},
11: J. A. Mu\~noz\altaffilmark{1}}
12: \affil{Instituto de Astrof\'\i sica de Canarias, E-38205 La Laguna,
13: Tenerife, Spain}
14: \email{aoscoz@ll.iac.es, dalcalde@ll.iac.es, mserra@ot.iac.es,
15: emg@ll.iac.es, jmunoz@ll.iac.es}
16:
17:
18: \begin{abstract}
19: We report an increase of more than 0.2 mag in the optical
20: brightness of the leading image (A) of the gravitational lens
21: Q0957+561, detected during the 09/2000 -- 06/2001 monitoring
22: campaign (2001 observing season). The brightening is similar to
23: or even greater than the largest change ever detected during
24: the 20 years of monitoring of this system. We discuss two
25: different provisional explanations to this event: intrinsic
26: source variability or microlensing (either short timescale
27: microlensing or cessation of the historical microlensing). An
28: exhaustive photometric monitoring of Q0957+561 is needed until
29: summer of 2002 and during 2003 to discriminate between these
30: possibilities.
31: \end{abstract}
32:
33: \keywords{gravitational lensing---quasars: individual (QSO 0957+561)}
34:
35:
36: \section{Introduction}
37:
38: The first gravitational lens system discovered, Q0957+561
39: \citep{wal79}, has become the most observed gravitational
40: mirage. This system has been the target of continuous
41: monitoring in optical and radio wavelenghts. The early works by
42: \citet{flo84,sch86,leh89,van89,sch90}; and \citet{rob91}, were
43: followed by other recent monitoring, as those by
44: \citet{bes92,kun95,osc96,kun97,osc97,haa99}; and \citet{ser99}.
45:
46: Three outstanding events can be noticed during these twenty
47: years of monitoring. Firstly, the existence of a large
48: timescale microlensing (of several years). The analysis made by
49: \citet{pel98} with data corresponding to the period 1979--1996
50: clearly shows its presence. Next, a strong 0.13--mag intrinsic
51: brightening of Q0957+561 in two months was detected by
52: \citet{kun95}. And finally, \citet{sch96} noticed a possible
53: microlensing event with maximum amplitude of 0.05 mag and a
54: timescale of 90 days. The sharp drop detected by Kundic
55: et al. allowed to solve the long-standing problem concerning
56: the ``short" ($\sim 410$ days) and ``long" ($\sim530$ days)
57: time delays between the A and B components of the system. The
58: observations confirmed that the short value was the correct one
59: \citep{osc96,kun97}, constraining the time delay between 410
60: and 440 days. Moreover, this feature allowed to obtain a first
61: accurate value for the delay (417$\pm$3 days, Kundic et al. 1997;
62: 424$\pm$3 days, Oscoz et al. 1997). This robust estimate lead
63: to search for the existence of possible microlensing events
64: \citep{gil98,goi98}, but no other event of the type reported by
65: \citet{sch96} has been detected (see Gil--Merino et al. 2001).
66:
67: The Instituto de Astrof\'\i sica de Canarias gravitational
68: lensing group started a long-term monitoring program on this
69: system in 1996, with the 0.82-m IAC80 telescope at the
70: Observatorio del Teide in Tenerife, Spain. Our set of almost
71: 500 individual observations in $R$ band, together with several
72: hundreds of points in $V$ band, constitutes one of the largest
73: photometric database of a gravitational lens system. The
74: application of a new data reduction method (to improve the
75: original aperture photometry), and the development of a new
76: procedure to estimate the time delay, gave a
77: value of $\Delta t_{Q0957} = (425\pm4)$ days \citep{ser99}. The
78: accuracy in the time delay was improved furthermore by
79: including data from other groups in the period 1984--1999 and
80: by using several statistical methods for the calculations. A
81: new value of $422.6\pm0.6$ days was derived \citep{osc01}.
