Presentations

From Projects

Jump to: navigation, search

Contents

  1. 2009 Fall AGU
    1. Abstract
    2. Travel Info
  2. 2009-09-18 EPSC
    1. ViRBO
    2. Travel Info
  3. 2009-08-03 ESSI UMBC
    1. FAST
    2. VxOware
  4. 2009-07-21 NASA HPDE Senior Review
  5. 2009-07-13 Wolfram|Alpha Review
  6. 2009-06-24/25 GEM
  7. 2009-05-01 Space Weather Workshop
  8. 2009-05-18 NCAR Visit to GMU
  9. 2009-05-08 GSFC Seminar
  10. 2006-12-15 Fall AGU
    1. Abstract
    2. Transcript
      1. Slide 1
      2. Slide 2
      3. Slide 3
      4. Slide 4
      5. Slide 5
      6. Slide 7
      7. Slide 8
      8. Slide 9
      9. Slide 10
      10. Slide 12
      11. Slide 13

[edit] 1 2009 Fall AGU

[edit] 1.1 Abstract

Solar wind density influence on the efficiency of geomagnetic response to the interplanetary electric field

Solar wind density has been suggested to have a strong effect on Earth's magnetosphere. Elevated solar wind density often exists during intervals of enhanced magnetospheric activity, which complicates the analysis required to make this conclusion. In contrast, statistical studies have consistently shown that the independent correlation between solar wind density and magnetospheric activity is small. These two seemingly contradictory results are resolved by showing that the solar wind density affects the interplanetary electric field geoefficiency in a way that is not captured by the standard correlation or epoch averaging approach. The solar wind density influence is quantified using statistical approaches that differ from the standard treatment, including (a) data-derived impulse response functions and (b) the ratio of the integrated response of geomagnetic activity to the integrated value of the interplanetary electric field during geomagnetically active time intervals. We show that (1) it is the solar wind density, not pressure, that is the mediating factor in the case for geomagnetic activity quantified by the $D_{st}$ index and (2) the geoefficiency depends on the latitude that geomagnetic activity is measured.

[edit] 1.2 Travel Info

[edit] 2 2009-09-18 EPSC

(Invited)

[edit] 2.1 ViRBO

ppt

The Virtual Radiation Belt Observatory (ViRBO) and tools for radiation belt science

Weigel, R. S. (1); Kihn, E. A. (2); Baker, D. N. (3); Frieidel, R. (4); Green, J. (5); Bourdarie, S. (6); Faden, J. (7); Zhizhin, M. (8); Mishin, D. (9)

ViRBO (http://virbo.org/) is one of the domain-specific virtual observatories that began operations in Fall, 2006 and is funded under the NASA Heliophysics Data Environment program. This presentation will cover three topics: (1) the data products available or exposed through ViRBO, (2) in-progress developments of data products, and (3) the future of domain-specific virtual observatories such as ViRBO within the international data environment. Data available through ViRBO include measurements from the SAMPEX, GOES, POES, LANL GEO, Polar, and GPS satellites. A number of new data sets, not previously openly available, include measurements from the HEO-1, HEO-3, CRRES, SCATHA, OV1-19, OV3-3, ICO, and S3-3 spacecraft along with scientist-contributed model and simulation data.

[edit] 2.2 Travel Info

uMFcIv <a href="http://oqhexbsejqcm.com/">oqhexbsejqcm</a>, [url=http://oibloeqvjudk.com/]oibloeqvjudk[/url], [link=http://ldsldquxgnid.com/]ldsldquxgnid[/link], http://cnvjjpuxpamd.com/

