SA51B-0235 0800h
Advanced Automated Solar Filament Detection and Characterization Code: Description, Performance, and Results
We have developed a code for automated detection and classification of solar filaments in full-disk H-alpha images that can contribute to Living With a Star science investigations and space weather forecasting. The program can reliably identify filaments, determine their chirality and other relevant parameters like the filaments area and their average orientation with respect to the equator, and is capable of tracking the day-by-day evolution of filaments while they travel across the visible disk. Detecting the filaments when they appear and tracking their evolution can provide not only early warnings of potentially hazardous conditions but also improve our understanding of solar filaments and their implications for space weather at 1 AU. The code was recently tested by analyzing daily H-alpha images taken at the Big Bear Solar Observatory during a period of four years (from mid 2000 until mid 2004). It identified and established the chirality of more than 5000 filaments without human intervention. We compared the results with the filament list manually compiled by Pevtsov et al. (2003) over the same period of time. The computer list matches the Pevtsov et al. list fairly well. The code results confirm the hemispherical chirality rule: dextral filaments predominate in the north and sinistral ones predominate in the south. The main difference between the two lists is that the code finds significantly more filaments without an identifiable chirality. This may be due to a tendency of human operators to be biased, thereby assigning a chirality in less clear cases, while the code is totally unbiased. We also have found evidence that filaments with definite chirality tend to be larger and last longer than the ones without a clear chirality signature. We will describe the major code characteristics and present and discuss the tests results.
SA51B-0236 0800h
Simulation of Solar Events Using the HAF Solar Wind and Magnetic Cloud Models
An interplanetary shock is often followed by a magnetic cloud. It is believed that magnetic clouds are related to halo-CMEs observed by SOHO/LASCO. Flare and erupting filament observations are used to locate the source on the solar surface. The Hakamada-Akasofu-Fry (HAFv2) solar wind model is used to calculate the shock arrival time, which can then be compared to shocks observed near Earth for the event on May 12, 1997. In order to fit the observed IMF configuration in the magnetic cloud that follows a shock, a magnetic cloud model is made and the force-free magnetic field model is applied inside the cloud. The velocity of the cloud is obtained on the basis of the average velocity of the shock from the HAF model. A three-dimensional configuration of interplanetary magnetic field lines, including the magnetic cloud, is obtained as a function of time. The simulated IMF using the combined models is in good agreement with solar wind observations near Earth. Several other examples are examined in the same way.
SA51B-0237 0800h
Predicting Solar Wind Conditions Beyond 1 AU Using the Hakamada-Akasofu-Fry Model
The Hakamada-Akasofu-Fry (HAF) solar wind model has been used to simulate the ambient solar wind during quiet solar periods, and the propagation of interplanetary shocks and coronal mass ejections during times of solar activity. For the past five years the HAF model has played an important part in the real-time forecasting of solar wind conditions and the arrival of interplanetary shocks at Earth. Our group is now using the model to characterize and predict the propagation of disturbances beyond 1 AU. The model also maps the IMF lines connecting the traveling shock region to the Earth, to interplanetary spacecraft, and to other planets such as Mars. This allows simulation results to be compared with, and used to interpret, solar wind and energetic particle observations at Earth and at a variety of spacecraft locations. We will compare predicted solar wind characteristics with observations at Earth, Mars, and the Voyager spacecraft.
http://gse.gi.alaska.edu/recent/
SA51B-0238 0800h
A New Method for Predicting the Dst Index Using Solar Wind Parameters
Previous methods of forecasting Dst from solar wind parameters have been limited by the need for continuous updating with observed Dst. The method proposed here eliminates this limitation by separating the empirical relation between the magnetosphere-solar wind power transfer function, epsilon and Dst into a driving term and a decay term and estimating the magnitude, delay time, and decay time of the Dst index using the epsilon function for the driving term. A postdiction of Dst for the period 1999-2003 showed a prediction efficiency of 91% and a linear correlation of 0.86 with the observed WDC-Kyoto Dst index.
