X-Sympa-To: ids.analysis.forum Received: from hector.cls.fr (hector.cls.fr [62.161.32.9]) by hermes.cls.fr (8.9.3/8.9.3) with ESMTP id VAA02798 for ; Thu, 8 Feb 2007 21:23:33 GMT Received: by hector.cls.fr (Postfix) id D265D3CB3F; Thu, 8 Feb 2007 21:49:58 +0000 (GMT) Delivered-To: ids.analysis.forum@cls.fr Received: from cluster-d.mailcontrol.com (orange-102.cls.fr [62.161.32.13]) by hector.cls.fr (Postfix) with ESMTP id 5BE793CD7A for ; Thu, 8 Feb 2007 21:49:58 +0000 (GMT) Received: from sol.csr.utexas.edu (sol.csr.utexas.edu [129.116.190.1]) by rly30d.srv.mailcontrol.com (MailControl) with ESMTP id l18LnSoX020551 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO) for ; Thu, 8 Feb 2007 21:49:31 GMT Received: from localhost (localhost.localdomain [127.0.0.1]) by sol.csr.utexas.edu (8.12.11.20060308/8.12.11) with ESMTP id l18LnR8E023710 for ; Thu, 8 Feb 2007 15:49:27 -0600 X-Virus-Scanned: amavisd-new at csr.utexas.edu Received: from sol.csr.utexas.edu ([127.0.0.1]) by localhost (sol.csr.utexas.edu [127.0.0.1]) (amavisd-new, port 10024) with LMTP id 27tqBa7kUWzW for ; Thu, 8 Feb 2007 15:49:24 -0600 (CST) Received: from [129.116.190.133] (ries.csr.utexas.edu [129.116.190.133]) by sol.csr.utexas.edu (8.12.11.20060308/8.12.11) with ESMTP id l18LnEww023678 for ; Thu, 8 Feb 2007 15:49:17 -0600 Mime-Version: 1.0 X-Sender: ries@sol.csr.utexas.edu Message-Id: In-Reply-To: <200702081908.l18J8Hr02939@cnes.fr> References: <200702081908.l18J8Hr02939@cnes.fr> Date: Thu, 8 Feb 2007 15:49:09 -0600 To: ids.analysis.forum@cls.fr From: "John C. Ries" Subject: RE : ITRF2000 orbits and SAA model Content-Type: multipart/Related; boundary="============_-1041151140==_mr============" ; type="text/html" X-Modified-HTML: 1 X-Mailcontrol-Inbound: c!PzNVdyE6pAgYrdZN4HGLIOLgW8bYa616COtuSiUsY= X-Spam-Score: 1.8 X-Scanned-By: MailControl A-07-06-80 (www.mailcontrol.com) on 10.68.0.140 X-Validation-by: laurent.soudarin@cls.fr --============_-1041151140==_mr============ Content-Type: multipart/alternative; boundary="============_-1041151140==_ma============" --============_-1041151140==_ma============ Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Hi Luca, Everyone, At 7:52 PM +0100 2/8/07, Cerri Luca wrote: >2) Long term frequency variations >We apply the SAA model AFTER having removed the long term frequency=20 >drift (difference between >on-board time and TAI is modeled as a second order polynomial). So=20 >we would expect to find >a zero mean in the short term frequency correction coming from the SAA mod= el. >Just as an example, If I calculate the mean of the short-term=20 >frequency correction over cycle >184, I obtain a non zero value (~ 0.013 Hz). this leads to a clock=20 >offset of about 5.5 microsec >after 10 days (on the 2GHz nominal frequency). In figure 3 I show=20 >the difference between two >doris only orbits, with and without SAA correction; the mean along=20 >track difference is about >1 cm, a bit less then what I was expecting. Nikita, do you find any=20 >difference in the along-track >>doris/laser bias when using the SAA model? Upon further thought, I guess I would expect the SAA model to restore=20 the frequency to the correct mean. Note that my bias-drift parameter, combined with editing the affected=20 data, has the same net result as the SAA model, which is to restore=20 the right frequency drift. The downside of my approach is that I have=20 to sacrifice a lot of the data. Borrowing a figure from the previous=20 email to illustrate what I do, I cut out the data in the vicinity of=20 the large frequency change, leaving a set of data that can be=20 approximated with the same slope. This likely works less well when=20 the SAA effect starts to become very serious. > >3) SAA and network degradation >Starting from around cycle 60, we see a degradation of the doris=20 >network (figures 4 and 5). >There are about 10 beacons disappearing, 8 of which are not in the=20 >SAA region. This should >have an impact in the comparison between SAA and non SAA orbits. Any=20 >difference brought by >the SAA correction would be stronger when the number of measurements=20 >outside the SAA is reduced. >I wonder if this can explain much of the different trend that is=20 >observed in the Z bias before >cycle 91 when using the SAA model. What do you think ? I found my earlier analysis, which I showed at the 2004 DORIS=20 workshop. What I found was essentially zero sensitivity to the SAA=20 effect for Z, whether I just left the 'bad' data in, edited it out,=20 or applied my global bias-drift parameter. This is consistent with my=20 other test where I scaled all the DORIS heights (by about 7 mm). The=20 RMS of the residuals improves as you remove the error, but the orbit=20 seems insensitive. However, an important caveat...I still use the 'dynamical' approach=20 for my processing. Only slow variations in the orbit can be=20 accommodated. 'Reduced dynamics' approaches, however, if the dynamics=20 are sufficiently relaxed, could accommodate the reference frame=20 variations that would occur within a day, and it is easy to show that=20 the 1/rev empiricals are correlated with the Z-centering. My question for Nikita would be how 'relaxed' the empirical=20 parameterization was for those SLR/DORIS orbits. Best regards, John Cliquez sur l'url suivante=20 https://www.mailcontrol.com/sr/gQmjxDBI7bKv5ZSS1R+8A5tvZm8g3AwC7J+BwZnpf34O= jQf9ynTEg0HGD3anHt0mji6H3I3Hz1rHbPHSCvVQsOJz6yK6oeqrqqL5ii1XrpWXX4wJ9u6ZWmR= hDnRnD3m!jHGVm95qJn19jlno+T8MQsXDwDrQDkwZQi9fZ0S7tZTuNrlEt18y3cSmxWN0o+tVLp= 0yc5Q0ja5dv9ojEBzCsJRS1djMmiDe=20=20 si ce message est ind=E9sirable (pourriel). --============_-1041151140==_ma============ Content-Type: text/html; charset="us-ascii" RE : ITRF2000 orbits and SAA model
Hi Luca, Everyone,

At 7:52 PM +0100 2/8/07, Cerri Luca wrote:
2) Long term frequency variations
We apply the SAA model AFTER having removed the long term frequency drift (difference between
on-board time and TAI is modeled as a second order polynomial). So we would expect to find
a zero mean in the short term frequency correction coming from the SAA model.
Just as an example, If I calculate the mean of the short-term frequency correction over cycle
184, I obtain a non zero value (~ 0.013 Hz). this leads to a clock offset of about 5.5 microsec
after 10 days (on the 2GHz nominal frequency). In figure 3 I show the difference between two
doris only orbits, with and without SAA correction; the mean along track difference is about
1 cm, a bit less then what I was expecting. Nikita, do you find any difference in the along-track
doris/laser bias when using the SAA model?


Upon further thought, I guess I would expect the SAA model to restore the frequency to the correct mean.

Note that my bias-drift parameter, combined with editing the affected data, has the same net result as the SAA model, which is to restore the right frequency drift. The downside of my approach is that I have to sacrifice a lot of the data. Borrowing a figure from the previous email to illustrate what I do, I cut out the data in the vicinity of the large frequency change, leaving a set of data that can be approximated with the same slope. This likely works less well when the SAA effect starts to become very serious.


3) SAA and network degradation
Starting from around cycle 60, we see a degradation of the doris network (figures 4 and 5).
There are about 10 beacons disappearing, 8 of which are not in the SAA region. This should
have an impact in the comparison between SAA and non SAA orbits. Any difference brought by
the SAA correction would be stronger when the number of measurements outside the SAA is reduced.
I wonder if this can explain much of the different trend that is observed in the Z bias before
cycle 91 when using the SAA model. What do you think ?

I found my earlier analysis, which I showed at the 2004 DORIS workshop. What I found was essentially zero sensitivity to the SAA effect for Z, whether I just left the 'bad' data in, edited it out, or applied my global bias-drift parameter. This is consistent with my other test where I scaled all the DORIS heights (by about 7 mm). The RMS of the residuals improves as you remove the error, but the orbit seems insensitive.

However, an important caveat...I still use the 'dynamical' approach for my processing. Only slow variations in the orbit can be accommodated. 'Reduced dynamics' approaches, however, if the dynamics are sufficiently relaxed, could accommodate the reference frame variations that would occur within a day, and it is easy to show that the 1/rev empiricals are correlated with the Z-centering.

My question for Nikita would be how 'relaxed' the empirical parameterization was for those SLR/DORIS orbits.

Best regards,
John


Cliquez ici si ce message est indésirable (pourriel).