82:
83: \section{The 2000 and 2001 monitoring campaigns}
84:
85: To date, we have only published the data corresponding to the
86: campaigns from 1996 to 1999 (Oscoz et al. 2001). A summary of
87: the last two observing campaigns (10/1999 -- 06/2000 and
88: 09/2000 -- 06/2001) are shown in Table 1. (full
89: data corresponding to all the observing campaigns --dates,
90: brightness and errors of the individual data-- can be found at
91: http://www.iac.es/project/quasar/mserra/meth.html). Each data
92: point is the result of averaging several individual measurements.
93: The reduction procedure was done by means of the {\it pho2com}
94: IRAF\footnote{IRAF is distributed by the National Optical
95: Astronomy Observatories, which are operated by the Association
96: of Universities for Research in Astronomy, Inc., under
97: cooperative agreement with the National Science Foundation.}
98: task (for a complete description of {\it pho2com}, see
99: Serra-Ricart et al. 1999). Once the final light curves were
100: obtained, the data were checked to eliminate inconsistent
101: measurements: some points are affected by systematic effects,
102: and show strong and {\it simultaneous} (non time-shifted)
103: variations in both components. These points are the result of
104: bad weather conditions or problems with the CCD and/or the
105: telescope. The number of discarded data points was always small
106: (18 out of 401).
107:
108: The final light curves of our monitoring campaign ranging from
109: 25.02.1996 to 06.06.2001 in the $R$ and $V$ bands are presented
110: in Figure 1. The apparent magnitudes of the A and B components
111: were derived by comparing the instrumental fluxes with those of
112: two reference stars (D and H, see Serra-Ricart et al. 1999).
113: From the scatter in the comparison star differential light curve
114: we estimate that the photometry is accurate to 2--3 per cent. In
115: Figure 1 a delay of 422.6 days has been applied to the B
116: component, but no magnitude correction has been applied to the
117: data set, both the A and B magnitude are the real ones. It is
118: obvious that the behavior of the light curves shows epochs in
119: which both components fade, followed by epochs in which they
120: brighten, in a quasi-periodic way. This is the general trend
121: observed during the 2000 campaign. However, a conspicuous
122: behavior can be seen in the 2001 campaign, where a brightening
123: of more than 0.2 mag in component A can be observed between day
124: $\sim$ JD2451500 and day $\sim$ JD2452065. This behavior is
125: evident when only the points corresponding to the 2001 campaign
126: are represented. This is shown in the upper panel of Figure 2,
127: where the B component is not delayed, and the data have been
128: averaged into 10-day bins to reduce the noise and to clearly
129: show the trend of both components.
130:
131: \section{Discussion}
132:
133: Every year, when the observing season for Q0957+561 is finished,
134: the obtained data are reduced by the IAC group together with
135: the data from previous campaigns. A possible explanation of the
136: trend appearing in the A component during the 2001 campaign is
137: that the data have been badly obtained and/or reduced, leading
138: to a wrong magnitude estimate. However, a mistake in the reduction
139: procedure would lead to changes in the whole data, not only in
140: the points corresponding to the latest year. Moreover, only the
141: A component points show this variation, while the B data remain
142: almost constant (see the upper panel of Fig. 2). These facts
143: demonstrate that a wrong reduction process or a failure in data
144: acquisition can not be the explanation for this trend in the
145: image A light curve.
146:
147: Differential photometry between both comparison stars (see
148: Section 2) has been performed in order to check for their
149: stability. No significant variability in the differential light
150: curve is observed, as can be seen in the lower panel of Figure 2.
151: So, the brightening of the A component is certainly not due to
152: any change of the reference stars.
153:
154: Two different explanations for the monotonous increase in the
155: brightness of the A component of Q0957+561 are proposed: (i) it
156: is intrinsic variability of the source; (ii) it is due to a
157: microlensing event, either a short-time one (months to a few
158: years) or the cessation of the historical microlensing (about
159: twenty years, see Pelt et al. 1998).
