1: \documentstyle[emulateapj,psfig]{article}
2: %\documentstyle[11pt,aaspp4]{article}
3: %\documentstyle[11pt]{kluwer}
4: %\documentstyle[12pt,aasms4]{article}
5:
6: \begin{document}
7:
8: \slugcomment{Published in {\bf Science in China}, Series A, 43 (2000), 439}
9:
10: \title{Are there real orthogonal polarization modes in pulsar radio
11: emission?}
12:
13: \author{R. X. Xu, G. J. Qiao\\
14: CAS-PKU joint Beijing Astrophysical Center
15: and Department of Astronomy, \\
16: Peking University, Beijing 100781, China}
17:
18: \altaffiltext{1}{email: rxxu@bac.pku.edu.cn}
19:
20: \altaffiltext{2}{This work is supported by NSFC (No. 19803001),
21: by Doctoral Program Foungation of Institution of Higher
22: Education in China and by the Younth Foundation of PKU.}
23:
24: \begin{abstract}
25:
26: The Orthogonal Polarization Modes (OPM) have been reported observationally
27: (see e.g. [1]) and accepted widely by pulsar researchers (see e.g. [2,3]).
28: However, no acceptable theory can show the origin of the OPM, which becomes
29: a mystery in pulsar research field. Here a possible way to solve this mystery
30: is presented. We ask a question: Does there exist any real so-called OPM
31: in pulsar radiation? It is proposed in this paper that the `observed OPM'
32: in individual pulses could be the results of depolarization of pulsar
33: radiation and the observational uncertainties
34: originated from polarimeter in observation.
35: A possible way to check this idea is suggested. If the idea is verified,
36: the pulsar research would be influenced significantly in theory and in
37: observation.
38:
39: \end{abstract}
40:
41: \keywords{pulsars ---
42: polarization ---
43: radiation mechanisms}
44:
45: \section{Introduction}
46:
47: Pulsars are effective astrophysical laboratories for quantum theory and
48: gravitation theory. However,
49: how to reproduce the observed radiation theoretically is still
50: one of the most essential challenges in pulsar study. It
51: is well known that the polarization observations are very important to
52: provide much information about pulsar physics, but there are still many
53: troubles in explaining the polarization data.
54:
55: One of the difficulties in understanding pulsar polarization observations
56: is the polarization position angle jumps in mean (or integrated) pulses as
57: well as in individual pulses [1-3]. {\it For mean pulses}, it is generally
58: found that position angles would have discontinuities about 90$^o$ at some
59: longitudes where the linear polarization intensities are near zero (totally
60: depolarized). {\it For individual pulses}, the position angles would be
61: dispersed or have two $\sim 90^o$ separated distributions at some
62: observational longitude bins where the linear polarization percentages are
63: remarkably small. A famous example to display the polarization position
64: angle jumps in individual pulses is shown in Fig.1 for PSR B2020+28[1]. In
65: case `A' and `C', the two position angle distributions are clear, and the
66: linear polarization percentages are obviously low. In case `B', there is
67: only one position angle distribution, and the emission in each longitude
68: point is highly linearly polarized. It means that such orthogonally
69: distributed position angles are usually observed only in the part of the
70: profiles where the linear polarization is low [4]. A conclusion from the
71: observation is that, {\it both for mean profiles and for individual pulses,
72: the position angle jumps are related to low linear polarization (percentage)
73: at all time}.
74:
75: Based on above observational facts, there should be two possibilities
76: logically: one is that `position angle jump' causes `low linear
77: polarization', another is that `low linear polarization' causes `position
78: angle jump'. Many authors believe, without justification, that the position
79: angle jumps should be attribute to the appearance of OPM [1-3], in-coherent
80: superposition of the OPM is the origin of depolarization. However, why the
81: another possibility is impossible? We investigate this possibility in this
82: paper.
83:
84: Previously, Stinebring et al. [1] concluded that most of pulsar emission
85: occurs in one orthogonal mode or the other, which is called Orthogonal
86: Polarization Modes (OPM). At a given longitude, the plane of polarization
87: can be of two perpendicular or nearly perpendicular states. Which one can
88: operate was governed by some variability as yet not understood. The OPM can
89: explain many things: the sharp jumps of position angles, the depolarization
90: of linear as well as circular polarization due to the existence of both modes
91: at the same time [2]. However, there is no acceptable theory to reproduce such
92: orthogonal modes, which is called as `OPM problem'.