 --============_-1041151140==_ma============-- --============_-1041151140==_mr============ Content-Type: image/gif; name="frequency_SAA_before.gif" ; x-mac-type="47494666" ; x-mac-creator="3842494D" Content-Disposition: attachment; filename="frequency_SAA_before.gif" Content-Transfer-Encoding: base64 R0lGODlhYwQ2AfcAAAAAAP///6qqqv9nbv+lqv/DyP+eqvh6jP9ujPy6yP/h 6P9QjP+qyOuxyOhuqvjV6P9u6OvQ6P8w//9J//9Q//9n//9u//+A//+F//+H //+M/+iA6P+e//+l//+q/+ie6P+8///D/8iZyP/I/+i+6P/h///o///6//h6 //iZ//i3//jV/+iA/+iM/+ie/+ux/8gw/8hQ/+i+/8hu/9zK6Kow/8h7/9ys /+jI/+vQ/6pQ/8iM/6pu6Iww/6pu//jz/4xQ/6p7/8iq/9zK/24w/4xn/6qM /4xu/87I5VAw/25Q/6qZ/4yA/wAA/xISqjAw/1BQ/25u/zAwblBQqoyM/6qq /1BQbsjI/4yMqujo/6qy/zBJ/1Bn/zBQ/+vu/1Bu/26H/26M/4yl/4yq/1CA /1CA6FCM/26e/9zo/zB7/6rI/xJu/zCA/1CZ/4y+/8/j/7zK3DCM/6rQ/8Dc /8jh/26x/zB7yFCs/4zI/zCe/4zL/4y+6DCq/xKq/27M/4zX/zDI/xLI/+j6 /4zo/4zp/263yFDe/W7m/6zp9cj1/zDe/Yzu/7H1/zDo/53L0Mj6/8D8/4z6 /xLy8jD//zD391D//1D9/W7//zBubm7k5Iz//47//4zp6Yzo6G6np6r//4zL y6ru7lBubrv//7z//77//8j//26MjM3//26Jic///4ytrcj19arQ0Nz//+j/ /+v//+j6+sjX1/j//4yOjm7/6Iz/7ozXyDD/yDCtjIz/17z/6Mj66G7/yIL1 yMj/6FDLjIz/vp7/yG7/qtn/6Kr/yFD/jKz/yG7pjIz/pVDkbnD8jFD/bp3/ ql/8bjD3MFD9UG7/bm7kboz/jKr/qqruqsj/yKrQquj/6Oj66MjXyKqyqozp bsj1qozLUP/hyPSyjOiAUP/Dqv+Hbv8wMP9QUP9nZ/9ubqpXV/+Hh/+MjOiA gOiMjP+lpf+qquienuiqqv/Dw//IyOi+vsiqqqqUlP/h4f/o6Ojo6MjIyIyM jG5ublBQUDAwMBISEv///yH5BAEAAP8ALAAAAABjBDYBAAj/AAMIHEiwoMGD CBMqXMiwocOHECNKnEixosWLGDNq3Mixo8ePIEOKHEmypMmTKFOqXMmypcuX MGPKnEmzps2bOHPq3Mmzp8+fQIMKHUq0qNGjSJMqXcq0qdOnUKNKnUq1qtWr WLNq3cq1q9evYMOKHUu2rNmzaNOqXcu2rdu3cOPKnUu3rt27ePPq3cu3r9+/ XfMJGJzPYr7CKg8rFIwYsETFHhmvFEw4IWTLjRNnRngZIT6BnQeGNvjZsenT qFNjzLevH4DX/fZtdpiPH7/ZI2vfRqjP9W7VBi/rxm2xN4DfJlm7hi07uG3i AYZPfr6YOm99h62L1h5cnz7g4MOL/0dt/LV5AP2+RxTwWkBK9gDcB1+OPLz0 6NxX04fusfx59OoNtM9r+yQ0IAAFqgSffAgtyFs/tzlIkIQH1ZbeeBhmqKFc +bjmzz6DCbCPP+jxdxCFJqE4EHz1hUfhaBexaCJHHQLwYYgjlijacv3gVqOO 77WnkIqguVagikQSNGCPGzbp5JNhHZiggASChphkJ76GXWUFUSbAZl42dhg+ nVGWjz5CdsnPa7+NWSaXBJlJ5pWZvfllnFfeGWeIYtbJX5hd8onnnFYSKhqc h3JZG5t07okoZFg6x2ihdupp5aWYGgRfP6UJ1BtyB1ZpUKgILgRpn4EiChph aMaXaGNJLv9ZGJJCeqmokVDmquuuUNXIZJyu9aibAMZdqOlr/vAYYAD+Gcts P8km+51u+1hXHrRpbnfebdRaqyxB1/YDYXZtegubesMWG2Cz01p3n0HsDhQv fvxUy61t9iJ27bLsCrCmeff+Nm+6wS4L2r+MdivwcgAezA+x4uqDT2/OAmsj iKgWedyav+64MZAV2gZxehNHLC/DFV9LoqvPnisQkTUmSGt8i26LmKy85qzz zj/BN6WS7W2aI8gr/gfblQwfHZ2H2PYIX7K7/fgfgwfb/PS4LZtnrNTnDZam g/41LPTKTHJdooRJml120jpeHSGyWIe9NduxIXycYLXSPStzZIP/affbNmKt 9t7oHehPvUrDi3JzBckac+MlPt4gc8giruPgGhvtntzfEYk2myEO9u+Xdn+d Lc+op646S0kuyKI+/p7+MrIg5pjggDfajl96P3p9HHYC4V77ylRXfXcArxdW 48OxMym8iMR/TrNvop/NJuyjB/B8a9ziqj3RVOaO9fYkHnk94fwAv3z1zl8s IoSkT+og+aW+HjvLcf67W/LBu6/76/rTR/Yq1BqtMW5ppfpexySHsyz9LoDZ o9/t/Ee89TUPb/grGsvgYzTzuIcy+iDRr5K0uhKa8IQYaV3QROU5USnwMN57 IfL2oY+h+W5KkvteBjUoH59pkHmjy+GB/3zXwxUep3rx8SHy2pNDyDguhhbD ofJiKCslLjGBVwQiE2MIGel9iYrWS6AKN+hCIYbRfPWTXaAK2LAsIpFBLHrj 5NJIxy9i8Yl3NKIWk6hG6XXwdBQLEAlRSMhCGtKBP6MSH8m4wyvC0YiPdFVv kuVBCllSjY6cHf6EVq3DYTCSRNSk72LTScAVcZHFE+UYU3lJVPJQg6TkhydX qcpsuc50jcykLnd5S1c5aJBxylH7OjglUp0nka/8JSR52MpR7qOUnzwWI5kX ogE+q2LAPKQ2t7k6X4EpWNHc5StrucktXmx0zRxnQbz4w9ARDpRejKOg4onJ DvljMMnCTQvL6f9KUbZznpikJz9DKc5xppOc7lFmLkczRLNpbYp/7NiEhKRQ gjqymfKszD5PWTwJBXKdmOSmSEe6MylBro7+BGkehQVGp1WplwwyYy7ZqUkp IrCYy9ThDUXzTlE2sTHc40dsOMNFiOLUpQPVoE1/mtIFmRGpHHVgVDV21GhW 1DkV02lG75dQ0FUzoBQN6/RWCsM8LtKmG03pKz+q0lSS9K1wfVKNbjQYYfa0 oFfMXfn6p1c67gMfQ0QR/WaKy5h6qHbjo6AeiRXEw77PlP7cHtbwMymESHY3 EuRltuaKWMz6D2tKnJ9i0TjVtqr1e30lbSZh5ti6ilBERIsZYGMLxWT/itU9 mUXt8MwpPsiatqBO9RDsLJXNuBr3uKkJm9bUQ9OJGu1XmJurbXKaualJ9bQh /E/U2GbOiy1ncysDWDgdhLn+YbFCekPgeZh0UE+Fd1LRBWcmMddc55ZWvQ9d 7W0Pkt3nwq62CiwgMhtoX/26qrwOleR775ZW4BrRZqdFroQnjBrlGFBM7srP 7sTlspMt92QfEmrANhOuyaqpXC261gFVJiR1+WthzFFehmHc4Qgr7sOeQhm6 tPOuHMfYwzX+lIxp3Mb79Hg7KCaxjh025CY/aL3NOTK98uWcFtHLyfeZl4/R Iy4ax8bJXfLekYdTM/H2T6IUTrOa+wKoLvmJ/zP5IJOqoiOoPcX5VBUijKEq 9ObgzNlMrZIPluzU50zRWVXFbXOiLBUdMPnoz3XmqaQXXehGYwZPeX5Uxixt GXdims97djOcJ83pVUH6S6E206dHBSQYQSp0esrhmmdN61orKKQiWRSabX0T WVOEwLwOtrCHbZHidqRmyCT2TIRsGKEaTNnQjra0r5ySRVl52i/xzowW4x1s e/vbwYYRSSIFbph0qiLnLre6183udrv73fCOt7znTe962/ve+M63vvfN7377 G0OqNlWd5vzvghv84BLeF2ceq69vIfzhEI/4Ibm26zIfOGm7lrjGN85xJwlP d2rqmm5B3vGSm/zkFf/2HrBX1dgErhzlMI+5zOFSX/vWfOY4z7nOwXJzhKbU VrAOutCHTvSiG/3oSE+60pfO9KY7/elQj7rUp071qlv96ljPuta3zvWue/3r YA+72MdO9rKbXenbvkvPL1rYAOBDRM+Mu9xFLPe62/3ueM+73vfO9777/e+A D7zgB0/4whv+8IhPvOIXz/jGO/7xkI+85CdP+cpb/vKYz7zmN1+tAwJm7TD9 edFFfPbSm/70qE+96lfP+ta7/vWwj73sZ0/72IvsNKDPqbFF5Nad+/73E+b9 aRgIPh4SP+OaAhHwl8985Ar/NB/fa6kjezGSL+T5zc++9rWJfcdQHMz+/L7/ Q7q//fKbP3XkB4zCqY3khteYIek/v/zn/6T4/yXghjY0/htif/r7///A0X+H JIAAWIAGyBcEWEgJeIAM2IBxsYAoBIEOOIEUaBYSqDNAJzoPU4Ec2IFrcYG8 8nZ3Jy6954EmeIJaAYK8koEvVoIo+IIwCBUqqDozGIM2eIM9UYOoo4M42IM+ KBM86Bb752eMZmpp9zLK94NKuIQ5mISOsX431kZbllVD4oRMeIVYCIRW6Bfi F2brpS/v5Q/P1iBbmIVmeIYnEYRrEX3nxVfPtFecZVf8V4ZoWId22BFqmBbH