0704.0303/ms.tex
1: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2: % NIM preprint of APF system.                                                 %
3: % --------------------------------------------------------------------------- %
4: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
5: 
6: \documentclass{elsart}
7: % \documentclass[reviewcopy]{elsart}
8: 
9: \usepackage{amsmath}                   % Use AMS math environments.
10: \usepackage{amssymb}                   % Use AMS symbols and fonts.
11: \usepackage{graphicx}                  % Graphics include environment.
12: \usepackage{subfigure}
13: \usepackage{rotate}
14: %\usepackage{lineno}                    % Package to number all lines.
15: %\usepackage{setspace}
16: % \setlength{\topmargin}{0.5in}          % Required: else margin is negative.
17: \bibliographystyle{elsart-num}
18: 
19: \begin{document}
20: 
21: %\linenumbers                           % Comment out for no line numbering.
22: 
23: 
24: % -----------------------------------------------------------------------------
25: \begin{frontmatter}
26: \journal{Astroparticle Physics}
27: 
28: \title{
29: Measurement of the Aerosol Phase Function at the 
30: Pierre Auger Observatory
31: }
32: 
33: \author[Columbia]{S.Y.~BenZvi},
34: \author[Columbia]{B.M.~Connolly},
35: \author[UNM]{J.A.J.~Matthews},
36: \author[Columbia]{M.~Prouza},
37: \author[Columbia,Carnegie]{E.F.~Visbal}, and 
38: \author[Columbia]{S.~Westerhoff}
39: 
40: \address[Columbia]{Columbia University, Department of Physics and Nevis Laboratories,
41:                    538 West $\it 120^{th}$ Street, New York, NY 10027, USA}
42: \address[UNM]{University of New Mexico, Department of Physics and Astronomy, Albuquerque, 
43:               NM 87131, USA}
44: \address[Carnegie]{Carnegie Mellon University, Department of Physics, Pittsburgh, 
45:                    PA 15213, USA}
46: 
47: % -----------------------------------------------------------------------------
48: 
49: \begin{abstract}
50: Air fluorescence detectors measure the energy of ultra-high energy cosmic rays by collecting
51: fluorescence light emitted from nitrogen molecules along the extensive air shower cascade.
52: To ensure a reliable energy determination, the light signal needs to be corrected for
53: atmospheric effects, which not only attenuate the signal, but also produce a non-negligible 
54: background component due to scattered Cherenkov light and multiple-scattered light.  The 
55: correction requires regular measurements of the aerosol attenuation length and the aerosol
56: phase function, defined as the probability of light scattered in a given direction.  At 
57: the Pierre Auger Observatory in Malarg\"ue, Argentina, the phase function is measured on an 
58: hourly basis using two Aerosol Phase Function (APF) light sources.  These sources direct a UV 
59: light beam across the field of view of the fluorescence detectors; the phase function can be 
60: extracted from the image of the shots in the fluorescence detector cameras.  This paper 
61: describes the design, current status, standard operation procedure, and performance of the 
62: APF system at the Pierre Auger Observatory.
63: \end{abstract}
64: 
65: 
66: % PACS codes here, in the form: \PACS code \sep code
67: \begin{keyword} 
68: Ultra-high energy cosmic rays; air fluorescence detectors; atmospheric monitoring; 
69: aerosol phase function
70: \PACS 42.68.-w \sep 42.68.Jg \sep 92.60.Mt \sep 92.60.Sz \sep 96.50.sd
71: \end{keyword}
72: 
73: \end{frontmatter}
74: 
75: 
76: % -----------------------------------------------------------------------------
77: \section{Introduction}\label{sec:intro}
78: % -----------------------------------------------------------------------------
79: 
80: The Pierre Auger Observatory in Malarg\"ue, Argentina, is designed to study the
81: origin of ultra-high energy cosmic rays with energies above $10^{18}$\,eV.  While 
82: still under construction, scientific data taking began in 2004, and first results
83: have been published~\cite{Mantsch:2005,Abraham:2006,Aglietta:2006}.
84: 
85: The Pierre Auger Observatory is a hybrid detector that combines two techniques 
86: traditionally used to measure cosmic ray air showers: surface particle 
87: detection and air fluorescence detection.  Both detector types
88: measure the cosmic ray primary indirectly, using the Earth's atmosphere as part of
89: the detector medium.  When the primary particle enters the atmosphere, it interacts
90: with air molecules, initiating a cascade of secondary particles, the so-called
91: extensive air shower.  Surface detectors in the form of ground arrays sample the 
92: shower front as it impacts the ground, whereas air fluorescence detectors make use 
93: of the fact that the particles in the air shower excite nitrogen molecules in the air, 
94: causing UV fluorescence.  Using photomultiplier cameras to record air shower UV
95: emission, we can observe showers as they develop through the atmosphere and obtain 
96: a nearly calorimetric estimate of the shower energy.
97: 
98: Upon completion, the surface detector (SD) array of the Pierre Auger Observatory 
99: will comprise 1600 water Cherenkov detector tanks, deployed in a hexagonal grid 
100: over an area of $3000~\mathrm{km}^{2}$, and four fluorescence detector (FD) stations 
101: overlooking the SD from the periphery.  An advantage of combining both detector 
102: types at the same site is the possibility to cross-calibrate.  Based on the subset 
103: of events seen with both detectors, the nearly calorimetric information of the FD 
104: provides the energy calibration of the SD.  
105: 
106: For the calibration to be meaningful, the properties of the calorimeter, {\it i.e.} 
107: the atmosphere, must be well-known.  At the Pierre Auger Observatory, this is achieved 
108: by an extensive program to monitor the atmosphere within the overall FD aperture 
109: and measure atmospheric attenuation and scattering properties in the 300 to 400~nm 
110: wavelength band recorded by the FDs~\cite{Cester:2005,BenZvi:2007,Fick:2006}.  