93:
94: For mean pulses, it is suggested that the depolarization and position angle
95: jumps might be attribute to the relative longitude shifts of pulsar beams
96: [5,6]. Such kind of longitude shifts of pulsar beams is natural in the
97: inverse Compton scattering (ICS) model [7]. For individual pulses, many
98: authors believe that there are two orthogonal modes at a given longitude.
99: Nevertheless, there may be in fact two possibilities to produce such
100: orthogonal modes. One is that the emission for a given frequency is emitted
101: at different heights, and another is that there is an unknown emission
102: mechanism to produce orthogonal modes at a same emission point [8]. The first
103: possibility has been studied already [6, 7]. But for the second one, no
104: acceptable theory has been found to produce such orthogonal modes hitherto
105: known. Thus, the `OPM problem' still confuses the pulsar world.
106:
107: We ask a question here: Are really such `orthogonal modes' the reason for
108: the `low linear polarization'? Otherwise, might the `low linear
109: polarization' be responsible for the observed `orthogonal modes'? Many
110: authors believe that the reduction in the percentage of linear polarization
111: is caused by in-coherent superposition of the OPM. Contrary to the above
112: idea, another possibility (i.e., the `low linear polarization' is the reason
113: for producing the observed `position angle jumps') is suggested in this
114: paper. {\it Our analysis and simulations show that, when the linear
115: polarization percentages are low enough, the position angles would be
116: distributed in two areas separated nearly ninety degrees}. A suggestion to
117: check this idea is presented.
118:
119: We show how `low linear polarization' causes so-called `position angle jumps'
120: in mean-pulses and in single-pulses in section 2 and 3, respectively. Some
121: troubles faced by OPM radiation mechanism are summed up in section 4. Finally,
122: conclusion and discussion are given in section 5.
123:
124: \section{Position angle jumps in mean pulses: depolarization?}
125:
126: Almost certainly, for observed mean-pulses, the smoothly changing position
127: angle curves will suddenly jump at some longitudes where the linear
128: polarization is highly depolarized. These facts can be understood under the
129: properties of Stokes parameters.
130: It could be verified mathematically that the position angle would jump
131: 90$^o$ when the line of sight travels across a {\it singular point} [5, 6]
132: where the linear polarization intensity is zero.
133:
134: The four Stokes parameters $\{I, Q, U, V\}$, from which one can obtain linear
135: polarization intensity $L=\sqrt{Q^2+U^2}$ and position angle $\chi$ (see
136: equ.(1) below), are functions of observational longitude $\phi$.
137: For the sake of simplicity, we let $V=0$, as the linear polarization is
138: focused on here. At
139: a singular point ($\phi=\phi_{\rm s}$), $L=0$ means
140: $
141: Q(\phi_{\rm s}) = 0,
142: U(\phi_{\rm s}) = 0.
143: $
144: Expanding $Q$ and $U$ near singular point, we come to
145: $$
146: Q(\phi_{\rm s} + \Delta) = {\partial Q\over \partial \phi}\Delta +
147: {1\over 2} {\partial^2 Q\over \partial \phi^2}\Delta^2 + {1\over 3!}
148: {\partial^3 Q\over \partial \phi^3}\Delta^3 + \cdot \cdot \cdot,
149: $$
150: $$
151: U(\phi_{\rm s} + \Delta) = {\partial U\over \partial \phi}\Delta +
152: {1\over 2} {\partial^2 U\over \partial \phi^2}\Delta^2 + {1\over 3!}
153: {\partial^3 U\over \partial \phi^3}\Delta^3 + \cdot \cdot \cdot.
154: $$
155: Assuming ${1\over q!} {\partial^q U\over \partial \phi^q}\Delta^q$ and
156: ${1\over u!} {\partial^u U\over \partial \phi^u}\Delta^u$ are the lowest
157: non-zero
158: power terms of $Q$ and $U$, respectively, and $\nu = {\rm min}[q, u]$, one
159: could find that $\chi(\phi_{\rm s} + \Delta) - \chi(\phi_{\rm s} - \Delta)$
160: should be $\pm 90^o$ as long as $\nu$ is an odd number
161: %
162: \footnote{
163: One might easily obtain this conclusion by inspecting the position
164: angles in Q-U plane (two dimension Poincare sphere for $V=0$).