Vycqd0bm5YIrQod3OIiEOBF5iBb15UV7OIeAWIiO//iIydeIdZGIbceGODR6 GwiJmriJJyKIeUGJjNRVXJYmIph3JMiJqIiKh3gWoBhVlrhoQnd7qTiLj7iK Fth2BmVOE4R86+SJtPiLWGiLZbGIHrOLGDRBuRSJwLiMdSiMZfGKnPaKcShC 2+aMzHiNAGiNYyF+0tGF/YUsY9iJkoiN5GiC2jgW65dlDjeF4SiO5fiOPXiO 26hpjjIbQ3h9vgiP+kiB8qgh/biPAIlz/4ghAxmQBnlyBSkeCXmQDKlxCwkY LCiLDTmR2ZiPqFOKdXeKFLmR9PeQ9xeLmciRIll+HqkaJTmSKElvJ4kaK5mS LuluLWkV91iPRHiEMzSOL/+ZkxsXk1QBhQXhkyqWdjypk0QpbUMZFV1oMfmF X5WFjzhZlFBZcEcJFdAINB8CcpdVjRYZlVz5b1PpFMRIVWdWVjbFiF15lhL3 lU3RiuS0JTaplmgZl8YFl0vBlmwXOFJ4aEMnkXLZl/xGl0phl73kD2SjHhg5 gv3wlH65mOsGmEkhmIt1Qaa2lyFJgabwCabAmFnomEgBmf20e1tpfvbACp/w CZdgCZagCZpQmpmpmT/ImUcRljc1lqGFaxMSmtk3C7AQAXDgCJdwmpZwCap5 mpfQmq6Jg7B5FNDoRNUHh47FMVqpmL6nm6oACZDwBkiACKWJmQJxmb+pCccZ j7j/ORfcSB3e+F5USIbSmXOzUJ3WCQmqAAuzoBCaIJzheYPJeRTpaC556ZNO KX+66QquYJ3xOZ8MYQqoaZz3iYL5GZv0+Co1CRENKm0B6gqqIKDyGRGn+QkL CoMT+oDjiXLtaZ0YaqASYQq2oAsd+oIfSnMhunG6CQuwIKAEaqIVEQzJYA0r eoIt+hY9umYjwAEv8AaqcKGukKEZQQ3RQA07ao4vuisRWZkxNwIeoAEaUAEp kAMyaqMZYQ3TMA1N6oE/ShaHGXcaKXMaQAEUYKUeMAIi8aU6GqYVOKZkEaXr WXAmkKYW4KYk4aVgKqf8+KToJ6gFNwJ8ahJwCqgTSKcf/0ioAKikTKqoDcio akGp9waphRSjWyqpFuiobTGTEHpp4+ep7CYP8pATfnpC1Emg7ukKnEoWlqoT /glk6fks1+aOXXGZ27mr3KkX8nAO5aAO95ATiYo6MeqeBCqjqsAIjOCqrxoW sYoTSekxSwkuxDOqdwoU8uAO3Nqt7iAPmlAJlfCb5Equq3kXlykL7VAOwaoT mKozqwqfyWqjrsCszvqsXhGtN1GV4fOGbegf46ivBkGlCMCu53CwCAsPnTCu vLqdq8mhcuGdqBkK8/CtO/GuuhKvyYqkBlGvkMCl+JoVAksTshmWFHOtZtkU urkCHQABC1AO5+Ct3CoPpqCgBv9Rn+D5FqYAnKhZnD6BsU8you9ZoA1RnbAQ sl0xsjPhmYJmJIGGrUtRoargBSDgAQxwqhNhCuJqs2ixs5ZACcHZqz6Rqk4y AiHgBQMKnxzbELMgoEjLFUoLhLjYVGlSRZuFidk6E/FaohlxCZUAsWhxBVXg BovQs1zrE8WaISNgARRwAStwpCDrEKhACq/wtilIqog4t7oENiYjIWWakYlp FEJboxxhCqppFoJbBVEABVAgBn8gtkMBtOKRphJAARhwAhcxCpswCpYrspjL ipqriMJ1P5dIdHwJFLOABkMwB2obuRihmoebFalbBVRABVGwulAQBVeQFLIL HnlKAXv/mhGmm7O9axVxGxNMy0H/wYu3mbcqEaMCOgRDkAPOqxHj6xVXgL3X W71V0L/bqxRkiyGGyhHQW77m+7tmUbJ9KCylkzj/2RNCa6Q/UAIlUcDSq7pQ 8ATZ679RkbjaZJqAa8BScb4xsZw305wJZCv/kmy4mhMVOq8oAcJWkbrYy7ra WxXdW0j3K8JTQcIwUZ7b9YXXlbI3UQJoQKKQuxIyLBWCq782XAX/i8NLOlIW zMMyiMBnsZ9eFo5Spp44EaQ3AJ/1WxI7DBVZgMEbDMVakcOFtMRW/BQ+HBP7 R2iLcYSAOQIdoAJjfBJV/BSpG8VbwcaE5MZvvJZYnCuOaQIz/0HIcRXAdaGr 0RvDlxDChVyXhwwlcawRjBxXHqyzpem3DFu5LlHGlWzJ7uskdtoWpGxcgty1 wAnKkxzJKNHHpdyZlxwen9saobsQ09u/alwR24q1P0HLcNXKZHGZqtmzrEkT m1zLRZHJVJHKBdHL+nu9+tu/QsAAFrsQ2wqs5eAOw/oTzUxSxhwWJSAHenAI qgm7zDzJzhyYt+wk2NfENny9T+zL1cwDDnAO6qAOMiuz3mywwuwTq1zMU5wW V+ADTBAEeiAHshwT4/zOQQHNWyEiWCC41XvNvgzIAjG92ewO/YywIo2wBrvN QkHM5HzQqLu6SVAES0DBPBHREt0z8f/cJAJgBU7Auvz7yxKxrf/srQM9FDK9 TY4cFiMgBKurwTdM0Kc700ZB0S4BqkYYoQ8hAFKQ00utFkO9TZ3cFQkNBEBg wxzdEyjt1D4B1Swxq+woGmxUqydiBVMw1l3b1MdVzlRxBUcABEQABEfA00Gx 1WaNE2idGBhnj4VNWevVjr14ylhR1iJl137cv9UL1j4g10AB2IFtE4OdEvzq hrpjISAimfBX00+B2RwhuJbtF5C9FBhtvVFQvUJwqEZR0JnNE5t9EgrscpcD VLYZiIx9FbrLuyfBC2DAulWQBalR1FIxAjOgBFCw06lNFI5d2zdx2ykSvJrL LNRIxG3xCqT/gAomYQ3RAAx2kL3R3Rdd/RSLCwNAkNVOYdrUDRPWXRJMa602 EiAsWE1SyhYCuscYId7Q8AzKUAjnrdoqvdwaYAE4MBXwHd8uMd8kUd/udd8D kctmustvAQvxGRIA/gzQEA1xGh6rfRSKTBUN7uCsQ9pfIeHZJYapQpm/fRVt e6+l2wvM4OEgniEjjjonjuIKouL5it2hqN0ULqFA7hT9Xbp+iwvGkOMasuM7 Yape0eM+jhIQnhsLzIe6zVJFbuQxLuNuK75+Ownj+gtOAuU48avBGs5bQeVV niJH3hUm7Nl79TTDVYQJceUUkeQWsbOVQOY++yTKLRRqTg7t2hWf/5AJrfDm OaHnuXbY3XjY6mtmo/3lVwEL8EkR3vnn4/rQ45HePlHoMBvUWvEJnlAPpM7o 6BvnXaHF7ldkDXyriz0Xs2Cd/l0Qfi6unZ4zaP4S29rPAh0Wn5AK6OAOqq7Z rB4YDzrVsEhwLRwXFkrjCaGrqimusbwzvc4Sah6s/pzqXGEKq0AL8XDsNeHo cJzsTeEKaPAA026a1r7Onn7mB74T6lAOhn4O3v4VnAAKrEDuNGHuTgHwEjEL DXAA6mAQuW7t7Mwz2Y4S8qAOwIrvaeHm/u4RAk8U0kwXCTAA/Oyt+XAJgL7w qTPoNXEPIK0O+S4WFF/xHHHxQpHLZ0oXD/8P8QhLD50Q6IUE6jQh5Wyx8iyv ES4vFBlvFz49s6J8SA2vK7T98++B7gHo9MyX9Loy3Uwf4VCfGkFPbCQ/yO5c 9Slu6Zh89cun8yZxBQXuFj7v9RSR9TTi7MxOk1Ut9sAn9RZBBc+N3HiR9hui m7fOo3JfFWqdNSmzjpWehXQ/EVdgvc+tF3ovHseKxH3fgWyfEdNaXa122Ny9 hIcPEfmbvVRw9m6x9DsTwQRqoZAg7TY4+RjR2SMnfazvxVi49R/R+RpMBX5B 9VBSAj9QnYygtkhar83qg6pvGFkeRWMpm3n+979H9hxh90oN+nHR+I4xCzkg v6Tbsb0f+Yuq/FH/kb7ZDftYuPkJkb9P8ATu3RfS7xcjoAJvMAdDgAZraxBG K55gPxbeP+R6CZL1/6jzDhABBA4kWNDgwQBXokB5AoUKQogRJU6kWNHiRYwG P136lNHjR5AhRY4sOMICBRRDVM2iOMuVK5IxZc6kWdPmTZw0BewTkNPnz5EC AADoGUAoUYJHiypVKrApvp37pE6d2q9fUaBZtW7l2tUrRmvTptk09WnQGChQ olz52tbrxo5u5c6F6EEDBQoYTlx8yZLuX8CBBePciXXwYYxNjQ41rJgpY6eQ A+QTUNnyZX78DCPm3NnzZ4NirdHUVKmSJjxV2IJmHRFua9giTYzQ4GFE/0ZY qmDF5t3bt8jCvz87lhwZ6WKixI9LDC7c+XPoFKlFoybT1KVJky6Zit76dXff I24TNAEy927w6dVzbr5+br5+APYJ3AegXz6C8OXTt59P//wA6ruPovbcM/DA wcIai6TSJjkNQcRM0UQTCDsziQINyiPJJZgq9PDDmgoEMav6/JHKn/0mwy9A AEzcB0UASzwxxYlEHPFGHGMSTaTrstsuR7om5A7Iti6UAMOZvHhgLyKbdNIp np7ESb+h+pssM/yoHGrAyeKrkssao5RyTDIHogaaaEC6rhIHKSyTq+/exIm2 uyiwYDyZCjCgADn7NNBGPz/Sx8t+9BEoHywFGv90S0MHWtS+RgkUM1BKPbQG GmhGw2hNB3+s1Kc4Pw1pBLs0MNU2m+RRRx1RW+0NUFcpokyAFQ+tdbLKbj00 14tgjfXX3qKBpjqLOK3EU2BpCjXZiei0wFQ8b1pVHmarDcxXa+fCNltuAVOw ImOR7VakZccV6MJnUf3JnXPcMffdrLaFVyt557X3p2miWTCiVz4xTdx7MSq3 WyOR1IpddwNWmKR6F7apYYcjDunSTCcqa0iJi50Q3oLv5ArhjEOuCGKRLZr1 