160:
161: \subsection{Intrinsic variability of the source}
162:
163: Figure 1 shows that there have been several epochs of remarkable
164: intrinsic variability in the last 5 years. However, these
165: changes are always less than 0.15 mag; for example $\sim 0.14$
166: mag between JD 2450300 and JD 2450500 or $\sim 0.12$ mag between
167: JD 2451100 and JD 2451300. Note that the sharp drop
168: detected by \citet{kun95} had an amplitude of 0.13 mag. So, if
169: the trend found in the 2001 observing campaign is the consequence
170: of intrinsic variability of the source, it would be the largest
171: intrinsic variation ever found, with an optical flux increase of
172: at least 0.2 mag. The large brightening now detected would make
173: it relatively easy to obtain a final confirmation of the time
174: delay between both components of Q0957+561. In addition, and
175: perhaps even more important, it would allow to obtain this delay
176: independently of the method selected, finally solving the
177: controversy of the last few years. Thus, a monitoring of Q0957+561
178: until 2004 would be crucial to improve our knowledge of both the
179: time delay and the robustness of several statistical methods.
180:
181: \subsection{Microlensing}
182:
183: As stated before, a microlensing event of more than ten years is
184: being produced in Q0957+561. \citet{pel98} made a statistical
185: analysis of the Q0957+561 light curves from the first 17 years
186: (1979--1996) with data from Schild et al.
187: (http://cfa-www.harvard.edu/$\sim$rschild), Princeton University
188: (Kundic et al. 1997) and the IAC group first observing campaign
189: (Oscoz et al. 1996). These observational data led them to obtain
190: a time delay of 416.3$\pm$1.7 days, and then to calculate the
191: differential light curve between both components of Q0957+561
192: (taking into account this value for the time delay). Pelt et al.
193: (1998) finally concluded the existence of a first variation of
194: 0.25 mag in about 6 years, followed by a quiet phase of about 8
195: years without variability over 0.05 mag. The historical differential
196: light curve is presented in Figure 3, where only data from Schild
197: et al. and the first four campaigns of the IAC group campaigns
198: have been used. A time delay of 422.6 days has been applied to
199: B component data, and only the annual averages are presented.
200: The analysis made by Pelt et al. (1998) shows that objects with
201: mass of $< 10^{-5} M_\odot$ can explain the 0.25--mag event. They
202: also stated that the existence of objects with a mass as high as
203: $1 M_\odot$ was possible, although they are quite unlikely. A
204: remarkable fact since the beginning of this event is that
205: component B remains brighter than component A. However, the
206: differential light curve does not clearly lead to the
207: interpretation of long term microlensing. The shape of this curve
208: does not match the one expected for a microlensing event, and
209: it is difficult to explain the dip between day 5000 and day 6000.
210: A point favouring the historical microlensing interpretation is
211: that component A is brighter than B in the line emission
212: (Angonin-Willaime \& Vanderriest 1995),
213: which is supposed to be not affected by microlensing.
214:
215: In any case, a point against the explanation of the observed
216: variability in the 2001 campaign as the end of the historical
217: microlensing is the fact that this large timescale microlensing
218: took six years ($\sim$ 1983-1988) to vary 0.25 mag, while about
219: the same variation has been measured now in only a year.
220:
221: Another interesting explanation to this large change in brightness
222: is that it can be produced by a short timescale (from several
223: months to a few years) microlensing event. Until now, no short
224: timescale microlensing event has been completely confirmed in
225: Q0957+561, although some observing campaigns with several
226: participating observatories have been carried out \citep{col02}.
227: Even the possible microlensing event reported by \citet{sch96}
228: is not entirely convincing. This author, with his own data and a
229: time delay of 404 days, found amplitude peaks of 0.05 mag and
230: 90 days long in the microlensing curves. This phenomenon was
231: interpreted as short timescale microlensing due to objects
232: with 10$^{-5} M_\odot$ mass.
233:
234: Refsdal et al. (2000) employed the microlensing light curve by
235: Pelt et al. (1998) to restrict the microlens mass. These authors
236: concluded that the lens mass could be restricted to values in the
237: interval 10$^{-6} M_\odot$-5$M_\odot$.