[edit] 3 2009-08-03 ESSI UMBC

See two talks at http://essi.gsfc.nasa.gov/schedule.pdf

[edit] 3.1 FAST

abstract | ppt | swf

FAST: A high-performance time series data server and data base

R.S. Weigel, D. Lindholm, A. Wilson, and J. Faden

Abstract

FAST (Fast Aggregated Server of Time series) is a high-performance time series server and data base that uses the OPeNDAP protocol and related software libraries. FAST was developed to respond quickly to requests for long time spans of data with very little network overhead and server-side processing. Data are served from a cache that was pre-generated, typically by aggregating granules from a remote data source. The data base was designed specifically for time series data in that it is optimized for requests where the longest dimension of the requested data is time. Scalar, vector, and spectrogram-type time series are supported. The user interacts with the server by requesting a time series, a date range, and a filter to apply to the data. Available filters include FFT, block average/min/max, and sort. Filters may be applied to the data on the client- or server-side. Also supported are constraint expressions, which allow the user to request data from one time series when a different time series satisfied a specified relationship. An example is provided in which FAST is used to allow a 1-second time series that spans 10 years to be browsed interactively across all time scales.

[edit] 3.2 VxOware

abstract | ppt

VxOware: Software for managing scientific metadata

R.S. Weigel, M. Zhizhin, D. Mishin, D. Kokovin, E. Kihn, J. Faden, and K. Borne

Abstract

The recent Heliophysics Virtual Observatory (VO) effort involves the development of separate observatories with a low overlap in physical domain or area of scientific specialization and a high degree of overlap in their metadata management needs. VxOware is a content and metadata management system that is intended for use by a VO or an entity that manages scientific metadata. VxOware has features such as system and user administration, metadata editing, user-editable content, version tracking, and an API for communicating its holdings and responding to search requests. Metadata can be contributed using a web browser or the upload API. We give examples of managing Earth Science and Space Science metadata and connecting a federation of VOs using VxOware.

[edit] 4 2009-07-21 NASA HPDE Senior Review

[edit] 5 2009-07-13 Wolfram|Alpha Review

Posted at: http://computationalanddatascience.blogspot.com/2009/07/wolfram-alpha-review.html

Wolfram Alpha was released with great fanfare recently. Their overarching objective is very broad:

"Wolfram|Alpha is the first step in an ambitious, long-term project to make all systematic knowledge immediately computable by anyone". [1]

This can be compared to Google’s corporate statement:

"To organize the world's information and make it universally accessible and useful." [2]

Many of the reviews deal with making side-by-side comparisons with results from different search engines. Even if Wolfram|Alpha succeeds in returning better results for a limited subset of data, it will fail to make a relevant fraction of “all systematic knowledge immediately computable” because doing so requires expert humans:

“… as the physicist sat, exhausted, immersed in the minutiae of food science. On the computer screen before him were raw tables of information from the U.S. Department of Agriculture, containing data on 7,000 foods, from blackberries to beef. He and a four-person team were "curating" the data, readying it for a new kind of online search.” [3]

Remember the old Yahoo where everything was placed into categories by humans? This is a much simpler problem than data “curation” described above. In fact, the act of organizing data and preparing it for systematic computation is a significant part of what a researcher does. Every field of science has their own way of organizing data and this requires specialists to organize and validate it.

The fundamental problem with Wolfram|Alpha is that it requires so much human intervention. What will happen a few years from now when all of the USDA’s data tables are updated or their data base format changes? Wolfram|Alpha needs to find and pay a team of experts to update and validate the new data.

My assessment of this fundamental flaw in Wolfram|Alpha’s approach is primarily based on what I know is required to deal with satellite orbit data. You can find satellite orbit data displayed in a comprehensive and clear manner at Wolfram|Alpha [4], which on the surface seems much more usable than what is available elsewhere [Start at http://sscweb.gsfc.nasa.gov/cgi-bin/sscweb/Query.cgi and then work your way through the menu to get equivalent data]. When a new satellite becomes available, who will update Wolfram|Alpha? And why would they be motivated to do it? If I have a detailed technical question about the implementation of an orbit calculation, who do I ask? If, 5-10 years in the future we see 100s of small satellites launched per year, who is going to update the Wofram|Alpha database?