111: 
112: Two primary forms of atmospheric light scattering need to be considered: molecular, 
113: or Rayleigh, scattering, mainly due to nitrogen and oxygen molecules; and aerosol 
114: scattering due to airborne particulates.  The 
115: angular distribution of scattered light in both types of scattering may be 
116: described by a phase function $P(\theta)$, defined as the probability per unit 
117: solid angle of scattering through an angle $\theta$.
118: 
119: Rayleigh scattering allows for an analytical treatment, and assuming isotropic
120: scattering, the Rayleigh phase function has the well known $1+\cos^{2}\theta$ 
121: angular dependence.  Matters are more complicated for
122: aerosols, because the scattering cross section depends on the size
123: distribution and shape of the scatterers.  Forward scattering typically
124: dominates in this case, but the fraction of forward-scattered light
125: varies strongly with aerosol type.  Moreover, a rigorous analytical
126: treatment is not possible, though the literature gives various
127: approximations.  For example, if one assumes spherical particles with a
128: known or estimated size distribution, then aerosol scattering can be
129: described analytically using Mie theory~\cite{Mie:1908}.  In practice, however,
130: aerosols vary a great deal in size and shape, and the aerosol content of
131: the atmosphere changes on short time scales as wind lifts up dust,
132: weather fronts pass through, or rain removes dust from the atmosphere.
133: 
134: The FD reconstruction of the primary cosmic ray particle energy must account 
135: not only for light that is ``lost'' between the shower and the camera due to 
136: scattering, but also for direct and indirect Cherenkov light contributing to 
137: the FD signal.  The amount of Cherenkov light seen by the FDs depends on the 
138: viewing angle, {\it i.e.} the angle between the shower axis and the FD line 
139: of sight, and can be calculated once the geometry of the air shower is determined.  
140: At small viewing angles, direct Cherenkov light dominates, while at viewing angles 
141: greater than $\sim 20^{\circ}$, the FDs detect mainly ``indirect'' Cherenkov 
142: light scattered into the FD field of view.  To calculate this scattered component, 
143: the aerosol phase function needs to be known.  Finally, a small multiple scattering 
144: component also adds to the contamination of the fluorescence light and must be 
145: removed~\cite{Roberts:2005}.
146: 
147: The Aerosol Phase Function (APF) light sources~\cite{Matthews:2001,Matthews:2003}, 
148: in conjunction with the fluorescence detectors at the Pierre Auger Observatory, are 
149: designed to measure the aerosol phase function on an hourly basis during FD data taking.  
150: The APF light sources direct a near-horizontal pulsed light beam across the field of 
151: view of a nearby FD.  The aerosol phase function can then be reconstructed from the 
152: intensity of the light observed by the FD cameras as a function of scattering angle.
153: Since the FD telescopes cover about $180^{\circ}$ in azimuth, the aerosol phase 
154: function is measured over a wide range of scattering angles.  
155: 
156: Currently, APF light sources are installed and operating at two of the FDs.
157: With their ability to measure the {\it angular 
158: distribution} of the scattered light, the APF light sources are meant to complement 
159: other atmospheric monitoring tools at the Auger site which measure the optical 
160: depth, and therefore the {\it amount} of attenuation due to aerosols.
161: 
162: This paper describes the design and performance of the APF light sources.
163: It is structured as follows.  Section 2 gives a description
164: of the APF facilities.  Section 3 describes how the aerosol phase function 
165: is determined from the APF data.  In Section 4, we show first results for
166: data taken between June and December 2006.  Section 5 summarizes the paper.
167:  
168: \begin{figure}[t]
169: \centering
170: \includegraphics[scale=.5]{fig1.eps}
171: \caption{\it Schematic layout of the Pierre Auger Observatory.  The shaded area 
172: indicates the shape and size of the surface detector area.  The fluorescence detectors
173: are placed at the periphery of the surface detector array.  The field of view of
174: the 6 bays of each fluorescence detector (FD) is indicated by the lines.
175: From the Central Laser Facility (CLF)~\cite{Fick:2006} in the center of the surface detector
176: array, a pulsed UV laser beam is directed into the sky, providing another test
177: beam which can be observed by the FDs.}
178: \label{auger_scheme}
179: \vskip1cm
180: \end{figure}
181: 
182: 
183: % -----------------------------------------------------------------------------
184: \section{APF Light Sources}
185: % -----------------------------------------------------------------------------
186: 
187: \subsection{Detector Buildings, Optics, and Electronics}
188: 
189: The Auger FD comprises four detector stations (see Fig.\,\ref{auger_scheme}).
190: At present, the sites at Los Leones, Coihueco, and Los Morados are completed 
191: and fully operational, while the fourth site at Loma Amarilla is under 
192: construction.  APF light sources are operating at the Coihueco and Los Morados
193: FD sites.  Both were built by the University of New Mexico 
194: group~\cite{Matthews:2003}.  Fig.\,\ref{apf_photo} shows a photograph of the 
195: APF container building at Los Morados.
196: 
197: Each APF building contains sources which operate at different wavelengths in the 
198: region of interest between 300~nm and 400~nm.  During the initial studies described 
199: in this paper, only one light source with a Johnson U-band filter of central
200: wavelength 350~nm was used.  However, in the near future, we plan to 
201: operate the light sources at several wavelengths to study the wavelength dependence 
202: of the phase function over the full range of the FD sensitivity.