MgEy26xklj/Kd9+BXmn5IyHH7TharVRldWaejZq054+goqqqq4A2uiJqqJmm mE+a1sQUmY+eaGBgTfD/AC+P25pWapFJ5nqgkzHT7GuyC1J6mFouuUSTuMo+ iOpPraFG0wBIxdkrkN1e2Gu9C+K774itYbppjAEfCO5AR/hgG2mowUawvA2f 92/DKZf8cpERl/PCCUS4hm7AIsd8XMtbnVXXgk43SHWLSh/99YAldPNTOrG+ +y/RYbfW9UofLRQh3yPNZx9CIw1zZd2Tb7lmSuk89XbAdFa+WnnkOUUU5MfV 0srUvdwyS36qZHTkn6c3P2TNm0RXA+gH2/r8VuU5pxx29sEk+25lfJHGgfSH 8cp+8CRl3GNO+eB3wIClD0ekysAGJmAw1uQOgW+SXznIUQ51XA9/2foPf8DU /6UUCShLtarPciLCuwmm0E8KBNEscpCCCmxgAx9oH2IkqMImycMdqypHOc5B LRT2STmNkYxiHIWiD0IkiDhk4pNYWKFZqAISc3gBB0KQjd7c8Gjiod78MKgO d1ALShvcXRGLg5ylnFEfKPKH8FCGMpU1UY7WeuKBoggJRkBiJcKRHtkulCFg VfCCPzTIEoEUNl4NMSlmXM4aWxQpoQ2NKlYh4xwt2SfZNWkWaKDBHPToF+e8 b4snOZIGaPK4A1mjG+OwoA/FWEgDxgpR4hsKPyjDSCIeRzGObGPq3ii2Sl5S mGRi3o3uiAQaRACUz9FiyEhVKtvR5FvrMUUvmOGNcP+48oSxdNUsaWlLRQ5E kbw0nqSCOUx0NqmO3XnAG/KIBgWkp5kOI9VJ7PQ8m1AMdM8pyyUqgQtjcEMc r1QiN7v5xizFJ0YEBOFC76OfXibGoOmkqDo5MiJVJYAGe1TPPO9FKlPZ0wLq womwiMVPf5rmWL9o3USz5b8U+Yc/LvrfUdpoGdQhxJAV5elz1snH+Z2DAExa j0c5Zs9TkdQnZ0qTcPrJpmN9onDmDNj2BoQoW4LwS/g5yjdzCstz9lSsB/pp bwSJQQT1kZ6ktFMNb6JP31wHEIHo1FRbGtZkBe9QiQqAXq9ESwBk9XhjJWyO MokgdbSSkBASpb3qSYEjZc3/KyblzZoa0Qc+HMKuvXIpt1hnq/zwCmwIJR9e C3ta6BRTPaoKKkERZFRmXQgvbZXLNEFzhSrgoTTHUsNqQrLTowEXtcM9TFk/ U8FzqMO1EILtr7KRAcjaSalu2VFncEsFKkRhDIOQqkyEC7TvEle8dDEuZ866 sxE1t1Vh0cYGaBsYpnJGIWnBrmpC1FndhXe8+/1KeQ+T2EEu10PqpZTcxDKN aoTArV+Bq2Dm25C1EAa/fUNkZeLIXwzzc2PQYa02gaRWZllDWNCoLmcoC5gr hMEMXVCLb2+iX4dFcmiUzHCNhaNas66qsUDacatmAYtdNANT0dgnYmzrFtwe AQhf/zBDHVwsYdOSrcIpG5uNrRwb/84lsckVcI4ILKU7QgIVx5gbbErsFVP4 oQ1cAAIQjGAKovoExi2b85XtHJMst6XD6JXSlzUpxTxyNDbTOSlXrmMJRdzh DEJYsE4m/Lo631nSIMlzV5Cr3Df5OUd3DPQyY3PkrMjOEpYAGL0enazPHiTV q/vqQSI9aVhfpNJbATBa5QTiPnH6k895WVZGEAJCkJptmzV1lF3lV4MguyCD EuxgY/1suhz2M6oER2sD1eMx6VrQzmnwnE7CAj1YYnZuefWNrIq6cycbRSbc prGh/e6Z4FgwchOWN8CBaUppGkIcgsSuwXNimxQsA4kgdv9Xyj0imAKIIAlf tvfOeXB4R1wgs7aJ0kbMjW+ISt/u+fFLXgILT0cH1DGp3XsBA3EPdZBFH1S5 CB1llXU//NQSpzm5LpqgaOQ8aUUO1MbTE+a+IOjMIgGpqWrTaIPPvFXhNE4a j/MffZyx3TWnOqhuDhg6IAMazyAysHweHWKUAo/+RlB8RyJbaHUG5RViOhqb 7sFbLmfKlrlw1e1OE2nThQplyAWaeP4pXI9IxMvwBdkh1G2P3Gw4Si9ThbmK y0XqEjKD+p1iZCxJ4hXt7puXiby/QoUnbKEQf28VthFk4GlAwxnCCDmEep2R okcTNGuPjjfFB07IixOXENXHAGn/NHcLV5nzww8JxUFyhew6hFtf//SIxVLm HA1dImg/OmxoDx3bVwn3kjfhELtKyyTqlPHEJ38AjO8RKqiFCk9mFvNZI+Kt +91JI0eI4nlzfeyTtuUM3f9DwwfY8HO18Su/4cs7rkC+hXiIcXE/z4C/Z5A/ KZG+gsiGDqgTybq/AawUhlORmZoRAAmb/1O4AnI3Ahw+z/OJ+YqC9XsXd/Ah oYu/riMT+hMI9iKB6hMO/DuQdMOqhBKfADSiESxBIaSI85uI9GsIBXwXeeih LuM2i4PAN0maQqNBYYmGaiiB6MjBA/ErHlSU4kGILqSqIRxDhChCiFCIJ3iC CLOX+UmY/+6APxKbBtKLQH2hwemIP5h5Di3UQdECrdGilYiQqdIiQ0IsCDM8 iCNUvoBhQLpQkCGbwzERMX0RixGLwe7YQ3jBxEIUKwO8CdBTQ/aDF0Z0C/jT F+gTFQV5vlMED000l1bcRJ46wZhAQEVcmMBrPjwMMUjUwwzUm1eERYo6xABI PxUMxXsxPdBIvQe0xPH6RVn6pboDxjH8BFLzxDRMQokZRa0IiyrcRcJyxla5 PKIhQWmEt0+QBEoouI+gRWzMxnb5tAOTQysDR9OBRuErRyE0BUqQhLYhCWJc wZLRxpwoRXm8MnqsloPERzm6BEvoR5H4xHYMmVs8DGUkMUlLyP9kwUiFxKFZ q4IE7Blk/Isr2AOtY0aD7EW30ciNTKFOzAjkqwKA5BmBlAnkowIxuAVv3C+V /I1V8yVA9MlWE0ByXMlYk0WKIMYqOJqZHIkUjElY28neUDaCUDbf2YegTAqU JEpJMy40fIKInJmlPD7soi94g8rYSLfu8cEV2Z7Ausox0soSbMmJwIMvQEKu cQdzeAcUWwgVtK+yzEo/2cCFaxEP7MCd6Idmmzq4JECjPAhNAIQ4WEOpkYd1 WIdYmIsHq8WIM8vW6L9b8UyV40AxXEzyq6ONqIRBcJt0WId4kAvQawJQpDrO ZI22876h0AfKcEusHErStDO5fJtPeBr/vYmHdUiHIjmCJ4DNr/xL3vST2sQl m4qP3/nDX6Ky5uxNK2vMAJCQpgGcWKjMJsSJV2gBHSCCFru72QwMx3M79nw7 NLKpdZtOcZwxzcPOAtwwiCiNq+ubdzAHN8wKTQiDJKiBIzBGiUvPv8i+Woo7 p8ulBj0KzRggLgE+67RP8pNFTciOceubsDSINIuDNVACIyA+BKULBW3L5+S+ ysi9ILTQ4aPGS0CIDJ2EDeXQdwSKSLiDODCDMDDQmitRE9U/hYK7/BjSlYs7 AAHCgrpOF8UwUxg1u5rRGu2biZyJR6gDNsiDHiVAIO2MDRREhoMongifAASr Jh2+S6CEGB0I/wnxpykFnJAcCTcggzQgAzEQwi7ljB1MFLR0pC8ppyU9U867 jrVpGn9aG3XUm6VEPjAgAzDwUfQETD/hQr6iyi8cREHdPNlRG9N4U8kRyKaE 1M3L085INUHclZ8MrVQdzUy9u7IgnOkZxcxczvIj1TKx1VbNVc+o0oxIRMkk Q1wdk2DVVWJ1H3wTiRbogST4VUIc1hGZsmgsVmk1l6+jDQuIgR0oR2f9kPnM PCadVnB9El6diLt4FqSrVUltGWi9x3BtV2aJU4jQAAmQAFNSyG1tknt1V331 CU3zgJOoV3tNV6nJ130t2JpowXPACEHoALs4VzwV2OCCWIOd2PVYwv9yCM+C kAEWcAEPgEuCfY8+VLWQRVXdFCeJpViU7Y42pAhTkIMlsIEOWMyPdQupfLnx sVlIuauU3dlMu9GI0ARLMAQtkAHSnNmvQMsiVUtFWbehiChn41modZJx1QRK UNNElUaj9QrB7B/C3J/5ENP9KVO/OdmoLdvWUIfkys+q9dSNzFqu8MykdSgk 5Q92E79vNVu8bQ2LFbCNEDcLddutSNEGhdunzVvDrZBzIIeEJQjuFM6/JVsn Edy3E0wKjdbDvdyKZcKBANr9tE/ArYn1lFy3OwqrkIxuHUfMTV0DQdt0YIWN WFtB/dyZONHta9DIs13Krc7gu1vV7V26cIf/eWgFhvTb2IXcCqFdBnVP91QK wm1R331e6LAH1+2uTJXd2RXSEOI/IxWho1iouhVK6A1f8Y0X422SL12RMI0P mmKoQB1f933fmbBeoNjTrOpTpn0kTIVf/d3fjJBfoKBUwbLUm2VV/i1gAwZf hTHVzxxZXFnVwj1gCD5g//2T8o1gC97XCXaPDL5gDr7ICk5ge+TdDh5hd93g 3+hWGiNhFX5eE+bJEF5hGFbdFmbFD45hGy7aGg6ZGb5hHqaoHc7CHO5hIcba ID4kBmY1VS3ZtxxiJp7YH66Jmt0V4gngSyXgJr5icH3i2fUe9v2rKukJpHVe LB5jadVimdjavRKf/55A4/YlYzfWVTMmieYdrf+jle3tYgR+Yz0uXhEmk7a7 3T+22z0e5CaN46BgUb9hjHCqXMTUXUd+ZEiOZEmeZEquZEu+ZEzOZE3eZE7u ZE/+ZFAOZVEeZVIuZVM+ZVROZVVeZVZuZVd+5VG2XFQjrUBuOkU6XamQTszb ZV7uZV/2ZX5AzF8eZmIuZqkIZn4wZmVe5qFBZmZ+ZmV2Zmie5l+WZmq+Zsyz Zmze5mMWZm7+Zm3+5msOZ3GeZnIu52c+Z3ReZnVeZ2NuZ3cmZniO52r2Znpm 5hS2FuSt5ffMPUaOUFh2ZGQO6EceaILWXYM+aGhsZIVeaIBuaMxgaIiO6P+H nuiUkWiLvuiKnuiEzuiOtuiP5miMBumRFumNhmhZ/pV9RmRAZmlBPgxnjOki zl/BkOk+DoqZtuK/sGnE4GmYzukHPjmgFuNqWc85vt2jfunByM2eHupAdGCh vmmRYOqfluqQoOrBcEasrmmnhoitvpauBkOo3umwFlkl5gpDHonzTWSkYGOx PmuuwAc9Heu/kGvE+OrAsOvDwGvA0Oulpmu68GvB4Ou6nmu43grBVk/AnovE BgzCNhf6vZWmCGOJaGxRsexPwexK0WxK4exA8Ww/Ae0+EW05IW1rAeDPrNQq JmTWbm3YUGAk/sPDdm3arm3bvm3czpie7Mn8aDX/U0WZ2Q5SqP7ty/BtBubt zjzukEUk415V5haO3V5u4JaIU23g4Fbs4Zbu4qbutZxuF85u5/ZukQXv6xZu dNNunPJq9D5i1oju8N5usSZv6Fbu907v+F5g+J4XZBue1Z5KxMwpSvUq3thv 4rlZBU1MnJ1OLxzg2CDwSz1wAH/w/9O+8qbZKuZvA5/wBRXZKV6RKG4NB89w AV+dw7QlCPeNEM9ZL6bw1SlwFV/wF2/wC3fx6Tzx1KHxRrFxGR9gDFdxHfdv Bv/wz0jxGtfwtjyIAL+9Cu8TP21aQ1FQBT8iqesrLu6l79tw2Ghy/IVyQ7ny I0/LrToULhZbztDyXuJy/+RQ8tUZ88cb8Qa/3zM38soDLARfcaSgbBCH8yeX 8y6nc3TT8BV18yzXczvPWS9HcDTXqu8ZcEJH80PXFUf3c0YXnzgHvz4XdBg3 cDaf9CqpdB+8dDVvuE9PcxZ/F7BFIv8IZgEi01t5FHbTD80I27gD6CV/W/Wd Ef8LIAvrDwi174LYQLcu1VuXdUTRdQmddV8fzPWdj15nb8849f4o9lXn9VpK drABdBZZ9t6A9lyf9vto9rHOvkCn9W0fdlSX9l3/9mo/blVPd/wIdj0192hv 92MH9z83dla3d97g9lTHd2oPLGsHoQhl9WwvzNeW9253d31Py4H/d3LPxKHw Xv8wJqHioLyYY+uiKKFxd3b2iHi6nXiuTY51zynQvGMyz2qPZ5E7p3iRB/hw N3lkX2y1S3mNF02V33jApow6TurhoHnIOFWNb3bjrmOh943u/XibD/qRB0OW t+PsPfma9vmVD3mc/yqgV+Slvz+pB/mPL/rV4SsR4vmOz96p7/qsTx2wd3iO L2qYXzZUF3P5iDp2s81RP/in1xVHUve6v11bdmlhv3u3//c/LSTotPRyB/yp RPUrj3LCb3nbPcu2T3zBZ/DGD/TB3/fIl3K9v3yIyPvkbU/QOGrPX3xADfzN p/zQz/ylnfwYB5t8wAf+nnMVtXu5Nf244/zehv3MA3X/1M+frv0fyXda7pU6 WN/1lqf00vfS3+efvmIj3ndyjO/72YcNNCYnUod+vu9nWnJa6l9+EWx+/PVy 7sd4fvaM6nf+68dfncL67U9+8/f+wLdywBr/4L/0x2+N8w//+Xd/6wcIAQAG EvSnLwDChAoXMmzo8CHEiPsA+Nu3zx+AfQv1YTQYQCDBgR4b5uM3sB++jwMF IATJMiLMmDJlTqx4MePGjgdBhhzJsOTJlDwLHpxp9CjSpEqXguzXb6VCjhSL BtDndCfUjU8pYgQgoGS/fQJu9su39CzTk1tfIpQ6EqxYsmYTumy5sq5Kr2j3 Im26NqpOhHDHYiyrEK/LwXL5/zKe6Tdr28ABFBeey9AlYsiNNz98rDeh26KU ARhumJhfWMKkLXNufVit5tCCUcet3FB23pd4XfP2zLaq5NGlc06dnXox796w P0cuPpm26uEKgRbPzDw5Z9+AnQtnnZC6x+7Yx5NfWtMiRo3NfeZ7alCASZwM rZLuyjKf5YnXy6M9f1M9cM4Jlp9m1glgIH+N+ZceaJINmJB+vyGIH4SaJdgf RejJF6BPkxG4H113ZbXbhXstuCFuDyIUoUN1UbiihSUqdSKAKXpYIYgpIijj WTQ2KOCNMOboYJABsMijeRn+92OHLxoZo2By7YhkUj6uR1WRR/60mJNaUvml a+3JN/+RYTYOFZJ0332lD1ROkggmTGKqR6ZZNhZJooFTwhmTnDCWSeSdBY7Y pmVv7hlRn0/+CWSgINrllZ6HxvnUnKtxiGWjv72ml5tQSkoSpX7WCWin+5k5 qKOfLpQonZeuWqiFdpaqqao/haroqIzO+iqOB6JKa63T3dqqrLDu1yWhIaYa LLOXDVSpV2J2+FxIBKWJbFkl8TOXl80+Cu1X7mGqLbeasbrauWl6mxe40o7L z7ZC/pYufvCWuyyzILUr7qr2ytviSvSua9qz8rrbb7xPpnpawt0OrK/B/E7n r8LzSvzdsJYOrOy+ulLs5cETN+xpvgVXHPJ3H5urcrQZq+v/LcQnXzzbyNeR azDLwG7crLQalgXSewfOlc+BB8anHoWjaQQSP9F5523PUoqkT9FmMe10kFZa ufNk7vlM9NRV59U0l+XaxODVyHEddWVAUy302Fg7+ShLW3PdtZJthw132rbN nVvfGq/ttdRTiR24YRS6LfaTZ2+4M9urLc73QGT7bfXeX8Wt9saR/5w55gBY bmnST41umN2dE6634ZSLLrdZYp4O9uuc3+2tVETlVW3Cm95nr5hoxr6V7nfn PvXuIW0bvO43M6/x8y+ve3x11VaOH/HIO5994tkzujP1Hp15/fPF3azsfdw/ DX5XyI8vOvbVhkcx3eVPy35PWFm//7z382/7fryit77pta96+4tf/p7zP+vB z36Y2lj49Mc7BDbPX/QpngBvFyDdAZCC2rNg9sLTvwdqkFkX7Ido4qO8pzkP hNaiygVJQ8KBnTCFB6wK8VBIs7nUMCpb0WEJe0itCeLwhTtsyw9hmMMZQjCJ s7lhDIF4viMWUYYlBI0Th7jCKlpRgay5WRSZSMMsUmeLYdxhGQmSMCFqUIhp vB4XpQg8FaqxTku8ohvpCMczehGJJ8ESG2+XRyjesY9VBGIcxXjFTxFNc2oq GtxIYhknNdKRj7SkBhv5RUhGcjKMU5FgOvnIAUJOlJX8ZCW/uElMelKUJdTk dDhpyVROh/yFqyTl2kwpy0l+skjf4aUrM6lLTgITky865TBxOTBYXhKVvTzm Li+pzHUxM5TElGYtY8nKVk4TasOEZDFVqU3v0HKR5jwnOtOpznWys53ufCc8 4ynPedKznva8Jz7zqc998rOf/vwnQAMq0IEStKAGPShCE6rQhTK0oQ59KEQj KtGJUrSiFr0oRjOq0Y1ytKMe/ShIQyrSkZK0pCY9KUpTqtKVsrSlLn0pTGMq 05nStKY2vSlOc6rTnfK0pz79KVCDKtShErWoRj0qUpOq1KUytalOfSpUoyrV qVK1qla9KlazqtWtcrWrXv0qWMMq1rGStaxmPSta07rOgAAAOw== --============_-1041151140==_mr============ Content-Type: image/gif; name="frequency_SAA_after.gif" ; x-mac-type="47494666" ; x-mac-creator="3842494D" Content-Disposition: attachment; filename="frequency_SAA_after.gif" Content-Transfer-Encoding: base64 R0lGODlhYwQ2AfcAAAAAAP///6qqqv+lqv9ujP/h6Pi3yP9QjP+qyOhuqvjV 6P9u6OvQ6P8w//9J//9Q//9n//9u//+A//+F//+H//+M/+iA6P+e//+l//+q /+ie6P+8///D/8iZyP/I/+i+6P/h///o///6//h6//iZ//i3//jV/+iA/+iM /+ie/+ux/8gw/8hQ/+i+/8hu/9zK6Kow/8h7/9ys/+jI/+vQ/6pQ/8iM/6pu 6Iww/6pu//jz/4xQ/6p7/8iq/9zK/24w/4xn/6qM/4xu/6qU/87I5VAw/25Q /6qe/4yA/wAA/xISqjAw/1BQ/25u/zAwblBQqoyM/6qq/1BQbsjI/4yMqujo /6qy/zBJ/1Bn/zBQ/+vu/1Bu/26H/26M/4yl/4yq/1CA/1CA6FCM/26e/9zo /zB7/6rI/xJu/zCA/26q/1CZ/4y+/8/j/7zK3DCM/6rQ/8Dc/8jh/zB7yFCl /263/4zI/zCe/4zL/4y+6FCy/zCq/27I/xKq/4zX/27Q/zDI/xLI/+j6/4zo /4zp/263yFDe/W7k/6zp9cj1/zDe/W7o/4zu/7H1/zDo/53L0Mj6/8D8/4z6 /zD//zD391D//1D9/W7//zBuboz//5D//5n//4zo6Kr//6zv71Bubr7//8j/ /26MjM3//8///8j19arQ0Nj//9z//93//+b//+j//+r//+v//8jX1+////X/ //f///j///v//27/6IzXyDD/yDCtjLz/6Mj66G7/yIL1yMj/6FDLjJ7/yG7/ qtn/6Kr/yFD/jKz/yG7pjFDkbnD8jFD/bp3/ql/8bjD3MFD9UG7/bm7kboz/ jKr/qqruqrf+t8P/w8j/yKrQqtb/1tz/3Oj/6Oz/7Oj66MjXyKqyqozpbsj1 qozLUOi+jP/hyOiAUP/Dqv+ljP+Hbv8wMP9QUP9ubv+MjOiMjP+lpf+qquiq qv/IyOi+vv/h4f/o6P/q6v319f/6+ujo6N3d3cjIyLu7u7m5uba2tq6uroyM