238: Another analysis was performed by Schmidt \& Wambsganss (1998)
239: with data in the $g$ band by Kundic et al. (1997, two observing
240: campaigns: 1994, December to 1995, May and 1995, November to
241: 1996, July) and a time delay of 417 days. No variation larger
242: than 0.05 mag was found in the differential light curve. Two
243: conclusions were derived: (i) MACHOs with masses in the interval
244: 10$^{-5} M_\odot$-10$^{-3} M_\odot$ can be excluded for a quasar
245: with a radius less than 10$^{-4} h_{60}^{-1/2}$ pc; and (ii)
246: there were no evidence of short timescale events. Lately,
247: Wambsganss et al. (2000) added to the previous light curves the
248: data obtained until 1998 in the same band and with the same
249: telescope, detecting again no microlensing with amplitude larger
250: than 0.05 mag. They could extend the previous limits, excluding
251: an halo only made by MACHOS with masses between 10$^{-6} M_\odot$
252: and 10$^{-2} M_\odot$ for a quasar with radius less than
253: 10$^{-4} h_{60}^{-1/2}$ pc.
254:
255: Finally, Gil-Merino et al. (2001) performed an exhaustive
256: analysis of the microlensing signal obtained with the IAC 1996,
257: 1997, and 1998 observing campaigns in the $R$ band. They
258: selected a delay of $\Delta t_{Q0957} = 425\pm4$ days
259: (Serra-Ricart et al. 1999). Gil-Merino et al. concluded that:
260: (i) no 3 months duration and 0.05 mag amplitude events are
261: found in the microlensing light curves, so these events do not
262: occur in a continuous way; (ii) from a conservative point of
263: view, the amplitude of any microlensing signal must be in the
264: interval [-0.05 mag, +0.05 mag] (the same limit found by
265: Pelt et al. 1998 and Wambsganss et al. 2000); and (iii) the
266: small variability observed in the differential light curves
267: could be originated, in a natural way, by observational noise
268: mechanisms.
269:
270: \section{Final remarks}
271:
272: We present in this letter a large, $|\Delta m| \geq 0.2$-mag,
273: brightening of component A of the gravitational lens system
274: Q0957+561. The event occurs between day $\sim$ JD2451500 and day
275: $\sim$ JD2452065, our last observing date, so its amplitude could be
276: even larger. Two different alternatives are offered to explain
277: this variation: intrinsic variability or microlensing.
278:
279: The historical light curve of Q0957+561 presents several large
280: variations in amplitude. Some of them are fast, as the 0.13--mag
281: sharp drop detected by \citet{kun95}, whereas others,
282: larger in magnitude (but always below 0.15 mag), are relatively
283: slow (see Figure 1). However, the detected 2001 variability in
284: component A, if intrinsic, would be the largest one ever reported
285: in this quasar, allowing so to obtain a confirmation of the time
286: delay independently of the method employed.
287:
288: Alternatively, the observed brightening could be due to a
289: microlensing event. As a possible explanation, it could
290: correspond to short timescale microlensing (months to years).
291: Microlensing events of this type have been detected in several
292: gravitational lenses, specially in Q2237+0305, where they are
293: almost routinely detected (a noticeable 0.15--mag microlensing
294: in component A of Q2237+0305 has been recently reported;
295: Wozniak et al. 2000; Alcalde et al. 2002). On the contrary, in
296: Q0957+561 it would be the first secure event of this type
297: detected (see Schild 1996; Gil--Merino et al. 2001). Another
298: possibility is that it could indicate the end of the historical
299: microlensing which started in 1983 (Pelt et al. 1996). However,
300: the variation seems too fast to correspond to the cessation of
301: such a microlensing event.
302:
303: In any case, the definitive answer will only come after the
304: observation of component B during 2002. So, an
305: exhaustive monitoring of Q0957+561 from several observing groups
306: is necessary from now until summer 2002 to study the behavior of
307: component B (if the variability is intrinsic, the same behavior
308: will appear in B component since $\sim$ JD2452000, April 2001,
309: until $\sim$ JD2452600, August 2002), and at least during 2003
310: to cover all the event.
311:
312:
313:
314: \begin{thebibliography}{}
315: \bibitem[Alcalde et al.(2002)]{alc02} Alcalde, D., et al. 2002,
316: \apj, in press
317: \bibitem[Angonin-Willaime \& Vanderriest(1995)]{1995tosm.conf..225A}
318: Angonin-Willaime, M.-C. \& Vanderriest, C.\ 1995, ASP Conf.
