In order to utilize solar imagery for real-time feature identification and large-scale data science investigations of solar structures, we need maps of the Sun where phenomena, or themes, are labeled. Since solar imagers produce observations every few minutes, it is not feasible to label all images by hand. Here, we compare three machine learning algorithms performing solar image classification using extreme ultraviolet and Hydrogen-alpha images: a maximum likelihood model assuming a single normal probability distribution for each theme from Rigler et al. (2012), a maximum-likelihood model with an underlying Gaussian mixtures distribution, and a random forest model. We create a small database of expert-labeled maps to train and test these algorithms. Due to the ambiguity between the labels created by different experts, a collaborative labeling is used to include all inputs. We find the random forest algorithm performs the best amongst the three algorithms. The advantages of this algorithm are best highlighted in: comparison of outputs to hand-drawn maps; response to short-term variability; and tracking long-term changes on the Sun. Our work indicates that the next generation of solar image classification algorithms would benefit significantly from using spatial structure recognition, compared to only using spectral, pixel-by-pixel brightness distributions.

We present the results from analyzing the physical and morphological properties of 154 dimmings (transient coronal holes) and the associated flares and coronal mass ejections (CMEs). Each dimming in our 2013 catalog was processed with the semi-automated Coronal Dimming Tracker using Solar Dynamics Observatory AIA 193 A observations and HMI magnetograms. Instead of the typically used difference images, we used our coronal hole detection algorithm to detect transient dark regions "directly" in extreme ultraviolet (EUV) images. This allowed us to study dimmings as the footpoints of CMEs - in contrast with the larger, diffuse dimmings seen in difference images that represent the projected view of the rising, expanding plasma. Studying the footpoint-dimming morphology allowed us to better understand the CME structure in the low corona. While comparing the physical properties of dimmings, flares, and CMEs we were also able to identify relationships between the different parts of this complex eruptive phenomenon. We found that larger dimmings are longer-lived, suggesting that it takes longer to "close down" large open magnetic regions. Also, during their growth phase, smaller dimmings acquire a higher magnetic flux imbalance (i. e., become more unipolar) than larger dimmings. Furthermore, we found that the EUV intensity of dimmings (indicative of local electron density) correlates with how much plasma was removed and how energetic the eruption was. Studying the morphology of dimmings (single, double, fragmented) also helped us identify different configurations of the quasi-open magnetic field.

We present a new tool, the Solar Eclipse Image Standardisation and Sequencing (SEISS), developed to process multi-source total solar eclipse images by adjusting them to the same standard of size, resolution, and orientation. Furthermore, by analysing the eclipse images we can determine the relative time between the observations and order them to create a movie of the observed total solar eclipse sequence. We successfully processed images taken at the 14 November 2012 total solar eclipse that occurred in Queensland, Australia, and created a short eclipse proto-movie. The SEISS tool was developed for the Eclipse Megamovie Project (EMP: https://www.eclipsemegamovie.org), with the goal of processing thousands of images taken by the public during solar eclipse events. EMP is a collaboration among multiple institutes aiming to engage and advance the public interest in solar eclipses and the science of the Sun-Earth connection.

Sunspots are a canonical marker of the Sun's internal magnetic field which flips polarity every ~22-years. The principal variation of sunspots, an ~11-year variation in number, modulates the amount of magnetic field that pierces the solar surface and drives significant variations in our Star's radiative, particulate and eruptive output over that period. This paper presents observations from the Solar and Heliospheric Observatory and Solar Dynamics Observatory indicating that the 11-year sunspot variation is intrinsically tied it to the spatio-temporal overlap of the activity bands belonging to the 22-year magnetic activity cycle. Using a systematic analysis of ubiquitous coronal brightpoints, and the magnetic scale on which they appear to form, we show that the landmarks of sunspot cycle 23 can be explained by considering the evolution and interaction of the overlapping activity bands of the longer scale variability.

We present a new approach to coronal dimming detection using the COronal DImming Tracker tool (CODIT), which was found to be successful in locating and tracking multiple dimming regions. This tool, an extension of a previously developed coronal hole tracking software, allows us to study the prop- erties and the spatial evolution of dimming regions at high temporal and spatial cadence from the time of their appearance to their disappearance. We use the Solar Dynamics Observatory/Atmospheric Imaging Assembly 193 A wavelength observations and Helioseismic and Magnetic Imager to study dimmings. As a demonstration of the detection technique we analyzed six recurrences of a dimming observed near AR 11305 between 2011 September 29 and October 2. The dimming repeatedly ap- peared and formed in a similar way, first expanding then shrinking and occasionally stabilizing in the same location until the next eruption. The dimming areas were studied in conjunction with the corre- sponding flare magnitudes and coronal mass ejection (CME) masses. These properties were found to follow a similar trend during the observation period, which is consistent with the idea that the mag- nitude of the eruption and the CME mass affect the relative sizes of the consecutive dimmings. We also present a hypothesis to explain the evolution of the recurrent single dimming through interchange reconnection. This process would accommodate the relocation of quasi-open magnetic field lines and hence allow the CME flux rope footpoint (the dimming) to expand into quiet Sun regions. By relating the properties of dimmings, flares, and CMEs we improve our understanding of the magnetic field reconfiguration caused by reconnection.