203: 
204: \begin{figure}[t]
205: \centering
206: \includegraphics[scale=.3]{fig2.eps}
207: \includegraphics[scale=.3]{fig3.eps}
208: \caption{\it Photos of the enclosure (left) and the light source (right) at the
209: Los Morados APF facility.  In the photograph on the left, the Los Morados FD can be seen 
210: on the horizon (to the left of the container).}
211: \label{apf_photo}
212: \vskip1cm
213: \end{figure}
214: 
215: The light beam is provided by a broad-band Xenon flash lamp source from Perkin Elmer
216: Optoelectronics (model LS-1130-4 FlashPac with FX-1160 flash lamp).  The Xenon flash 
217: lamps were chosen because of their excellent stability in intensity and pulse shape.
218: A Johnson/Cousins (Bessel) U-band filter from Omega Optical Inc. (part number XBSSL/U/50R)
219: selects a central wavelength of $\sim 350$~nm, FWHM 60~nm) from the broad flash 
220: lamp spectrum.  The beam is focused using a 20.3~cm
221: diameter UV enhanced aluminum spherical mirror (speed f/3) 
222: from Edmund Scientific Co. (part number R43-589).
223: All optical components are assembled on a commercial optical plate.
224: We use Thor optical table parts, assembled from Nomex 
225: Epoxy/Fiberglass 1.91~cm panels from TEKLAM (part number N507EC).
226: 
227: The Xenon lamps rest inside refurbished 6.1~m
228: shipping containers, and the light is sent through a 0.749~cm
229: thick acrylite UV transmitting window (Cyro Industries acrylite OP-4 UVT acrylic).  
230: Each light source provides a nearly horizontal beam of divergence $\leq 10~\mathrm{mrad}$ 
231: pulsed across the field of view of the nearby fluorescence detector.
232: Computer control occurs from the corresponding FD building.  A serial radio link 
233: (YDI Wireless, model 651-900001-001 (TranzPoint ESC-II Kit)) connects the computer 
234: to a commercial ADC/relay system (model ADC-16F 16 channel 8 bit ADC and RH-8L 8-relay 
235: card from Electronic Energy Control Inc.) at the light source.  
236: 
237: Once during each hour of FD data taking, the ADC/relay system enables a 1~Hz GPS pulser 
238: (CNS Systems Inc., model CNSC01 with TAC32 software) and a 12~V to 24~V inverter 
239: to power the Xenon flash lamps.  Each lamp fires a set of 5 shots, pulsed at 2 second 
240: intervals.  The APF events are flagged by the FD data acquisition system and the 
241: corresponding FD data are stored on disk in especially designated APF data files.
242: 
243: When the light sources are not operating, only the radio link and the ADC board are powered.
244: The total current draw is therefore only $\sim 0.2\,\,\mathrm{A}$ at 12~V, and the whole 
245: system can be powered by batteries recharged during the day with 12~V solar panels 
246: (two Siemens SP75 75~W solar modules with Trace C35 controller).
247: 
248: \subsection{APF Signals in the Fluorescence Detectors}
249: 
250: The light beam produced by the APF sources is observed by the cameras of the corresponding FD
251: site.  The FD detectors of the Pierre Auger Observatory are described in detail 
252: elsewhere~\cite{Bellido:2005}.  Here, we only give a short summary of the main 
253: characteristics relevant for the analysis of APF shots.
254: 
255: Each Auger FD site contains six bays, and each bay encloses a UV telescope
256: composed of a spherical light-collecting mirror, a photomultiplier camera at the focal
257: surface, and a UV transmitting filter in  the aperture.  The mirrors have a radius
258: of curvature of 3.4~m and an area of about $3.5\times 3.5~\mathrm{m}^{2}$.  The camera
259: consists of 440 photomultipliers with a hexagonal bialkaline photocathode, arranged in a
260: $20\times 22$ array.  Each camera has a field of view of $30.0^{\circ}$ in azimuth and
261: $28.6^{\circ}$ in elevation, covering an elevation angle range from $1.6^{\circ}$ to
262: $30.2^{\circ}$ above horizon.  To reduce optical aberrations, including coma, the FD 
263: telescopes use Schmidt optics with a circular diaphragm of diameter 2.2~m placed at 
264: the center of curvature of the mirror, and a refractive corrector ring at the telescope 
265: aperture.
266: 
267: Fig.\,\ref{apf_shot} shows an APF shot as seen by the Coihueco FD.  Five out of the 6 bays
268: of the Coihueco FD site observe light from the Coihueco APF facility.  In this figure,
269: the light travels from right to left.
270: Fig.\,\ref{apf_geo} shows the relative positions of the APF source and 
271: the FD at the Coihueco site.  The geometry is in part dictated by the local topography, and 
272: consequently is slightly different for the Los Morados site.  
273: 
274: 
275: \begin{figure}[t]
276: \centering
277: \includegraphics[scale=.7]{fig4.eps}
278: \caption{\it The APF pulse as seen by the Coihueco FD.  The light travels from right to left,
279: and each PMT Cluster observes $30^{\circ}$ in azimuth.  Note that the projection of the 
280: approximately horizontal APF beam onto the spherical FD surface results in a curved track.}
281: \label{apf_shot}
282: \includegraphics[scale=1.]{fig5.eps}
283: \caption{\it Scheme of the location of the Coihueco APF light source relative to the 
284: Coihueco FD.  Located at the center is the Coihueco FD with its field of view indicated.
285: The value of $\alpha$ is $26^{\circ}$ and $\beta$ is $38^{\circ}$, measured from the North.
286: The shot direction $\gamma$ is about $24^{\circ}$.}
287: \label{apf_geo}
288: \vskip1cm
289: \end{figure}
290: 
291: 
292: % -----------------------------------------------------------------------------
293: \section{Determination of the Aerosol Phase Function}\label{sec:method}
294: % -----------------------------------------------------------------------------
295: 
296: The signal from the APF light source observed by the $i^{th}$ pixel of a fluorescence detector 
297: can be expressed as
298: 
299: \begin{equation}
300: S_i = I_0 \cdot T_i \cdot \left[ \frac{1}{\Lambda_m} 
301: \left(\frac{1}{\sigma_m} \frac{d \sigma_m}{d\Omega}\right)+\frac{1}{\Lambda_a}
302: \left(\frac{1}{\sigma_a} \frac{d\sigma_a}{d\Omega}\right)\right] _i \cdot 
303: \Delta z_i \cdot \Delta \Omega_i \cdot \epsilon_i~~.