SA51B-0239 0800h
Continuous Time Model for the Dst Index
We have used the NARMAX based system identification approach to deduce, directly from measurements of upstream magnetic field and solar wind bulk velocity, a continuous time differential relationship that governs the evolution of the Dst index. This relation has been used to deduce the analytical dependence of the ring current decay rate upon values of VBs and Dst. It is shown that this depends upon a number of previous values of the solar wind parameters and Dst. Therefore the assumption often used previously in empirical models that the magnetosphere is a first order Markov's system is not valid. We show that the continuous time relation derived can be used for a reliable forecast of the Dst index.
SA51B-0240 0800h
Solar wind low-energy energetic ion enhancements: a tool for forecasting large geomagnetic storms
The use of energetic ion enhancements (EIEs) as a tool for forecasting large geomagnetic storms was investigated in a previous study. The study was based on a data set in the rise and maximum of solar cycle 23 (Feb. 1998 - Dec. 2000). An excellent correlation was found between large geomagnetic storms (storms with Kp $>$6) and the peak flux of large energetic ion enhancements. As there are many more substantial EIEs than large geomagnetic storms, other characteristics were sought to help determine which of the EIEs are likely to be followed by large storms. These characteristics included the nature of the IP driver (whether transient or co-rotating high speed stream), and whether the EIEs result from complex solar and/or interplanetary activity. Here we present a forecasting technique using an additional parameter, the magnitude of the total magnetic field at the time of the shock arrival. This improves the identification of the EIEs that are likely to be followed by large storms. The success of this forecasting technique, and the warning times obtained, are presented for the time period of the original study and compared to the declining phase of the solar cycle. Note that the original study used data from the ACE/EPAM/LEMS-30 instrument. This was replaced in 2002 by the LEMS-120 instrument. We identify the equivalent EIE flux thresholds for the LEMS-120 instrument.
SA51B-0241 0800h
A Statistical Model of Electron Fluxes into the Earth's Atmosphere
We have developed a statistical model of precipitating electron fluxes based on 10 years of NOAA-12 Space Environment Monitor (SEM) data. For each hour of universal time, the electron spectral shape is defined for solar activity (using F10.7), geomagnetic activity (using Dst, Kp, and PC indices), invariant latitude (invariant latitudes equal to or $>$ 40\deg in both hemispheres), and magnetic local time. The climatology includes the precipitating differential fluxes determined from the SEM high-energy integral detectors and its low-energy differential electrostatic analyzers. In addition to total global energy flux input, the spectral information contained in the model enables computations of ionization rate profiles and Hall and Pederson conductivities. Model output is provided to the community interactively via a website where the fluxes may be obtained both graphically and as ASCII files. Strengths and limitations of the model and its applicability for use within an operational environment are illustrated.
SA51B-0242 0800h
Kp forecast models
Magnetically active times, e.g., Kp > 5, are notoriously difficult to predict, precisely when the predictions are crucial to the space weather users. Taking advantage of the routinely available solar wind measurements at Langrangian point (L1) and nowcast Kps, Kp forecast models based on neural networks were developed with the focus on improving the forecast for active times. In order to satisfy different needs and operational constraints, three models were developed: (1) model that inputs nowcast Kp, solar wind parameters, and predict Kp 1 hr ahead; (2) model with the same input as (1) and predict Kp 4 hr ahead; and (3) model that inputs only solar wind parameters and predict Kp 1 hr ahead (the exact prediction lead time depends on the solar wind speed and the location of the solar wind monitor). Extensive evaluations of these models and other major operational Kp forecast models show that while the new models can predict Kps more accurately for all activities, the most dramatic improvements occur for moderate and active times. The evaluations of the models over 2 solar cycles, 1975-2001, show that solar wind driven models predict Kp more accurately during solar maximum than solar minimum. This result, as well as information dynamics analysis of Kp, suggests that geospace is more dominated by internal dynamics during solar minimum than solar maximum, when it is more directly driven by external inputs, namely solar wind and IMF.
SA51B-0243 0800h
An Integrated Program to Forecast Geostorms
We have developed several operational products and automated tools for assessing the helicity content of solar regions and their probability of launching a geoeffective coronal mass ejection. These include detection of active region sigmoids, measurement of magnetic helicity injection in active regions, measurement of the sense of helicity in solar filaments, and the estimate of magnetic helicity content of active regions from vector magnetogram observations. In this presentation we discuss a new program to integrate the separate products and tools into a single product that provides a quantitative mid-term forecast of solar activity that results in geomagnetic storms.