In recent years the role of space weather forecasting has grown tremendously as our society increasingly relies on satellite dependent technologies. The forecasting of flare and CME related transient geomagnetic storms has become a primary initiative, however, minor magnetic storms caused by coronal holes (CHs) have also proven to be of high importance due to their long lasting and recurrent geomagnetic effects. In order to study CH properties, the author developed an automated CH detection method (CHARM), which uses local intensity histograms to identify CH boundaries. An additional algorithm package (CHEVOL) was developed to study individual CHs by tracking their boundary evolution. It is widely accepted that the short-term changes in CH boundaries are due to the interchange reconnection between the CH open field lines and small loops. In order to test the interchange reconnection model, the magnetic reconnection rate and the diffusion coefficient at CH boundaries were determined using observed CH boundary displacement velocities. The results were found to be in agreement with those determined by the theory. The MIST algorithm was developed by the author to build on the CHARM package, providing a fast and consistent way to link CHs to high-speed solar wind periods detected at Earth. This allowed us to carry out a long-term analysis (2000-2009) to study the relationship between CHs, the corresponding HSSW properties, and geomagnetic indices. The relationship between CH related high-speed solar wind streams and the electron flux enhancements in the Van Allen radiation belt was confirmed. The research presented in this thesis includes the small-scale analysis of individual CHs on time scales of days, which is complemented with large scale analysis of CH groups on time scales of years. This allowed us to further our understanding of CH evolution as a whole.

The solar surface and atmosphere are highly dynamic plasma environments, which evolve over a wide range of temporal and spatial scales. Large-scale eruptions, such as coronal mass ejections, can be accelerated to millions of kilometres per hour in a matter of minutes, making their automated detection and characterisation challenging. Additionally, there are numerous faint solar features, such as coronal holes and coronal dimmings, which are important for space weather monitoring and forecasting, but their low intensity and sometimes transient nature makes them problematic to detect using traditional image processing techniques. These difficulties are compounded by advances in ground- and space- based instrumentation, which have increased the volume of data that solar physicists are confronted with on a minute-by-minute basis; NASA's Solar Dynamics Observatory for example is returning many thousands of images per hour (~1.5 TB/day). This chapter reviews recent advances in the application of images processing techniques to the automated detection of active regions, coronal holes, filaments, CMEs, and coronal dimmings for the purposes of space weather monitoring and prediction.

The interaction of open and closed field lines at coronal hole boundaries is widely accepted to be due to interchange magnetic reconnection. To date, it is unclear how the boundaries vary on short timescales and at what velocity this occurs. Here, we describe an automated boundary tracking method used to determine coronal hole boundary displacements on short timescales. The bound- ary displacements were found to be isotropic and to have typical expansion/contraction speeds of \leq2 km s^-1, which indicate magnetic reconnection rates of \leq 3 \times 10^-3. The observed displacements were used in conjunction with the interchange reconnection model to derive typical diffusion coeffi- cients of \leq 3 \times 10^13 cm^2 s^-1. These results are consistent with an interchange reconnection process in the low corona driven by the random granular motion of open and closed fields in the photosphere.

We identify coronal holes using a histogram-based intensity thresholding technique and compare their properties to fast solar wind streams at three different points in the heliosphere. The thresholding technique was tested on EUV and X-ray images obtained using instruments onboard STEREO, SOHO and Hinode. The full-disk images were transformed into Lambert equal-area projection maps and partitioned into a series of overlapping sub-images from which local histograms were extracted. The histograms were used to determine the threshold for the low intensity regions, which were then classified as coronal holes or filaments using magnetograms from the SOHO/MDI. For all three instruments, the local thresholding algorithm was found to successfully determine coronal hole boundaries in a consistent manner. Coronal hole properties extracted using the segmentation algorithm were then compared with in situ measurements of the solar wind at 1 AU from ACE and STEREO. Our results indicate that flux tubes rooted in coronal holes expand super-radially within 1 AU and that larger (smaller) coronal holes result in longer (shorter) duration high-speed solar wind streams.