165: }
166: %
167: . It is very possible that $\nu=1$, thus, position angle {\it naturally}
168: jumps $90^o$
169: if $L=0$ [6]. Therefore, {\it the reason that position angle jumps in
170: integrated profiles might be why the beamed radiation is depolarized}.
171: Depolarization should be the cause of position angle jumps in mean-pulses.
172:
173: There are many ways to cause depolarization. First of all, depolarization
174: may have an intrinsic origin. As emission beams are formed in
175: different heights, and each of them has different position angle,
176: depolarization must take place by incoherent superposition of such emission
177: beams. In the ICS model [7], different emission beams are formed in
178: different heights, hence, the retardation and aberration effects could make
179: the apparent emission beams be superposed incoherently [5,6]. Secondly,
180: depolarization might be originated from propagation process, such as the
181: scattering by interstellar medium or magnetospheric plasma [4], and the
182: propagating properties of different radiation modes in plasma. The third
183: way might be the result of observational effect. Since the Stokes parameters
184: are added from many frequency channels after de-dispersion, the emissions
185: in each frequency channel are incoherent superposed. Such kind of treatment
186: in observation should also depolarize the original radiation.
187:
188: \section{Position angle jumps in single-pulses: observational
189: uncertainty?}
190:
191: There are many factors to reduce the precision of observational results,
192: such as noises from the observational system and the sky background. Usually,
193: we put thresholds for the total intensity ($I$) and the linear polarization
194: intensity ($L$) in each longitude bin in order to exclude fake polarization
195: due to the observational error. For example, we select observational data
196: whose $I$ and $L$ are greater than 5 to 10 times of off-pulse rms. However,
197: as will be discussed in this section, some fake polarization data, which
198: may responsible for the observed `position angle jumps' in individual pulses
199: [5], do survive from such selection.
200:
201: Some observational uncertainties can cause the observed position angles to
202: `jump' in individual pulses, such as the `error transference' (section 3.1),
203: the unequal rms of Stokes parameter $Q$ and $U$ (section 3.2), and the fake
204: linear polarization (section 3.3). All these uncertainties can bring wrong
205: polarization information.
206:
207: \subsection{Position angle `jumps' in individual pulses due to the error
208: transference}
209:
210: If $x$ is a random number, then the function $y=f(x)$ is also random. The
211: random distribution of $y$ is known as long as the distribution of $x$ is
212: given. For example, let the distribution function of $x$ as a gaussian
213: distribution, with the expectation value $x_0$. The distribution function
214: of $y$ depends on the function $f(x)$. If $f(x)$ is a monotonous
215: function near $x_0$, the distribution function of $y$ is approximately a
216: gaussian. Whereas, if $f(x)$ is a very complex function near
217: $x_0$, the distribution function of $y$ is also complicated.
218:
219: It is known that the position angle $\chi$ is a function of the Stokes
220: parameters $Q$ and $U$
221: %
222: \footnote{
223: Usually $\chi = {1\over 2}\tan^{-1}{U\over Q}$, depending on the signs of
224: $Q$ and $U$. From
225: $
226: \left\{ \begin{array}{lll}
227: \sin 2\chi & = & U/L\\
228: \cos 2\chi & = & Q/L
229: \end{array} \right.
230: $
231: , one can get a general expression for $\chi$ , where the value region of
232: $\chi$ is from $0^o$ to $180^o$.
233: }
234: %
235: $$
236: \chi={1\over 2} [{\rm sign}U \cos^{-1}{Q\over L} + \pi (1-{\rm sign}U)],
237: \eqno(1)
238: $$
239: which is `singular'(unusual) near $L=\sqrt{Q^2+U^2}=0$. Here ${\rm sign}U
240: = +1$ (${\rm sign}U = -1$) if ${\rm sign}U > 0$ (${\rm sign}U < 0$). This
241: property of singularity of $\chi$ would cause two error distribution peaks
242: (see Fig.2), which will be discussed later in section 3.2. In a word, the
243: observational uncertainty (error) of $Q$ and $U$ could bring an error
244: distribution of $\chi$ in two regions with 90$^o$ separation by the error
245: transference effect. Such observational results might be mistakenly
246: considered as position angle `jumps' in a real beamed radiation.