319: Ser. 71: IAU Colloq. 149: Tridimensional Optical Spectroscopic
320: Methods in Astrophysics, 225
321: \bibitem[Beskin \& Oknyanskij(1992)]{bes92} Beskin, G. M., \&
322: Oknyanskij, V. L. 1992, in Lectures Notes in Physics 406,
323: Gravitational Lenses, ed R. Kayzer, T. Schramm, \&
324: S. Refsdal, Springer--Verlag (Heidelberg), 67
325: \bibitem[Colley et al.(2002)]{col02} Colley, W. N. et al. 2002,
326: \apj, 565, 105
327: \bibitem[Florentin-Nielsen(1984)]{flo84} Florentin-Nielsen, R.
328: 1984, \aap, 138, 119
329: \bibitem[Gil-Merino et al.(2001)]{gil01} Gil-Merino, R., Goicoechea,
330: L. J., Serra-Ricart, M., Oscoz, A., Alcalde, D., \& Mediavilla,
331: E. 2001, \mnras, 322, 397
332: \bibitem[Gil-Merino et al.(1998)]{gil98} Gil-Merino, R., Goicoechea,
333: L. J., Serra-Ricart, M., Oscoz, A., Mediavilla, E., \&
334: Buitrago, J. 1998, \apss, 263, 47
335: \bibitem[Goicoechea et al.(1998)]{goi98} Goicoechea, L. J., Oscoz,
336: A., Mediavilla, E., Buitrago, J., \& Serra-Ricart, M. 1998,
337: \apj, 492, 74
338: \bibitem[Haarsma et al.(1999)]{haa99} Haarsma, D. B., Hewitt, J. N.,
339: Leh\'ar, J., \& Burke, B. F. 1999, \apj, 510, 64
340: \bibitem[Kundic et al.(1997)]{kun97} Kundic, T., et al. 1997, \apj,
341: 482, 75
342: \bibitem[Kundic et al.(1995)]{kun95} Kundic, T., et al. 1995, \apjl,
343: 455, L5
344: \bibitem[Leh\'ar, Hewitt, \& Roberts(1989)]{leh89} Leh\'ar, J.,
345: Hewitt, J. N., \& Roberts, D. H. 1989, in Gravitational Lenses,
346: ed J. M. Moran, J. N. Hewitt, \& K. Y. Lo, Reidel (Dordrecht), 84
347: \bibitem[Oscoz et al.(2001)]{osc01} Oscoz, A., et al. 2001, \apj,
348: 552, 81
349: \bibitem[Oscoz et al.(1997)]{osc97} Oscoz, A., Mediavilla, E.,
350: Goicoechea, L. J., Serra-Ricart, M., \& Buitrago, J. 1997,
351: \apjl, 479, L89
352: \bibitem[Oscoz et al.(1996)]{osc96} Oscoz, A., Serra-Ricart, M.,
353: Goicoechea, L. J., Mediavilla, E., \& Buitrago, J. 1996,
354: \apjl, 470, L19
355: \bibitem[Pelt et al.(1998)]{pel98} Pelt, J., Schild, R., Refsdal,
356: S., \& Stabell, R. 1998, \aap, 336, 829
357: \bibitem[Refsdal et al.(2000)]{ref00} Refsdal, S., Stabell, R.,
358: Pelt, J., \& Schild, R. 2000, \aap, 360, 10
359: \bibitem[Roberts et al.(1991)]{rob91} Roberts, D. H., Leh\'ar, J.,
360: Hewitt, J. N., \& Burke, B. F. 1991, \nat, 352, 43
361: \bibitem[Schild(1990)]{sch90} Schild, R. 1990, \aj, 100, 1771
362: \bibitem[Schild(1996)]{sch96} Schild, R. 1996, \apj, 464, 125
363: \bibitem[Schild \& Cholfin(1986)]{sch86} Schild, R. E., \&
364: Cholfin, B. 1986, \apj, 300, 209
365: \bibitem[Schmidt \& Wambsganss(1998)]{sch98} Schmidt, R., \&
366: Wambsganss, J. 1998, \aap, 335, 379
367: \bibitem[Serra-Ricart et al.(1999)]{ser99} Serra-Ricart, M.,
368: et al. 1999, \apj, 526, 40
369: \bibitem[Vanderriest et al.(1989)]{van89} Vanderriest, C.,
370: Schneider, J., Herpe, G., Chevreton, M., Moles, M., \&
371: Wl\'erick, F. 1989, \aap, 215, 1