SA51B-0244 0800h
Real-time Kp and b2i Nowcast from GOES Magnetometer Estimates of Magnetotail Stretching
One key characterization of the global state of the magnetosphere is the extent to which the Earth's magnetotail is stretched from a dipolar configuration. When the geosynchronous GOES satellites are on the nightside of the Earth, they directly measure that stretching, at least at one particular local time. GOES magnetometer data has been made available in real time by NOAA's Space Data Center. We have correlated the degree of magnetotail stretching (specifically, T=atan(v/h), where v and h are the usual magnetic components) with Kp, and the low-latitude extent of ion precipitation, b2i, to provide real time nowcasts of the latter two quantities. The correlation between the real-time GOES measurements and Kp ranges between 0.4 and 0.75, depending on the local time of GOES. We emphasize that the degree of magnetotail stretching is a more fundamental characterization of the state of the magnetosphere than is Kp, and thus the moderate correlation likely says more about the extent to which Kp can characterize the magnetosphere than the limitations of the GOES measurements. The major limitation of the magnetotail stretching index is that it varies with local time, but is measured at only one local time. By converting the GOES measurements to an equivalent Kp, therefore, the major purpose is merely to translate the level of activity into a scale familiar to most operational space weather forecasters. The real time measurements of magnetotail stretching are available graphically on our web site at http://sd-www.jhuapl.edu/Aurora/goes_realtime/index.cgi along with movies showing how the magnetotail stretching has varied recently. The result is a representation of the magnetotail stretching data in a format convenient for space weather forecasters.
http://sd-www.jhuapl.edu/Aurora/goes_realtime/index.cgi
SA51B-0245 0800h
Aurora Boundaries Quantified by Geomagnetic Index
Various operational systems require information on the location and intensity of the aurora. A statistical model of the aurora is given using global images from the Ultraviolet Imager (UVI) on the Polar satellite. The equatorward (EQ), poleward (PO) and peak (PK) boundaries of the auroral oval are determined. using UVI images averaged into $1\deg$x$1\deg$ spatial bins according to common geomagnetic indices such as Kp, AE, AL, and PCI. From these bin-averaged images, latitude intensity profiles at 1 hour MLT intervals are constructed by interpolation. A background is subtracted for each profile, and the EQ, PO, and PK boundary latitudes are found from the corrected profile. (The PK boundary is the maximum, and the EQ and PO boundaries are threshold locations of fixed irradiances such as 1, 2, or 4 photons/cm$^{2}$s.) Several months of images during the winter and summer of 1997 were used to statistically quantify the boundaries at various levels of geomagnetic activity given by the several indices. As expected, the higher the level of activity, the wider and more expanded the oval. More importantly, the boundaries are functionally related to the indices at any local time. These functional relations can then be used to determine the auroral location at any level of geomagnetic activity given by the indices. Thus, given a level of geomagnetic activity, one can find the boundaries of the oval as defined on the basis of intensity. By monitoring the relevant geomagnetic index, an operational system can then easily compute the expected oval location and judge its impact on performance. The optimum indices that best define the oval will be discussed.
SA51B-0246 0800h
Auroras Now! - Auroral nowcasting service for Hotels in Finnish Lapland and its performance during winter 2003-2004
European Space Agency is currently supporting 17 Service Development Activities (SDA) within its Space Weather Pilot Project. Auroras Now!, one of the SDAs, has been operated during November 2003 - March 2004 as its pilot season. The service includes a public part freely accessible in Internet (http://aurora.fmi.fi) and a private part visible only to the customers of two hotels in the Finnish Lapland through the hotels' internal TV-systems. The nowcasting system is based on the magnetic recordings of two geophysical observatories, Sodankyl\"a (SOD, MLAT ~64 N) and Nurmij\"arvi (NUR, MLAT ~57 N). The probability of auroral occurrence is continuously characterised with an empirically determined three-level scale. The index is updated once per hour and based on the magnetic field variations recorded at the observatories. During dark hours the near-real time auroral images acquired at SOD are displayed. The hotel service also includes cloudiness predictions for the coming night. During the pilot season the reliability of the three-level magnetic alarm system was weekly evaluated by comparing its prediction with auroral observations by the nearby all-sky camera. Successful hits and failures were scored according to predetermined rules. The highest credit points when it managed to spot auroras in a timely manner and predict their brightness correctly. Maximum penalty points were given when the alarm missed clear bright auroras lasting for more than one hour. In this presentation we analyse the results of the evaluation, present some ideas to further sharpen the procedure, and discuss more generally the correlation between local auroral and magnetic activity.