247:
248: \subsection{Position angle jumps due to unequal errors of $Q$ and $U$}
249:
250: Usually, the rms of the Stokes parameter $Q$ and that of $U$ are not equal
251: for an astronomical polarimetry, i.e., $\sigma_{\rm Q} \neq \sigma_{\rm U}$.
252: The difference between $\sigma_{\rm Q}$ and $\sigma_{\rm U}$ can be as large
253: as several percentages in observations. The observed linear polarization
254: position angles could `jump' during different observing time as long as
255: $\sigma_{\rm Q} \neq \sigma_{\rm U}$. As demonstrated in Fig.3, we see that
256: $\chi_B - \chi_A$ is about $0^o$, and $\chi_B - \chi_C$ is about $90^o$.
257:
258: The reason that $\sigma_{\rm Q}\neq \sigma_{\rm U}$ might be diversity. For
259: example, for some polarimetry, the Stokes parameters $Q=S_0-S_{90}$ and
260: $U=S_{45}-S_{135}$ are computed from the hybrid networks' out-put signals.
261: Here $S_0$ and $S_{90}$ are the observed intensity from two orthogonal
262: dipole antenna, $S_{45}$ and $S_{135}$ are the intensity received from a
263: system which has been rotated 45 degrees.
264: For a dipole antenna, $S_0$ and $S_{90}$
265: are obtained directly. However, the $S_{45}$ and $S_{135}$ are yielded
266: through a turnstile junction where the phase mis-alignment can make the
267: rms of $S_{45}$ (and $S_{135}$) larger than the rms of $S_0$ (and $S_{90}$).
268: So, the rms of $Q$ and that of $U$ can not be equal because of the imperfection
269: of the turnstile junction.
270:
271: Simulations of this kind of polarimetry are given in Fig. 4
272: (see the Appendix), from which we see that
273:
274: \begin{itemize}
275:
276: \item The observed percentages of polarization $\Pi'$ are much greater than
277: the true percentage $\Pi$, if $\Pi$ is small enough. Observed linear
278: polarization may be larger than that of the true value.
279:
280: \item The position angle `jumps' takes place when linear polarization
281: percentage $\Pi_l \leq 0.1\%$. When $\Pi_l \geq 1\%$, there is few
282: possibilities to make position angle jump.
283:
284: \end{itemize}
285:
286: Because the observational uncertainty is of random, position angle jumps
287: that come of this kind of errors discussed above can be avoided by more time
288: observation. It is almost impossible that position angle jump due to
289: observational uncertainty appears in integrated profiles.
290:
291: \subsection{The observational noise responsible for fake polarization}
292:
293: For a telescope with an effective detection area $A$, a frequency bandwidth
294: $\delta \nu$, a time constant $\tau$, and a systematical noise temperature
295: $T_{\rm sys}$, then the systematical noise flux $S_{\rm sys}$ is
296: $
297: S_{\rm sys} = k T_{\rm sys}/A
298: $,
299: where $k=1.38\times 10^{-23}{\rm JK}^{-1}$ is the Boltzmann's constant, and
300: the off-pulse rms $\sigma_{\rm off}$ is $$
301: \sigma_{\rm off} = {S_{\rm sys} \over \sqrt{\delta \nu \ \tau}}.
302: $$
303: Whereas there is a signal with intensity flux $S_{\rm i}$, the rms of the
304: signal flux is $\sigma_{\rm on}$ (the on-pulse rms)
305: $$
306: \sigma_{\rm on} = {S_{\rm sys} + S_i \over \sqrt{\delta \nu \ \tau}},
307: $$
308: where $S_{\rm i}$ can be one of the out-put signals from the hybrid networks,
309: such as $S_0$, $S_{90}$, $S_{45}$, $S_{135}$, $S_R$, $S_L$, which are
310: correspondent to the linearly polarized components of the input signals at
311: position angle $0^o$, $90^o$, $45^o$, and $135^o$, and to the right-hand
312: and left-hand circularly polarized components.