372: \bibitem[Walsh, Carswell, \& Weymann(1979)]{wal79} Walsh, D.,
373: Carswell, R. F., \& Weymann, R. J. 1979, \nat, 279, 381
374: \bibitem[Wambsganss et al.(2000)]{wam00} Wambsganss, J., Schmidt,
375: R. W., Colley, W., Kundic, T., \& Turner, E. L. 2000, \aap,
376: 362, L37
377: \bibitem[Wozniak et al.(2000)]{woz00} Wozniak P. R., Udalski,
378: A., Szymanski, M., Kubiak, M., Pietrzynski, G., Soszynski, I.,
379: \& Zebrun, K. 2000, \apjl, 540, L65
380: \end{thebibliography}
381:
382:
383:
384: \clearpage
385:
386: \begin{deluxetable}{ccccc}
387: \tabletypesize{\scriptsize}
388: \tablecaption{Summary of the 2000 and 2001 observing campaigns of
389: Q0957+561.
390: See http://www.iac.es/project/quasar/mserra/meth.html for more
391: details on all the observations.\label{tbl-1}}
392: \tablewidth{0pt}
393: \tablehead{
394: \colhead{Campaign} & \colhead{Filter} & \colhead{Number of points}
395: & \colhead{Number of nights} & \colhead{Points removed}}
396: \startdata
397: &$R$&138&74&8\\
398: 10/1999-06/2000&&&&\\
399: &$V$&88&68&3\\
400: \hline\\
401: &$R$&92&76&5\\
402: 09/2000-06/2001&&&&\\
403: &$V$&83&77&2\\
404: \enddata
405: \end{deluxetable}
406:
407: \clearpage
408:
409: \begin{figure}
410: \caption{Upper panel: $R$ band real magnitude light curves of the
411: A (filled circles) and B (open squares) data from the Instituto
412: de Astrof\'\i sica de Canarias monitoring of Q0957+561. The B data
413: have been delayed by 422.6 days, but no magnitude shift has been
414: applied between A and B components. The ``real" years (mid date of
415: each observing campaign) appear in the top x-axis.
416: Second panel: $R$ band data averaged into 10-day bins in order to
417: reduce the scatter.
418: Third panel: same as in the upper panel but in $V$ band.
419: Lower panel: $V$ band data averaged into 10-day bins.
420: \label{fig1}}
421: \end{figure}
422:
423: \begin{figure}
424: \caption{Upper panel: The 2000 and 2001 campaigns of Q0957+561 in
425: the $R$ band averaged into 10-day bins. A (filled circles) and B
426: (not shifted in time, open squares). Note the almost monotonous
427: brightening of the A component between days 2500 and 3000. Lower
428: panel: Difference light curve of the two comparison stars (H and
429: D) to check their stability.\label{fig2}}
430: \end{figure}
431:
432: \begin{figure}
433: \caption{Differential light curve Q0957+561A (t) - Q0957+561B
434: (t-$\Delta$ t) of Q0957+561 from Schild's
435: and IAC's data. B component data is delayed by 422.6 days. The
436: data corresponding to each year have been averaged.
437: \label{fig3}}
438: \end{figure}
439:
440: \clearpage
441:
442: \begin{figure}
443: \plotone{f1.eps}
444: \end{figure}
445:
446: \clearpage
447:
448: \begin{figure}
449: \plotone{f2.eps}
450: \end{figure}
451:
452: \clearpage
453: \begin{figure}
454: \plotone{f3.eps}
455: \end{figure}
456:
457:
458:
459:
460:
461:
462: \end{document}
463:
464: