From brideout at haystack.mit.edu Mon Nov 3 09:44:34 2003 From: brideout at haystack.mit.edu (William Rideout) Date: Wed Jul 21 10:29:05 2004 Subject: [gps-developers] GPS movie for Oct 29, 2003 available Message-ID: <3FA669D2.1020002@haystack.mit.edu> The first GPS movie of our ongoing solar storm is now available on my web site. This movie covers only Oct.29, 2003, but later days will be added soon. Note that this movie was processed with corrected svbias data, and looking at tec versus time plots shows that the data has improved significantly. My web site is http://www.haystack.mit.edu/address/brideout.html. The raw tec files are available at: /usr/oasis/experiments/gps/algo/tec_302_2003 Please let me or Ching know if you need help accessing these files. Bill -- Bill Rideout MIT Haystack Observatory Email: brideout@haystack.mit.edu Phone: 781 981-5624 From brideout at haystack.mit.edu Tue Nov 4 12:19:33 2003 From: brideout at haystack.mit.edu (William Rideout) Date: Wed Jul 21 10:29:06 2004 Subject: [gps-developers] Re: GPS movie for Oct 29, 2003 available References: <3FA669D2.1020002@haystack.mit.edu> Message-ID: <3FA7DFA5.4080008@haystack.mit.edu> The movie and tec files now cover Oct 29 and 30th. You'll see in the Oct 30th movie that 120 TEC is way too low for the upper limit, since half of the US is pegged at that limit. I'm redoing data for both days by manually checking data (which will increase coverage), so I'll pick a higher upper tec limit then. Bill William Rideout wrote: > The first GPS movie of our ongoing solar storm is now available on my > web site. This movie covers only Oct.29, 2003, but later days will be > added soon. Note that this movie was processed with corrected svbias > data, and looking at tec versus time plots shows that the data has > improved significantly. > > My web site is http://www.haystack.mit.edu/address/brideout.html. > > The raw tec files are available at: > > /usr/oasis/experiments/gps/algo/tec_302_2003 > > Please let me or Ching know if you need help accessing these files. > > Bill > -- Bill Rideout MIT Haystack Observatory Email: brideout@haystack.mit.edu Phone: 781 981-5624 From brideout at haystack.mit.edu Wed Nov 12 14:48:06 2003 From: brideout at haystack.mit.edu (William Rideout) Date: Wed Jul 21 10:29:06 2004 Subject: [gps-developers] Comments on Pat Doherty's receiver bias algorithm Message-ID: <3FB28E76.20303@haystack.mit.edu> Anthea, I think what Pat Doherty proposed for an algorithm to find the receiver bias of an isolated receiver is coupled to my suggestion about how to find the optimal mapping function. Here's why: 1. Assume Z(el) is a perfect mapping method (for the moment). Also assume, as Pat did, that at a certain time of day the ionosphere is uniform within the field of view of the isolated receiver. Then we have: True Slant TEC -------------- = constant Z(el) for all observations. Now if True Slant TEC = Measured Slant TEC + recBias, then: Measured Slant TEC recBias ----------------- = constant - ------- Z(el) Z(el) and the best fit recBias can be solved for which minimizes the differences in measurements. 2. Now assume Z(el) is not optimal. Then what you are minimizing depends on the error in Z. However, there's no reason not to allow Z to also vary to minimize error even more (for example, by having an adjustable scale height). Again, the point I'm making is that these errors are coupled, so I think a two parameter fit is justified (and would still be fast). This would also pull out effective scale height information on a global scale for quiet conditions. 3. Finally, if I drop the uniform ionosphere condition (as as during the day), but assume only that a given mapping function is valid for a given time frame, and assume I have enough data so that there shouldn't be any difference in vertical TEC for low elevation measurements versus high, I can still use the above technique to get both rec bias and optimal Z. It may take more data due to the added noise of the non-uniform ionopsphere (or would have larger error bars for the same data size), but I still think it would yield valuable information about effective scale height that we aren't pulling out now, and could provide a better mapping method that the single one we use now. Bill -- Bill Rideout MIT Haystack Observatory Email: brideout@haystack.mit.edu Phone: 781 981-5624 From brideout at haystack.mit.edu Mon Nov 17 11:50:16 2003 From: brideout at haystack.mit.edu (William Rideout) Date: Wed Jul 21 10:29:06 2004 Subject: [gps-developers] Revised version of March 21, 2001 TEC movie Message-ID: <3FB8FC48.30004@haystack.mit.edu> John and Anthea, In an earlier email I said that we had a bug in the code that corrected for satellite biases; and that this bug was effectively a source of noise. As an example, I reran the movie that you asked for once that shows tec over North America on March 21, 2001. As you may recall, you wanted a map with TEC from 10 to 30 TEC, and that map looked quite noisy. See my web site (http://www.haystack.mit.edu/address/brideout.html) for both versions of the movie. Note that the newer, less noisy version starts two hours later since I only re-processed the 21st. Some of the plans that Anthea and I have to improve receiver bias and mapping function corrections should have the effect of reducing this noise level even further. I think a good metric of our noise might be the standard deviation of the TEC over a ceratin area of North America during a quiet night. So I think that our request for funding for improved TEC processing should focus on: 1. Allowing automated 24x365 processing of TEC data. 2. Minimizing noise of TEC values to see ever smaller TEC features. 3. Longer term extraction of vertical information via John Holt's approach. Bill -- Bill Rideout MIT Haystack Observatory Email: brideout@haystack.mit.edu Phone: 781 981-5624