jG5ublBQUDAwMBISEv///yH5BAEAAP8ALAAAAABjBDYBAAj/AAMIHEiwoMGD CBMqXMiwocOHECNKnEixosWLGDNq3Mixo8ePIEOKHEmypMmTKFOqXMmypcuX MGPKnEmzps2bOHPq3Mmzp8+fQIMKHUq0qNGjSJMqXcq0qdOnUKNKnUq1qtWr WLNq3cq1q9evYMOKHUu2rNmzaNOqXcu2rdu3cOPKnUu3rt27ePPq3cu3r9+/ XesJGFzPYr3CKg8rFIwYsETFHhmvFEw4IWTLjRNnRngZ4TyBnQeGNvjZsenT qFNjrLevH4DX/fZtdliPH7/ZI2vfRqjP9W7VBi/rxm2xN4DfJlm7hi07uG3i AYZPfr6YOm99h62L1h5cnz7g4MOL/0dt/LV5AP2+RxTwWkBK9gDcB1+OPLz0 6NxX04fusfx59OoNtM9r+yQ0IAAFqgSffAgtyFs/tzlIkIQH1ZbeeBhmqKFc 9bjmzz6DCbCPP+jxdxCFJqE4EHz1hUfhaBexaCJHHQLwYYgjlijacv3gVqOO 77WnkIqguVagikQSNGCPGzbp5JNhHZiggASChphkJ76GXWUFUSbAZl42dtg8 nVFWjz5CdsnPa7+NWSaXBJlJ5pWZvfllnFfeGWeIYtbJX5hd8onnnFYSKhqc h3JZG5t07okoZFg6x2ihdupp5aWYGgRfP6UJ1BtyB1ZpUKgILgRpn4EiChph aMaXaGNJLv9ZGJJCeqmokVDmquuuUNXIZJyu9aibAMZdqOlr/vAYYAD+Gcts P8km+51u+1hXHrRpbnfebdRaqyxB1/YDYXZtegubesMWG2Cz01p3n0HsDhQv fvxUy61t9iJ27bLsCrCmeff+Nm+6wS4L2r+MdivwcgAezA+x4uozT2/OAmsj iKgWedyav+64MZAV2gZxehNHLC/DFV9LoqvPnisQkTUmSGt8i26LmKy85qzz zj/BN6WS7W2aI8gr/gfblQwfHZ2H2PYIX7K7/fgfgwfb/PS4LZtnrNTnDZam g/41LPTKTHJdooRJml120jpeHSGyWIe9NduxIXycYLXSPStzZIP/affbNmKt 9t7oHehPvUrDi3JzBckac+MlPt4gc8giruPgGhvtntzfEYk2myEO9u+Xdn+d Lc+op646S0kuyKI+/p7+MrIg5pjggDfajl96P3p9HHYC4V77ylRXfXcArxdW 48OxMym8iMR/TrNvop/NJuyjB/B8a9ziqj3RVOaO9fYkHnk94fwAv3z1zl8s IoSkT+og+aW+HjvLcf67W/LBu6/76/rTR/Yq1BqtMW5ppfpexySHsyz9LoDZ o9/t/Ee89TUPb/grGsvgYzTzuIcy+iDRr5K0uhKa8IQYaV3QROU5USnwMN57 IfL2oY+h+W5KkvteBjUoH59pkHmjy+GB/3zXwxUep3rx8SHy2pNDyDguhhbD ofJiKCslLjGBVwQiE2MIGel9iYrWS6AKN+hCIYbRfPWTXaAK2LAsIpFBLHrj 5NJIxy9i8Yl3NKIWk6hG6XXwdBQLEAlRSMhCGtKBP6MSH8m4wyvC0YiPdFVv kuVBCllSjY6cHf6EVq3DYTCSRNSk72LTScAVcZHFE+UYU3lJVPJQg6TkhydX qcpsuc50jcykLnd5S1c5aJBxylH7OjglUp0nka/8JSR52MpR7qOUnzwWI5kX ogE+q2LAPKQ2t7k6X4EpWNHc5StrucktXmx0zRxnQbz4w9ARDpRejKOg4onJ DvljMMnCTQvL6f9KUbZznpikJz9DKc5xppOc7lFmLkczRLNpbYp/7NiEhKRQ gjqymfKszD5PWTwJBXKdmOSmSEe6MylBro7+BGkehQVGp1WplwwyYy7ZqUkp IrCYy9ThDUXzTlE2sTHc40dsOMNFiOLUpQPVoE1/mtIFmRGpHHVgVDV21GhW 1DkV02lG75dQ0FUzoBQN6/RWCsM8LtKmG03pKz+q0lSS9K1wfVKNbjQYYfa0 oFfMXfn6p1c67mMeQ0QR/WaKy5h6qHbjo6AeiRXEw77PlP7cHtbwMymESHY3 EuRltuaKWMz6D2tKnJ9i0TjVtqr1e30lbSZh5ti6ilBERIsZYGMLxWT/itU9 mUXt8MwpPsiatqBO9RDsLJXNuBr3uKkJm9bUQ9OJGu1XmJurbXKaualJ9bQh /E/U2GbOiy1ncysDWDgdhLn+YbFCekPgeZh0UE+Fd1LRBWcmMddc55ZWvQ9d 7W0Pkt3nwq62CiwgMhtoX/26qrwOleR775ZW4BrRZqdFroQnjBrlGFBM7srP 7sTlspMt92QfEmrANhOuyaqpXC261gFVJiR1+WthzFFehmHc4Qgr7sOeQhm6 tPOuHMfYwzX+lIxp3Mb79Hg7KCaxjh025CY/aL3NOTK98uWcFtHLyfeZl4/R Iy4ax8bJXfLekYdTM/H2T6IUTrOa+wKoLvmJ/zP1IJOqoiOoPcX5VBUijKEq 9ObgzNlMrZIPluzU50zRWVXFbXOiLBUdMPnoz3XmqaQXXehGYwZPeX5Uxixt GXdims97djOcJ83pVUH6S6E206dHBSQYQSp0esrhmmdN61orKKQiWRSabX0T WVOEwLwOtrCHbZHidqRmyCT2TIRsGKEaTNnQjra0r5ySRVl52i/xzowW4x1s e/vbwYYRSSIFbph0qiLnLre6183udrv73fCOt7znTe962/ve+M63vvfN7377 G0OqNlWd5vzvghv84BLeF2ceq69vIfzhEI/4Ibm26zIfOGm7lrjGN85xJwlP d2rqmm5B3vGSm/zkFf/2HrBX1dgErhzlMI+5zOFSX/vWfOY4z7nOwXJzhKbU VrAOutCHTvSiG/3oSE+60pfO9KY7/elQj7rUp071qlv96ljPuta3zvWue/3r YA+72MdO9rKbXenbvkvPL1rYAMxDRM+Mu9xFLPe62/3ueM+73vfO9777/e+A D7zgB0/4whv+8IhPvOIXz/jGO/7xkI+85CdP+cpb/vKYz7zmN1+tAwJm7TD9 edFFfPbSFx0f9Ei96lfP+ta3/h6mj73sZ0/72tv+9rjPve53z/ve+77pIjsN 6HNqbBG5defIT/6EjX8aBoKPh87PuKZApPzqWx+5zD/Nx/da6shejOQLyf7/ 9cdPfm2K3zEUB7M/0++Q85f//fBPnfsBo3BqI7nhNWbI/OPP//4/af9/EXCG ZmgC2BAA6H8ImIDAcYCHxIAK+IAQyBcOWEgTGIEWeIFxUYEopIEY2IEeaBYc qDNAJzoP84EmeIJrEYK88nZ3Jy7Hh4IwGINaoYK8MoIv9oIymIM6CBU0qDo9 uINAGIQ98YOoQ4RCeIRIKBNG6BYF6GeMZmpp9zLUl4RUWIVDOIWOUX831kZb llVDgoVWGIZiqIRg6BfsF2brpS/v5Q/P1iBlOIZwGIcnsYRrsX3nxVfPtFec ZVcG+IZy+IeA2BF0mBbRVycqd0bmhYMr4oeB2IiO/zgRg4gW9eVFhdiHiviI mJiJ03eJdTGJbWeHODR6JaiJpFiKJ8KIeeGJjNRVXJYmLJh3LmiKsiiLkXgW qhhVoLhoQhd8s9iLmViLINh2BmVOEyR964SKvpiMYgiMZVGJHlOMGDRBubSJ yliNf8iMZZGLnJaLeyhC24aN1hiOCgiOY8F+0nGG/YUsbXiKnCiO7giD5DgW 9ZdlDteF68iO75iPRxiP5ahpjjIbTRh+yKiPBOmB/KghB1mQColzCYkhDbmQ EHlyDykeExmRFqlxFQkYNsiLF9mR4ziQqPOKdReLHlmS/peRAbiLo2iSLPl+ KKkaL9mSMklvMYkaNTmTOP/pbjdpFQH5j04YhTPUjjk5lBu3k1ShhQWBlCqW dkZJlE4pbU0ZFWdoMfmFX5UlkEL5lFpZcFEJFdoINB8Ccpf1jSC5lWb5b13p FM5IVWdWVjZliWcZlxKXlk1xi+S0JUBJl3K5l8all0thl2wXOFx4aEPHkXx5 mPzml0oBmL3kD2SjHiLZgv2QlYhZmeummEnBmIt1QaZWmCvpgaDwCpYJh5iJ FJrZT8VXlvDnDqTACZxgCZWwCqM5hqV5FGt5U20ZWrg2Iao5frHACgzQBo5g CbBZCa4wm8vYm3OhjU70fXroWBxDlpSJfL85CpAACWxABIfgmpwACsiZnNMJ F+b/SB3o+F5e6IbhiXOxYJ3XCQmjwAqx8J1yWJtIMY/mMphIiZX895uncArX +Z7xKZ/XqJx0UYCE9pMQQZ/Yxp+nMAr9CZ8C2ogKOhcTCm3reZ0PGqARKqEE ujoVymu/yQqs0J//qaEb6ogfSnMdGnEecAEqwAaj4KCnAKEn+osrqisb+Zkx 5wEZUAEVAAEkQAMiaqI1iokpahaRGXckKXMV8AAP4KMZ4AFFOotHahY5mp7/ FgJNGgFSOqW9WKVtAabI5QFd6qVfeqNFiKZmuqbbJKYpqKZsqhMhygqyEKdl 4aY90ZOvshhRiKerww7ssG/V+Z+jIJp2OhZ+qhP5CWTn//ks14aPXQEK3MkJ nwAY7GAO5XAO72BvIcqe/ymisHCoiAqnajGVHlOV4EI87UeqPMEO6vCqsKoO 7IAJlEAJxGkJmdAXktoK6VAOmUpvg+qen0qkohoWiYoTXxk+eXiH/tGOx3oQ PEoAvmoO1Fqt67AJtsqdlZoXklqcndAOsipvwfqpNFqsIMiqZ3Gba0kxqgqX TfGbJoABC3AA5WAOsfqq7AAK3skXoFCclVAJlrCv8Xah7Qmg5vqmWJqCwthU