-
Efficient labeling of solar flux evolution videos by a deep learning model
Authors:
Subhamoy Chatterjee,
Andrés Muñoz-Jaramillo,
Derek A. Lamb
Abstract:
Machine learning (ML) is becoming a critical tool for interrogation of large complex data. Labeling, defined as the process of adding meaningful annotations, is a crucial step of supervised ML. However, labeling datasets is time consuming. Here we show that convolutional neural networks (CNNs), trained on crudely labeled astronomical videos, can be leveraged to improve the quality of data labeling…
▽ More
Machine learning (ML) is becoming a critical tool for interrogation of large complex data. Labeling, defined as the process of adding meaningful annotations, is a crucial step of supervised ML. However, labeling datasets is time consuming. Here we show that convolutional neural networks (CNNs), trained on crudely labeled astronomical videos, can be leveraged to improve the quality of data labeling and reduce the need for human intervention. We use videos of the solar magnetic field, crudely labeled into two classes: emergence or non-emergence of bipolar magnetic regions (BMRs), based on their first detection on the solar disk. We train CNNs using crude labels, manually verify, correct labeling vs. CNN disagreements, and repeat this process until convergence. Traditionally, flux emergence labelling is done manually. We find that a high-quality labeled dataset, derived through this iterative process, reduces the necessary manual verification by 50%. Furthermore, by gradually masking the videos and looking for maximum change in CNN inference, we locate BMR emergence time without retraining the CNN. This demonstrates the versatility of CNNs for simplifying the challenging task of labeling complex dynamic events.
△ Less
Submitted 28 August, 2023;
originally announced August 2023.
-
The best of both worlds: Using automatic detection and limited human supervision to create a homogenous magnetic catalog spanning four solar cycles
Authors:
A. Munoz-Jaramillo,
Z. A. Werginz,
J. P. Vargas-Acosta,
M. D. DeLuca,
J. C. Windmueller,
J. Zhang,
D. W. Longcope,
D. A. Lamb,
C. E. DeForest,
S. Vargas-Dominguez,
J. W. Harvey,
P. C. H. Martens
Abstract:
Bipolar magnetic regions (BMRs) are the cornerstone of solar variability. They are tracers of the large-scale magnetic processes that give rise to the solar cycle, shapers of the solar corona, building blocks of the large-scale solar magnetic field, and significant contributors to the free-energetic budget that gives rise to flares and coronal mass ejections. Surprisingly, no homogeneous catalog o…
▽ More
Bipolar magnetic regions (BMRs) are the cornerstone of solar variability. They are tracers of the large-scale magnetic processes that give rise to the solar cycle, shapers of the solar corona, building blocks of the large-scale solar magnetic field, and significant contributors to the free-energetic budget that gives rise to flares and coronal mass ejections. Surprisingly, no homogeneous catalog of BMRs exists today, in spite of the existence of systematic measurements of the magnetic field since the early 1970's. The purpose of this work is to address this deficiency by creating a homogenous catalog of BMRs from the 1970's until the present. For this purpose, in this paper we discuss the strengths and weaknesses of the automatic and manual detection of BMRs and how both methods can be combined to form the basis of our Bipolar Active Region Detection (BARD) code and its supporting human supervision module. At present, the BARD catalog contains more than 10,000 unique BMRs tracked and characterized during every day of their observation. Here we also discuss our future plans for the creation of an extended multi-scale magnetic catalog combining the SWAMIS and BARD catalogs.
△ Less
Submitted 22 March, 2022;
originally announced March 2022.