304: \label{eq:signal}
305: \end{equation}
306: 
307: In this equation, $I_0$ is the light source intensity; 
308: $T_i$ is the transmission factor $e^{-r_i/\Lambda_{tot}}$ which 
309: accounts for light attenuation from the beam to the pixel; 
310: $r_i$ is the distance from the beam to the detector; 
311: $\Lambda_{tot}$, $\Lambda_m$, and $\Lambda_a$ are the total, molecular, 
312: and aerosol extinction length, respectively; 
313: and $\sigma_m^{-1} d\sigma_m/d\Omega$ and $\sigma_a^{-1} d\sigma_a/d\Omega$ are the 
314: normalized differential molecular and aerosol scattering cross sections, respectively,
315: which are identical to the phase functions $P_m(\theta)$ and $P_a(\theta)$.  The integral 
316: of $P_m(\theta)$ and $P_{a}(\theta)$ over all solid angles is equal to 1.  Finally, 
317: $\Delta z_i$, $\Delta\Omega_i$, and $\epsilon_i$ are the track length, detector solid 
318: angle, and the efficiency for the $i^{th}$ pixel of the detector.  
319: 
320: The data come in the form of total PMT signal per pixel from a particular shot.  Those 
321: data are binned as a function of azimuth and averaged between the five shots taken within 
322: 10 seconds.  In this analysis, $5^{\circ}$ bins are used, although the fit is relatively
323: insensitive to the number of bins.  Each FD pixel is hexagonally shaped, so for those
324: lying at the boundary of two azimuth bins, the fractional area of the hexagon in each bin 
325: is used to properly distribute the signal. The signal in each pixel is divided by 
326: $\Delta z_i$, $1/r_i^2$ and $\epsilon_i$ to correct for the geometry of the beam 
327: and pixel calibration.  Note that in the roughly cylindrical geometry of the FD-APF beam,
328: the $\Delta z_i$ and $1/r_i^2$ corrections almost completely cancel out.
329: 
330: Typical values for the aerosol extinction length in dry atmospheres are between
331: 10~km and 20~km, reaching 40~km for very clear conditions.  Since the perpendicular 
332: distance from the beam to the FD is only on the order of a few hundred meters, it 
333: is reasonable to assume full atmospheric transmission $(T_i=1)$ over the length of 
334: the beam.  In reality, this assumption does not hold well for the most distant beam 
335: points, so these points are not used in the present study.  In the near future, 
336: measurements of the extinction length from the Auger lidar stations~\cite{BenZvi:2007}
337: will be used to improve the APF analysis.  In another approximation, we assume that 
338: the extinction lengths are identical for each pixel for single measurements and do 
339: not require an index $i$. In principle, the extinction length depends on the number 
340: density of scatterers and is therefore a function of the density (temperature, 
341: pressure) of the air. 
342: 
343: Given corrections for geometry, attenuation, and pixel efficiency,  Eq.\,\ref{eq:signal}
344: reduces to 
345: \begin{equation}
346: S_i = C \cdot\left[ \frac{1}{\Lambda_m} \left(\frac{1}{\sigma_m}
347: \frac{d \sigma_m}{d\Omega}\right)+\frac{1}{\Lambda_a}
348: \left(\frac{1}{\sigma_a}\frac{d\sigma_a}{d\Omega}\right)\right]~~,
349: \end{equation}
350: where $C$ is a constant whose value is unimportant because arbitrary units are sufficient 
351: in determining the phase function.  
352: 
353: \begin{figure}[t]
354: \centering
355: \includegraphics[scale=1.]{fig6.eps}
356: \caption{\it Schematic of track seen by $i^{th}$ pixel.}
357: \label{beamgeo}
358: \vskip1cm
359: \end{figure}
360: 
361: From the theory of Rayleigh scattering it is known that the Rayleigh phase 
362: function is 
363: \begin{equation}
364: P_{m}(\theta) = \frac{3}{16\,\pi (1+2\,\gamma)}
365: \left[(1+3\,\gamma)+(1-\gamma)\cos^{2}\theta\right]
366: \label{eq:rayleigh1}
367: \end{equation}
368: where $\gamma$ accounts for the effect of molecular anisotropy on Rayleigh scattering. For
369: isotropic scattering, $\gamma=0$, this reduces to the familiar 
370: \begin{equation}
371: P_{m}(\theta) =  \frac{3}{16\,\pi} (1+\cos^2\theta)~~.
372: \label{eq:rayleigh2}
373: \end{equation}
374: The effect of the anisotropy is small and wavelength-dependent.  Bucholtz~\cite{Bucholtz:1995} 
375: estimates $\gamma\simeq 0.015$ at 360~nm and concludes that the correction leads to
376: a $\sim 3\,\%$ systematic increase in the Rayleigh scattering cross section, and a fractional
377: change $\leq 1.5\,\%$ from the approximate $(1+\cos^{2}\theta)$.  In our analysis, only the
378: shape of the function is relevant, and we use Eq.\,\ref{eq:rayleigh2} as an approximation
379: of Eq.\,\ref{eq:rayleigh1}.