The most basic problem is that if they are going to keep up with every special data set from every scientific subspecialty that has data that is computable, they need a community that will contribute. Wolfram’s approach is not community oriented, however. If I have an article about physics that could fit in http://scienceworld.wolfram.com/ or Wikipedia, I would choose Wikipedia. Many would justify this choice by saying that Wolfram Research is a corporation, and why should I give free contributions to a corporation if I don’t get anything in return? I have a more pragmatic reason; Wolfram Research is a corporation, and corporations come and go. When they go, their intellectual assets tend to follow. Releasing intellectual assets from a company that is dying or being swallowed requires money, which is not typically plentiful for a company in such a state.

[edit] 6 2009-06-24/25 GEM

See three links at http://virbo.org/GEM_NGRSC_2009

[edit] 7 2009-05-01 Space Weather Workshop

pdf | odp

Presentation about http://swxcontest.gmu.edu/. Presented by Brian_Curtis.

[edit] 8 2009-05-18 NCAR Visit to GMU

ppt

The UCAR team will be visiting GMU on Monday, May 18. As a part of presentations to the UCAR team I would like to invite a member of CDS faculty to make a short (15 minute) presentation on the academic and research programs of CDS including your department's activities related to atmospheric and Earth sciences.

[edit] 9 2009-05-08 GSFC Seminar

(Invited)

ppt

The Virtual Radiation Belt Observatory (ViRBO) and the Future of the VxO Environment

ViRBO (http://virbo.org/) is one of the domain-specific virtual observatories that began operations in Fall, 2006 and is funded under the NASA Heliophysics Data Environment program. This presentation will cover three topics: (1) the data products available or exposed through ViRBO, (2) our experience in developing a VxO over the past three years, and (3) our perspectives and predictions about the future of the VxO Environment based on this experience. Data available through ViRBO include measurements from the SAMPEX, GOES, POES, LANL GEO, Polar, and GPS satellites. A number of new data sets, not previously openly available, include measurements from the HEO-1, HEO-3, CRRES, SCATHA, OV1-19, OV3-3, ICO, and S3-3 spacecraft along with scientist-contributed model and simulation data. Data are served in a number of ways, including a basic FTP site and an OPeNDAP server. As part of this project, we have developed or extended a number of existing software codebases. These codebases have cross-VxO uses, and we are developing them to be usable by other virtual observatories.

[edit] 10 2006-12-15 Fall AGU

ppt

[edit] 10.1 Abstract

http://adsabs.harvard.edu/abs/2006AGUFMSM53B..06W

Magnetospheric activity during undriven recovery

Weigel, R. S.; Wiltberger, M.

A significant fraction of magnetospheric and geomagnetic activity is a direct result of variations in the external solar wind energy input, which fluctuates on time scales that are similar to the time scales of internal magnetospheric phenomena. To isolate the part of the magnetospheric activity signal that is due to internal magnetospheric processes, we consider the magnetosphere during undriven recovery. A recovery event is defined to occur when the magnetosphere is first in an excited state after which time the solar wind energy input is suddenly reduced for at least thirty minutes or longer. The statistics of these events are compared with that from non--events. For the events we find power--law behavior in both the lifetime of bursts above a decay curve and the power spectrum of geomagnetic variations. The functional form of the probability of large changes in the ground magnetic field is also found to follow a power law. The relaxation time for the events has a broad distribution that is not correlated with the initial activity level, and the power law exponent of the undriven recovery events is near that of the non--event intervals. The results from the actual events are compared to magnetospheric and geomagnetic activity signals extracted from MHD simulations of relaxation events.

[edit] 10.2 Transcript

A partial transcript of what I intended to say in the talk.

[edit] 10.2.1 Slide 1

Today I am going to discuss a special magnetospheric state: that of undriven recovery. This week we have seen a number of sessions that cover "nonlinear" magnetospheric phenomena, for example, the role of "preconditioning" mechanisms were covered. In this talk I am going to consider a special magnetospheric state which is when the solar wind has pushed a lot of magnetic and plasma energy into the magnetosphere and then this driving force is suddenly reduced. I will then look at a large ensemble of ground magnetometer records in order to assess the influence of the solar wind during undriven recovery and to better understand the internal system dynamics that are not attributable to the solar wind.

[edit] 10.2.2 Slide 2

Dynamical Systems Approaches

There are many ways to study a dynamical system. In magnetospheric physics we look at such things as filter models and epoch studies. The framework under which I view the problem of ground magnetic activity is using dynamical systems framework, which can be used to both understand and predict time evolution of a large class of systems.

When we talk about predictability, to determine how much of a in-situ signal can be causally connected to a time history of solar wind measurements ... this has a long tradition and we are at the point where we have reached a near limit of redictability and there is an interest of better understanding the intrisic dynamics that are not trivially related to the solar wind, and cannot be modeled by a circuit-type set of linear ordinary differential equations. 

 x dot (t) = F(x) + S(t)
 x(t) = int H(tau) S(t-tau) dtau
 x(t) = x(t-1) + F(x) + S(t)
 x(t) = sum h_tau S_t-tau

There are a few points to make about this represenation: (1) It does not capture very nonlinear events (2) It does account for dynamics - you often hear discussions about "coupling functions". Those coupling functions are derived assuming that you can drop the time history. It is not clear what this will do in terms of model bias. Show visual of removing these terms.

Now, even if you do keep these terms, you only get 50-60% predictability. Why is that? To better understand this, let's look at system in its most simple state.

[edit] 10.2.3 Slide 3

Motivation

  • Understand the non-predictable portion. In the auroral zone, we can predict about 50% of the variations on the ground based on solar wind measurements alone. What about the other 50%? What does it tell us about the system? At this conference we have seen many studies of such phenomena, for example, there was a session on the role of "preconditioning" on the resulting geomagnetic activity.
  • Understand the system in its most "nonlinear" ionspheric state. During active times, the ionosphere lights up and there are order-of-magnitude changes in the ionospheric conductance. On one side is the expansion phase, but equally dynamic (in terms of how fast the magnetic field decreases) is the initial part of the recovery phase.
  • Understand signature of midnight sector. We usually talk about "substorm current wedge". By comparing dynamics with EEJ, we can look for a unique midnight sector signature. (Refer to image and compare the fact that average EEJ and WEJ perturbation look similar, but this hides some major facts, such as how dynamic they are.)
  • Provide a reference for data-assimilative approaches.

[edit] 10.2.4 Slide 4

Relaxation

Recovery can be driven or undriven. In the magnetosphere we have the advantatge that sometimes the input shuts off. This gives us an opportunity to isolate the solar wind component. Another way of isolating the solar wind component is to subtract off the linear prediction, which is an approach I will take later.

I also want to emphasize the importance of dynamics - in correlative studies of coupling we typically look at single hour intervals. But a given solar wind perturbation can have a lasting influence of many hours. Some of the results from coupling functions may just be artifacts of chopping of time influence.

The recovery phase

  • Opgenoorth et al.
  • Weimer et al. Weimer, who developed an epoch study of substorms ... an

objection to this result was that ...

  • Kamide et al. Comment
  • Kamide et al. The two component electrojet - used Assimilative Mapping of Ionospheric Dynamics to get a "snapshot" of convection patterns. In comparison, I will provide an average time evolution of the pattern based on data without interpolation. In addition, difference is that I will use solar wind data in order to remove the "directly driven" component.

[edit] 10.2.5 Slide 5

Data

The data set is comprised of measurements from the FMI set of ground magnetometers [ref] The solar wind data is from the ACE and WIND satellites. The approach I will take is "stroboscopic", that is, using the Earth's rotation to sample a region once per day.

Note that on the scales in this plot I am looking at magnetometer perturbations from their quiet day values. It is very important to note that I am not looking at indices, which have the problem of not always being in the same place In this study I wanted fine-grain knowlege of the magnetic local time.

T5ZUrj <a href="http://npzkczznaamj.com/">npzkczznaamj</a>, [url=http://tvlgzfzlsbkr.com/]tvlgzfzlsbkr[/url], [link=http://juwmloiyudwe.com/]juwmloiyudwe[/link], http://qwwwlbyjgqpf.com/

[edit] 10.2.6 Slide 7

Event statistics

In this slide I show the results from using the above algorithms on geomagnetic measurements G and -G with a threshold of 300~nT. The top panel shows the number of events as a function of magnetic local time (MLT) for G and the bottom panel shows that for -G. The largest number of events for G is for the EEJ region, as may be expected, and the largest number of events for -G is in the pre-midnight sector (and not the WEJ region, so right there we note a lack of symmetry with respect to the electrojects.) The number of events is about double for the WEJ than EEJ region. It is important to compare EEJ and WEJ regions to make sure we can compare between convection and substorm phenomena. Now let's zoom in on these locations.

Compare these results with those of substorm onsets (show image from .pdf file).

[edit] 10.2.7 Slide 8

Events in WEJ vs. Events in EEJ

The results of this slide are the easiest to explain ...

[edit] 10.2.8 Slide 9

The last slide was easy to interpret, this slide is not. I have sorted the relaxation intervals into two categories: high and low, corresponding to how strong the solar wind driver was after the peak and I have keyed the data on when it crosses threshold in downward direction.

What I expect is that if the solar wind continues to provide reconnection energy, the relaxation would be slower, because there would be continued convection. Instead, what we see is that when the current systems that drive this perturbation want to relax, they do, and they essentially ignore what is happening in the solar wind: it is as if reconnection efficiency is suddenly set to zero.

The same thing happens in the EEJ region. How does on interpret this from a mechanical system perspective? It is quite easy if we think in terms of dipolarization and the storage-unloading metaphor. Think about the tipping of a large rock vertically. You get it vertical and it rocks back and forth and then crushes you. You try to resist it with the same force that you used to push it up but its momentum overtakes you.

Pulkinen: Larger IMF drives relatively lower activity. In her case she was looking at extreme events, in this case I am looking at typical events over a limited portion of their time. The result is similar: there are times when the magnetosphere just wants to ignore the energy input.

We usually talk about perturbations in the high latitude magnetic field in terms of a directly driven and storage-unloading component. This result is in line with that, but it makes a very clear picture that the first 60 minutes of decay are decoupled.

[edit] 10.2.9 Slide 10

In this slide I compare the decay rate as a function of thresholding.

As I mentioned earlier, the latitude and longitude of these perturbations are right under the "hot spots" of precipitation, so we may expect that if the local conductance structure is greatly altered, so would be the decay time, which, thinking in terms of circuits, depends on the effective system resistance among other things.

One can also think about ionospheric conductance that are parameterized by the ground magnetic field amplitude ...

In this first figure I zoom in on the relaxation time and plot the log of the amplitude as a function of time. What we see is that the lines are nearly parallel indicating that the speed of decay constant does not strongly depend on initial amplitude.

What this means is that the initial rapid decay is not due to local phenomena and conductance enhancements, but rather due to large scale current systems that are feeding the system.

Another way of looking at this is that we often talk about the magnetosphere as acting as either a constant current or constant voltage source, At these points in time, it is appears that the currents that are acting as the source.

[edit] 10.2.10 Slide 12

Dependence on local time or latitude?

[edit] 10.2.11 Slide 13

  • Initial part of relaxation (or recovery phase, but I have not isolated myself to substorm intervals only) is a special magnetospheric state in the same way as expansion
  • Generally in agreement with Kamide's two-component electrojet ideas.
  • Prediction models should probably take into account if substorm is currently occuring.
Retrieved from "http://aurora.gmu.edu/projects/index.php/Presentations"