-
Measurements of Solar Differential Rotation and Meridional Circulation from Tracking of Photospheric Magnetic Features
Authors:
Derek A. Lamb
Abstract:
Long-lived rotational and meridional flows are important ingredients of the solar cycle. Magnetic field images have typically been used to measure these flows on the solar surface by cross-correlating thin longitudinal strips or square patches across sufficiently long time gaps. Here, I use one month of SDO/HMI line-of-sight magnetic field observations, combined with the SWAMIS magnetic feature tr…
▽ More
Long-lived rotational and meridional flows are important ingredients of the solar cycle. Magnetic field images have typically been used to measure these flows on the solar surface by cross-correlating thin longitudinal strips or square patches across sufficiently long time gaps. Here, I use one month of SDO/HMI line-of-sight magnetic field observations, combined with the SWAMIS magnetic feature tracking algorithm to measure the motion of individual features in these magnetograms. By controlling for perturbations due to short-lived flows and due to false motions from feature interactions, I effectively isolate the long-lived flows traced by the magnetic features. This allows me to produce high-fidelity differential rotation measurements with well-characterized variances and covariances of the fit parameters.I find a sidereal rotational profile of $(14.296\pm0.006)+(-1.847\pm0.056)\sin^{2}b+(-2.615\pm0.093)\sin^{4}b$, with units of $\textrm{ deg d}^{-1}$, and a large covariance $σ_{BC}^{2}=-4.87\times10^{-3}(\textrm{ deg d}^{-1})^{2}$. I also produce medium-fidelity measurements of the much weaker meridional flow that is broadly consistent with previous results. This measurement shows a peak flow of $16.7\pm0.6\text{ m s}^{-1}$ at latitude $b=45^\circ$ but is insufficiently characterized at higher latitudes to ascertain whether the chosen functional form $2\cos b\sin b$ is appropriate. This work shows that measuring the motions of individual features in photospheric magnetograms can produce high precision results in relatively short time spans, and suggests that high resolution non-longitudinally averaged photospheric velocity residual measurements could be produced to compare with coronal results, and to provide other diagnostics of the solar dynamo.
△ Less
Submitted 10 January, 2017;
originally announced January 2017.
-
Development of Technique to Detect and Classify Small-Scale Magnetic Flux Cancellation and Rapid Blueshifted Excursions
Authors:
Xin Chen,
Na Deng,
Derek A. Lamb,
Ju **g,
Chang Liu,
Rui Liu,
Sung-Hong Park,
Haimin Wang
Abstract:
We present a set of tools for detecting small-scale solar magnetic cancellations and the disk counterpart of type II spicules (the so-called Rapid Blueshifted Excursions (RBEs)), using line-of-sight photospheric magnetograms and chromospheric spectroscopic observations, respectively. For tracking magnetic cancellation, we improve the Southwest Automatic Magnetic Identification Suite (SWAMIS) so th…
▽ More
We present a set of tools for detecting small-scale solar magnetic cancellations and the disk counterpart of type II spicules (the so-called Rapid Blueshifted Excursions (RBEs)), using line-of-sight photospheric magnetograms and chromospheric spectroscopic observations, respectively. For tracking magnetic cancellation, we improve the Southwest Automatic Magnetic Identification Suite (SWAMIS) so that it is able to detect certain obscure cancellations that can be easily missed. For detecting RBEs, we use a normalized reference profile to reduce false-positive detections caused by the non-uniform background and seeing condition. Similar to the magnetic feature tracking in SWAMIS, we apply a dual-threshold method to enhance the accuracy of RBE detection. These tools are employed to analyze our coordinated observations using the Interferometric BIdimensional Spectrometer at Dunn Solar Telescope (DST) of the National Solar Observatory (NSO) and Hinode. We present the statistical properties of magnetic cancellations and RBEs, and explore their correlation using this data set.
△ Less
Submitted 6 January, 2015;
originally announced January 2015.