313:
314: Because $S_{\rm i} = (S_{\rm sys} + S_{\rm i}) - S_{\rm sys}$, the rms of
315: $S_{\rm i}$ is
316: $
317: \sigma_{\rm i} = \sqrt{\sigma_{\rm off}^2 + \sigma_{\rm on}^2}
318: $.
319: If we assume that $S_{\rm sys}$ is accurate enough (i.e., we have enough
320: time to measure $S_{\rm sys}$), then
321: $
322: \sigma_{\rm i} \approx \sigma_{\rm on}
323: $,
324: which will be used in the following discussion.
325:
326: If we let $A = 40^2 \pi$ square meters, $\delta \nu = 10$MHz, $\tau=0.3$ms,
327: and $T_{\rm sys} = 40$K, then $\sigma_{\rm off}$ is about 0.2 Jy. For
328: $S_i=30$Jy[9], then $\sigma_i=0.8 {\rm Jy}$ which is more than three times
329: that of $\sigma_{\rm off}$, and the fake linear polarization percentage
330: could be as large as $3\%$. Thus, some data, whose linear polarization
331: is originated from such uncertainty, {\it can also exceed} the observational
332: threshold level with an un-negligible possibility.
333:
334: \section{ Are real orthogonal polarization modes in pulsar radiation?}
335:
336: We might be in a dilemma if there are really so-called orthogonal
337: polarization modes in pulsars' beams. First, how does the OPM radiative
338: mechanism produce? No reasonable theory has been appeared in literature.
339: Furthermore, the two orthogonal modes should be in-coherent, which makes
340: the OPM more difficult to be set up. The suggestion for observed `orthogonal
341: modes' in mean-pulses by Xu at al. [6] is not a real one. (In their
342: calculations, two components are emitted in-coherently from different
343: regions.) Furthermore, If the two modes are coherent, the total radiation
344: is elliptically polarized, thus no position angle `jump' appears.
345:
346: Secondly, why haven't we seen that single-pulses are highly polarized, but
347: the position angle distribution is till separated by 90$^o$ (like Fig.5)?
348: Individual pulses, which could be highly polarized, are generally
349: conjectured to be from single radiation elements. Since two orthogonal
350: modes are incoherent, a radiation element might emit only one of the OPR
351: modes at one time. Therefore, it is possible to observe some highly linearly
352: polarized individual pulses, while the position angles of which are 90$^o$
353: separated. Unfortunately, observation result similar to that of Fig.5 has
354: {\it never} been found.
355:
356: Thirdly, how to explain the non-orthogonal separation of position angles in
357: the regime of OPM? Non-orthogonal emission modes have proverbially been
358: found in observation [1]. These facts are rigorous for anyone to theorize
359: OPM models.
360:
361: \section{Conclusion and discussion}
362:
363: From analysis above some conclusions and discussions are reached:
364:
365: 1. Another possible way to solve the problem of position angle `{\it jumps}'
366: in pulsars' beamed radio emission was proposed. There might be {\it no real}
367: `orthogonal polarization modes' in the emission at all.
368:
369: 2. Position angle jumps due to the observational uncertainties could appear
370: in observed individual pulses when the linear polarization {\it percentages}
371: are small (not only the linear polarization intensity to be small). At least
372: part of the observed position angle jumps in individual pulses and
373: mean-pulses can be explained by depolarization and observational
374: uncertainty.
375:
376: For a real pulsar, we must put together these two possible factors to
377: investigate the position angle variation in the individual pulses as well
378: as in the integrated pulses. For example, observational uncertainty might
379: be the main reason of position angle separation near point `A' in Fig.1.
380: Nevertheless, near point `C', orthogonal and non-orthogonal separations are
381: clear, which might be the result of the relative longitude
382: shifts of pulsar beams
383: [6] and the observational uncertainties.
384:
385: Rathnsree \& Rankin[10] pointed out that, for PSR B1929+10, lower degree
386: of polarization is seen simultaneously with the presence of `orthogonal' modes
387: whereas the polarized power is not seen to be highly correlated with the
388: position angle flip. Also, they have got dynamic pictures of the orthogonal
389: polarization mode changes for PSR B2110+27 at 430 MHz, and they found the
390: transition from the dominant mode to the other orthogonal one
391: and back are {\it rapid}.