380: 
381: The aerosol phase function is often parameterized by the Henyey-Greenstein 
382: function~\cite{Henyey:1941}:
383: \begin{equation}
384: P_{a}(\theta) = \frac{1-g^2}{4\pi} \frac{1}{(1+g^2-2g\mu)^{3/2}}~~,
385: \label{eq:henyey1}
386: \end{equation}
387: where $\mu=\cos\theta$ and $g$ is an asymmetry parameter equal to
388: the mean cosine of the scattering angle: $g=\langle\cos\theta\rangle$.  The parameter 
389: $g$ is a measure of how much light is scattered in the forward direction; a greater
390: $g$ means more light is forward-scattered.  Values for $g$ range from $g=1$ (total 
391: forward scattering) to $g=-1$ (total backward scattering), with $g=0$ indicating isotropic 
392: scattering.  
393: 
394: The Henyey-Greenstein function works well for pure forward scattering,
395: but it cannot describe realistic aerosol conditions, which typically
396: give rise to non-negligible backscattering.  
397: Following~\cite{Fishburne:1976,Riewe:1978}, we modify Eq.\,\ref{eq:henyey1}
398: so that
399: \begin{equation}
400: P_{a}(\theta) =
401: \frac{1-g^2}{4\pi}\left(\frac{1}{(1+g^2-2g\mu)^{3/2}}
402: +f\frac{3\mu^2-1}{2(1+g^2)^{3/2}}\right)~~.
403: \label{eq:henyey2}
404: \end{equation}
405: The new term in this expression is proportional to the second Legendre
406: polynomial, and it is introduced to describe the extra backscattering
407: component.  The value $f$ is a fit parameter used
408: to tune the relative strength of forward to backward scattering.
409: 
410: The binned APF signal observed in the FD is therefore subjected to a
411: 4-parameter fit:
412: \begin{equation}
413: S_i = A \cdot (1+\mu_i^2) + B \cdot (1-g^2) \left(\frac{1}{(1+g^2-2g\mu_i)^{3/2}}
414: +f\frac{3\mu_i^2-1}{2(1+g^2)^{3/2}}\right)~~, 
415: \label{eq:fit}
416: \end{equation}
417: where $A$, $B$, $g$ and $f$ are the fit parameters.  
418: 
419: In principle, the parameters $A$ and $B$, which describe the relative amount of
420: Rayleigh and Mie scattering, can be determined from measurements of the extinction 
421: lengths $\Lambda_m$ and $\Lambda_a$ and assumptions about the particle albedo, 
422: {\it i.e.} the ratio of light scattered by the aerosol particle in all directions to 
423: the amount of incoming light.  The albedo is close to one if the particle is mostly 
424: reflective.  Since local information on the extinction lengths was not available for 
425: this analysis, we use $A$ and $B$ as additional fit parameters.  We find that the 
426: distinct shapes of the two phase functions does allow a determination of $A$ and $B$ 
427: from the data themselves.
428: 
429: \begin{figure}
430: \centering
431: \includegraphics[scale=.5]{fig7.eps}
432: \includegraphics[scale=.5]{fig8.eps} 
433: \caption{\it Two examples for APF data fits on different days.  In the upper plot
434: (June 28, 2006, 5:12 am local time) aerosols are visible.  Data are fit to the
435: function given in Eq.\,\ref{eq:fit}.
436: The phase function in the lower plot (July 2, 2006, 3:12 am local time)
437: is consistent with pure Rayleigh scattering.  Data are fit to 
438: Eq.\,\ref{eq:fit}, with $B=0$, $f=0$, and $g=0$.
439: Error bars for both plots are the standard deviation of the 5 APF events.}
440: \label{apf_fits}
441: \vskip1cm
442: \end{figure}
443: 
444: At Coihueco, the APF signal is seen in 5 out of the 6 mirrors, so the track is visible
445: over $\sim 150^{\circ}$ in azimuth.  At the boundary between each mirror there is some 
446: overlap in the fields of view of pixels.  This overlap produces a double counting 
447: of signal resulting in the value of bins at boundaries being too large.  These bins are 
448: simply ignored in the fit.  The values of the other bins and their errors are obtained 
449: from the mean and standard deviation of the five APF shots in each shot sequence.
450: 
451: On clear nights with few or no aerosols, the fit to Eq.\,\ref{eq:fit} returns unphysical
452: values for the parameters $B$, $f$, and $g$.  In those cases, we re-fit the data to a 
453: pure Rayleigh function by setting $B$, $f$, and $g$ equal to zero.  Two examples of
454: fits, one for a night with aerosol content, and one for a night with pure 
455: Rayleigh scattering, are shown in Fig.\,\ref{apf_fits}.  The aerosol, molecular, and 
456: total phase functions are shown.  The aerosol phase function is obtained by subtracting 
457: the molecular component determined by the fit. 
458: 
459: We fit the data only over a subrange of the available scattering angles,
460: from $\theta_{min}\simeq 32.5^{\circ}$ to $\theta_{max}\simeq 147.3^{\circ}$.  
461: As Fig.\,\ref{apf_fits} indicates, the data deviates
462: from the theoretical prediction for scattering angles below $\theta_{min}$ and above 
463: $\theta_{max}$.  At smaller and larger angles, several effects corrupt the 
464: signal and make it unusable for the fit to the phase function.  Due to the local geometry 
465: at the Coihueco site (see Fig.\,\ref{apf_geo}), the APF shot is not visible for 
466: $\theta<24^{\circ}$, and below $30^{\circ}$, the signal is incomplete because 
467: the beam is still partially beneath the detector field of view.
468: At large scattering angles, the beam is at a rapidly increasing distance to the 
469: corresponding FD bay, and attenuation of light from the beam to the detector becomes 
470: important.  As mentioned earlier, because local measurements of the optical depth are 
471: not yet available, we simply assume $T=1$.  As measurements of $T$ become available, 
472: the attenuation of light scattered at large angles can be used to correct the data.