aRIzgbaqCVsTDDoKWrABGYAAgYoa/VoJkwCw3SlvHsABWuCf7lmuB6sWz1oT pykfVbRZohixMRGsGaoaUxAFa/+wCP8asPPmARHwABJgAjNKrCebFilLEyt7 TZ3jinCHd0s6FARboqlRs1HQBEzABF7QBx87b03aAA8wASIwtBmIrsG4igXm SHOFPWX0skQRC2TgA3BQskK7F1IbBVAABU1AtUzQBFNgb1r6AFwKtnJRtDOx shz0H8bImzCLEiHanz7gAzQQt3KLt3dbt1FQuXuLb2QKuBQqts14iBJViBaH ql+YuCVBsDKqAyBgGjVLtUuQt5aruS0puDPBnDfjnAlkK/+SbJCaEww6rI4h tXhbtXoLuzgpuzIxntuVhtflrjcBAmSAoUH7u1MrvMIbBZdLvDNpvDJhn162 jlKGnjj/0aIy4J6QqxdVML2ua73YS5Tae7z+uKeXBrHhiwElUL58IbXXu77s y7m5gpkhwJeSmrX66xft6xUFTJABbAm1agmcoAoDTMD8Ox5X+sANsbEAu8AC TMF6ccBTkaTP1LQIMbeVq74V4aoZi5iSigmYkLOuKbAaLIERLB4TXBAiLLl3 K7mV2wMIEK4L4aqYWg7qsKl8CQJvcAeGoMIZ/MIQTLq7In6rK7x3W70jbMM3 kADmcA7ncK/3+sPTesJxOQU5gAQ8cAdv4MJK/Hkx7CQiQgU1W7c4PML5KxBz q8PqgMXVesfVOq08LJdTQLVFAARHkLpnbJNp3CQCIAVKULWU/0vCEuGqWhyr XnyWHtADrJu3cTzIjsHBL6GnUIigDyEATpDIw0u8YLwDOyC8l4zJmVzIYrGo 9igabNSoJyIFT5DKJzsFQrADP7ADQsDIqpwamswSplpdl1M6snyMTDyaUlu3 ppwDtvzLphHMK5GseKg7FgIinKl/rCy3vrxvbWy3TVC3PVCm0Awe0lxtnmuI Lnc5QLWbi5jMf4ELXFC1UVAF/OYBLmAETLDIz1zOhAzPZHG0qfp8u7sr0rAM vCAHlnzPEbACOzDK/oyQ22zAC6tLKJJdbKiLQWeYUHLQypAMxEAI/UxvHlAB ETADEa3GE90VAs0sJJLRAuHBcgfCTf/i0cmgDMsgDQb3vymt0gCNqBVNUxht MDZYTTqqIaCQC8Zw0znd02t6zihxtEO9HivNFv1KCbUQDE2tb4Dq1DoD1Seh rul8thIB1jRx1ZJgq7uwb5eaqULs1bpi1iZBu9W8V08zXE+YEHINE2htq2Zs b21NDr8K101c1VyBvBCVX4VrZtr802vRrZSQ1jrL1Zgq2OYQyYT9f4bNFdyL f0VWOsfT2OBx1bXq11xdx+fQxZmdM3udHO/byRqd1+B7GgGswhj81/TW1pma xZi92lDS2lEB3CEB2baKxLhdb6lt2b3t2/272ebs3FZhCgpc2gx83PXGDueA qZfN3Gnq2OD/McN9oQmSncT69g6ovdzcHdfQ3RcyTdN50QmTbXBdnd7yt95s ppLe7RXHSd+wK9xP4d/8zd0AXpf2HeAGDhMDzhQJjnNTMNIHvjMLHhkEB78+ +ckF/odQsM/2jIS/ab85GOEc4cpZkzL1KNo4OQV2u89C2KkYmgpVCOIaMcxW CV0Yx5QXLoZ9nLdQ4OAeaLr/eQqo8OI3PhXUPHLcV+SzbZI53rpQAIQgoAPW yQglS6MuToUwrh/r/Lljnc7M65EZ3roQLYOxQAONC7VxeOUxEtSFtXbIXJJ9 vARLEOYx6AElwAZw4ANkYLKkOeTBrearSFNFTYL5/YdLruI6yLMP/zACPjAK Hi7kg07RZDuMqCkkMq2kkxmRkioIX4DKO9ijTuq1RsrnPOjnHAXoaruQtEoJ mFAH3RyDIVDSUaqJaF5spO5zeFXQ+divkiAJ8Q2DmUsQPE2Ks14RYp3lbtmW Xf6OtJrWmJCDiF4BwU6lou6VtpsgzRmWR17t8pvrlrDrvX6CiM61FaCMw07s NU5tZyjjSR6Ofd3sc+6jTvq35D7tUNHZr8yo90iN7K7Akm3dFsijPgql5JyM 5W4Yr31oADnh6+6Lff3tHljSJu2jA1+NBR8UFc9uDe/vEYjoEVABsZ6PF98z 9I6AqsAJC6zxEBjuT6qQIe8TLV9ukoryD/+o8vJekC9PE+CtiQ68gzQ/8fp4 8zLR3pfuiNQADNyJCaCw8zIYAhkQ7z5PkEDvvht91HJIDb4wC8SJCZwAg9Lg DDotEDz69DY/8nsR9bV2DUbfwijoARqgDczgDNYgk2aPE3P/4CuB6A7QAdPw 9bFL9mqp8Ajvydtu96sD8U6fk3WfECKuhRY2mPpJ+KkD8QKfk4AaCp7w6F4h 4+NpuPmOuJDPMxxfAWLvkW2NDvtwCZjfFUXOjUKFzdGZ7J//JDxKARbgACtP +ZWdqZaf+ltR7NBYakME+7GvIWNOAhBgARagAaNvka6Kxaqd+NJ0XzWXXYf7 zsPvJBcKBypwARz/gA399ut0odtYvMfQ/1soUl9T3Zn4ff3EH+Xk62/PHu1v Edj1itlmP4J3df4Lm/5uBxAC9g0kWLBfPwEBFC5k2NDhQ4gRJU6kWNHiRYwZ NW7k2NHjR5AhRY4kWdLkSZQpK8Yi8wIOpFGxVM6kWdPmTYceIjxo8KACTqAo pXEbV45cOXPsIApMGNRpynr8AEylyq+egKlNsQJoqnCr1qwK9fkD4E9fw6sC 1K5dy49f16dx5c6lW9fuXbx59ZqMNQoSkRcMZO4lXFivhwwZKlR48CCCB8N6 QeUy1g0cUqURmUbmHJVq1athA3xlSHp02LFlz2LczNn1a9ixZc+m/VoB/xtG kMgUqN3b90XEOx0vzgD5901QnCxRqhVsW7jMElsfl5uWbb0A9foB2KdwH4B+ 2Bdq5+4dfD3yZjVOp97e/Xv48eWrZHfOwIuY8/VzRrxYeITi9jMJlOUoMdCS XS5iT0CnvvNnILK6y048ByEsbyuz1hKPogUZ9PBDEEMU0Sl2zEFqABFGVBEn nRojLsAVN0puOUkosYQTUDLqMMaTyKMqvOzcws7HqYDc6jMArKpoRx6bdPJJ KOMr0ahyzonySo9a7OkxLCsi8A9AarQkx42Y7LIjfbYDb7WolAwgzSLZlApJ Nycy80w889RzT5POoTIpPgNVKDieHDNOUIUIpP+kET70UIRMju5EdKK0NpyQ oUoxZUstS6XbB65JQxV1VDzrMxFQUrtssTFDQ50iijowOdCMKUCSNFW6bsV1 V1579W1Kc86JzlcesaGgUAAPDfRVKKBo4gtBcBxJV2JxorZabLPVNqgpj7Jy 2xGlaSYbC1qddIommGCi2ShqLelacFGCN15667X3Ij+PQvVe/aRxphmAn+FA 2WXTXYKJJtw9aV5+MbKOLbdAbXhiiis2FbOK45NmGWWUAVgaUafoQowsEFZY 3k8zHmkegQoa6CCJVZZ55l7rO+fmYWmuLRZWbjmm42VAPjcKIXbYQow0Tk6J YZ3R2lStiJuWempE/Qz/NmeqX+sLEkhEEcYZoREFxQ81sNhhhyAgpYnprB9i u224427v4m/l5mzr3PILlcBKEsljjB4IXjtluyMlvHDEE3cPWGEVJwxvmAab FBRMMKmkkjGfelvQTCPqHCL0FDzccdJLdy3fKk3HC3K9EfWAg0Ewx0TauDbn E841IcK9n9UcSrNOO0dXfXji4xpKnFOxLh4o1iUXtMUT7qgEk7ps15PI8xzC HsiGUuPKIuuXF398ifzluBtvGic/qFhOOYXr1p/fqScKEFFbrvDxrHCfCB3a v3+G4O57SxLe+gx4QIz8i2PK2MY3EGgTnrnPfaxwXqA8sJjGcAkv+esSeST0 /x3uKcSD5glhmvpBlgFyqIAPZGELA7CxZSzDGWBzoUq2Bgn3VZBP/VlMBWCU Fw5iyTSm8YpoTONBfYhGhTGrYROXF4dhKCMZQXMiSn7xCUjkTYd7WtViBLfB FfZqiEo8zQBNA8LQpPBhT4taFd1YPCiEwRbKoOIbR7KxYugicqHSkrkME0QV rVEA2BljCssIlu+Z8CymYZnLXAYzO0aSdFBYwhUIETZJdsRfAFMGMnqxxTzx MINf3AsgReQZOqURkXAp5CC3k6E5SSg7T4PYWzJ5y7hNwVnrwqUmFwgwGrpO OMQhJWFMGSJUfsYqrSyNEbNyJCSF0FNM7GU1KQYFhP9BQWnWpMjGpEhHTFpw fn6MzTGR+bQhbeeD2RuPOkmInjlFs1NLCSM37RkvXaYLCve0iDeTAc5QYQMD jCGnbMz5pP+V51IBSKiEHhZL8NWTnxPtFbqyuU2KNqQZ36wjosSVjQ/4sJiv OaiTtjckIWVHTUWa5yEjSs2MxlRU2DzYPmXqkE1yrKOC8uYyngGC9pTUSbuT k5uICrqUEhCmN2VqntC1hCUkrKkL8abHmhHOQEXjm814j1BNyilMWepzDwnd S6d6Vj7RlJdnFddGAToqaLzVPV7NGF3RetfZUDKqGM1oT5sRTFJVA6vUsSvF CotXxBomn2ttqj89FtPDNiz/soml7F2w2QRtnnWj/9zpRCebLUEKoI2VJa2H 9GrTxjZDp4Pl52er1UhHQrK0s53PYlF707Z+bKqurVZoR0tb4FLnspllql9Z m1He0iu5wWVuSE471c0+9q7LBRd1m3tdjURBn02dAh6i2NndSpR41sVueSei yygQV6a6hIIXaHFcppJ3P2PVHlidNkjWiNe8+33KZaPAVItilq9nle98jvqQ A4tFTf3YR0vdpl/+RrgmT13CbSnKXnVZOLEFjs9JtbdSdmIvSQ5uCIcl3Nw6 bKGm600XZttFWxO/p6H+K4uF1vkggfQDeJqB8Il9TBJM/MENUs2oRVcc3Bi3 Z4QM/2WnCN3J5PAseaFL/HGVaaIcSghCppRMwl6bm2TCOtOQrdyKPq5C4gcv 1cpr5ggnOIGJ+93TA0JYQpc1DOMej4qZCyHzVPyxHd5pipZQsyWbDQ0Syrk5 pqpAQQ1+YLLyghk2giSkmFlpaQyhMNABgK0jX4aQQ4eaI7K6UUwx0YUiwEAI A8azmveUTNDsuYhmfOZU3iLaEA+a0K4Wda8VgglJSIJ6FAXFHtxwBiMEIcKS dg2sba3KWV+a1lyR9TR9fW2JAFvYGY3EHNwghi6wGsl5zhOlVapQNGLqyWjc ioSIaG1sx5sh2h42Px9BBzTYIdw+ZvZvZrzQGafnU1KRJv+PeS3vH1NuOfW+ 5xrAUAYweKHK/faNh9uUzs9wz3txMivCe/0JLFsCzhPVJRfAwAVxY5fivjnw xRXM8YUkmMoeF3UmCsRwbgZYvRMnN6nGWlYR2lfQaHbIymleQ5DTzp5GZuya jR6qpx9d6thSK5ENHXVEYX3qW8cVCnBQBKsfWutY8m2huX72B14wAiywwbXH HqVOGwTUaKf7+hgTgQqM1Mpvj1LZD153wMetAg1owE/izXc9IT7wi19RBnZi +MP3XHGKZ3zlPRQIDChG71eXfOIob3nQy6cFJ0hBBo7+ecPQ977zVD28Q/96 moHiDUOIAQakjnrCyDyAC+7dm3j/L7q/w1742LJcIazQgqnjXi8eRguINY5C P/fe9cOnvr0wMYlJZC75nUfUvxkS8FdaqOD0DH71zS+q608C56fnfqCWnO52 ovs87TaPIQ1+fvxnSznTQ7vyN2jpZpo2V5K/8Uuz/DtAXkm0keu/9uOTamMm 7wuttjA7BKxAPomHhbCcUqs7/wsKc3tAANyKgxCNuIutubNAFMwTeCAF5cC+ 9eO6DsQJZxsxEBRAJnsQ/lGoWdK130pBH4ySeCgFS7icF4TBBnySGVwmAOQz S3u/Jpu+H4zCJpEHFlQ6wItBGUSnc1snaXLC8KA/JrM/8pNCMizDoDrCPfm3 sgK/GsvB/wIsOjQ0QzmcQ6DAwrqwuJRivo1TjY6jQz/8Q72ww7pouaSSOd2D QkBMREUMCkG8Q6Gbsh1kvUecuUWsREscnPLzvDi8RE7sxNLYxKxpRE8cRV8T RQHxO1JMRVXsQ9UpwU/LxFWMRUU0xfmipR6URVxMRVoUkV3MRV9krl4EkWD8 RWIkrWH0kGMsRmWcLlB8tUkkK7F6RkRcRmqUw2TEiUMMun3QMfHIxjGsRnCc w2u0CebDlHj6nnIMHlgMR3aEvXGsCe8bj3NMiHicxna8xwN8x5nwQkmckwHk QqLzimbER4LcvnUUlGoLQLVYQnUsSIfMP31UiYRkQmpbQgmEGv8d0zWN3EiO 7EiP/EiQDEmRHEmSLEmTPEmUTEmVXEmWbEmXfEkBuAd7mEmarEmbvEmcpMl8 gEme7Emf/EmgDEqhHEqiLMqhvEWf08KJjDZmckWDAA9Pi0qpnEqqpEp+0LGq zEqt3MqBuEp+4EqwDEuX8UqxLEuwJEuzTMuqREu1bEtPY0u3jMuuxEq5rEu4 rMu2vEu8TEu93Muy7Eu/DEvADEyuHEzC1ErDPMy1pEvFFEvZ2pUkhDaXcpta s8EdHDSvNMqOzEzN3EjO7EwezEjQDM1bG01bFE3TZCPUTE2IWU3WxMjSfE3Y lM3WjE3Z/EzaxM3bdM3X1M3e5E3WREr/UYnMpTyk4iyxgZyW5BSJJGvO5QwJ 5zxIiXxOW6HOj4hOY7JOj8DOUtLOjuBOXjE3flyKrBhPSsyLM8tO6YQK/OrO 9USJ9HRPwojPQPROjqBPILLPjcBPMHrPHmnP+vRPk+DPu4hIlVDDTiGNevQc ANWLeSgMAr2LB53PBs2LCd2LCLWLC12+CsWLDUXPDpVQCA1RDR3RgIyLD8WL DO0t50Mp4CGNdJyIFEWUGRWUGg2UG+WTHN2THdWTHs2TH8WTIOUVQgQel/M9 mHtIJV3SFfk51otGEmVSKZ1SKq1SK6Uo1Wu98SCxn9uUE8XQZ+zS62DQDtXS 18jSR1wjLk1T/y+tuDBl0zGVCKCLxONA0wZVU0oRDzztDTuFUg0hU0mM0mZ7 0zttU9CB0z+tjT4VtEQlK0L9UgotU0SVxqAr1DjVlgSrh21M0u7hRgT7vSBJ pdnI1E3NnVBVppYyRFAd1VXVVFB1th17OVOFVUjNC1J91XMcMUfdRqNaVdq4 VZij1V3lRmH91VYt1U0rVkxB1t7zRs4A1lnN1Vh11WCV1lq9C2hNVmv9VE51 1sjIVjkRVd/BVXHFlj1UD2fbtABCIYkRID6EpmeTjXM9i3RdJCTR1ebLOD1t 0VGFPj6s1zJC1Q/T14AFjV/1V3TN1UCDV3w1x89IiBiFjXk91R+x1/9ylceH ZdhYdY2JBViN7RSAPbcfudZBRFh6VViLFViHzbiUNVh5NVmK5biPHVeWdTKC fdnPSNhoatl47dSanVlsETg3hCcGg5onFCBQIY+3GNqtKE2SnQuhJYsou8pP wbUvtLVG/b42BKAFjYyoPY+oKFqrTSOnfYgZa9qsnbTwG9qwrVqCI9u0FaF5 vEEbo42vnVqxfVu0lUZYW8gkidtmW1upJVq3nT+sDVOqNVog6drUE1ywTdyx 3dsyhVy9PVxBnU/HxdvCvdq/DdPtuLW3pdsclKUzzVzCVVy49dzOrdzOvVxS AcPvQMcNid2uMCF2LTHRoN29fdq5gN2wADr/3T3cluJH8ySpqVgn2V2I4O3c 4V03w2Xe2vDd5K0/v721NfXH4oUN6YXY2a1M62VQf9zd6D1e6p2y5f1eaFTe 8nTeNzQm8g1D7lXfioTe9KVe8Z2N7TVf7+XbpEKj7OWM/AXe/Q3RI2U34d2W /00N7kEiMprMr4BXdS1dAuwUBU6jii06TGPIwJ3g7hlcCJY+B67MC56NBPbg nZWOylwlEmZfCjbhmtUMEeZUteXgdX3eEY6ICj5ODGXhDrZhGfYdF77hGQbI HuZcId7SeXDVhdXgzuDhGjbiH56QJN7GJX7hbKnH1FCPd3q3cwPdGI4+2cBi slAPhtViilwlHd4L/zF+13s142hjyjYGYddYYzKOYxieXxWeDTrmWT4kz/nN WTnmjD0uWDDWnTG22DyGjUEu40B+k0Mm5D6OjUW2Y7Kak3540DTWi0lGEjde 2UuG5E72FRFUE7jI4t5RpMncPT9j17YViMEtpyIh5RrW4lYeWoWEY8vU3lgm I1OW26K15TOOtlp+ZYPaZUPq5SD5ZWLGYDx+Y/w15lJ+5GT+FGBm5kFK3GrW ZfCQ5ZiT5mF+wm7u40wGImieZTlRZnAWC2/G5mU23m3m5XVG5wJEpTrOZXce wWOOZ2puZ4z9V3ZOZ17xPmQ+N1h6X589od/tXjE0DIGWZhFSaGlDYyYW5P+t 1cGB1t8xA0ABXujIaOhIxmgmeuCJnuOKlqWL3uiQTmj5FdCb8Gg3RmkgjuRx xguX7j2Y7uGXhugwLulwzumVBh2mHWmG5ml1/uib1h5bPmpfGc+LZliWctR6 SKLpTeUm5uCLnrJ36zN7rmoivmqgy2qN3hAu5up3cuSPxuoGPuOZVlEe9mqx Tus3/mq4blyrduiHDmZzvmWqJmsoww63xuuelj655mjMreuzHuykhei1vsO2 tmu0tj+YRmxs2d70EOxcfeqwUt+pdRPa1WbkHcBOLmAyIl4ndufPrmy06N/R ZmHRJmz3lT+uQG1z5Oy5/or/dQ3KfiXLpm2Otm3/1XbtUnpf2pVteeRtuCBu myViz6Ze5A4S41a3w/5tll4b4V7f6H7u4qYQ61Zu02Zu3U5t7JZb427t6TYp 08UQfUjUh3rf0Pnm7kDbbPba846+P4XvV15DomZcCJ1v1ahvrE1q7cbB/rFv gKZrAX/eDAErAgeSOXUp/TZw8UujBMevBR+SBN3f+K7qA+fcCa+0vwXwMupw 8zhw0t3gDZfw9Fbw/75vDxfxCo+Nu0Vx//5wFu9iV86eB9/hiubwFKfwFc+e 9v7cG+fcpeXnXplYp45VGBUSEWNwEDvrZ4XZJEePJ09YJWG+iI1yQIbkERMx K8e4kRVZqghljpXye+3y/yo/2Rf9XX492C2fci9Xc0ux7TR38zHnYzTnZDmf cjEvcMNA8jO3ijh3bgtG1UHH2TvnckGvc5dzV1pu89gA9HtddD0ndCRN9EPP FkKcdNbLw5QSIHUF9Ub+c3IVVVEX7270VW8l9Wrl9EuP1l5N0lOvjU039ZXS ViMtRF9l1Va39RE+UlR/9QiW11IX2Fm/uMiU1WGX2GI32GNn8stWklnn9Wh1 9WePdVNV9lFn9Wr39WrF9lC/9W2HzEmlVKA797S1Dt51REu9VHXP7PF4VDdt 90Z99y0duvtyXa8tdwC197tGCyhd97sQU3fP2gafMn9XVH7fkIQvK3OLd30f UWx6F7qG11MtrNSIj1Q/rXeDD/j6yngwnfh+73h8Z1SBv1KUT3mVX3mWb3mX f3mYj3mZn3mar3mbv3mcz3md33me73mf/3mgD3qhH3qiL3qjP3qkT3qlX3qm b3qnf3qoj3qpn3qqr3qrv3r3CAgAOw== --============_-1041151140==_mr============--