392: Most of the transition is achieved over time scales of a individual
393: period, and the change of modes does not seem to be any periodicity in time
394: evolution (like a {\it stochastic} process). All this observational facts
395: have the properties of observational uncertainties discussed in section 3.
396:
397: For PSR B0525+21[1] at 1404 MHz, the position angle sweep is S-shaped
398: in the averaged profile. The polarization position angles in individual
399: pulses are also an `S' shape distribution, but two weak patches of
400: `orthogonal modes' on the outside edges of the profile where
401: more individual pulses have very small percentages of linear polarization.
402: These two patches' appearance should come of the observational
403: uncertainty according to our analysis.
404: This statement {\it can be checked} by
405: future expriment of observation.
406:
407: 3. From simulations, we see that the jumped position angles are distributed
408: near $45^o$ and $135^o$. In fact, there are observational data which does
409: show that the position angles distribute near $45^o$ and $135^o$ in the
410: scatter plots, such as Fig.26, Fig.37 in Stinebring et al. [1]. If the rms
411: of $U$ less than that of $Q$, the jumped position angles should be near $0^o$
412: and $90^o$, like PSR 0525+21 (Fig.2 in [1]) in observations. There are
413: observational data where the jumped position angles are not distributed near
414: $0^o$, $45^o$, $90^o$, and $135^o$, which could be intrinsic in polarimetry
415: or resultant from the longitude shift of beam phases [6].
416:
417: 4. The idea suggested in this paper can be checked experimentally. We can
418: input the polarimetry a simulated lower polarized and pulsed signal to see
419: if two $90^o$ separated position angle distribution can appear in the output.
420: If such distribution can also be obtained, the OPM in pulsar emission should
421: be doubted.
422:
423: 5. If OPM does not exist in pulsar radio emission, we should develop our
424: instruments to avoid observational uncertainties.
425:
426: \acknowledgements{We thank our pulsar group for discussions.}
427:
428: \begin{center}
429: \Large{{\bf Appendix} A Simulation}
430: \end{center}
431:
432: For the kind of polarimetry discussed in section 3, let's study a partially
433: polarized wave, with the total intensity $I$, the un-polarized intensity
434: $I^{unp}$, the percentage of linear polarization $\Pi_l$, the percentage
435: of circular polarization $\Pi_c$, and the position angle $\chi$. If we
436: measure this wave by a telescope with an effective area $A$, systematic noise
437: temperature $S_{\it sys}$, a bandwidth $\delta\nu$, a time constant $\tau$,
438: the angle between two the dipole antenna (parasitism polarization) $\alpha$
439: (whose expectation value is $\pi\over 2$, $\delta\alpha = \alpha -{\pi\over
440: 2}$, the rms of $\delta\alpha$ is $\sigma_{\alpha}$), and the phase
441: misalignment $\delta\phi$ (the rms of $\delta\phi$ is $\sigma_{\phi}$),
442: then the six intensity for the Stokes parameters can be deduced as
443: $$
444: \begin{array}{lll}
445: S_0 & = & {1\over 2}X^2 + {1\over 2}I^{unp},\\
446: S_{90} & = & {1 \over 2}(Y^2 \cos^2 \delta\alpha +\\
447: & & X^2 \sin^2 \delta\alpha +2XY\cos\delta\sin\delta\alpha\cos\delta\alpha) +\\
448: & & {1\over 2}I^{unp},\\
449: s_{45} & = & {1\over 4} [X^2 + Y^2\cos^2\delta\alpha +\\
450: & & X^2\sin^2\delta\alpha +\\
451: & & 2XY\cos(\delta-\delta\phi)\cos\delta\alpha + \\
452: & & 2XY\sin\delta\alpha\cos\delta\alpha\cos\delta +\\
453: & & 2X^2\sin\delta\alpha\cos\delta\phi] + {1\over 2}I^{unp},\\
454: S_{135} & = & {1\over 4} [X^2 + Y^2\cos^2\delta\alpha +\\
455: & & X^2\sin^2\delta\alpha +\\
456: & & 2XY\cos(\delta-\delta\phi-\pi)\cos\delta\alpha+\\
457: & & 2XY\sin\delta\alpha\cos\delta\alpha\cos\delta +\\
458: & & 2X^2\sin\delta\alpha\cos(\delta\phi+\pi)] + {1\over 2}I^{unp},\\
459: S_R & = &{1\over 4} [X^2 + Y^2\cos^2\delta\alpha +\\
460: & & X^2\sin^2\delta\alpha +\\
461: & & 2XY\cos(\delta+\delta\phi+{3\over 2}\pi)\cos\delta\alpha+\\
462: & & 2XY\sin\delta\alpha\cos\delta\alpha\cos\delta +\\
463: & & 2X^2\sin\delta\alpha\cos(\delta\phi+{3\over 2}\pi)]+{1\over
464: 2}I^{unp},\\
465: S_L & = &{1\over 4} [X^2 + Y^2\cos^2\delta\alpha + X^2\sin^2\delta\alpha
466: +\\
467: & & 2XY\cos(\delta+\delta\phi+{\pi\over 2})\cos\delta\alpha+\\
468: & & 2XY\sin\delta\alpha\cos\delta\alpha\cos\delta +\\
469: & & 2X^2\sin\delta\alpha\cos(\delta\phi+{\pi\over 2})]+{1\over
470: 2}I^{unp},
471: \end{array}
472: $$
473: here,
474: $$
475: \begin{array}{lll}
476: \delta & = & \tan^{-1}{\Pi_c\over \Pi_l\sin2\chi},\\ X & = & \sqrt{I} \times
477: \sqrt{\sqrt{\Pi^2_l + \Pi^2_c} + \Pi_l\cos2\chi},\\
478: Y & = & \sqrt{I} \times \sqrt{\sqrt{\Pi^2_l + \Pi^2_c} - \Pi_l\cos2\chi}.\\
479: \end{array}
480: $$
481: The observed Stokes parameters should be $$
482: \begin{array}{lll}
483: I' & = & S_0 + S_{90},\\
484: Q' & = & S_0 - S_{90},\\
485: U' & = & 2S_{45} - I',\\
486: V' & = & 2S_R - I'.\\
487: \end{array}
488: $$
489: So that, the observed linear polarization intensity $L'$, the observed
490: percentages of linear polarization $\Pi'_l$, the observed percentages of
491: circular polarization $\Pi'_c$, and the observed linear polarization
492: position angle $\chi'$ would be $$
493: \begin{array}{lll}
494: L' & = & \sqrt{Q'^2+U'^2},\\
495: \Pi'_l & = & {L'\over I'},\\
496: \Pi'_c & = & {V'\over I'},\\
497: \chi' & = & {1\over 2} [{\rm sign}U' \cos^{-1}{Q'\over L'} + \pi (1-{\rm
498: sign}U')].
499: \end{array}
500: $$
501:
502: Considering this kind of observational uncertainty, we have obtained some
503: simulation results to show the position angle `jumps' in individual pulse
504: observations. One of the simulations is shown in Fig.4, where we have chosen
505: $$
506: \begin{array}{lll}
507: I & = & 50\;{\rm \;Jy},\\
508: \sigma_{\alpha} & = & 5^o,\\
509: \sigma_{\phi} & = & 5^o,\\
510: A & = & \pi\;40^2 \;{\rm \;square\;meters},\\ S_{sys} & = & 40\;{\rm \;K},\\
511: \delta\nu & = & 10\;{\rm \;MHz},\\
512: \tau & = & 0.3\;{\rm \;ms},\\
513: \chi & = & 45^o,\\
514: \Pi_l & = & \Pi_c=\Pi=0.01\%.
515: \end{array}
516: $$
517: The scatter plots in Fig.4 are resemble to observations, especially the
518: $L'-\chi'$ plot, which is similar to the observed linear polarization versus
519: position angle scatter plots for position angle jumps at a fixed longitude.
520: Based on this simulation and other simulations for different parameters,
521: we found that the position angles `jump' if $\Pi_l \leq 0.1\%$, whereas,
522: there is few possibilities of position angle jump if $\Pi_l \geq 1\%$.