473: 
474: In order to apply geometrical corrections when binning the data, the angle at which the 
475: APF light source shoots ($\gamma$ in Fig.\,\ref{apf_geo}) with respect to the FD and the 
476: elevation angle of the shot direction needs to be known.  We determined these values from the 
477: data themselves.  The elevation angle was determined from a reconstruction of APF shots 
478: with the FD offline reconstruction~\cite{Argiro:2007}, and $\gamma$ was determined from 
479: the analysis of APF shots on nights where aerosol scattering was negligible.  The data 
480: from these nights were fit to the Rayleigh component of the phase function, with the 
481: position of the minimum (nominally at $90^{\circ}$ scattering angle) as a free parameter.  
482: The fit value of this angle was then used to deduce the direction which the APF light 
483: source shoots relative to the FD ($\sim 24^{\circ}$ at Coihueco).
484: 
485: \begin{figure}[t]
486: \centering
487: \includegraphics[scale=.7]{fig9.eps} 
488: \caption{\it {\it Top:} Distribution of the asymmetry parameter $g$ (top left) and 
489: the backscatter parameter $f$ (top right) for all measurements performed between June
490: and December 2006.  Values of $g=0$ (and $f=0$) indicate that 
491: the phase function can be described with pure Rayleigh scattering.  {\it Bottom:} 
492: Asymmetry parameter $g$ as a function of time.}
493: \label{gf_param}
494: \vskip1cm
495: \end{figure}
496: 
497: \begin{figure}[t]
498: \centering
499: \includegraphics[scale=.4]{fig10.eps} 
500: \caption{\it Distribution of the Rayleigh normalization parameter $A$
501: for all measurements performed between June and December 2006.}
502: \label{A_param}
503: \vskip1cm
504: \end{figure}
505: 
506: 
507: % -----------------------------------------------------------------------------
508: \section{First Results}
509: % -----------------------------------------------------------------------------
510: 
511: We have applied the analysis described in Section\,\ref{sec:method} to data recorded 
512: between June and December 2006 at the Coihueco site.  Since the APF light sources
513: operate during all nights of FD operation, this data set includes all moonless nights,
514: with the exception of nights with rain or strong winds when the FDs remain closed.
515: Fig.\,\ref{gf_param} shows the 
516: distribution of the asymmetry parameter $g$ (left) and the backscatter parameters 
517: $f$ (right).  For most nights with aerosol contamination, the value of $g$ at the 
518: experiment site in Malarg\"ue is $\sim 0.6$, with an average of 0.59 and a standard 
519: deviation of 0.07 for the data period analyzed here.  Values of $g=0$ indicate hours where 
520: the measured phase function can be described by pure Rayleigh scattering, so the 
521: aerosol phase function is effectively negligible.  Fig.\,\ref{gf_param} also shows 
522: the asymmetry parameter as a function of time for the analyzed period.  
523: With the limited amount of data taken so far, no conclusions concerning seasonal 
524: variations can be drawn.  The asymmetry parameter appears to be stable during the 
525: observed time period.  With more data becoming available over the next
526: few years, we plan to monitor the month-to-month variation in $g$ and
527: analyze possible correlations with other weather measurements.
528: 
529: One of the main tasks of the APF, in addition to providing the {\it in situ} aerosol 
530: phase function for every hour of FD data taking, is the identification
531: of ``clear'' nights with small aerosol contamination.  These nights play an 
532: important role in the calibration of other atmospheric monitoring devices such as
533: the Central Laser Facility (CLF)~\cite{Fick:2006}.  On clear nights, the measured 
534: phase function can be described by pure Rayleigh scattering (measurements where this
535: is the case appear as $g=0$ in Fig.\,\ref{gf_param}).  
536: 
537: To confirm the reliability of the fit where both the normalization of the Mie and 
538: the Rayleigh contribution are fit parameters, Fig.\,\ref{A_param} shows the Rayleigh 
539: normalization factor $A$ for the same data set.  One might expect the molecular
540: contribution to be rather stable, and in fact this parameter does not change much 
541: with time.
542: 
543: It is instructive to compare the average asymmetry parameter obtained from the APF
544: with model expectations and measurements at comparable locations.  Typically, 
545: measurements are performed at optical wavelengths and cannot be directly compared 
546: to measurements at UV wavelengths.  However, a compilation at different wavelengths 
547: from 450~nm to 700~nm~\cite{Fiebig:2006} shows that the wavelength dependence 
548: of $g$ is small; values at 450~nm are a few percent larger than at 550~nm.  
549: 
550: To first order, $g=0.7$ is often used as a generic value for $g$ in radiative transfer 
551: models.  A smaller value for $g$ is expected at dry locations.  A parameterization of 
552: aerosol optical properties by d'Almeida et al.~\cite{dAlmeida:1991} suggests values for $g$ 
553: between 0.64 and 0.83 at 550~nm depending on aerosol type and season, with higher 
554: averages for high relative humidity.  
555: 
556: The Pierre Auger Observatory is located east of the Andes in the Pampa Amarilla, 
557: an arid high plateau at 1420~m a.s.l., so values around 0.6 are within expectations.
558: For comparison, recent measurements carried out in the 
559: Southern Great Planes of the US~\cite{Andrews:2006} yield values for $g$ at 550~nm 
560: of $0.60\pm0.03$ for dry conditions and $0.65\pm0.05$ for ambient conditions.
561: 
562: \begin{figure}[t]
563: \centering
564: \includegraphics[width=4in]{fig11.eps}
565: \caption{\it Comparison of the Longtin aerosol phase function (desert atmosphere
566: simulated with a wind speed of $10$~m/s) with the default phase
567: function used in the Auger atmospheric database ($f=0.5$, $g=0.7$)
568: and the typical phase function measured by the APF ($f=0.4$, $g=0.6$).}
569: \label{phasefunctions}
570: \vskip1cm
571: \end{figure}
572: 
573: The aerosol phase function most commonly used in fluorescence detector data analysis, 
574: both for the High Resolution Fly's Eye (HiRes) Experiment~\cite{Thomson:2004}, which operated 
575: in Utah between 1997 and 2006, and the Pierre Auger FD detectors, is the function
576: obtained from a desert aerosol simulation by Longtin~\cite{Longtin:1988}.