-
Chromospheric Rapid Blueshifted Excursions Observed with IBIS and Their Association with Photospheric Magnetic Field Evolution
Authors:
Na Deng,
Xin Chen,
Chang Liu,
Ju **g,
Alexandra Tritschler,
Kevin P. Reardon,
Derek A. Lamb,
Craig E. Deforest,
Carsten Denker,
Shuo Wang,
Rui Liu,
Haimin Wang
Abstract:
Chromospheric rapid blueshifted excursions (RBEs) are suggested to be the disk counterparts of type II spicules at the limb and believed to contribute to the coronal heating process. Previous identification of RBEs was mainly based on feature detection using Dopplergrams. In this paper, we study RBEs on 2011 October 21 in a very quiet region at the disk center, which were observed with the high-ca…
▽ More
Chromospheric rapid blueshifted excursions (RBEs) are suggested to be the disk counterparts of type II spicules at the limb and believed to contribute to the coronal heating process. Previous identification of RBEs was mainly based on feature detection using Dopplergrams. In this paper, we study RBEs on 2011 October 21 in a very quiet region at the disk center, which were observed with the high-cadence imaging spectroscopy of the Ca II 8542 A line from the Interferometric Bidimensional Spectrometer (IBIS). By using an automatic spectral analysis algorithm, a total of 98 RBEs are identified during a 11 minute period. Most of these RBEs have either a round or elongated shape, with an average area of 1.2 arcsec^2. The detailed temporal evolution of spectra from IBIS makes possible a quantitative determination of the velocity (~16 km/s) and acceleration (~400 m/s^2) of Ca II 8542 RBEs, and reveal an additional deceleration (~-160 m/s^2) phase that usually follows the initial acceleration. In addition, we also investigate the association of RBEs with the concomitant photospheric magnetic field evolution, using coordinated high-resolution and high-sensitivity magnetograms made by Hinode. Clear examples are found where RBEs appear to be associated with the preceding magnetic flux emergence and/or the subsequent flux cancellation. However, a further analysis with the aid of the Southwest Automatic Magnetic Identification Suite does not yield a significant statistical association between these RBEs and magnetic field evolution. We discuss the implications of our results in the context of understanding the driving mechanism of RBEs.
△ Less
Submitted 12 December, 2014;
originally announced December 2014.
-
Spatial Nonlocality of the Small-Scale Solar Dynamo
Authors:
Derek A. Lamb,
Timothy A. Howard,
Craig E. DeForest
Abstract:
We explore the nature of the small-scale solar dynamo by tracking magnetic features. We investigate two previously-explored categories of the small-scale solar dynamo: shallow and deep. Recent modeling work on the shallow dynamo has produced a number of scenarios for how a strong network concentration can influence the formation and polarity of nearby small-scale magnetic features. These scenarios…
▽ More
We explore the nature of the small-scale solar dynamo by tracking magnetic features. We investigate two previously-explored categories of the small-scale solar dynamo: shallow and deep. Recent modeling work on the shallow dynamo has produced a number of scenarios for how a strong network concentration can influence the formation and polarity of nearby small-scale magnetic features. These scenarios have measurable signatures, which we test for here using magnetograms from the Narrowband Filter Imager (NFI) on Hinode. We find no statistical tendency for newly-formed magnetic features to cluster around or away from network concentrations, nor do we find any statistical relationship between their polarities. We conclude that there is no shallow or "surface" dynamo on the spatial scales observable by Hinode/NFI. In light of these results, we offer a scenario in which the sub-surface field in a deep solar dynamo is stretched and distorted via turbulence, allowing the field to emerge at random locations on the photosphere.
△ Less
Submitted 12 April, 2014;
originally announced April 2014.
-
Solar Magnetic Tracking. IV. The Death of Magnetic Features
Authors:
Derek A. Lamb,
Timothy A. Howard,
Craig E. DeForest,
Clare E. Parnell,
Brian T. Welsch
Abstract:
The removal of magnetic flux from the quiet-sun photosphere is important for maintaining the statistical steady-state of the magnetic field there, for determining the magnetic flux budget of the Sun, and for estimating the rate of energy injected into the upper solar atmosphere. Magnetic feature death is a measurable proxy for the removal of detectable flux. We used the SWAMIS feature tracking cod…
▽ More
The removal of magnetic flux from the quiet-sun photosphere is important for maintaining the statistical steady-state of the magnetic field there, for determining the magnetic flux budget of the Sun, and for estimating the rate of energy injected into the upper solar atmosphere. Magnetic feature death is a measurable proxy for the removal of detectable flux. We used the SWAMIS feature tracking code to understand how nearly 20000 detected magnetic features die in an hour-long sequence of Hinode/SOT/NFI magnetograms of a region of quiet Sun. Of the feature deaths that remove visible magnetic flux from the photosphere, the vast majority do so by a process that merely disperses the previously-detected flux so that it is too small and too weak to be detected. The behavior of the ensemble average of these dispersals is not consistent with a model of simple planar diffusion, suggesting that the dispersal is constrained by the evolving photospheric velocity field. We introduce the concept of the partial lifetime of magnetic features, and show that the partial lifetime due to Cancellation of magnetic flux, 22 h, is 3 times slower than previous measurements of the flux turnover time. This indicates that prior feature-based estimates of the flux replacement time may be too short, in contrast with the tendency for this quantity to decrease as resolution and instrumentation have improved. This suggests that dispersal of flux to smaller scales is more important for the replacement of magnetic fields in the quiet Sun than observed bipolar cancellation. We conclude that processes on spatial scales smaller than those visible to Hinode dominate the processes of flux emergence and cancellation, and therefore also the quantity of magnetic flux that threads the photosphere.