523:
524: \begin{references}
525:
526: \reference{}
527: [1] Stinebring D.R., Cordes J.M. et al., Pulsar Polarization Fluctuation
528: I. 1404 MHz Statical Summaries, {\it ApJS}, 1984, 55:247$\sim $277
529:
530: \reference{}
531: [2] Mckinnon, M.M., Stinebring D.R., A Statistical Model for the Orthogonal
532: Modes of Polarization in Pulsars Radio Emission, {\it ApJ}, 1998,
533: 502:883$\sim$897
534:
535: \reference{}
536: [3] Gangadhara, R.T., Orthogonal Polarization Mode Phenomenon in Pulsars,
537: {\it A\&A}, 1997, 327:155$\sim $166
538:
539: \reference{}
540: [4] Hankins, T.H., Microstructure: A Review, In: Johnston, S.
541: et al. eds. Pulsars: Problems and Progress, ASPC Vol.105, 1996, 197$\sim
542: $202
543:
544: \reference{}
545: [5] Xu, R.X., Qiao, G.J., Two Possibilities of Observed Position Angle Jumps
546: in Pulsar Radio Emission, {\it In}: Cheng K.S. (ed), Proc. of the 21st
547: Chinese Astronomy Conf., World Scientific, 1997, 197$\sim $300
548:
549: \reference{}
550: [6] Xu, R.X., Qiao, G.J., Han, J.L., Depolarization and position angle jumps
551: due to relative longitude shift of pulsar beams, {\it A\&A}, 1997,
552: 323:395$\sim $398
553:
554: \reference{}
555: [7] Qiao,G.J., and Lin,W.P.,
556: An inverse Compton scattering(ICS) model of pulsar emission I. Core and
557: conal emission beams,{\it A\&A}, 1998, 333,172$\sim $180
558:
559: \reference{}
560: [8] Manchester,R.N. and Taylor, J.H., 1977, Pulsars, USA: W.H.Freemen and
561: Company, p.202
562:
563: \reference{}
564: [9] Backer, D.C, Rankin, J.M., Statistical Summaries of Polarized Pulsar
565: Radiation, {\it ApJS}, 1980, 42:143$\sim $173
566:
567: \reference{}
568: [10] Rathnsree, N., Rakin, J.M., On `Orthogonal' Polarization Modes in
569: Pulsars: A Study of PSR 2120+27, {\it Astro. Lett. \& Commu.}, 1996,
570: 35:281$\sim $ 288
571:
572: \end{references}
573:
574: {\bf Figures:}
575:
576: \vspace{0.2cm}
577: \centerline{}
578: \centerline{\psfig{file=F1.eps,width=20cm,height=15cm}}
579: \figcaption{
580: Polarization distribution of PSR B202+28. The top plot shows the
581: position angle distribution; the middle plot shows the linear polarization
582: percentage of individual pulses; the lowest plot gives the integral pulse
583: profile. The observation is done by Stinebring et al. [1].
584: \label{Fig.1} }
585:
586: \centerline{}
587: \centerline{\psfig{file=F2.eps,width=15cm,height=10cm}}
588: \figcaption{
589: A sketch picture for the possibility of the linear polarization
590: position angle `jump' due to the error transferring from $Q$ and $U$.
591: \label{Fig.2} }
592:
593: \centerline{}
594: \centerline{\psfig{file=F3.eps,width=10cm,height=12cm}}
595: \figcaption{
596: A demonstration of position angle `jumps' which come from
597: observational uncertainty. Position angles $\chi_A$, $\chi_B$ and $\chi_C$
598: are for points A, B, and C, respectively.
599: \label{Fig.3} }
600:
601: \centerline{}
602: \centerline{\psfig{file=F4.eps,width=20cm,height=14cm}}
603: \figcaption{
604: The simulated scatter plots of the observed polarization degrees of
605: linear polarization $\Pi'_l$, and circular polarization $\Pi'_c$. The lower
606: one is the $L'-\chi'$ (linear polarization intensity vs. the position angle)
607: plot, where the solid horizontal line shows a possible threshold level for
608: linear polarization. The parameters in the simulation are given in the text.
609: \label{Fig.4} }
610:
611: \centerline{}
612: \centerline{\psfig{file=F5.eps,width=7cm,height=7cm}}
613: \figcaption{
614: A possible observational result predicted by OPM models.
615: In the figure, the lowest points present the linear polarization
616: persentages of individual pulses; the upper two distributions are
617: for the position angles.
618: \label{Fig.5} }
619:
620:
621: \end{document}