577: Longtin's desert model is based on Mie scattering theory and assumes that the
578: desert atmosphere has three major components: carbonaceous particles,
579: water-soluble particles, and sand. For each aerosol component, the model
580: assumes a characteristic log normal size distribution and refractive index.
581: Longtin performed his calculations for several wavelengths and wind speeds;
582: those made at 550~nm with a wind 
583: speed of 10~m/s most closely match the 300~nm to 400~nm nitrogen 
584: fluorescence band observed by the FDs and have therefore been traditionally 
585: used in air fluorescence data analysis.
586: 
587: Fig.\,\ref{phasefunctions} compares the Longtin aerosol phase function at 550~nm 
588: to the modified Henyey-Greenstein function of Eq.\,\ref{eq:henyey2} with two sets 
589: of $f$ and $g$: $f=0.5$ and $g=0.7$, the default values used by the Auger
590: atmospheric database; and $f=0.4$ and $g=0.6$, the values determined in this
591: study to be more typical of the detector location.  The comparison shows that,
592: on average, the difference between the Longtin function and the measured phase 
593: function is small for those scattering angles relevant in fluorescence measurements
594: --- $\sim 30^\circ$ to $150^\circ$.  Only at the largest scattering angles above 
595: $160^\circ$ do the phase functions differ notably.  This region is outside the 
596: current range of validity of our measurement.
597: 
598: \begin{figure}[t]
599: \centering
600: \includegraphics[width=4.5in]{fig12.eps}
601: \caption{\it Differences in the energies of golden hybrid events reconstructed with
602: default phase function values $(E_\text{std})$ and those reconstructed
603: using phase function fit parameters determined from APF measurements
604: $(E_\text{apf})$.  The red (bold) histogram represents data taken during
605: nights with measurable aerosols; the blue (light) histogram depicts events
606: observed on purely molecular nights.}
607: \label{energyspread}
608: \vskip1cm
609: \end{figure}
610: 
611: Our primary interest in aerosol scattering is its effect on the air shower reconstruction,
612: most notably the determination of the shower energy.  However, it is not straightforward to
613: estimate the extent to which the use of measured rather than averaged values of $f$ and $g$
614: changes the energy reconstruction, as this depends strongly on other atmospheric parameters,
615: for example the aerosol optical depth.  Rather than singling out the phase function
616: measurement, we need to study the effect of the combined measurement of all atmospheric 
617: parameters, a task which is beyond the scope of this paper.
618: 
619: We can, however, get an estimate of the relevance of the phase function measurement     
620: by studying its effect on the energies of events that are of particular importance 
621: for the energy calibration of the detector, the ``golden hybrid events.''  These are 
622: events observed by one or more fluorescence detectors and three or more surface array 
623: tanks.  For ``golden hybrid events'' observed by the Coihueco FD site between June and
624: December 2006, we performed the reconstruction twice: first, using the default
625: parameters $f=0.5$ and $g=0.7$ to estimate aerosol scattering; and second,
626: using the fit parameters determined from APF measurements.  In both cases
627: atmospheric extinction was simulated using an average aerosol profile model
628: representative of the Malarg\"ue site~\cite{BenZvi:2007a,Prouza:2007}.  
629: 
630: Fig.\,\ref{energyspread} depicts the relative differences in energies
631: caused by reconstructing showers with the default phase function and the
632: measured phase function.  The red (bold) histogram represents data taken during
633: nights with aerosol contamination ($f>0$, $g>0$) while the blue (light) histogram
634: represents data taken during nights where according to the APF analysis
635: scattering is purely molecular.  The correction is typically of order one percent.
636: However, on those nights when aerosol loading is extremely low,
637: so that atmospheric scattering may be characterized as purely molecular, the
638: use of the default scattering parameters causes
639: larger errors in the shower reconstruction.  Under such conditions, the
640: total phase function lacks the strong forward-scattering component typical of
641: aerosols.  During these periods, incorrectly accounting for aerosol scattering 
642: starts to impact the energy calibration of the detector.  A correct determination 
643: of the phase function on a regular basis is
644: therefore an important part of the atmospheric monitoring efforts at the site.
645: 
646: 
647: % -----------------------------------------------------------------------------
648: \section{Conclusions and Outlook}
649: % -----------------------------------------------------------------------------
650: 
651: As part of the atmospheric monitoring program at the Pierre Auger Observatory, 
652: the aerosol phase function at 350~nm is routinely measured
653: at two of the four FD sites.  A first analysis of data taken from June to December 2006
654: shows that values of $g=\langle\cos\theta \rangle\simeq 0.6$ for the mean cosine of the
655: scattering angle $\theta$ are typical for aerosols at the site of the experiment.  
656: Over the next several years, the APF light sources will
657: produce a data set of unprecedented size of the scattering properties of aerosols.
658: This data set will enable us to carefully study any seasonal change in the 
659: aerosol content.  The APF light sources and the other atmospheric monitoring instruments
660: at the Auger site will accumulate one of the largest sets of continuous measurements in the 
661: 300~nm to 400~nm range ever recorded for a single location.
662: 
663: The APF light sources are currently operating at a wavelength of 350~nm only.  In the near 
664: future, we will add regular measurements at 330~nm and 390~nm to study the dependence 
665: of the phase function on the wavelength of the scattered light.
666: 
667: \ack
668: We are grateful to the following agencies and organizations for financial support:
669: The APF light sources were built by a grant from the 
670: Department of Energy (DOE) Office of Science (USA) (DE-FG03-92ER40732).