△ Less
Submitted 15 July, 2013;
originally announced July 2013.
-
Coronal hole boundaries at small scales: IV. SOT view Magnetic field properties of small-scale transient brightenings in coronal holes
Authors:
Z. Huang,
M. S. Madjarska,
J. G. Doyle,
D. A. Lamb
Abstract:
We study the magnetic properties of small-scale transients in coronal hole. We found all brightening events are associated with bipolar regions and caused by magnetic flux emergence followed by cancellation with the pre-existing and newly emerging magnetic flux. In the coronal hole, 19 of 22 events have a single stable polarity which does not change its position in time. In eleven cases this is th…
▽ More
We study the magnetic properties of small-scale transients in coronal hole. We found all brightening events are associated with bipolar regions and caused by magnetic flux emergence followed by cancellation with the pre-existing and newly emerging magnetic flux. In the coronal hole, 19 of 22 events have a single stable polarity which does not change its position in time. In eleven cases this is the dominant polarity. The dominant flux of the coronal hole form the largest concentration of magnetic flux in terms of size while the opposite polarity is distributed in small concentrations. In the coronal hole the number of magnetic elements of the dominant polarity is four times higher than the non-dominant one. The supergranulation configuration appears to preserve its general shape during approximately nine hours of observations although the large concentrations in the network did evolve and were slightly displaced, and their strength either increased or decreased. The emission fluctuations seen in the X-ray bright points are associated with reoccurring magnetic cancellation in the footpoints. Unique observations of an X-ray jet reveal similar magnetic behaviour in the footpoints, i.e. cancellation of the opposite polarity magnetic flux. We found that the magnetic flux cancellation rate during the jet is much higher than in bright points. Not all magnetic cancellations result in an X-ray enhancement, suggesting that there is a threshold of the amount of magnetic flux involved in a cancellation above which brightening would occur at X-ray temperatures. Our study demonstrates that the magnetic flux in coronal holes is continuously recycled through magnetic reconnection which is responsible for the formation of numerous small-scale transient events. The open magnetic flux forming the coronal-hole phenomenon is largely involved in these transient features.
△ Less
Submitted 6 October, 2012;
originally announced October 2012.
-
Solar Magnetic Tracking. I. Software Comparison and Recommended Practices
Authors:
C. E. DeForest,
H. J. Hagenaar,
D. A. Lamb,
C. E. Parnell,
B. T. Welsch
Abstract:
Feature tracking and recognition are increasingly common tools for data analysis, but are typically implemented on an ad-hoc basis by individual research groups, limiting the usefulness of derived results when selection effects and algorithmic differences are not controlled. Specific results that are affected include the solar magnetic turnover time, the distributions of sizes, strengths, and li…
▽ More
Feature tracking and recognition are increasingly common tools for data analysis, but are typically implemented on an ad-hoc basis by individual research groups, limiting the usefulness of derived results when selection effects and algorithmic differences are not controlled. Specific results that are affected include the solar magnetic turnover time, the distributions of sizes, strengths, and lifetimes of magnetic features, and the physics of both small scale flux emergence and the small-scale dynamo. In this paper, we present the results of a detailed comparison between four tracking codes applied to a single set of data from SOHO/MDI, describe the interplay between desired tracking behavior and parameterization of tracking algorithms, and make recommendations for feature selection and tracking practice in future work.
△ Less
Submitted 23 April, 2007;
originally announced April 2007.