SA51B-0247 0800h
The utility of auroral image-based activity metrics
Auroral activity indices such as Hemispheric Power and Auroral Boundary are currently key data products used for space weather predictions and nowcasting. However, these products are necessarily based on limited observations which must be extrapolated to provide global coverage. The advent of routine space-based auroral imaging in the last decade offers the seeming advantage of more detailed measures of auroral activity. Examples of image-derived products include energy deposition maps, oval location, cap size, and morphological classification. However, activity metrics derived from auroral images have shortcomings, as well. For example, limited fields-of-view and orbital motion prevent full coverage of the auroral regions. This paper will examine the utility of activity metrics derived from auroral images for operational purposes. The eight-year collection of Polar UVI images databased in the UVI Online Search Tool (OST) will be used to illustrate the advantages and shortcomings of auroral activity metrics. The potential role of other currently-active imaging missions will also be examined and correlative studies to date using auroral imaging will be summarized.
http://csds.uah.edu/uvi-ost/
SA51B-0248 0800h
Updated Millstone Hill and Sondrestrom Incoherent Scatter Radars Convection Model
High latitude convection electric fields play a central role in linking the solar wind and magnetosphere with ionosphere and thermosphere processes. The characteristics of the convection has been extensively studied, and empirical or statistical models of the convection have been created and are under steady revision over the years with new data and novel analysis techniques becoming available. These models include those based on OGO-6, DE-2 and DMSP low altitude satellite measurements of electric fields. Assimilation of various types of data (e.g., AMIE) provides also a global mapping of the convection pattern. Ground based instruments such as incoherent scatter radar (ISR) and SuperDARN measure the line of sight velocity and their long-term observations make important contributions to the modeling effort. Holt et al. (1987) described the technique and model using Millstone Hill ISR data, and the technique was then applied to combining Millstone Hill and Sondrestrom ISR data to form a single model (Foster et al., 1987), MHS. The earlier Millstone Hill model provides potentials and velocities at 3 levels of Kp (or 9 levels of Power Precipitation index) for IMF "away" and "from" directions. MHS is however, for 4 combinations of By and Bz signs. Since MHS was published more than 1 decade ago, additional data from both sites have been collected, and it is now the right time to update them with more data and finer By and Bz dependence. This paper will describe the updated model, and compare it to the old version as well as to models from other techniques.
SA51B-0249 0800h
Ionospheric Tomography Application to Model Validation at High Latitude: A Trough Study
Data sets of ionospheric trough observations obtained in the Alaska and Western European sectors have been used for validation of ionospheric models. The dataset of over 1300 cases (selected from more than three years of observations) includes the total electron content (TEC) characteristics of the troughs that were obtained by integrating vertically through electron density profiles produced by ionospheric tomography. The well-known great variability exhibited by the trough, in terms of its latitudinal location, latitudinal extent, longitudinal variation, boundaries, and gradients is indicative of the challenge of ionospheric modeling in the high latitudes. Present and future models that seek accurate characterization in the higher latitudes must represent the trough region. However, failure to accurately locate the latitudinal position of the trough can have a worse impact than excluding it altogether. In this paper, we report on application of this tomography-derived database to a validation study of the Parameterized Real-Time Ionospheric Specification Model (PRISM). PRISM was developed for the purpose of using near real-time observations to provide a global specification of ionosphere. In particular, PRISM uses the energetic particle and ion drift measurements available from the DMSP satellites to identify several high latitude boundaries including the equatorial and poleward boundaries of the trough. In this study, we focused on assessing PRISM's ability to specify the F region trough. The effort included a statistical study of PRSIM climatology, and case studies where DMSP particle data was used to provide PRISM with definition of the location of the troughs. Case studies also performed runs that were designed to invoke PRISM's alternative empirical F-region trough model. Details of these studies, and implications for further PRISM studies and for the next generation of ionospheric assimilation models will be presented.
SA51B-0250 0800h
Seasonal and Solar-Cycle Variation of Polar Cap Patches
Polar cap patches are regions of enhanced plasma density that form in the dayside cusp or equatorward of the cusp during periods of southward IMF and then convect in an antisunward direction across the polar cap. The density enhancement in a patch relative to the background ionospheric density varies over a wide range, from a few percent to a factor of 100. A horizontal cross-section of a polar cap patch can appear roughly circular or cigar-shaped, and patches are typically from about 200 to 1000 km across. We have used a time-dependent, 3-dimensional, fluid model of the coupled ionosphere-polar wind system to model the evolution of a representative polar cap patch and to study its effect on the polar wind for a range of seasonal and solar-cycle conditions. This model calculates the plasma dynamics in a large number (about 1500) of flux tubes as they move horizontally across the polar region under the influence of corotational and magnetospheric convection electric fields. The large number of flux tubes included in a simulation makes it possible to model mesoscale features of the ionosphere, such as plasma patches.
SA51B-0251 0800h
Comparison of the UAF Ionosphere Model with Incoherent-Scatter Radar Data
The UAF Eulerian Parallel Polar Ionosphere Model (UAF EPPIM) is a first-principles three-dimensional time-dependent representation of the northern polar ionosphere (>50 degrees north latitude). The model routinely generates short-term (~2 hours) ionospheric forecasts in real-time. It may also be run in post-processing/batch mode for specific time periods, including long-term (multi-year) simulations. The model code has been extensively validated (~100k comparisons/model year) against ionosonde foF2 data during quiet and moderate solar activity in 2002-2004 with reasonable fidelity (typical relative RMS 10-20% for summer daytime, 30-50% winter nighttime). However, ionosonde data is frequently not available during geomagnetic disturbances. The objective of the work reported here is to compare model outputs with available incoherent-scatter radar data during the storm period of October-November 2003. Model accuracy is examined for this period and compared to model performance during geomagnetically quiet and moderate circumstances. Possible improvements are suggested which are likely to boost model fidelity during storm conditions.
http://www.arsc.edu/SpaceWeather
SA51B-0252 0800h
Ionosphere dynamics study of Sun-Earth Connection Events of October-November 2003 with the UAF Polar Ionosphere Model
The UAF Polar Ionosphere Model has been used to simulate the spatial and temporal changes of the high-latitude northern ionosphere during the geomagnetically active period of October-November 2003. The results illustrate the highly variable ionospheric structure and dynamics. Model results have also been compared with 2-dimensional tomography data obtained in the Alaskan meridian. Previous work with this model during geomagnetically quiet times has shown very good agreement between model and data. This study has provided an opportunity to compare the model with data during a more disturbed period.
SA51B-0253 0800h
Quantitative Description of Monthly Ionospheric Variability in the IRI Model
The International Reference Ionosphere (IRI) model provides an empirical specification of the ionospheric climatology at the level of monthly averages. Operational use of the IRI model often requires an estimate of the monthly variability, so that an operator not only knows the expected monthly average value of an ionospheric parameter but also the expected variation around this monthly average. A special IRI Task Force Activity at the International Center for Theoretical Physics (ICTP) in Trieste, Italy has worked on this modeling goal during the last few years using ionosonde data from many stations worldwide focusing primarily on the electron density in the region below the F peak. Other IRI team members have looked at the variability at different heights and have studied the variability seen for the plasma temperatures. We will report on the status and progress of this activity and will discuss the different parameter used for describing ionospheric variability (mean, median, standard deviation, quartiles, deciles) and the planned model implementation. First results will be reported based on the ICTP meetings and the 2003 IRI Workshop in Grahamstown, South Africa.
http://nssdc.gsfc.nasa.gov/space/model/ionos/iri.html
SA51B-0254 0800h
Extending the IRI Topside Electron Density Profiles to Several Re using IMAGE/RPI Measurements
The radio plasma imager (RPI) onboard the IMAGE spacecraft measures electron density distributions along the magnetic field; these density profiles can be converted to the density distribution as a function of height. For many individual events, the vertical profiles from the spacecraft down to about 2000 km were derived in the polar cap. These profiles were extrapolated downward to join smoothly with the IRI profile at a few hundred km above the F2 layer peak. We use Chapman functions [Reinisch and Huang, 2004], simple exponential functions, and a power law [Nsumei et al., 2003] to test what best describes the vertical profiles from the height of the F2 peak to ~4 Re. The resulting profiles are compared with the topside electron densities computed with the international reference ionosphere (IRI) model. The preliminary comparisons show considerable differences between the RPI driven profiles and the topside IRI model. Although only individual RPI profiles were used, the results confirm that the IRI topside model at high latitudes may not be realistic above ~ 600 km. Our results demonstrate the potential of using the RPI measurements, combined with the IRI model, to construct an empirical electron model that covers the altitude range from the ionosphere to several Re in the polar cap. A similar method can then be developed for lower latitudes, i.e., the plasmasphere.
SA51B-0255 0800h
Investigation of the Accuracy of Ionospheric Models at Mid-Latitudes: Examining Ionospheric Metrics
The electron density specification of the ionosphere is the key parameter supporting many operational products. To assess the accuracy of tools based on space weather models of the ionosphere one must know the accuracy of the underlining models. We are developing a software/database package to assess the accuracy of ionospheric models. The package will be placed at the Community Coordinated Modeling Center (CCMC). Initial focus is on the mid-latitude ionosphere as observed by the Arecibo Incoherent Scatter Radar (ISR). This ISR database has extensive ionospheric coverage over variability in solar cycle, season, local time, and geomagnetic activity. The assessments of models need to be based on careful constructed metric definitions to compare the model specifications with the ISR "ground truth." Our goals for the assessment tool are (1) to provide reliable, metric assessment of models for users represented by agencies of the Nation Space Weather Program and, (2) to provide the scientific community with an assessment of conditions when models are adequate and inadequate. The second implementation plan of the NSWP (2000) has established the priority of metrics and has specified these metrics. We begin with the NSWP ionospheric metrics as a reasonable starting place, but examine other strategies to assess ionosphere weather specifications through several new metric definitions for the F region. We present our initial studies on the weaknesses and benefits of several different metric definitions for F region profile accuracy. Three models will be use in the metric assessment (1) the physics-based Ionospheric Forecast Model (IFM), (2) the physics-based and Coupled Thermospheric-Ionospheric-Plasmasphere-electrodynamics Model (CTIPe), and (3) the empirical International Reference Model (IRI). Central to creating reliable metric results is the need to quantify the quality and accuracy of the "ground truth" ISR database. Metric issues associated with ISR operational modes, data gaps, instrument noise, and representation errors will be addressed using an initial demonstration period, January 24 to 28, 1990.
SA51B-0256 0800h
Validation of the Objective Analysis Algorithm IDA3D
We have developed a set of metrics and skill scores to analyze the performance of the objective analysis algorithm Ionospheric Data Assimilation Three-Dimensional (IDA3D). The objective of the skill score is to quantify the improvement in model predictions after either assimilation of a new data source, or improvement in the model itself. The metrics chosen to be evaluated by the skill score include electron density in the E-regin, F-region, and topside of the ionosphere, total electron content (TEC), F-region peak height and F-region layer thickness. The metrics can be sub-binned into different geographic regions, solar and magnetic activity levels and time of day. For a first principles data assimilation model that is capable of forecasting, the metric can also be a function of prediction time beyond the analysis time. The skill score we have developed begins with a pure climate prediction of the metric values. Then we run the data assimilative model and produce assimilative prediction of the metric values. The skill score is then given as \begin{equation} \label{eqn:skill} S = 1.0 - \frac{<|\vec D - \vec T|>}{<|\vec M - \vec T|>}. \end{equation} Where $\vec D$ is the array of metric values ouput from the data assimilation model, $\vec T$ is the independent measurements of the metric that serve as ``truth'', and $\vec M$ is the array of climate predictions of the metric values. The skill is a function of time in the sense that every time the data assimilative model is improved we generate new skill scores. The values of the skill score range from $1.0$ for when the data assimilative model is perfect (that is agrees with the measurements), to $0.0$ when the data assimilative model is no better than the climate predictions, to negative infinity as the data assimilative model does much worse than climate. We will present preliminary results for several days of analysis with incoherent scatter observatins of electron density serving as truth.
SA51B-0257 0800h
Real-time Determination of F-region Critical Frequencies Using SuperDARN
In addition to detecting backscatter from ionospheric field-aligned irregularities, SuperDARN radars are sensitive to ground scatter returns. The profiles of these returns vary in a systematic manner with frequency and can be used to determine HF propagation characteristics and ionospheric critical frequencies. The large spatial coverage and good time resolution provided by the SuperDARN network make it possible to produce maps of high-latitude critical frequencies approximately every 15 minutes. Recent upgrades to the SuperDARN operating software provide the capability to receive the ground scatter profiles and angle of arrival information in real-time via the internet and we use these data to construct near real-time maps of F-region critical frequencies. In this paper, the technique will be described and representative data will be discussed.
SA51B-0258 0800h
Comparison of DMSP F13 Cross-Track Ion Drift Velocities With AMIE Results
The Assimilative Mapping Of Ionospheric Electrodynamics (AMIE) model [Richmond et al., J. Geophys. Res., 93, 5741, 1988] has been used in a wide range of studies pertaining to the magnetosphere, ionosphere and thermosphere. In these studies historical data from several different data sources (electric fields from radars and satellites, electric currents from satellites, magnetic perturbations from satellites and magnetometers) have been assimilated. In its real-time mode (rt-AMIE), only data from an array of ground-based magnetometers are assimilated in and convection patterns are produced in one-minute increments. However, the reliability of these real-time patterns for applications involving ionosphere-thermosphere specifications and forecasts has never been systematically tested. To address this issue, a comparison of one year of DMSP F13 cross-track ion drift velocities with AMIE convection patterns based on data from 80 ground-based magnetometers has been conducted. First, for each high-latitude DMSP velocity observation, the corresponding AMIE value was calculated. Then, the measured and calculated cross-track ion drift velocities along the high-latitude pass were compared and criteria were established to determine whether or not the AMIE patterns fit the measurements adequately. The comparisons were done for one year (1998) of satellite crossings of the northern polar region (4300 consecutive satellite crossings). The results indicate that the AMIE patterns adequately represented the DMSP observations about 32% of the time, which is a significant improvement over statistical convection patterns (6% of the time) [Bekerat et al., J. Geophys. Res., 108, 1413, 2003]. This is particularly impressive in view of the fact that only a limited number of ground-based magnetometers are included in the AMIE patterns used in this study.
SA51B-0259 0800h
Geomagnetic Heating Modifications to Neutral Atmospheric Density Models
We present modifications of the geomagnetic input to the Jacchia 1970 neutral atmospheric density model based on neutral density estimates for 18 satellites. Our research uses a Joule heating proxy for geomagnetic heating, instead of Ap which was the sole input used in the original Jacchia 1970 model. We compute the difference between the original Jacchia 1970 model, and our modified model, for 18 satellites in a variety of low earth orbits. The results of this research will allow better operational forecasting of atmospheric neutral drag. To this end we compare the "forecast" of atmospheric neutral density over a one year interval during 2001-2002, the peak of the most current solar cycle.
SA51B-0260 0800h
Satellite Ballistic Coefficients and the Lower Thermosphere
The observed variations in the 'estimated' ballistic coefficients (B') for low-perigee satellites in the lower thermosphere (about 200 km) during the year 2001, have been used to study their implications for neutral atmospheric densities with respect to changes in the solar activity and geomagnetic activity. The changes in B' reflect the unmodeled corrections in the atmospheric density models. It is shown that statistically, on the average, based on the 'true' ballistic coefficients (Bt), the atmospheric density models predict the 'true' density within 10 percent as a function of solar activity for quiet and mild geomagnetic periods. The observed variations in the ballistic coefficients with days exhibit a semiannual variation (SAV). The 'scaling' factor required to correctly predict the SAV in the 'true' densities from the model densities, ranges from a maximum of about 1.13 in the spring (equinox) to a minimum of about 0.93 in the summer (solstice), implying a semiannual amplitude(SAA) of (max/min) 1.2. The calculations based on the CIRA-86 models show that changes in B' during some major magnetic storms cannot be explained, as they are probably contaminated by 'high density neutral cells' observed by the S85-1 satellite at 200 km and predicted by the NCAR-TIGCM models by Crowley, et al (1995). Changes in B' due to these cells all the more require temperature/density corrections of the steady-state High Accuracy Satellite Drag Model (HASDM). It is suggested that the NCAR-TIGCM model should be used to see if it performs better in bringing the B' values closer to Bt than the steady-state models.