671: Parts of the APF analysis were performed during the 2006 REU (Research Experience for
672: Undergraduates) program at Columbia University's Nevis Laboratories which is supported
673: by the National Science Foundation (USA) under contract number NSF-PHY-0452277.
674: 
675: \begin{thebibliography}{99}
676: 
677: \bibitem{Mantsch:2005}   P.~Mantsch (for the Pierre Auger Collaboration), 
678:                          \textit{The Pierre Auger Observatory Status and Progress}, 
679:                          in Proc. $29^{\mathrm{th}}$ Int. Cosmic Ray
680:                          Conference, Pune, India (2005) (astro-ph/0604114); {\tt www.auger.org}.
681: \bibitem{Abraham:2006}   J.~Abraham et al. (Pierre Auger Collaboration), 
682:                          Astroparticle Phys. 27\,(2007)\,155.
683: \bibitem{Aglietta:2006}  J.~Abraham et al. (Pierre Auger Collaboration),
684:                          Astroparticle Phys., in press (2007) (astro-ph/0607382).
685: \bibitem{Cester:2005}    M.~Mostaf\'a (for the Pierre Auger Collaboration),
686:                          \textit{Atmospheric Monitoring for the Pierre Auger 
687:                          Fluorescence Detector}, in Proc. $28^{\mathrm{th}}$ 
688:                          Int. Cosmic Ray Conference, Tsukuba, Japan (2003),
689:                          2\,(HE\,1.3), 465.
690: \bibitem{BenZvi:2007}    S.Y.~BenZvi et al., Nucl. Instr. Meth. A, in press (2007) 
691:                          (astro-ph/0609063).
692: \bibitem{Fick:2006}      B.~Fick et al., Journal of Instrumentation 1\,(2006)\,P11003.
693: \bibitem{Mie:1908}       G.~Mie, Ann. Phys. 25\,(1908)\,377.
694: \bibitem{Roberts:2005}   M.D.~Roberts, J. Phys. G: Nucl. Part. Phys 31\,(2005)\,1291.
695: \bibitem{Matthews:2001}  J.A.J.~Matthews, R.~Clay, for the Pierre Auger Collaboration,  
696:                          \textit{Atmospheric Monitoring for the Auger Fluorescence Detector},
697:                          in Proc. $27^{\mathrm{th}}$ Int. Cosmic Ray Conference, Hamburg, Germany (2001),
698:                          2\,(HE\,1.8), 745.
699: \bibitem{Matthews:2003}  J.A.J.~Matthews et al., 
700:                          \textit{APF Light Sources for the Auger Southern Observatory},
701:                          in Proc. $28^{\mathrm{th}}$ Int. Cosmic Ray Conference, Tsukuba, Japan (2003),
702:                          2\,(HE\,1.5), 873.
703: \bibitem{Bellido:2005}   The Pierre Auger Collaboration,
704:                          \textit{Performance of the Fluorescence Detectors of the 
705:                          Pierre Auger Observatory}, in Proc. $29^{\mathrm{th}}$ Int. Cosmic Ray
706:                          Conference, Pune, India (2005) (astro-ph/0508389).
707: \bibitem{Bucholtz:1995}  A.~Bucholtz, Appl. Opt. 34\,(1995)\,2765.
708: \bibitem{Henyey:1941}    L.~Henyey and J.~Greenstein, Astrophys. Journal 93\,(1941)\,70.
709: \bibitem{Fishburne:1976} E.S.~Fishburne, M.E.~Neer, and G.~Sandri, {\it Voice Communication via Scattered
710:                          Ultraviolet Radiation}, Report 274, Vol. 1,  Aeronautical Research Associates
711:                          of Princeton, Princeton, NJ (1976).
712: \bibitem{Riewe:1978}     F.~Riewe and A.E.S.~Green, Applied Optics 17\,(1978)\,1923.
713: \bibitem{Argiro:2007}    S.~Argir\`o et al., submitted to Computer Physics Communications (2007).
714: \bibitem{Fiebig:2006}    M.~Fiebig and J.A.~Ogren, J. Geophys. Res. 111\,(2006)\,D21204.
715: \bibitem{dAlmeida:1991}  G.A.~d'Almeida, P.~Koepke, and E.P.~Shettle, \textit{Atmospheric Aerosols: 
716:                          Global Climatology and Radiative Characteristics}, A. Deepak Publishing, 
717:                          Hampton, Virginia (1991).
718: \bibitem{Andrews:2006}   E.~Andrews et al., J. Geophys. Res. 111\,(2006)\,D05S04.
719: \bibitem{Thomson:2004}   G.B.~Thomson et al. (for the HiRes Collaboration),
720:                          Nucl. Phys. Proc. Suppl. 136\,(2004)\,28; {\tt www.cosmic-ray.org}.
721: \bibitem{Longtin:1988}   D.R.~Longtin, {\it A Wind Dependent Desert Aerosol Model: Radiative Properties},
722:                          Air Force Geophysics Laboratories, AFL-TR-88-0112, 1988.
723: \bibitem{BenZvi:2007a}   S.Y.~BenZvi et al. (for the Pierre Auger Collaboration),
724:                          \textit{Measurement of Aerosols at the Pierre Auger Observatory},
725:                          in Proc. $30^{\mathrm{th}}$ Int. Cosmic Ray
726:                          Conference, M\'erida, M\'exico (2007).
727: \bibitem{Prouza:2007}    M.~Prouza (for the Pierre Auger Collaboration),
728:                          \textit{Systematic Study of Atmosphere-Induced Influences and 
729:                          Uncertainties on Shower Reconstruction at the Pierre Auger Observatory},
730:                          in Proc. $30^{\mathrm{th}}$ Int. Cosmic Ray
731:                          Conference, M\'erida, M\'exico (2007).
732: 
733: \end{thebibliography}
734: 
735: \end{document}
736: 
737: